minerva anestesiologica vol. 82 no. 4 pages 375 · pdf filemechanical resuscitation devices...

ADV SIAARTI/2 10-03-2016 12:13 Pagina 1

Colori compositi

C M Y CM MY CY CMY K

V O L U M E 8 2 · N o. 4 · A P R I L 2 0 1 6

MIN

ER

VA

AN

EST

ESIO

LOG

ICA

. V

OL. 8

2

. N

o. 4

.

PAG

ES 3

75

-49

7

. AP

RIL 2

01

6

389How should ethical committees promote research in critically ill patients?Caironi P., Fumagalli R., Tognoni G.

392ORIGINAL ARTICLESItalian Registry of Complications associated with Regional Anesthesia (RICALOR). An incidence analy-sis from a prospective clinical surveyAllegri M., Bugada D., Grossi P., Manassero A, Pinciroli R. L., Zadra N., Fanelli G., Zarcone A., Cataldo R., Danelli G., Borghi B., on Behalf of RICALOR Group

403Postural effects on pulmonary gas exchange abnor-malities in severe obesity before and after bariatric surgeryRivas E., Arismendi E., Agustí A., Gistau C., Wagner P. D., Rodriguez-Roisin R.

411Treatment of chronic cervicobrachial pain with peri-radicular injection of meloxicamAurini L., Borghi B., White P. F., Tognù A., Rossi B., Fini G., Fusco P., Mosca M., Borghi R.

375EDITORIALSMessages from the RICALOR StudyApan A., Apan O. C.

377Pulmonary pathophysiology in obesity: did we miss something?Magni F., Fumagalli R.

380Chronic cervical radicular pain: time to tackle a new horizonGómez-Ríos M. Á., Freire-Vila E.

383Mechanical resuscitation devices under special cir-cumstances in the out of hospital settingGreif R.

386The heart of the artCarassiti M., Cataldo R.

MINERVA ANESTESIOLOGICA

Vol. 82 April 2016 No. 4

CONTENTS

Vol. 82 - No. 4 MINERVA ANESTESIOLOGICA I

OFFICIAL JOURNAL OF ITALIAN SOCIETY OF ANESTHESIOLOGY, ANALGESIA,RESUSCITATION AND INTENSIVE CARE (SIAARTI)

ITALIAN JOURNAL OF ANESTHESIOLOGY AND ANALGESIAMONTHLY JOURNAL FOUNDED IN 1935 BY A. M. DOGLIOTTI

OFFICIAL JOURNAL OF ITALIAN SOCIETY OF ANESTHESIOLOGY, ANALGESIA,RESUSCITATION AND INTENSIVE CARE (S.I.A.A.R.T.I.)

CONTENTS

II MINERVA ANESTESIOLOGICA April 2016

419Diagnostic accuracy of hemoconcentration for pulmo-nary edema as the cause of weaning failurePottier V., Valette X., Seguin A., Masson R., Parienti J.-J., Sauneuf B., Du Cheyron D., Terzi N.

429Mechanical LUCAS resuscitation is effective, reduces physical workload and improves mental performance of helicopter teamsRehatschek G., Muench M., Schenk I., Dittrich W., Schewe J.-C., Dirk C., Hering R.

438REVIEWSEffectiveness of benzodiazepine premedication on recovery in day-case surgery: a systematic review with meta-analysisMijderwijk H., Van Beek S., Duivenvoorden H. J., Stolker R. J.

465Non-pharmacological interventions to reduce ICU-related psychological distress: a systematic reviewWade D. M., Moon Z., Windgassen S. S., Harrison A. M., Morris L., Weinman J. A.

479POINT OF VIEWHow to protect incompetent clinical research subjects involved in critical care or emergency settingsZamperetti N., Piccinni M., Bellomo R., Citerio G., Mistraletti G., Gristina G., Giannini A.

486JOURNAL CLUB CRITIQUEAnother failed attempt of neuroprotection: proges-terone for moderate and severe traumatic brain injuryMeyfroidt G., Taccone F. S.

492LETTERS TO THE EDITORSpinal epidural abscess: stay focused, stay tuned! A clinical report with negative neurological outcome from the “Italian Registry of Complications associa-ted with Regional Anesthesia - RICALOR”Manassero A., Ugues S., Coletta G., Bugada D.

493Impact of AxotrackTM for ultrasound-guided central venous catheter insertion: a randomized controlled study conducted on inanimate manikiAlves M., Bigé N., Pichereau C., Baudel J.-L., Galbois A., Ait-Oufella H., Maury E.

495Tetanus and Tako-Tsubo: is there a relationship?Brunelli G., Tesoro S., Trotta A., Dentini N., Peduto V. A.

497TOP 50 MINERVA ANESTESIOLOGICA REVIEWERS

About the cover: cover shows a magnetic resonance highlighting an epidural abscess with inner cystic organization of multiple air fluid level. Spinal canal is markedly narrowed and spinal cord is compressed, resulting in neurologic symptoms. For more information, see article by Allegri M. et al. beginning on page 392.

V O L U M E 8 2 · N o. 4 · A P R I L 2 0 1 6

Vol. 82 - No. 4 MiNerVa aNestesiologica 375

estly reported their complications. However, since this is a web-based study, missing data entry is an expected limitation. this limitation may cause incomplete reporting of the inci-dence of some procedures and complications, or incorrect interpretations of the causes and management of a certain complication.

according to our opinion, studies conducted across the country offer advantages for identi-fying insufficient centers in terms of training and technical facilities. Based on the results of these studies, the centers can be supported and ra-related complications can be reduced.

in the study by allegri et al.,3 local an-esthetic systemic toxicity (last) was ob-served most frequently in axillary block. the researchers indicated that further investiga-tions are required concerning the use of Usg for preventing the development of last and post-operative neurologic symptoms (PoNs). in cases with complication following Usg-guided block, the type of the block used should be reviewed in terms of safety. We suggest that peripheral nerve block techniques with re-duced complication rate should be preferred in Usg-guided procedures in the future.5

Despite the enrollment of patients taking anticoagulant therapy in this study3, no hemor-rhagic complication was reported. This finding indicates the benefits of the guidelines related to ra and anticoagulant therapy. it is also em-phasizes that participants have been increas-ingly aware of ra-related complications in

Nowadays, regional anesthesia (ra) tech-niques used for both anesthesia and post-

operative analgesia procedures have become increasingly widespread. While the introduc-tion of ultrasonography (Usg) for use as a part of these techniques has increased the interest in ra, visualizing anatomic structures during the procedure has also provided a considerable advantage regarding the safety of ra tech-niques. However, it does not seem possible to advocate that the use of Usg can completely eliminate complications.1, 2 For this reason, it is important to report complications associated with ra in order to safely perform future pro-cedures and to develop novel technologies.

the study entitled “italian registry of com-plications associated with regional anesthe-sia (ricalor). an incidence analysis from a prospective clinical survey” by allegri et al.,3 published in this issue of Minerva Anes-tesiologica, is a web-based prospective study comprising 17 centers from italy. in this study, complications of 117.182 procedures per-formed for primary anesthesia technique or postoperative analgesia were reported on a volunteer basis. considering that previous re-ports about this issue have usually been con-ducted on a retrospective basis,4 the study of allegri et al. 3 is a remarkable work. it is also appreciated that long-term follow-up after ra has been performed and the participants hon-

E D I T O R I A L

Messages from the ricalor studyalparslan aPaN *, ozgun cUVas aPaN

Department of anesthesiology and intensive care Medicine, giresun University Medical Faculty, giresun, turkey*corresponding author: alparslan apan, Department of anesthesiology and intensive care Medicine, giresun University Medical Faculty. Nizamiye Yerleskesi, Orhan Yılmaz Cad. Mumcular Sok. No: 1, Giresun, Turkey. E-mail: [email protected]

anno: 2016Mese: aprilVolume: 82No: 4rivista: Minerva anestesiologicacod rivista: Minerva anestesiol

lavoro: titolo breve: Messages FroM tHe ricalor stUDYprimo autore: aPaNpagine: 375-6citazione: Minerva anestesiol 2016;82:000-000

comment on p. 392.

Minerva anestesiologica 2016 april;82(4):375-6© 2016 eDiZioNi MiNerVa MeDicathe online version of this article is located at http://www.minervamedica.it

aPaN Messages FroM tHe ricalor stUDY

376 MiNerVa aNestesiologica april 2016

study investigating ra complications is a pio-neering work in this field.

References 1. Barrington MJ, Kluger r. Ultrasound guidance reduces

the risk of local anesthetic systemic toxicity follow-ing peripheral nerve blockade. reg anesth Pain Med 2013;38:289-99.

2. lewis sr, Price a, Walker KJ, Mcgrattan K, smith aF. Ultrasound guidance for upper and lower limb blocks. cochrane Database syst rev 2015;sep 11;9:cD006459, doi: 10.1002/14651858 cD006459.pub3.

3. allegri M, Bugada D, grossi P, Manassero a, Pinciroli rl, Zadra N, et al. italian registry of complications as-sociated with regional anesthesia (ricalor). an inci-dence analysis from a prospective clinical survey. Min-erva anestesiol 2016;82:392-402.

4. Brull r, Mccarney cJl, chan VWs, el-Beheiry H. Neu-rological complications after regional anesthesia: con-temporary estimates of risk. anesth analg 2007;104:965-74.

5. Kessler J, Marhofer P, Hopkins PM, Hollmann MW. Peripheral regional anaesthesia and outcome: lessons learned from the last 10 years. Br J anaesth 2015;114:728-45.

6. Horlocker tt. regional anaesthesia in the patient receiv-ing antithrombotic and antiplatelet therapy. Br J anaesth 2011;107:i96-i106.

7. Manassero a, Ugues s, coletta g, Bugada D. spinal epidural abscess: stay focused, stay tuned! a clinical report with negative neurological outcome from the “italian registry of complications associated with regional anesthesia-ricalor”. Minerva anestesiol 2016;82:492-3.

such patients. these guidelines should be up-dated according to the novel anti-thrombotic agents.3, 6

a spinal-epidural abscess was reported as an infectious complication in only one patient.3 this case report was also published in this is-sue of Minerva Anestesiologica.7 it warns cli-nicians about compliance of aseptic technique during epidural catheter placement for ra procedures or postoperative analgesia. Fur-thermore, it also emphasizes the importance of consultation of the febrile patients with lo-cal and laboratory signs of infection before the emergence of neurological deficit.

although the overall incidence of major complications of ra is low, they may cause permanent neurological sequela when these complications occur and are not noticed until too late. safe and successful applications of ra techniques depend on increased aware-ness, training, and technical facilities. Detailed reports about techniques, tools, habits, medi-cations, equipment, and complications related to ra techniques will allow performing such procedures more safely in the future. allegri’s et al.3 nationwide multicenter prospective

Conflicts of interest.—The authors certify that there is no conflict of interest with any financial organization regarding the material discussed in the manuscript.Article first published online: March 9, 2016. - Manuscript accepted: March 8, 2016. - Manuscript revised: February 15, 2016. - Ma-nuscript received: November 25, 2015.(Cite this article as: apan a, cuvas apan o. Messages from the ricalor study. Minerva anestesiol 2016;82:375-6)


in a cohort of healthy obese female subjects before and after bariatric surgery (Bs). they hypothesized that prior to Bs, supine position would worsen pulmonary gas exchange abnor-malities and that this effect would vanish after the surgery-induced weight loss. interestingly, they found an unexpected result analyzing the overall population because in obese individu-als oxygenation did not deteriorate when su-pine before BS, in spite of finding of previous series. conversely, one year after Bs in recum-bent position oxygenation significantly de-creased while alveolar-arterial po2 difference increased. Furthermore, the authors intrigu-ingly found a distinctive effect between non-hypoxemic and hypoxemic patients whereas postural oxygenation alterations mainly oc-curred in the latter. indeed in this subset, su-pination one year after Bs results in a note-worthy unforeseen pulmonary gas exchange deterioration.

Reason of these findings should be neces-sarily sought in changes of lungs volume and mechanic, in breathing pattern or in hemo-dynamic. Unfortunately, the authors failed to show a strong causative physiopathological mechanism, analyzing neither intrapulmonary shunt nor dead space. Measurements of lungs volumes and mechanics might have provided other interpretative hypothesis.

arguably, hemodynamics seems to be a

obesity is a complex condition involv-ing an excessive amount of body fat and

represents a major risk factor for a number of chronic diseases, including type 2 diabetes, cardiovascular disorder and certain types of cancer.

Due to the high prevalence, obesity is nowa-days considered a public health concern in the majority of high-income countries. How-ever, it is spreading globally and becoming an emerging issue also in low- and middle-income countries.1

in the United state, one-third of adults are obese (BMI≥30 kg/m2) and the medical costs for these individuals are about $ 1500 higher than those for normal weight subjects. in italy, although prevalence is lower than in Us (11% of adults are obese), childhood obesity is one of the most serious ongoing challenges for pe-diatricians.2

therefore, treating obese subjects is increas-ingly common both in the operating room and in intensive care units. the care of obesity is often a challenge for the anesthesiologist due to the complex pathophysiology changes oc-curring in these patients.

in this issue of Minerva Anestesiologica, ri-vas et al.3 compare data regarding pulmonary gas exchange in upright and supine position

E D I T O R I A L

Pulmonary pathophysiology in obesity: did we miss something?

Federico MagNi 1 *, roberto FUMagalli 1, 2

1Department of Health sciences, Università degli studi di Milano Bicocca, Monza, italy; 2 s.c. anestesia e rianimazione 1, ospedale Niguarda ca’ granda Hospital, Milan, italy*corresponding author: Federico Magni, Department of Health sciences, Università degli studi di Milano Bicocca, Via cadore 48, 20900, Monza, Italy. E-mail: [email protected]

Anno: 2016Mese: aprilVolume: 82No: 4rivista: Minerva anestesiologicacod rivista: Minerva anestesiol

lavoro: titolo breve: PUlMoNarY PatHoPHYsiologY iN oBesitYprimo autore: MagNipagine: 377-9citazione: Minerva Anestesiol 2016;82:377-9

Comment on p. 403

Minerva Anestesiologica 2016 April;82(4):377-9© 2015 EDIZIONI MINERVA MEDICAThe online version of this article is located at http://www.minervamedica.it

MagNi PUlMoNarY PatHoPHYsiologY iN oBesitY

378 MiNerVa aNestesiologica April 2016

ume should be set using predicted body weight and not the actual weight 14 in order to reduce postoperative pulmonary complication. again, in these patients low tidal volume ventilation may have a greater protective role in compari-son to normal weight patients. indeed, a sup-posedly safe tidal volume might be not safe enough and produce a cyclic dynamic over-strain in an obese subject with reduced func-tional residual capacity.15

relevant data about mechanical ventila-tion strategy during surgery in obese patients will come from the ongoing ProBese trial (NCT02148692). The investigators aim to compare a ventilation strategy using higher levels of PeeP with recruitment maneuvers with one using lower levels of PeeP without recruitment maneuvers in obese patients at an intermediate-to-high risk for postoperative pulmonary complications.

Moreover, another key issue should be mentioned regarding the critically ill obese patients. in fact, in recent year the revival of interest in the transpulmonary pressure mea-surement 16 to guide mechanical ventilation 17 raises a number of questions about safety lim-its of mechanical ventilation especially in the obese patients. rib cage fat and increased in-tra-abdominal pressure usually result in a re-duced chest wall compliance and, accordingly, in an early exceeding of the safety plateau pressure threshold (30 cmH2O) defined by the arDsnet trial.18

this issue was highlighted analyzing data of partitioned respiratory mechanic in a cohort of H1N1 arDs patients suitable for extracorpo-real membrane oxygenation (ecMo) support due to unsafe high plateau pressure.19 in this study the authors showed that titrating PeeP neglecting abnormalities of chest wall me-chanics may overestimate the incidence of hy-poxemia refractory to conventional ventilation and, consequently, lead to inappropriate use of ecMo. therefore obese patients, often hav-ing a stiff chest wall, may benefit of increas-ing PeeP until a physiologically reasonable transpulmonary plateau pressure limit (sup-posedly 25 cmH2o) irrespective of total respi-ratory system plateau pressure. this approach

conceivable player in determining the study results. Prior Bs orthodeoxia would be in-duced by a decreased cardiac output coupled with an impaired hypoxic pulmonary vasocon-striction. this phenomenon seems to be more critical than the improvement of lung volumes in determining oxygenation when moving from supine to upright position in hypoxemic obese patients.

the observation that after weight loss oxy-genation worsens in supine position is quite as-tonishing. the authors put forth the hypothesis that the abnormal pulmonary vascular adapta-tion after Bs determines this conduct. altera-tions observed in Miget-derived V

.a/Q

. des-

criptors reinforced this assumption.it should be emphasized that study popula-

tion consists of obese otherwise healthy female patients and therefore data are not soiled by the possible coexistence of any lungs comor-bidities, which are frequent in obese subjects. Hence, a note of caution is required to general-ize the findings of this study.

Previous studies have shown that in healthy subjects total respiratory system compliance decreases when shifting from the sitting to supine position.4 it has been mainly attributed to the reduction in functional residual capac-ity resulting from expiratory reserve volume drop.5 indeed, gravitational forces in horizon-tal position cause a cephalic displacement of diaphragm due to increased abdominal pres-sure. this phenomenon is well known in an-esthesia setting and has been investigated in several physiological studies in non-obese patients.6 instead, limited data7-10 are available on partitioned respiratory mechanics and lung volumes of obese patients in the anesthesiol-ogy and intensive care setting.

Hence, ventilation strategy in obese patients is a matter of broad debate. there is some evidence that recruitment maneuver added to PeeP compared with PeeP alone improves intraoperative oxygenation and compliance without adverse effects. currently, no substan-tial evidence exists to recommend a specific ventilatory mode.11 intraoperative low tidal volume ventilation should be implemented even in these patients.12, 13 Note that tidal vol-

PUlMoNarY PatHoPHYsiologY iN oBesitY MagNi


used during anesthesia. Anesth Analg 1991;73:422-6. 7. Hakala, Maasilta, sovijarvi. Upright body position and

weight loss improve respiratory mechanics and daytime oxygenation in obese patients with obstructive sleep ap-noea. Clin Physiol 2000;20:50-5.

8. Vaughan rW, Bauer s, Wise l. effect of position (semi-recumbent versus supine) on postoperative oxygenation in markedly obese subjects. Anesth Analg 1976;55:37-41.

9. Hakala K, Mustajoki P, Aittomäki J, Sovijärvi AR. Effect of weight loss and body position on pulmonary function and gas exchange abnormalities in morbid obesity. Int J Obes Relat Metab Disord 1995;19:343-6.

10. Pelosi P, Croci M, Calappi E, Mulazzi D, Cerisara M, Vercesi P, et al. Prone positioning improves pulmonary function in obese patients during general anesthesia. Anesth Analg 1996;83:578-83.

11. aldenkortt M, lysakowski c, elia N, Brochard l, tramèr Mr. Ventilation strategies in obese patients un-dergoing surgery: a quantitative systematic review and meta-analysis. Br J Anaesth 2012;109:493-502.

12. anzueto a, Frutos-Vivar F, esteban a, Bensalami N, Marks D, Raymondos K, et al. Influence of body mass index on outcome of the mechanically ventilated patients. Thorax 2011;66:66-73.

13. Mascia L, Zavala E, Bosma K, Pasero D, Decaroli D, An-drews P, et al. High tidal volume is associated with the development of acute lung injury after severe brain in-jury: an international observational study. crit care Med 2007;35:1815-20.

14. Lellouche F, Dionne S, Simard S, Bussières J, Dagenais F. High tidal volumes in mechanically ventilated patients increase organ dysfunction after cardiac surgery. anes-thesiology 2012;116:1072-82.

15. gommers D. Functional residual capacity and absolute lung volume. Curr Opin Crit Care 2014;20:347-51.

16. Chiumello D, Carlesso E, Cadringher P, Caironi P, Valen-za F, Polli F, et al. lung stress and strain during mechani-cal ventilation for acute respiratory distress syndrome. Am J Respir Crit Care Med 2008;178:346-55.

17. talmor D, sarge t, Malhotra a, o’Donnell cr, ritz r, lisbon a, et al. Mechanical ventilation guided by esophageal pressure in acute lung injury. N Engl J Med 2008;359:2095-104.

18. the acute respiratory Distress syndrome Network. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. the acute re-spiratory Distress Syndrome Network. N Engl J Med 2000;342:1301-8.

19. Grasso S, Terragni P, Birocco A, Urbino R, Del Sorbo L, Filippini c, et al. ECMO criteria for influenza A (H1N1)-associated arDs: role of transpulmonary pressure. in-tensive Care Med 2012;38:395-403.

allowed to improve arterial oxygenation and to avoid an inappropriate use of ecMo.

to conclude, we necessarily must strive to investigate the obese respiratory pathophysiol-ogy because this topic is becoming increasingly part of our daily lives and many uncertainties remain.

For instance, rivas et al. reveals unexpect-ed data. We believed it is just one example of what we currently ignore about the obese lung physiology. The study adds to the field the knowledge that, in obese patients, supine positioning may alter gas exchange in a differ-ent manner before and after bariatric surgery. surprisingly, in the former case arterial oxy-genation may improve; in the latter conversely oxygenation may worsen.

In conclusion, findings of this paper add an unexpected piece to the multifaceted puzzle of obese lung pathophysiology. the complexity of this field results in several unanswered issue regarding our daily practice.

References 1. WHo: obesity. World Health organization; [internet].

Available from: http://www.who.int/topics/obesity/en/ [cited 2015, Jul 16].

2. Ministero della salute: obesità; [internet]. available from: http://www.salute.gov.it/portale/salute/p1_5.jsp?lingua=italiano&id=175&area=Malattie_endocrine_e_metaboliche/ [cited 2015, Jul 16].

3. rivas e, arismendi e, agustí a, gistau c, Wagner PD, rodriguez-roisin r. Postural effects on pulmonary gas exchange abnormalities in severe obesity before and after bariatric surgery. Minerva Anestesiol 2016;82:403-10.

4. Behrakis PK, Baydur A, Jaeger MJ, Milic-Emili J. Lung mechanics in sitting and horizontal body positions. chest 1983;83:643-6.

5. Damia g, Mascheroni D, croci M, tarenzi l. Periopera-tive changes in functional residual capacity in morbidly obese patients. Br J Anaesth 1988;60:574-8.

6. Lumb AB, Nunn JF. Respiratory function and ribcage contribution to ventilation in body positions commonly

Conflicts of interest.—The authors certify that there is no conflict of interest with any financial organization regarding the material discussed in the manuscript.Article first published online: October 1, 2015. - Manuscript accepted: September 29, 2015. - Manuscript revised: September 28, 2015. - Manuscript received: July 21, 2015.(Cite this article as: Magni F, Fumagalli r. Pulmonary pathophysiology in obesity: did we miss something? Minerva anestesiol 2016;82:377-9)

380 Minerva anestesiologica april 2016

(IL1α,IL1β,IL6,IL17andTNF)thatinducechemokine and metalloproteinases produc-tion,adhesionmoleculesonendothelialcells,chemoattraction of neutrophils, stimulationof phagocytosis, andproductionofPGE2bymacrophages amplifying the inflammatorycascade, thusestablishingaviciouscircle.8-10 Furthermore,neurogenicfactors,generatedbyboth disc and immune cells, sensitize nerveendingsandinduceexpressionofnociceptivecationchannelsinthedorsalrootganglion.2,8 Depolarizationof these ionchannelscontrib-utestothegenerationofradicularpain.The diagnosis ofCRPmust bemadewith

caution in order to direct the treatment withetiologictarget.Intheabsenceofagoldstan-dard,thediagnosisisbasedonacombinationof history, clinical examination, and comple-mentarystudies thatcan includemedical im-agingtechniques,beingMRIthetechniqueofchoice, electrophysiologic tests or diagnosticselectivenerverootblocks.3Thenaturalhistoryof radicularpainhas a

favorable prognosis with self-limited symp-tomatologyanditresolvesspontaneouslyevendespite treatment. Nevertheless, in a propor-tionofpatientsbecomeinachronicdisorderdespiteanappropriateetiologictherapy.ThetreatmentofCRPconstitutesachalleng-

ing problem sincemost of the recommenda-

cervicalradicularpainisacommoncondi-tionamong themultiplepresentationsof

cervicobrachialpain,withanannualincidenceof83per100,000population.1Itconstitutesamajorpublichealthproblembeingoneofthemain causes of disability, missed workdays,andinvolvessubstantialhealthcarecosts.It is described as a pain perceived in the

upperextremitycausedby the stimulationordysfunction of the nerve root or dorsal rootganglia.2Thus,thepaincanbeperceivedinallstructures that are innervated by the affectednerverootsuchasmuscles,joints,ligaments,andtheskinwithoutfollowingadermatomaldistribution.3,4 The most common causes ofcervicalradicularpainarecervicalspondylosisencroachingon the foramen, disc protrusión,andcervicalspinalstenosis.1Although the pathophysiology of radicular

pain continues to be unclear,2 the proposedmechanisms include neural compression, in-flammation and biochemical mediators, vas-cular compromise, and intraneural edema.5,6 Compression alone is not enough to causepain,beingnecessaryaninflammatorychemi-calcomponent.7Degenerationof thecervicalintervertebral disc plays a major role, sinceits cells secrete proinflammatory cytokines

E D I T O R I A L

Chroniccervicalradicularpain:timetotackleanewhorizon

ManuelÁngelGÓMEZ-RÍOS1,2*,EnriqueFREIRE-VILA1,2

1DepartmentofAnaesthesiologyandPerioperativeMedicine,ComplejoHospitalarioUniversitariodeACoruña,ACoruña,Spain; 2AnesthesiologyandPainManagementResearchGroup,InstituteforBiomedicalResearchofACoruña(INIBIC),ACoruña,Spain*Correspondingauthor:ManuelÁngelGómez-Ríos,ComplejoHospitalarioUniversitariodeACoruña,AsXubias15006,ACoruña,Spain.E-mail:[email protected]

Anno:2016Mese:AprilVolume:82No:4Rivista:MinervaAnestesiologicaCodRivista:MinervaAnestesiol

Lavoro:titolobreve:Chroniccervicalradicularpainprimoautore:GÓMEZ-RÍOSpagine:380-2citazione: MinervaAnestesiol2016;82:380-2

Commentonp.411.

MinervaAnestesiologica2016April;82(4):380-2©2016EDIZIONIMINERVAMEDICATheonlineversionofthisarticleislocatedathttp://www.minervamedica.it

CHRONICCERVICALRADICULARPAIN GÓMEZ-RÍOS

Vol.82-No.4 Minerva anestesiologica 381

complicationsandintensedebate.3,17IthasledtheFDAtoissueawarningfortheuseofepi-duralcorticosteroids.18Pulsedradiofrequencytreatment adjacent to the cervical dorsal rootganglion is another recommended treatmentfor chronicCRP.3 In selectedpatients refrac-tory to other treatment options, spinal cordstimulationmaybeconsidered.3Amultimodal aproach based on the com-

bination of drugs and techniques is appliedin clinical setting 2, 5 as it is probably moreeffective thaneithermonotherapy,11 althoughitseffectivenessandsafetyshouldbeformallyevaluated in further extensión. Nonetheless,onethirdofpatientscanhavepersistentsymp-toms.5Surgical treatment is considered when ra-

dicularpainbecomesrefractoryandlonglast-ingdespiteofmultimodalanalgesictherapy.15 Althoughsomepatientsmaybenefitfromsur-gery,thereismoderateevidencefortheeffec-tivenessofcervicalspinalsurgeryforradicularpain.2Thus,mostrandomizedstudiesevaluat-ingsurgeryforneuropathiccervicalpainhavefoundminimal long-termbenefit inmostpa-tients.19,20Likewise,surgicalinterventionscanresultinpostcervicalsurgerysyndrome.21

Taking into account all these data, thesearchfornewnonsurgicalmodalitiesoftreat-ment takes special relevance. In this issueofMinerva Anestesiologica,Auriniet al.22pres-ent a new target specific treatmentmodality,the ultrasound-guided cervical periradicularmeloxicam injection, as a promising tool forthe controlof chronicCRP.Thispreliminarystudyindicatedthatallpatientswhocompletedthe study reported significant improvementsin cervicalbrachial pain and disability, with94%of the patients showed complete remis-sionduring the90daysof follow-up,havinganexcellentinfluenceonworkstatus.Nosideeffectswere reported.Theseexcellent resultscouldbeexplainedby thesuppressionof theproductionofPGE2,proinflammatorymedia-tor thatplaysan important role to induce ra-dicularpain,duetocyclooxygenaseinhibitoryeffectofmeloxicamdeliveredinthesiteofpa-thology.However, it isnecessary toconsiderthattheresultsofthestudybyBorghiet al.are

tionsarebasedonweakorabsentevidence.11 Pharmacotherapy, rehabilitation and physicaltherapy,interventionaltechniquesandsurgeryconstitute the different alternatives of treat-ment.Pharmacologic therapy is indicated as a

first-line treatment,mainly in acute phase ofCRPaschronicdrugtherapyhasbeenshowntobe ineffective.2Thus, nonsteroidal anti-in-flammatorydrugsareprimarilyrecommendedfor short-term treatment in the acute phase.Oral steroids have shown efficacy only on ashort-term basis although is not supportedbyanyhigh-quality study,7 and its long-termuseshouldbeavoidedbecauseofsideeffects.Similarly, there are insufficient evidence torecommendopioidsinthetreatmentofneuro-pathic pain beyond twomonths and it is as-sociatedwithmultipleadverseeffects.12Anti-depressants(tricyclicsandselectiveserotoninreuptake inhibitors) and antiepileptic agentssuchasgabapentinoidshavebeenconsideredforchronicCRP,howeverdifferenttrialsfailedtodemonstrate its efficacy inchronic lumbarradicular pain.13 Other commonly prescribeddrugs includebenzodiazepines, skeletalmus-cle relaxants or anti-TNF agents. However,mostofthemhavenotbeenrigorouslyexam-inedinrandomized,placebo-controlledtrials.Physicaltherapyandrehabilitationarerec-

ommended to restoring the range of motionand pain relief due to noninvasiveness, al-thoughthereisnoqualityevidencetosupportthisattitude.11Moreover,modalitiesascervi-calimmobilizationorspinalmanipulationarenotrisk-free.14When conservative treatment fails, inter-

ventionaltreatmentisconsidered.Fluoroscop-icallyguidedcervicalepiduralinjectionswithsteroidsand/orlocalanesthetics arethemain-stay of the interventional techniques.1 Bothdrugsactdecreasinginflammationattheleveloftheaffectednerveroot.1,15,16Thereisgoodevidence for the effectiveness of cervical in-terlaminarepiduralinjectionsinmanagingcer-vicalradicularpain.However,theevidenceispoor for cervical transforaminal aproach.15,16 Both aproaches, fundamentally the last one,are associated with catastrophic neurologic

GÓMEZ-RÍOS CHRONICCERVICALRADICULARPAIN


andu-PAinexperimentalcervicalspondyloticmyelopa-thy.Spine(PhilaPa1976)2013;38:4-10.

10. KangJD,GeorgescuHI,McIntyre-LarkinL,Stefanovic-RacicM,DonaldsonWF,EvansCH.Herniated lumbarintervertebraldiscsspontaneouslyproducematrixmetal-loproteinases,nitricoxide,interleukin-6,andprostaglan-dinE2.Spine(PhilaPa1976)1996;21:271-7.

11. CohenSP,HayekS,SemenovY,PasquinaPF,WhiteRL,Veizi E, et al. Epidural steroid injections, conservativetreatment,orcombinationtreatmentforcervicalradicularpain: amulticenter, randomized, comparative-effective-nessstudy.Anesthesiology2014;121:1045-55.

12. McNicol ED, Midbari A, Eisenberg E. Opioidsfor neuropathic pain. Cochrane Database Syst Rev2013;8:CD006146.

13. BaronR,FreynhagenR,TölleTR,CloutierC,LeonT,MurphyTK,et al. The efficacy and safetyof pregaba-lin in the treatmentofneuropathicpainassociatedwithchronic lumbosacral radiculopathy. Pain 2010;150:420-7.

14. Malone DG, Baldwin NG, Tomecek FJ, Boxell CM,GaedeSE,CovingtonCG,et al. Complicationsof cer-vical spine manipulation therapy: 5-year retrospec-tive study in a single-group practice. Neurosurg Focus2002;13:ecp1.

15. WooJH,ParkHS.Cervicaltransforaminalepiduralblockusing low-dose local anesthetic: a prospective, rand-omized,double-blindstudy.PainMed2015;16:61-7.

16. ManchikantiL,NampiaparampilDE,CandidoKD,Bak-shi S, Grider JS, Falco FJ, et al. Do cervical epiduralinjections provide long-term relief in neck and upperextremity pain? A systematic review. Pain Physician2015;18:39-60.

17. Manchikanti L,Hirsch JA.Neurological complicationsassociated with epidural steroid injections. Curr PainHeadacheRep2015;19:482.

18. Manchikanti L, Candido KD, Singh V, Gharibo CG,Boswell MV, Benyamin RM, et al. Epidural steroidwarningcontroversystilldoggingFDA.PainPhysician2014;17:E451-74.

19. EngquistM,LöfgrenH,ÖbergB,HoltzA,PeolssonA,SöderlundA,et al. Surgeryversusnonsurgicaltreatmentof cervical radiculopathy: a prospective, randomizedstudycomparingsurgeryplusphysiotherapywithphysi-otherapyalonewitha2-yearfollow-up.Spine(PhilaPa1976)2013;38:1715-22.

20. NikolaidisI,FouyasIP,SandercockPA,StathamPF.Sur-geryforcervicalradiculopathyormyelopathy.CochraneDatabaseSystRev2010:CD001466.

21. VeeravaguA,ColeT,JiangB,RatliffJK.Revisionratesandcomplicationincidenceinsingle-andmultilevelan-teriorcervicaldiscectomyandfusionprocedures:anad-ministrativedatabasestudy.SpineJ2014;14:1125-31.

22. AuriniL,BorghiB,WhitePF,TognùA,RossiB,FiniG,MoscaM,et al. Treatmentofchroniccervicobrachialpainwithperiradicularinjectionofmeloxicam.MinervaAnestesiol2016;82:411-18.

notreproducibleastheproposedtechniqueisnotcomparedwithgoodstandardneitherpla-cebo. Furthermore, periradicular meloxicaminjections continue to being off-label at thismoment.A prospective, randomized, double-blindstudyeliminatingtheposibleplaceboef-fectcomponentismandatorytoconfirmthesepromisingresults.The debate currently remains over CRP

treatmentbecauseof thelackofevidencere-garding the different management strategies.Futurehigh-quality trials shouldexpande thebodyofevidenceabout theeffectivenessandsafety of this and others interventions. ThearticleofAuriniet al. could represent anewstimulustolookatthehorizontofindanswerstothemultipleunresolvedquestionsraisedbythecervicalradicularpain.

References 1. ManchikantiL,FalcoFJ,DiwanS,HirschJA,SmithHS.

Cervicalradicularpain:theroleofinterlaminarandtrans-foraminal epidural injections.Curr PainHeadacheRep2014;18:389.

2. Manchikanti L, Hirsch JA. Clinical management ofradicularpain.ExpertRevNeurother2015;15:681-93.

3. VanZundertJ,HuntoonM,PatijnJ,LatasterA,MekhailN, vanKleefM,PracticeP. 4.Cervical radicular pain.PainPract2010;10:1-17.

4. BogdukN.The anatomy and pathophysiology of neckpain.PhysMedRehabilClinNAm2011;22:367-82,vii.

5. EubanksJD.Cervicalradiculopathy:nonoperativeman-agementofneckpainandradicularsymptoms.AmFamPhysician2010;81:33-40.

6. CooperRG,FreemontAJ,HoylandJA,JenkinsJP,WestCG,IllingworthKJ,et al. Herniatedintervertebraldisc-associatedperiradicularfibrosisandvascularabnormali-ties occur without inflammatory cell infiltration. Spine(PhilaPa1976)1995;20:591-8.

7. YoonSH.Cervicalradiculopathy.PhysMedRehabilClinNAm2011;22:439-46,viii.

8. RisbudMV,Shapiro IM.Roleofcytokines in interver-tebraldiscdegeneration:painanddisccontent.NatRevRheumatol2014;10:44-56.

9. KaradimasSK,KlironomosG,PapachristouDJ,Papan-ikolaouS,PapadakiE,GatzounisG.Immunohistochemi-calprofileofNF-κB/p50,NF-κB/p65,MMP-9,MMP-2,

Conflicts of interest.—Theauthorscertifythatthereisnoconflictofinterestwithanyfinancialorganizationregardingthematerialdiscussedinthemanuscript.Articlefirstpublishedonline:March11,2016.-Manuscriptaccepted:March9,2016.-Manuscriptrevised:February15,2016.-Manuscriptreceived:January5,2016.(Cite this article as: Gómez-Ríos MA, Freire-Vila E. Chronic cervical radicular pain: time to tackle a new horizon.MinervaAnestesiol2016;82:380-2)


ropean recommendations on resuscitation is-sued in october 2015 3, 11 do not recommend the general use of these resuscitation devices. Both resuscitation councils consider its use in very special situations, “where sustained high-quality manual chest compressions may not be practical.”9 examples described are prolonged cPr in patients with refractory VF or pulse-less Vt with an onset and offset of return of spontaneous circulation (rosc) “where pro-vider fatigue may impair high-quality manual compressions (e.g., hypothermic arrest), dur-ing interventions in the cardiac catheter lab during coronary angiography, or “cPr in a moving ambulance where provider safety is at risk”.

The valued reader will find in this issue of Minerva Anestesiologica a report by Dr. re-hatschek et al. from the University of Bonn, germany, about a randomized controlled man-ikin study during simulated helicopter flight comparing the lUcas device with manual cPr.12 they found in this setting that lU-CAS delivered significantly better adherence to the guideline-correct compression depth. the 2-mm difference on average might not be of clinical importance but comparing every single compression the lUcas device was in 36% better than manual compression. that is what we expect from machines and why we apply machines — constant high quality de-livery of work.

cardiopulmonary resuscitation (cPr) es-pecially for longer time is strenuous and

exhausting for rescuers because the proper compression depth of 5-6 cm with a frequency of 100-120 compressions per minute 1-3 is hard muscular effort. therefore resuscitation guide-lines recommend now for years that chest-compression-performing rescuers should change every 2 minutes to provide optimal and to avoid suboptimal chest compression result-ing in low cardiac output. the medical device industry also tried to build resuscitation ma-chines to overcome these physical challenges for humans with the final aim to provide con-tinuous chest compression. the ultimate aim of all these exertions is an increased rate of survival after cardiac arrest. Besides some rare developments, 2 devices entered the mar-ket.4, 5 Despite earlier mostly observational reports about improved cPr with the use of these devices, large studies published in the last years could not show superior survival rates for the lUcas device 6, 7 as well as for the autopulse device 8 in the out of hospital setting.

the 2015 consensus of science and treat-ment recommendation (costr) document 9, 10 took that into account in the als Pico ques-tion number 782.9 Derived from the costr 2015 both, the North-american and the eu-

E D I T O R I A L

Mechanical resuscitation devices under special circumstances in the out of hospital setting

robert greiF*

Department of anesthesiology and Pain therapy, Bern University Hospital, University of Bern, Bern, switzerland*corresponding author: robert greig, Department of anesthesiology and Pain therapy, Bern University Hospital, University of Bern, inselspital, 3010 Bern, switzerland. e-mail: [email protected]


lavoro: titolo breve: Mechanical resuscitation devices under special circumstances in the out of hospital settingprimo autore: greiFpagine: 383-5citazione: Minerva anestesiol 2016;82:383-5

comment on p. 429.


greiF MecHaNical resUscitatioN DeVices UNDer sPecial circUMstaNces iN tHe oUt oF HosPital settiNg


i have my doubts, but that is the advantage of the scientific discussion by publishing research results. The scientific community starts to think about new ways of providing better care.

important and i would like to stress that out, is the last sentence in the conclusion and also in the key messages: the implementation of such devices needs constant training in a simu-lation setting to increase the performance of the team. that fact was even recognized in the educational chapter of the 2015 erc guide-lines.14 But before implementing cost-inten-sive devices we need solid results on improved patient outcome, also in these settings.

What could be the future direction even in such special settings? if we bring in highly educated and equipped resuscitation teams to victims of cardiac arrest, does it really make sense to apply mechanical devices with little proven evidence? on the other hand we see the emergence of extracorporeal life support or e-cPr out of hospital.15, 16 extracorporeal membrane oxygenation (ecMo) might be an option and is suggested by the ilcor costr als Pico 723.9 Nowadays emergency phy-sicians are trained in the use of ultrasound and fast and safe vascular access. Ultrasound is available for out-of-hospital use. to place proper vascular access for an ecMo device is a training issue but thinkable. certainly we need to wait for properly designed studies to apply these strategies broadly, but for special patients, logistic and transport circumstances that might be an approach that could really im-prove survival for prolonged resuscitation ef-fort during cardiac arrest.

References

1. Monsieurs Kg, Nolan JP, Bossaert ll, greif r, Macon-ochie iK, Nikolaou Ni, et al. european resuscitation council guidelines for resuscitation 2015: section 1. executive summary. erc guidelines 2015 Writing group. resuscitation 2015;95:1-80.

2. Perkins gD, Handley aJ, Koster rW, castrén M, smyth Ma, olasveengen t, et al. european resuscitation coun-cil guidelines for resuscitation 2015: section 2. adult basic life support and automated external defibrillation. Adult basic life support and automated external defibril-lation section collaborators. resuscitation 2015;95:81-99.

3. soar J, Nolan JP, Böttiger BW, Perkins gD, lott c, carli

Unfortunately this application of a mechani-cal resuscitation device was not translated into better overall cPr performance, even in the simulation setting. the authors do not report a difference in other cPr quality parameters, dosing and application intervals of drugs and defibrillation between both groups. This is in accordance with the mentioned human studies that also could not improve survival after out-of-hospital cardiac arrest.6, 7

interestingly the authors also looked into the physical activity of the rescue helicopter crew and their cognitive performance after the simulation training with both cPr methods. Nice idea because that is a reason why we in-stall machines in our daily life and also during work, that is to liberate muscle force for better use of our brain resulting in overall better per-formance. Heart rate of the person performing manual cPr was about 20 beats/min higher than the average 80 beats per min while using lUcas. Not a big surprise as we know cPr is demanding but on the other hand that increase was not that much. No difference in heart rate was found with all other time points measured. the claimed enhanced cognitive performance was based on a questionnaire about the pres-ence of “relevant medical information” and an “irrelevant word” memory test. Medical information was remembered better in 6.6%, the “irrelevant words” in 11% in the lUcas group.

the question now is what to do with these results, if the overall performance of the resus-citation during simulated flight was more or less the same. gässler et al.13 also found that the lUcas device complied with the erc-guidelines but the other tested devices worked consistently during their simulated helicopter flight scenarios. The modest increase of heart rate during manual cPr in the current study did not decrease decision making. the fact that in the current simulation study the crew could not remember all information or irrelevant words after the flight also did not affect CPR performance during the flight. The authors claim that lUcas is more effective and less physically demanding and also enhances cog-nitive performance compared to manual cPr.

MecHaNical resUscitatioN DeVices UNDer sPecial circUMstaNces iN tHe oUt oF HosPital settiNg greiF


kins gD. Mechanical chest compression for out of hos-pital cardiac arrest: systematic review and meta-analysis. resuscitation 2015;94:91-7.

11. Brooks sc, anderson Ml, Bruder e, Daya Mr, gaffney a, otto cW, et al. Part 6: alternative techniques and an-cillary Devices for cardiopulmonary resuscitation: 2015 american Heart association guidelines Update for car-diopulmonary resuscitation and emergency cardiovas-cular care. circulation 2015;132(18 suppl 2):s436-43.

12. rehatschek g, Muench M, schenk i, Dittrich W, schewe Jc, Dirk c, et al. Mechanical lUcas resuscitation is ef-fective, reduces physical workload and improves mental performance of helicopter teams. Minerva anestesiol 2016;82:429-37.

13. gässler H, Kümmerle s, Ventzke MM, lampl l, Helm M. Mechanical chest compression: an alternative in heli-copter emergency medical services? intern emerg Med 2015;10:715-20.

14. greif r, lockey as, conaghan P, lippert a, De Vries W, Monsieurs Kg, european resuscitation council guide-lines for resuscitation 2015: section 10. education and implementation of resuscitation. education and imple-mentation of resuscitation section collaborators; col-laborators. resuscitation 2015;95:288-301

15. Maekawa K, tanno K, Hase M, Mori K, asai Y. extra-corporeal cardiopulmonary resuscitation for patients with out-of-hospital cardiac arrest of cardiac origin: a propen-sity-matched study and predictor analysis. crit care Med 2013;41:1186-96.

16. sakamoto t, Morimura N, Nagao K, asai Y, Yokota H, Nara s, et al. saVe-J study group. extracorporeal car-diopulmonary resuscitation versus conventional cardi-opulmonary resuscitation in adults with out-of-hospital cardiac arrest: a prospective observational study. resus-citation 2014;85:762-8.

P, et al. european resuscitation council guidelines for resuscitation 2015: section 3. adult advanced life sup-port. adult advanced life support section collaborators. resuscitation 2015;95:100-47.

4. steen s, liao Q, Pierre l, Paskevicius a, sjöberg t. eval-uation of lUcas, a new device for automatic mechani-cal compression and active decompression resuscitation. resuscitation 2002;55:285-99.

5. timerman s, cardoso lF, ramires Ja, Halperin H. im-proved hemodynamic performance with a novel chest compression device during treatment of in-hospital car-diac arrest. resuscitation 2004;61:273-80.

6. rubertsson s, lindgren e, smekal D, Östlund o, silfver-stolpe J, lichtveld ra, et al. Mechanical chest compres-sions and simultaneous defibrillation vs conventional cardiopulmonary resuscitation in out-of-hospital cardiac arrest: the liNc randomized trial. JaMa 2014;311:53-61.

7. Perkins gD, lall r, Quinn t, Deakin cD, cooke MW, Horton J, et al. ParaMeDic trial collaborators. Me-chanical versus manual chest compression for out-of-hos-pital cardiac arrest (ParaMeDic): a pragmatic, cluster randomised controlled trial. lancet 2015;385:947-55.

8. Wik l, olsen Ja, Persse D, sterz F, lozano M Jr, Brou-wer Ma, et al. Manual vs. integrated automatic load-distributing band cPr with equal survival after out of hospital cardiac arrest. the randomized circ trial. re-suscitation 2014;85:741-8.

9. soar J, callaway cW, aibiki M, Böttiger BW, Brooks sc, Deakin cD, et al. advanced life support chapter col-laborators. Part 4: advanced life support: 2015 interna-tional consensus on cardiopulmonary resuscitation and emergency cardiovascular care science with treatment recommendations. resuscitation 2015;95:e71-e120.

10. gates s, Quinn t, Deakin cD, Blair l, couper K, Per-

Conflicts of interests.—erc Board Director education and training (Dte); ilcor task Force education, implementation, team (eit); editor-in-chief trends in anesthesia and critical care (tacc).Article first published online: December 16, 2015. - Manuscript accepted: December 11, 2015. - Manuscript received: November 22, 2015.(Cite this article as: greif r. Mechanical resuscitation devices under special circumstances in the out of hospital setting. Minerva anestesiol 2016;82:383-5).


again, discussed topic.2-5 in this context some questions arise.

Modern anesthesia techniques have led to a reduction in the need of premedication initially developed to contrast the side effects of ether and chloroform that were widely used;6 actual-ly the question has to be about what we want to achieve with the premedication.

reducing anxiety can be the operator’s main goal, but at least 50% of the patients refuse premedication upon request.7

a positive anxiety can help the patient to better assimilate informations about the peri-operative period, to “share decisions” 8 about anaesthesia and surgery. this is important in particular if the drug is prescribed at home, be-fore patient hospitalization, as often in case of day surgery.9 it is important to remember the BZD induced amnesia that can be interpreted as one of their “beneficial effect”, rather than a complication,4 but amnesia can even reduce patient postoperative satisfaction due to the impossibility to remember the attentions given by doctors.5

another topic can be the choice of a benzo-diazepine: Midazolam is the most prescribed, lorazepam or alprazolam are better suggested as home prescriptions, as well as, recently, Melatonine.10

the choice of the drug can be guided by the onset and offset time, but also on the timing

in this issue of Minerva Anestesiologica, Mijderwijk et al. address the use of benzo-

diazepines (BZD) for premedication for one day surgery patients.1 the well conducted systematic review and meta-analysis includes nineteen studies selected without time interval restriction. the authors focused especially on data regarding time to recovery, psychological phenomena and postoperative somatic symp-toms.

the results show that time to recovery is significantly higher in patients premedicated with BZD (particularly the time for eye open-ing), but the discharge time is not significantly compromised in the BDZ group. Moreover, premedication reduces postoperative headache and nausea, but collected data are inconclusive on psychological phenomena with premedica-tion.

in conclusion, the authors send three mes-sages: removing BDZ from our clinical prac-tice in day case patients to prevent delayed discharge is not justified; BDZ are linked to a 53% reduction of post-operative side effects; more studies are needed to show any benefits on physiological post-operative sequelae with the use of BDZ.

certainly we can recognise that the authors have highlited an interesting and, recently

E D I T O R I A L

the heart of the artMassimiliano carassiti *, rita catalDo

1Unit of anesthesia, intensive care and Pain Management, Department of Medicine, University campus Bio-Medico of rome, rome, italy*corresponding author: Massimiliano carassiti, University campus Bio-Medico of rome, via a. Del Portillo 200, 00128 rome, italy. e-mail: [email protected]

anno: 2016Mese: aprilvolume: 82no: 4rivista: Minerva anestesiologicacod rivista: Minerva anestesiol

lavoro: titolo breve: tHe Heart oF tHe artprimo autore: carassitipagine: 386-8citazione: Minerva anestesiol 2016;82:000-000

comment on p. 438.

Minerva anestesiologica 2016 april;82(4):386-8© 2016 eDiZioni Minerva MeDicathe online version of this article is located at http://www.minervamedica.it

tHe Heart oF tHe art carassiti

vol. 82 - no. 4 Minerva anestesiologica 387

ive relationship instead of a simple informa-tive and skillful preoperative visit may have a positive benefit, improving the anxiety and the confidence. Lawrence DE clearly thought so in 1963 in his paper, a milestone on Humanism in anesthesia.16

Furthermore we strongly think that “medi-cine is the most scientific of the humanities and the most humanistic of the sciences”, and that the anesthesiologist is the internist of the op-erating theatre 17 and certainly the major spe-cialist of sedatives. in the current times of cost saving and high performance pressures, he/she might provide a more humanistic relationship with patients rather than just prescribe a drug. this approach can represent the Heart of the anesthetic art.16

References

1. Mijderwijk H, van Beek s, Duivenvoorden HJ, stolker rJ effectiveness of benzodiazepine premedication on recovery in day-case surgery; a systematic review with meta-analysis. Minerva anestesiol 2016;82:438-64.

2. Maurice-szamburski a, auquier P, viarre-oreal v, cu-villon P, carles M, ripart J, et al. Premed� study in-Premed� study in-vestigators. effect of sedative premedication on patient experience after general anesthesia: a randomized clini-cal trial. JaMa 2015;313:916-25.

3. Fujita t. Presurgery sedation and Patient experience. JaMa 2015;313:2379.

4. Heller ar, Koch t. Presurgery sedation and Patient ex-perience. JaMa 2015;313:2379-80.

5. Maurice-szamburski a. Presurgery sedation and Patient experience-reply. JaMa 2015;313:2380-1.

6. sheen MJ, chang Fl, Ho st. anesthetic premedication: new horizons of an old practice. acta anaesthesiol tai-wan 2014;52:134-42.

7. van den Berg aa. towards needleless induction of an-aesthesia. anaesthesia 2003;58:806.

8. Kon aa. the shared decision-making continuum. JaMa 2010;304:903-4.

9. steeds c. orme r. Premedication. anaesthesia & inten-sive care Medicine 2006;7:393-6.

10. Hansen Mv, Halladin nl, rosenberg J, gögenur i, Møller aM. Melatonin for pre- and postoperative anxiety in adults. cochrane Database syst rev. 2015;4:cD009861.

11. Walker KJ, smith aF. Premedication for anxi-ety in adult day surgery cochrane Database syst rev. 2009;7:cD002192.

12. Bilotta F, lauretta MP, Borozdina a, Mizikov vM, rosa g. Postoperative delirium: risk factors, diagnosis and perioperative care. Minerva anestesiol 2013;79:1066- 76.

13. radtke FM, Franck M, Hagemann l, seeling M, Wer-necke KD, spies cD. risk factors for inadequate emergence after anesthesia: emergence delirium and hypoactive emergence Minerva anestesiol 2010;76:394- 403.

14. Struys M1, Versichelen L, Rolly G. Influence of pre-an-aesthetic medication on target propofol concentration us-

of the scheduled surgery and on the in or out-patient. In fact patients significantly sedated before or after anesthesia may have different problems, needing frequently to be sit on beds or trolleys, requiring a more careful transport and often monitoring;11 furthermore we should not forget that BDZ premedication is risk fac-tor for postoperative delirium.12, 13

the effects of BDZ are surely not well pre-dictable in two types of patients increasingly scheduled for day surgery: the elderly and obese, whose absorption distribution and wash out of drugs are very peculiar. this case de-serves a tailor-made approach.

at this point, it would be interesting to evaluate which type of anesthesia is planned. Patients scheduled for loco-regional anesthe-sia rather than for a general anesthesia clearly require a different approach to the periop-erative sedation. nevertheless, this popula-tion of patients, with some possible residual motor and sensory deficit, would benefit of a fully recovered cognitive status at the dis-charge time. Pharmacokinetic and administra-tion techniques of modern anesthetics make it possible, whereas the response to the BZD, especially when orally taken, is not so easily predictable.

It is sure that premedication modifies the ef-fect of propofol administered by a target con-trolled infusion system, by reducing the dose at which the anesthesia is induced, but for ex-ample increasing the apnea incidence.14 the exact dose-effect relationship between pre-medication and tci is not known and thus not calculable.

Having made these comments it appears that patient’s drowsiness is achieved more eas-ily than his/her calmness. Factors influencing a highly apprehensive state needing premedica-tion have to be identified: personality, family environment, disease and type of surgery, care and attitude of the medical and nurse staff dur-ing hospital stay.

the paper by Mijderwijk et al. provided in-teresting but not final conclusions.

We believe that the anesthesiologist pres-ence can deeply understand and influence the patient 15and that an emotional and support-

carassiti tHe Heart oF tHe art


value of the preoperative visit by an anesthetista study of doctor-patient rapport. JaMa 1963;185:553-5.

17. Egbert LD, Jackson SH. Therapeutic benefit of the an-esthesiologist-patient relationship. anesthesiology 2013; 119:1465-8.

ing a ‘Diprifusor’ tci system during ambulatory surgery anaesthesia 1998;53 (suppl 1):68-71.

15. leigh JM, Walker J, Janaganathan P. effect of preopera-tive anaesthetic visit on anxiety. Br Med J 1977;2:987-9.

16. egbert lD, Battit ge, turndorf H, Beecher HK. the

Conflicts of interest.—The authors certify that there is no conflict of interest with any financial organization regarding the material discussed in the manuscript.Acknowledgments.—We acknowledge Dr lawrence De, for how topical and enlightening is your “old “ article.Article first published online: January 5, 2016. - Manuscript accepted: January 4, 2016. - Manuscript revised: December 23, 2015. - Manuscript received: october 29, 2015.(Cite this article as: carassiti M, cataldo r. the heart of the art. Minerva anestesiol 2016;82:386-8)


septic shock. in these cases, most of the time the patient cannot have expressed her/his “in-formed” consent prior the development of the acute pathological process. Moreover, if the patient was previously competent, there is no legal representative at the time of the manifes-tation of the pathological condition. Finally, such acute conditions make not feasible to nominate a legal representative, especially when the potential benefit of the intervention tested implies its rapid application.4 the re-spect of their human life, dignity, and integ-rity is assumed to coincide with the capacity of providing the available (though not always accurately evidence-based) responsible care.

While awaiting for the application of the novel european Union regulations,5 it could be useful to remind some concepts, which are recurrent in the above debate.

First, the leading term is “emergency”. Whenever, in routine care, the incompetence of a patient is associated with a major clini-cal risk, any due intervention is assured, le-gitimately and legally, without information or consent. this procedure is a common clini-cal practice well accepted both on a national and international level, as long as such in-

We read with interest the article by Zam-peretti et al.,1 published in this issue of

Minerva Anestesiologica, which offers a timely and updated review on the issue of the “informed consent” in emergency situation, providing the opportunity to underline more explicitly some of the points which make this area a recurrent object of both legal and ethical discussion.2 on one side, this debate is an exemplary case of at-tention to a very sensitive issue, i.e. the protec-tion of incompetent patients, who become even more fragile when exposed to “experiments”. on the other, the daily clinical reality documents that in acute severe clinical conditions, when patients are obviously incompetent, the last and least concern is their direct or indirect consent.

Very often, patients in emergencies are not in an appropriate status to be adequately in-formed on a clinical experimentation, since they are either incompetent or under a highly stressed clinical condition.3 think about a pa-tient with dyspnea for acute heart failure and pulmonary edema, an aphasic patient after an acute ischemic stroke, or a patient with dys-pnea and obnubilation for the development of

E D I T O R I A L

How should ethical committees promote research in critically ill patients?

Pietro cairoNi 1, 2 *, roberto FUMagalli 3, gianni togNoNi 4

1Dipartimento di Fisiopatologia Medico-chirurgica e dei trapianti, Fondazione irccs ca’ granda, ospedale Maggiore Policlinico, Università degli studi di Milano, Milan, italy; 2Dipartimento di anestesia, rianimazione ed emergenza Urgenza, Fondazione irccs ca’ granda, ospedale Maggiore Policlinico, Milan, italy; 3Department of anesthesia and intensive care Medicine, school of Medicine and surgery, Università Milano-Bicocca, Niguarda ca’ granda Hospital, Milan, italy; 4Department of cardiovascular research, irccs istituto di ricerche Farmacologiche Mario Negri, Milan, italy*corresponding author: Pietro caironi, Department of Medical and surgical Pathophysiology and of transplants, Fondazione irccs ca’ granda, Università degli studi di Milano, Maggiore Policlinico Hospital, Milan, Via F. sforza 35, 20122, Milan, italy. e-mail: [email protected].


lavoro: titolo breve: ProMotioN oF researcH iN criticallY ill PatieNtsprimo autore: cairoNipagine: 389-91citazione: Minerva anestesiol 2016;82:000-000

comment on p. 479


cairoNi ProMotioN oF researcH iN criticallY ill PatieNts


ence of the albumin italian outcome sepsis (alBios) trial (#Nct00707122) 8 has been an important test of the relevance of this ques-tion. the trial, funded by the italian Medicines agency, has seen the participation of about 100 centers across italy, with the enrollment of more than 1800 patients admitted to icU with severe sepsis, and randomized to two arms to test the comparative benefit profile of intrave-nous albumin replacement as compared to the use of only crystalloids. the prevalent position of the majority of ecs coincided with the rec-ognition that the undefined legislation in rela-tion with the “informed consent” could have been solved by including the suboptimal tool of a “deferred consent”, which is internation-ally employed and accepted as a compromise.2 a minority of ecs, however, requested (after endless discussions and eventually a denial of approval) the impossible compliance with a formal pre-randomization informed consent. the alBios trial has responded to a critical question,9 and generated a further and criti-cally important hypothesis to be verified on the most severe patients, i.e. those with septic shock.10, 11 it is clear that only the coincidence and the articulation of a controlled care and a controlled experimentation can be the “ethi-cal”, i.e. responsible and transparent, strategy to respect the patients’ needs.

Confronted with the permanent undefined position of regulatory authorities, it is up to those who care for incompetent patients in emergency conditions the duty to translate as many as possible situations of uncertain care into experimental research projects. let us hope that also ecs recognize that the safe-guard of the important patients’ rights is hier-archically more mandatory than the respect of formal administrative procedures.

References

1. Zamperetti N, Piccinni M, Bellomo r, citerio g, Mis-traletti g, gristina g, et al. How to protect incompetent clinical research subjects involved in critical care or emergency settings. Minerva anestesiol 2016;82:479-85.

2. Kompanje eJ, Maas ai, Menon DK, Kesecioglu J. Medi-cal research in emergency research in the european Un-ion member states: tensions between theory and practice. intensive care Med 2014;40:496-503.

tervention is a treatment generally accepted to reverse the health-threatening condition.6 second, the key ambiguity of this debate co-incides with the misleading concept according to which “experimental clinical research” is a separate domain of care from the usual clinical treatment. in contrast, clinical research coin-cides with a “mandatory expression of care”, whenever there is a documented uncertainty on one of its component, and there is a plausi-ble, documented, and independently assessed alternative aimed at improving the outcome of patients who do not present any contraindica-tions to the experimental intervention.4 in fact, how has Medicine generally improved over the years? First, by increasing the knowledge of patient physiology and pathophysiology, and second, by gathering evidence on the po-tential benefit of novel clinical treatments ap-plied in specific pathological conditions, based on a solid scientific rationale. Both cases imply clinical experimentations, and correspond to the main aims of the Helsinki Declaration,7 as clearly pointed out by Zamperetti et al..1 third, the protection of patients’ rights is assured pro-vided that a competent and documented evalu-ation has verified that nothing of the due care is withdrawn from the patients assigned, in the randomization process, to the treatment arm, where the uncertain and promising, yet un-proven, novel therapy is tested. in this frame-work, a controlled protocol is certainly a better protection of the rights of all patients enrolled, as compared to the routine and uninformed ex-posure to uncertain interventions. Fourth, the acceptance or the refusal of a protocol by an ec must be based exclusively on the careful evaluation of the methodological coherence of the study protocol with its declared aims.

Despite the evident importance of clinical experimentations, however, in italy different ecs often apply discrepant criteria on how to deal with incompetent patients, with the ex-cuse of the unclear legislation, and the untold reason of not taking responsibility. conse-quently, in doing so, they discriminate de facto critically ill patients from being subjects of clinical research. is really this action ethical?

The recent and highly significant experi-

ProMotioN oF researcH iN criticallY ill PatieNts cairoNi


critically ill. the consensus for Worldwide end-of-life Practice for Patients in intensive care Units (WelPi-cUs) study. am J respir crit care Med 2014;190:855-66.

7. World Medical association Declaration of Helsinki: ethi-cal principles for medical research involving human sub-jects. JaMa 2013;310:2191-2194.

8. caironi P, tognoni g, Masson s, Fumagalli r, Pe-senti a, romero M, et al. albumin replacement in pa-tients with severe sepsis or septic shock. N engl J Med 2014;370:1412-21.

9. caironi P, gattinoni l. the clinical use of albumin: the point of view of a specialist in intensive care. Blood transfus 2009;7:259-67.

10. caironi P, tognoni g, gattinoni l. albumin replace-ment in severe sepsis or septic shock. N engl J Med 2014;371:84.

11. Caironi P, Gattinoni L. Proposed benefits of albumin from the alBios trial: a dose of insane belief. crit care 2014;18:510.

3. Halperin H, Paradis N, Mosesso V Jr, Nichol g, sayre M, ornato JP, et al. recommendations for implementation of community consultation and public disclosure under the Food and Drug administration’s “exception from informed consent requirements for emergency research”: a special report from the american Heart association emergency cardiovascular care committee and coun-cil on cardiopulmonary, Perioperative and critical care: endorsed by the american college of emergency Phy-sicians and the society for academic emergency Medi-cine. circulation 2007;116:1855-63.

4. tognoni g. the challenged but indispensable role of ethi-cal committees for human clinical experimentation. cor-tex 2015;71:420-2.

5. Mentzelopoulos sD, Mantzanas M, van Bg, Nichol g. evolution of european Union legislation on emergency research. resuscitation 2015;91:84-91.

6. sprung cl, truog rD, curtis Jr, Joynt gM, Baras M, Michalsen a, et al. seeking worldwide professional consensus on the principles of end-of-life care for the

Conflicts of interest.—The authors certify that there is no conflict of interest with any financial organization regarding the material discussed in the manuscript.Article first published online: January 28, 2016. - Manuscript accepted: January 25, 2016. - Manuscript revised: January 15, 2016. - Manuscript received: November 15, 2015.(Cite this article as: caironi P, Fumagalli r, tognoni g. How should ethical committees promote research in critically ill patients? Minerva anestesiol 2016;82:389-91)


O R I G I N A L A R T I C L E

italian registry of complications associated with regional anesthesia (ricalor). an incidence

analysis from a prospective clinical surveyMassimo allegri 1-3 *, Dario BUgaDa 2, 3, Paolo grossi 4, alberto Manassero 5,

rosa l. Pinciroli 6, nicola ZaDra 7, guido Fanelli 1, 2, alberto Zarcone 8, rita catalDo 9, giorgio Danelli 10, Battista BorgHi 11, on Behalf of ricalor group

1Department of anesthesiology, intensive care and Pain therapy, University Hospital of Parma, Parma, italy; 2Department of surgical sciences, University of Parma, Parma, italy; 3siMPar group (study in Multidisciplinary Pain research); 4Department of regional anesthesia and Pain therapy, irccs Policlinico san Donato, san Donato Milanese, Milan, italy; 5Department of anesthesiology and intensive care, a.o. “s: croce e carle”, cuneo, italy; 6Department of anesthesiology and intensive care, ospedale civile di legnano, Milan, italy; 7Department of anesthesia and reanimation, azienda ospedaliera, Padova, italy; 8Unit of Day surgery, cliniche Humanitas gavazzeni, Bergamo, italy; 9Department of anesthesia and intensive care University campus Bio-Medico of rome, rome italy; 10Department of anesthesia and Perioperative Medicine, istituti ospitalieri di cremona, cremona, italy; 11Department of Biomedical and neuromotor sciences, University of Bologna, research Unit of anesthesia and intensive care, rizzoli orthopedic institute, Bologna, italy*corresponding author: Massimo allegri, centro di terapia del Dolore, Uo 2ª anestesia, rianimazione e terapia antalgica, azienda ospedaliero-Universitaria di Parma, viale gramsci 14 – 43126 - Parma, italy. e-mail: [email protected]


lavoro: titolo breve: ricalor: an inciDence analYsis FroM a ProsPective clini-cal sUrveYprimo autore: allegripagine: 392-402citazione: Minerva anestesiol 2016;82:392-402

a B s t r a c tBacKgroUnD: regional anesthesia (ra) is associated with many advantages, but side effects also occur. several registries were developed to investigate such complications in many countries, which produced conflicting results. In consideration of the ongoing evolution and improvements in ra, and its widespread diffusion in italy in the last decade (with increasing experience by anesthesiologists), a reappraisal of the incidence and the characteristics of major compli-cations are useful to improve patient’s safety.MetHoDs: a web-based prospective registry was developed in italy with: 1) quarterly report of total anesthetic acts and ra procedures performed; and 2) voluntary registration of complications on dedicated forms. We evaluated incidence of complications, describing their characteristics and outcomes.resUlts: Participants (n.=17 hospitals) registered 117,182 procedures, including 63,692 with ra (54.3%, both as primary anesthetic technique and for postoperative analgesia). a total of 34,147 neuraxial blocks (4954 epidurals/cse, 29,193 subarachnoid blocks) and 29,545 peripheral (single shot and continuous) blocks were registered. total incidence of complication was 4.6/10.000; incidence was 4.1/10,000 for central blocks and 5.1/10,000 for peripheral blocks, long-term neurologic deficit (at 6 months) was observed after an epidural abscess, while other complications did not lead to any long-term adverse outcomes. no hemorrhagic events or other infections have occurred. incidence of major complica-tions was 0.07/1000, while minor complications presented in 0.38/1000 cases.conclUsions: We confirmed RA as generally safe, but monitoring and diagnosis, together with further research ef-forts, are needed to improve patients’ care and clarify potential risk factors.(Cite this article as: allegri M, Bugada D, grossi P, Manassero a, Pinciroli rl, Zadra n, et al. italian registry of complica-tions associated with regional anesthesia (ricalor). an incidence analysis from a prospective clinical survey. Minerva anestesiol 2016;82:392-402)Key words: anesthesia, local - complications - neurologic manifestations - epidemiology.

comment in p. 375.


ricalor: an inciDence analYsis FroM a ProsPective clinical sUrveY allegri


Materials and methods

We performed a prospective observational study, designed according to stroBe guide-lines.38 We received irB approval in the sponsoring center (s. Matteo Hospital - Pa-via, italy) in 2009. the project was registered on clinicaltrials.gov (registration number nct02038491, Principal investigator Massi-mo allegri, MD). advertising of the study was performed in all national anesthesia meetings from 2009 through 2012, as well at the esra italian chapter annual Meeting. any center willing to participate was provided with all the documents needed for local irB approval, and a “local representative” was nominated to co-ordinate the project and to be responsible for data collection.

centers that received irB approval col-lected data from May 2009 to March 2013; both University and non-University Hospitals within national territory enrolled patients un-dergoing different types of procedures includ-ing adult major and minor surgery, pediatric surgery, obstetric anesthesia, and chronic pain management. Data were collected on a pro-spective fashion on a universal form used by all centers, which was sent every three months to the sponsoring center along with a report that included specific details of every side ef-fect occurred.

on the universal form, each local represen-tative was asked to register the total number of interventions and the number of central neur-axial blocks (cnBs), peripheral (PnBs) and continuous peripheral nerve blocks (cPnBs) performed during the study period (from enroll-ment start after irB approval to March 2013). CNBs were classified into two groups: epidur-als or combined spinal–epidurals (cses) and subarachnoid blocks; PnBs and cPnBs were divided into ultrasound (Us)-guided, electri-cal nerve stimulation (ens)-guided and blind techniques. every three months each center had to send the completed form (including all this data) to the coordinating center. non-an-esthesiology physicians often perform regional anesthesia by themselves, and occasionally the published literature contains case reports of

regional anesthesia (ra) is considered safe and beneficial for the patients, but as in all

interventional therapies, the possibility of side effects and complications still exists.1-11 a risk-benefit analysis should be carefully evaluated, with the emphasis being placed on complica-tions both related to technique and drugs used during regional procedures;1, 12-17 furthermore, the incidence of side effects associated with ra in patients receiving new anti-thrombotic drugs is still unknown.18 lack of data and of exhaus-tive guidelines hinder physicians to have clear protocols for their clinical practice, leading to potential personal/hospital-based approaches, resulting in increased patient risk.19

even though knowledge about complica-tions’ incidence is essential for the clinical decision-making and consent processes, there are few prospective trials that can support such discussions.20-24 as side effects related to ra are so rare, neither rct nor meta-analysis of-fers a useful approach to investigate their real incidence. to overcome this problem, national registries in different countries have been de-veloped to explore ra related complications in recent decades,20-34 focusing both on central and peripheral blocks. Unfortunately, the regis-tries have delivered conflicting results 20-36 and have so far provided poor evidence about risk factors for complications. the literature still underlines the importance of new surveys, in-cluding a description of each block’s technical features when side effects do occur, in order to better elucidate possible risk factors.37 impor-tant clinical outcomes are often multifactorial by nature, which makes it difficult to distin-guish surgical, anesthetic or patient etiology as the underlying reason. new knowledge is always needed to update ongoing changes in clinical practice.

in accordance with this data, we designed a prospective web-based multicenter popu-lation-based registry, involving teaching and non-teaching italian hospitals, to detect ra side effects in different clinical settings and their outcome in a long term (six months) follow-up. For each complication we also reg-istered data about the technique, equipment, medications, and patient.

allegri ricalor: an inciDence analYsis FroM a ProsPective clinical sUrveY


“pencil point” or other needles for spinal, soft-tip or sharp needles for PnBs, ultrasound vs nerve stimulation), drug doses (type and dos-age of local anesthetics, as well as type and dosage of adjuvants), performers’ experience in ra and number of attempts to successfully place the block, presence/absence of tourni-quet Finally, all applicable adverse effects (as defined above) were registered. Follow-up was planned at 3 and 6 months for patients with neurological complications, and electromyog-raphy and electroneurography were performed only on patients with persisting neurologic symptoms at 3 or 6 months.

Statistical analysis

our primary endpoint was to assess cumula-tive incidence of adverse effects; incidence of such events was also calculated according to different types of blocks.

the study population was estimated based on previous studies of major complications after cnBs26, which indicated an incidence of major adverse effects in 4.2 of every 100.000 patients. a sample size of 25.000 produced a two-sided 95% confidence interval with a width from 1x10-6 to 2.2 x 10-4.

Quality assessment

a common concern about all registries is that one cannot be totally certain that all data (especially complications) are accurately cap-tured and recorded. We sought to improve ad-herence through ongoing data checks, frequent contacts with local representatives and updates on the status of the study during meetings.

Monthly data integrity checks were made (equivalence between total number of pro-cedures and the sum of all naB, PnB and cPnB), together with a check of complica-tions’ reports (to ensure that no duplication of data, incomplete data, or any issue potentially accounting for a reduction in the validity of data occurred). any issue was immediately ad-dressed through contacts with local representa-tives, in order to ensure completeness and reli-ability of data.

various kinds of difficulties and complications. therefore, this survey includes only local an-esthetic blocks performed by anesthesiologists.

each patient who eventually developed a potentially related complication (as defined below) after ra was recorded as a “case” for this study, which required additional data to be recorded on a separate form. Parents or guard-ians were asked for consent in case of under-age or legally impaired patients. the report of a complication was voluntary; each represen-tative, once aware of a complication in his/her institution, was in charge of reporting to the sponsoring center/data manager.

Specific data — report for complications

all investigators participating to the study were provided by a list of signs to be regis-tered, trying to assure a uniform definition of “complication” between all institutions. events registered as “complications” were:

— death; — cardiac failure/arrest; — respiratory failure/arrest; — major cardiac arrhythmias; — central nervous system complications

(convulsions; coma); — signs of local anesthetic systemic tox-

icity (last) - i.e. circumoral and/or tongue numbness, metallic taste, lightheadedness, diz-ziness, visual and auditory disturbances like difficult focusing and tinnitus, disorientation, drowsiness);

— Postoperative neurologic symptoms (Pons) - i.e. paresthesia after block place-ment, anesthesia, motor impairment;

— signs of infections; — bleeding; — Post Dural Puncture Headache (PDPH).

Data collected in “case” forms included: age, asa physical status class, height and weight, and information on surgical technique (type of surgery, open/video-laparoscopic/robotic approach, scheduled/emergency sur-gery), and on thromboprophylaxis (drugs, tim-ing of suspension before/after blockade).

another section of the report form addressed the performed technique: equipment (needle:



29,193 subarachnoid blocks; a total of 29,545 PnBs/cPnBs were collected: 10,080 Us-guided, 18,506 ens-guided and 959 landmark-based blocks. indications included different types of surgery (visceral major surgery, minor surgery, orthopedics, pediatric surgery, obstet-ric anesthesia, chronic pain management).

We registered 8562 procedures on pediatric patients, with 3474 ra techniques (40.6%).

We collected 5 cases of major complications (death, hemodynamic/respiratory failures, hemorrhagic or infectious complications, per-manent neurologic deficits), with an incidence of 0.07/1000. Minor complications (PDPH, last, transient neurologic symptoms, minor hemodynamic involvement) were registered in 24 cases (0.38/1000).

Epidural blocks/CSE

a total of 2 complications were registered, with a cumulative incidence of 0.4/1000; 100% of the forms were fully completed. one (0.2/1000) of these side effects, an epidural ab-

every time a complication report was sent, we asked for a double check by the local rep-resentative, to make sure that all data about the complications were recorded, to remember about follow-up (in patients with last or at risk for negative outcome), according to the study criteria. We also asked local representa-tives for a check on their institution’s activity, to ensure accuracy on the number of recorded procedures.

Results

General data

seventeen teaching and non-teaching hospi-tals, located in northern (12), central (2) and southern italy (3) participated to the study. We received 88.2% of expected reports from par-ticipants, with 100% of them being returned fully completed.

We registered a total of 117,182 procedures; 63,692 regional anesthesia techniques (54.3%) were performed. a total of 34,147 cnBs were registered: 4,954 epidurals/cses blocks and

Figure 1.—Magnetic resonance showing the epidural abscess reported in our survey. spinal canal is markedly narrowed and spinal cord is compressed, resulting in neurologic symptoms. Multiple airfluid levels with cystic organization are evident inside the abscess.vb: vertebral body; spc: spinal canal; ea: epidural abscess.



among them, two patients (0.07/1000) presented major side effects: in one patient hemodynamic signs were associated with loss of consciousness (without any long-term damage once treated with fluids, oxygen and ephedrine), in one case a high thoracic (t3-t4 dermatomes) sensitive-motor block was regis-tered, with respiratory failure, severe hypoten-sion and bradycardia (table i). three cases of lower limb dysesthesia/paresthesia were reg-istered (0.1/1000); none of them resulted in a permanent neurologic deficit at 6 months. Sev-en cases of PDPH (0.2/1000) were reported; one was in pediatric patient, but none of them resulted in chronic symptomatic dysfunctions worthy of treatment by neurosurgical approach or epidural blood patches.

Peripheral nerve blocks

a total of 15 adverse events, related to pe-ripheral nerve blocks (both single injection or continuous) were registered in 14 patients (one patient had multiple complications, table

scess, occurred with long lasting neurological deficit. The abscess developed 13 days after a thoracic epidural block (8 days after catheter removal) with neurologic symptoms (worsen-ing paraplegia): an MRI revealed the definitive diagnosis, and emergency laminectomy was performed (Figure 1). after rehabilitation, the patient regained sphincters’ control, but bilat-eral sensory (paresthesia) and motor deficit of the lower limb persisted at 6 months, worsen-ing with walking and with need of crutches support. no other exam was performed as fol-low-up because of patient refusal, and a claim was filed to the established authority.

one case of PDPH (0.2/1000) was regis-tered after an epidural block.

Spinal blocks

a total of 12 complications related to sub-arachnoid blocks were registered (table i) with a cumulative incidence of 0.4/1000; 91.7% of the forms was fully completed, with minor de-tails missing.

Table I.—�Description of complications associated to subarachnoid blocks.

surgery type* local anesthetic % volume

(ml) tourniquetexperience of

anesthesiologist (years)

complication

#1 inguinal hernia repair

s HB Bupi 0.5 2.5 n/a - Pons

#2 Femoral synthesis

s HB Bupi 0.5 2 yes 10-30 PDPH

#3 Proctologic surgery

s HB Bupi 0.5 2 n/a < 10 PDPH


s HB Bupi 0.5 2.2 n/a 10-30 PDPH

#5 cesarean section e HB Bupi 0.5 2.3 n/a 10-30 Pons#6 Proctologic

surgerys HB Bupi 0.5 2.2 n/a 10-30 PDPH


s HB Bupi 0.5 2.2 n/a <10 severe hypotension/bradycardia-respiratory

failure#8 ankle surgery e HB Bupi 0.5 1.1 Yes 10-30 PDPH#9 cesarean section e HB Bupi 0.5 2 n/a 10-30 Pons#10 inguinal hernia

repairs HB Bupi 0.5 2.2 n/a < 10 PDPH


s HB Bupi 0.5 2.2 n/a <10 PDPH

#12 open Prostatic adenoma resection

s HB Bupi 0.5 2.2 n/a <10 arrhythmia-loss of consciousness

*s: scheduled; e: emergency; HB Bupi: hyperbaric bupivacaine.



ratory failure. the patient was intubated and then monitored in the post anesthesia care unit until discharge to surgical ward, with no other symptoms or long-term deficits. In one patient a combined involvement of the recurrent la-ryngeal nerve, phrenic nerve and sympathetic cervical chain with claude-Bernard-Horner syndrome was registered during an intersca-lene brachial plexus block, leading to respira-tory failure and need for general anesthesia, but no long-term damage was observed.

three cases were registered as Pons or pro-longed sensory-motor block (0.1/1000): in one case a quadriceps muscle motor dysfunction was observed after a femoral nerve block for anterior cruciate ligament repair, persisting for 20 days with a further complete resolution; in another patient a sensory dysfunction was ob-served after an upper limb block for hand sur-gery, without motor involvement. Finally, one patient presented with a prolonged duration of sensory-motor block (sciatic nerve block - 26 hours) after a combined femoral and sciatic nerve block for total knee replacement, and all symptoms had complete resolution, without any identifiable deficit at 6 months. Unfortu-

ii), with a cumulative incidence of 0.5/1000. 100% of the reports were fully completed.

among the 10 registered cases of last (0.3/1000), three presented minor symptoms (metallic taste, tingling and mouth numbness), while seven patients presented major signs of cardiovascular and neurologic toxicity (table ii).

among the major symptoms, 4 seizures oc-curred, and 2 patients presented severe brady-cardia and cyanosis followed by respiratory arrest and loss of consciousness. the last case was an isolated arrhythmia (av block) without any other symptoms; the patient had already been examined by a cardiologist in the preop-erative setting, establishing that no pacemaker stimulation was required during surgery. after surgery, the patient was admitted to the icU and treated with external pace-making and close monitoring, without any further conse-quence. none of these 7 patients presented any residual deficit at 6 months.

two cases of major hemodynamic and re-spiratory events were observed (0.07/1000): in one case epidural/spinal block spread oc-curred, leading to neurologic symptoms and loss of consciousness, bradycardia and respi-

Table II.—�Description of complications associated to PNB/CPNB.

Block local anesthetic % volume (ml) guidance tourniquet

(yes/no)experience of

anesthesiologist (years)

complication

#1 axillary Mepivacaine 1 30 Us Yes 10-30 last (seizures, arrhythmia)#2 axillary Mepivacaine 2 20 Us - 10-30 last (seizures, arrhythmia)#3 axillary Mepivacaine 2 20 Us Yes <10 last (arrhythmia, respiratory

failure)#4 interscalene Bupivacaine 0.75 75 ens no >30 Phrenic nerve palsy

(respiratory failure)#5 axillary Mepivacaine +

Bupivacaine- 10 ens no > 30 last (minor symptoms)

#6 interscalene ropivacaine 0.75 25 ens no 10-30 last (arrhythmia, respiratory failure)

#7 Femoral ropivacaine 0.75 20 ens Yes >30 Pons#8 axillary levobupivacaine 0.5 25 ens Yes >30 Pons + last (minor

symptoms)#9 sciatic ropivacaine +

Mepivacaine0.5 25 ens Yes >30 Pons

#10 Femoral + sciatic

Mepivacaine 2 - ens Yes <10 last (minor symptoms)

#11 interscalene ropivacaine 0.75 - ens no <10 last (seizures)#12 lumbar

plexusropivacaine 1 20 ens no <10 last (third grade av block)

#13 lumbar plexus

ropivacaine 1 20 Us no 10-30 epidural spread - respiratory failure

#14 axillary ropivacaine 0.75 30 Us no <10 last (seizures)



fection. We did not perform microbiological surveillance, but no patients presented signs of infection in our study. We therefore confirm that PnBs’ associated symptomatic infections are uncommon, but strict attention towards asepsis can potentially further reduce or erase the risk.

Ponss’ incidence in our study has been the lowest reported to date;20, 23, 24 although the in-cidence of permanent injury is very rare, other researchers 36 have noted that mild transient neurological deficits can be quite high (almost 3%). this evidence argues for methodological discrepancies as a causal factor for the high Ponss’ incidence variability that has been re-ported in the literature. in a large meta-anal-ysis, the incidence of 3%, was mainly based on studies conducted in the last years of the nineties (1997-1999): new equipment (soft tip needles, nerve stimulation, ultrasound) have developed, and together with increased experi-ence these may account for a reduction in nerve injury. Further, in those studies PONS defini-tion included also cauda equine syndrome, which is now to be considered as anecdotal (and thus reducing the incidence of neuro-logic deficits). Otherwise, our incidence is the lowest even considering more recent works.23 theoretically, some cases of Pons may not be identified (once patients are discharged and lost to follow-up). the current policy in italy is still to keep the patient in the hospital until the block wares off, and none of the institutions involved in the study do discharge patients at home with indwelling catheters. thus, in the case of our study, the risk of under-reporting of in-hospital Pons should be considered as neg-ligible. otherwise, we underline that we regis-tered those cases of Pons which were already evident during hospital stay, and we did not pursue any systematic post-discharge follow-up (like in the study by Barrington et al.) 23 to identify those patients who develop symptoms of Pons once at home. Future studies should address this issue during study design, since it may justify a lower reported rate of Pons.

as other investigators have suggested,22 we deliberately did not attempt to distinguish be-tween a primary surgical-, positional-, or nerve block - related etiology because it is often im-

nately, basing on our data, we cannot rule out whether these cases were related to ra (intra-neural or intrafascicular injection, nerve injury) or not (surgical tourniquet, patient positioning).

Discussion

through this prospective study involving 17 different italian hospitals, we recorded a total of 63,692 ra procedures (54.3% of the total surgical procedures performed) in differ-ent clinical settings. the incidence of adverse effects after ra technique was 4.6/10,000 patients, supporting the perception of global safety with regional anesthesia.20-40

the incidence of severe complications in epidural blocks matched what was previ-ously found in other studies.20, 25, 30-33 no epi-dural hematomas were recorded: a previous survey 25 reported a higher incidence, but was retrospective, and covered a 10 year period which saw a substantial increase in epidural procedures along with simultaneous improve-ment in guidelines accuracy (about the manage-ment of thrombo-profilaxis, aseptic technique, pre-existing neurologic deficits). The major-ity of spinal hematomas were registered in the earlier years and steadily decreased with time, suggesting that an evolution in technique and awareness of risk factors was instrumental in minimizing this complication. the negative long-term outcome after the one case of epidur-al abscess reminds practitioners that strict vigi-lance on asepsis41 is imperative, and that obser-vation should be maintained even after removal of catheters. We cannot provide an explanation for the lower incidence of severe events during subarachnoid anesthesia, as compared to previ-ous studies,20-26 but a decade has passed since the publication of the paper by auroy et al.,20 and we could argue that improved awareness, as well as better protocols for proper management, may have contributed to this improvement.

the cumulative incidence of complications during PnB and cPnB is similar to what has been reported in previous literature.20, 21 con-sidering infections, previous paper 21 reported catheter contamination rates of 29%, with 3% of patients demonstrating clinical signs of in-



ing to ens guidance (table iii); four cases of last occurred despite ultrasound use (as previously reported in other papers) 47, 48 and, incidentally, all cases of Pons occurred after ens-guided procedures (without any statisti-cally significant correlation) and, interestingly, no nerve injury was observed with Us. Despite the potential for improved safety using Us guidance,49 evidence about reduced incidence of unintentional intravascular needle place-ment and neural damage is not yet available,43 as the effectiveness of Us guidance in improv-ing block safety is inherently limited by its own dependence on operator skill. Finally, data on sedation and LAST are not significant because of the small number of cases, but future sur-veys on the role of benzodiazepines dosages in masking initial symptoms of last could be interesting.

the prospective nature of our survey im-plies that we have a precise assessment of the total number of performed ra techniques (not based on calculations or approximations). all complications were immediately registered and a follow-up was pursued. noteworthy, the complications were reported voluntarily; thus, the reported incidence may not reflect the actual incidence of complications, and the risk of underreporting of complications is still present: despite this, we believe that the present design promoted good quality report-ing; participants often expressed spontaneous interest in improving the quality of data reg-

possible to determine or to rule out multiple causes. We wanted to provide a conservative estimate for associated risks of ra in the con-text of surgery, which we consider, beyond the medico-legal implications, more useful in a patient’s outcome perspective.

additionally, the incidence of last is the lowest reported to date.20-23, 29, 35, 36 We decided to consider also minor symptoms, which could be associated with lower amounts of absorbed la. small amounts can be harmful in patients with preexisting cardiac conduction defects,43 and from a clinical, decision-making perspec-tive, we pursued an inclusive approach taking into account the global incidence of last.

respiratory failure was infrequent in our study (0.005/1000), and was associated with specific higher risk procedures such as sub-arachnoid block, lumbar plexus block and interscalene block. as previously reported,20 these results warn physicians to manage lPB with the same attention of a neuraxial block 5 to always take into account the dangers of per-forming proximal brachial plexus blocks in patients with pre-existing respiratory comor-bidities, because of high reported incidence phrenic nerve involvement (PnP).44, 45 Despite this, the clinical impact of PnP can widely vary, and as we documented in this study, life-threatening events should be probably sup-posed to occur less frequently than expected.

cumulative incidence of complications fol-lowing the use of Us is not different compar-

Table III.—�List of events considered as “complications” and incidence according to the type of block. Data are expressed as number and incidence of events every 1000 procedures, with their 95% confidence interval (in square brackets).

epidural/cse (n.=4.954)

spinal(n.=29.193)

PnB/cPnB(n.=29.545)

ens-guided(n.=18.506)

Us-guided(n.=10.080)

Blind(n.=959)

total(n.=63.692)

Death n/(‰) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0)Major hemodynamic/respiratoryside effects* n/(‰)

0 (0) 2 (0.07)[0.02-0.26]

2 (0.07)[0.02-0.26]

1 (0.05)[0.01-0.36]

1 (0.1)[0.02-0.64]

0 (0) 4 (0.06)[0.02-0.16]

Hemorragic complications n/(‰) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0)infectious complications n/(‰) 1 (0.2)

[0.04-1.16]0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 1 (0.02)

[0-0.11]last n/(‰) 0 (0) 0 (0) 10 (0.34)

[0.20-0.56]6 (0.32)

[0.17-0.63]4 (0.4)

[0.19-0.85]0 (0) 10 (0.16)

[0.08-0.28]Pons/sensory-motor alterations n/(‰)

0 (0) 3 (0.1)[0.04-0.30]

3 (0.1)[0.03-0.30]

3 (0.16)[0.06-0.46]

0 (0) 0 (0) 6 (0.09)[0.04-0.20]

PDPH n/(‰) 1 (0.2)[0.04-1.16]

7 (0.24)[0.13-0.46]

n/a n/a n/a n/a 8 (0.13)[0.07-0.24]

total n/(‰) 2 (0.40)[0.14-1.18]

12 (0.41)[0.27-0.63]

15 (0.51)[0.36-0.72]

10 (0.54)[0.35-0.82]

5 (0.49)[0.26-0.92]

0 (0) 29 (0,46)[0.35-0.60]



compromise in some patients) were reported. although the risk of underreporting is always present, we believe that the practice of regional anesthesia has undergone significant improve-ment in terms of safety and operator’s aware-ness of complications in these 20 years.20 Un-fortunately, catastrophic complications are still reported, but long-term patient injuries may be declining in prevalence. Future, regular sur-veys will be useful to keep pace with the on-going significant changes occurring in current clinical practice.

Key messages

— the use of regional anesthesia re-sulted in a low incidence of major com-plications, despite such complications can easily lead to persistent disability when not properly diagnosed (and treated).

— Monitoring is mandatory for patient’s safety; anesthesiologists, as well as all the personnel involved in patients’ care (nurs-es, surgeons) should always investigate early signs of complications to promptly adopt effective interventions, without any delay. long-term follow up (even after pa-tient’s education to report negative signs) is worth, since complications can be delayed after hospital discharge.

— Pace should be kept with new anti-thrombotic drugs; no hemorrhagic compli-cations were observed in our survey, but new specific guidelines for drug timing ac-cording to regional anesthesia are needed to maintain high standard of safety.

— More studies are needed to define the role of ultrasound in avoiding last and Pons. Despite the low incidence of last, such event is still possible despite Us-guid-ance, and need to be promptly recognized and treated to avoid negative outcomes. since ruling out the cause for Pons is not always easy, anesthesiologist should put in place all strategies to avoid intra-fascicu-lar injection or nerve damage: despite not significant, a slight favorable effect of US-guidance emerged in our survey.

istration (frequent contacts were made asking questions or requesting missing registration forms), and data about total ra procedures is similar to what has been previously reported in retrospective surveys by the same group of investigators.2, 19

additionally, a selection bias may be pres-ent: centers more inclined to practice ra could be preferentially interested in participating in this type of registry, and have greater operator skills and experience. However, we observed that centers performing more ra procedures have a higher incidence of complications, sug-gesting that the greater level of experience does not account for a reduction in compli-cations and possibly could be ruled out as a confounder; otherwise, the higher rate of com-plications in centers that performed more ra may also be due to better patient evaluation for potential complications and patient follow-up.

Finally, a detail is to be considered when rating the risk associated with different ap-proaches to the same plexus. the higher rate of complications associated with interscalene and axillary brachial plexus block is also de-pendent on whether or not the other techniques were used in the same rate (or underused). Un-fortunately, we did not differentiate between approaches to each plexus, but we guess that their increased risk is also related to the over-use of these two approaches in the clinical practice in our country.

Conclusions

We present data from a large prospective clinical registry with immediate registration of ra-associated complications. the major contributions are, however, the report of the absence of hemorrhagic complications both in central and peripheral blocks (despite the increased use of anti-thrombotic agents and the introduction of new anticoagulants drugs during last years), the lack of infections dur-ing peripheral techniques, and the absence of persistent dysfunctions following peripheral nerve blocks. no cases of cauda equina syn-drome, transient neurologic symptoms, or cardiac arrest (despite severe hemodynamic



20. auroy Y, Benhamou D, Bargues l, ecoffey c, Falissard B, Mercier FJ, et al. Major complications of regional an-esthesia in France: the sos regional anesthesia Hotline service. anesthesiology 2002;97:1274-80.

21. capdevila X, Pirat P, Bringuier s, gaertner e, singelyn F, Bernard n, et al. continuous peripheral nerve blocks in hospital wards after orthopedic surgery: a multicenter prospective analysis of the quality of postoperative anal-gesia and complications in 1,416 patients. anesthesiol-ogy 2005;103:1035-45.

22. sites BD, taenzer aH, Herrick MD, gilloon c, anton-akakis J, richins J, et al. incidence of local anesthetic systemic toxicity and postoperative neurologic symptoms associated with 12,668 ultrasound-guided nerve blocks: an analysis from a prospective clinical registry. reg an-esth Pain Med 2012;37:478-82.

23. Barrington MJ, Watts sa, gledhill sr, thomas rD, said sa, snyder gl, et al. Preliminary results of the australa-sian regional anaesthesia collaboration: a prospective audit of more than 7000 peripheral nerve and plexuss blocks for neurologic and other complications. reg an-esth Pain Med 2009;34:534-41.

24. Polaner DM, taenzer aH, Walker BJ, Bosenberg a, Krane eJ, suresh s, et al. Pediatric regional anesthesia network (Pran):a multi-institutional study of the use and incidence of complications of pediatric regional an-esthesia. anesth analg 2012;115:1353-64.

25. Moen v, Dahlgren n, irestedt l. severe neurological complications after central neuraxial blockades in swe-den 1990-1999. anesthesiology 2004;101:950-9.

26. cook tM, counsell D, Wildsmith Ja; royal college of anaesthetists third national audit Project. Major com-plications of central neuraxial block: report on the third national audit Project of the royal college of anaes-thetists. Br J anaesth 2009;102:179-90.

27. Jacob aK, Mantilla cB, sviggum HP, schroeder Dr, Pagnano MW, Hebl Jr. Perioperative nerve injury after total hip arthroplasty: regional anesthesia risk during a 20-year cohort study. anesthesiology 2011;115:1172-8.

28. Jacob aK, Mantilla cB, sviggum HP, schroeder Dr, Pagnano MW, Hebl Jr. Perioperative nerve injury after total knee arthroplasty: regional anesthesia risk during a 20-year cohort study. anesthesiology 2011;114:311-7.

29. orebaugh sl, Kentor Ml, Williams Ba. adverse out-comes associated with nerve stimulator-guided and ul-trasound-guided peripheral nerve blocks by supervised trainees: update of a single-site database. reg anesth Pain Med 2012;37:577-82.

30. Pöpping DM, Zahn PK, van aken HK, Dasch B, Boche r, Pogatzki-Zahn eM. effectiveness and safety of post-operative pain management: a survey of 18 925 consecu-tive patients between 1998 and 2006 (2nd revision):a da-tabase analysis of prospectively raised data. Br J anaesth 2008;101:832-40.

31. Wang lP, Hauerberg J, schmidt JF. incidence of spinal epidural abscess after epidural analgesia. anesthesiology 1999;91:1928-36.

32. green lK, Paech MJ. obstetric epidural catheter-related infections at a major teaching hospital: a retrospective case series. int J obstet anesth 2010;19:38-43.

33. cameron cM, scott Da, McDonald WM, Davies MJ. a review of neuraxial epidural morbidity: experience of more than 8,000 cases at a single teaching hospital. an-esthesiology 2007;106:997-1002.

34. schulz-stübner s, Kelley J. regional anesthesia surveil-lance System: first experiences with a quality assessment tool for regional anesthesia and analgesia. acta anaesthe-siol scand 2007;51:305-15.

35. neal JM. Ultrasound-guided regional anesthesia and patient safety: an evidence-based analysis. reg anesth Pain Med 2010;35(2 suppl):s59-67.

References

1. lirk P, Hollmann M. outcome after regional anesthe-sia: weighing risks and benefits. Minerva Anestesiol 2014;80:610-8.

2. allegri M, niebel t, Bugada D, coluzzi F, Baciarello M, Berti M, et al. regional analgesia in italy: a survey of current practice. eur J Pain suppl 2010;4:219-26.

3. Block BM, liu ss, rowlingson aJ, cowan ar, cowan JA Jr, Wu CL. Efficacy of postoperative epidural analge-sia: a meta-analysis. JaMa 2003;290:2455-63.

4. carli F, Kehlet H, Baldini g, steel a, Mcrae K, slinger P, et al. evidence basis for regional anesthesia in multi-disciplinary fast-track surgical care pathways. reg an-esth Pain Med 2011;36:63-72.

5. enneking FK, Wedel DJ: the art and science of periph-eral nerve blocks. anesth analg 2000;90:1-2.

6. Borgeat a, tewes e, Biasca n, gerber c. Patient-con-trolled interscalene analgesia with ropivacaine after major shoulder surgery: Pcia vs Pca. Br J anaesth 1998;81:603-5.

7. Borgeat a, schappi B, Biasca n, gerber c. Patient-con-trolled analgesia after major shoulder surgery: Patient-controlled interscalene analgesia versus patient control-led analgesia. anesthesiology 1997;87:1343-7.

8. capdevila X, Barthelet Y, Biboulet P, ryckwaert Y, ru-benovitch J, d’athis F. effects of perioperative analgesic technique on the surgical outcome and duration of reha-bilitation after major knee surgery. anesthesiology 1999; 91:8-15.

9. singelyn FJ, Deyaert M, Joris D, Pendeville e, gou-verneur JM. effects of intravenous patient-controlled analgesia with morphine, continuous epidural analgesia, and continuous three-in-one block on postoperative pain and knee rehabilitation after unilateral total knee arthro-plasty. anesth analg 1998;87:88-92.

10. Heil JW, ilfeld BM, loland vJ, sandhu ns, Mariano er. Ultrasound-guided transversus abdominis plane catheters and ambulatory perineural infusions for outpatient in-guinal hernia repair. reg anesth Pain Med 2010;35:556-8.

11. grossi P, allegri M. continuous peripheral nerve blocks: state of the art. curr opin anaesthesiol 2005;18:522-7.

12. Mcleod ga, Dell K, smith c, Wildsmith JaW. Measur-ing the quality of continuous epidural block for abdomi-nal surgery. Br J anaesth 2006;96:633-9

13. Buggy DJ, smith g. epidural anaesthesia and analgesia: better outcome after major surgery? growing evidence suggests so. Br Med J 1999;319:530-1.

14. Wildsmith JaW. state of the art: regional anaesthesia. anaesthesia 2003;58:1200-3.

15. Wildsmith JaW. no sceptic me, but the long day’s task is not yet done: the 2002 gaston labat lecture. reg anesth Pain Manag 2002;27:503-8.

16. Allegri M, Delazzo MG, Grossi P, Borghi B. Efficacy of drugs in regional anesthesia: a review. eur J Pain 2009;3:41-8.

17. Horlocker t, Kopp s. epidural hematoma after epidural blockade in the United states: it’s not just low molecular heparin following orthopedic surgery anymore. anesth analg 2013;116:1195-7.

18. Horlocker tt, Wedel DJ, rowlingson Jc, enneking FK, Kopp sl, Benzon Ht, et al. regional anesthesia in the patient receiving antithrombotic or thrombolytic therapy:american society of regional anesthesia and Pain Medicine evidence-Based guidelines (third edi-tion). reg anesth Pain Med 2010;35:64-101.

19. Bugada D, allegri M, ricalor group investigators, Zadra n, Braschi a, Borghi B, et al. regional anesthesia and anticoagulant drugs: a survey of current italian prac-tice. eur J Pain suppl 2011;5:335-43.



visory on local anesthetic systemic toxicity. reg anesth Pain Med 2010;35:152-61.

44. Urmey WF, talts KH, sharrock ne. one hundred percent incidence of hemidiaphragmatic paresis associated with interscalene brachial plexusanesthesia as diagnosed by ultrasound. anesth analg 1991;72:498-503.

45. Mak PH, irwin Mg, ooi cg, chow BF. incidence of dia-phragmatic paralysis following supraclavicular brachial plexus block and its effect on pulmonary function. an-aesthesia 2001;56:352-6.

46. lee JH, cho sH, Kim sH, chae Ws, Jin Hc, lee Js, et al. ropivacaine for ultrasound-guided interscalene block: 5 ml provides similar analgesia but less phrenic nerve paralysis than 10 ml. can J anaesth 2011;58:1001-6.

47. Baciarello M, Danelli g, Fanelli g. real-time ultrasound visualization of intravascular injection of local anesthetic during a peripheral nerve block. reg anesth Pain Med 2009;34:278-9.

48. Zetlaoui PJ, labbe JP, Benhamou D. Ultrasound guid-ance for axillary plexus block does not prevent intravas-cular injection. anesthesiology 2008;108:761-2.

49. Barrington MJ, Kluger r. Ultrasound guidance reduces the risk of local anesthetic systemic toxicity follow-ing peripheral nerve blockade. reg anesth Pain Med 2013;38:289-97.

36. Brull r, Mccartney cJ, chan vW, el-Beheiry H. neuro-logical complications after regional anesthesia: contem-porary estimates of risk. anesth analg 2007;104:965-74.

37. Barrington MJ. international registries of regional anesthesia: are We ready to collaborate in virtual Departments of anesthesiology? reg anesth Pain Med 2012;37:467-9.

38. von elm e, altman Dg, egger M, Pocock sJ, gøtzsche Pc, vandenbroucke JP; stroBe initiative. the strengthening the reporting of observational stud-ies in epidemiology (stroBe) statement: guidelines for reporting observational studies. J clin epidemiol 2008;61:344-9.

39. Horlocker tt. complications of regional anesthesia. eur J Pain suppl 2010;4:227-34.

40. staender se, Mahajan rP. anesthesia and patient safe-ty: have we reached our limits? curr opin anaesthesiol 2011;24:349-53.

41. Hebl Jr, niesen aD. infectious complications of region-al anesthesia. curr opin anaesthesiol 2011;24:573-80.

42. turnbull DK, shepherd DB. Post-dural puncture head-ache: pathogenesis, prevention and treatment. Br J anaesth 2003;91:718-29.

43. neal JM, Bernards cM, Butterworth JF 4th, Di gregorio g, Drasner K, Hejtmanek Mr, et al. asra practice ad-

Authors’ contributions.—Massimo allegri and Dario Bugada equally contributed to the paper.ricalor group.—laura albonico, Department of anesthesia and intensive care, azienda ospedaliera san Paolo, Milano, italy; andrea l. ambrosoli, Day surgery Unit - azienda ospedaliera ospedale di circolo e Fondazione Macchi - Polo Universitario, varese, italy; vincenzo avallone, Department of anesthesia and intensive care, Presidio ospedaliero cto/Maria adelaide, torino, italy; Marco Baciarello, Department of anesthesiology, intensive care and Pain therapy, University Hospital of Parma, Parma, italy; astrid U. Behr, Department of Medicine, anesthesiology and intensive care, University of Padua, Padua, italy; elisa Bertoldi, Department of Medicine, anesthesiology and intensive care, University of Padua, Padua, italy; chiara Borromeo, Department of regional anesthesia and Pain therapy, irccs Policlinico san Donato, san Donato Milanese, Milan, italy; Barbara Bucci sabatini, Department of anesthesiology and intensive care, azienda ospedaliera Universitaria “Maggiore della carità”, novara, italy; gi-anluca cappelleri, Department of anesthesia, istituto ortopedico g. Pini, Milan, italy; Daniela caristi, Department of anesthesiol-ogy, intensive care and emergency, azienda ospedaliero-Universitaria “ospedali riuniti”, trieste, italy; carlo ceffa, Department of anesthesiology and intensive care, azienda ospedaliera Universitaria “Maggiore della carità”, novara, italy; christian angel compagnone, Department of anesthesiology, intensive care and Pain therapy, University Hospital of Parma, Parma, italy; Pasquale De negri, Department of surgical oncology and Pain Medicine, irccs centro di riferimento oncologico Basilicata, Potenza, italy; Piero Di Mauro, Department of anesthesia and intensive care, azienda ospedaliera san Paolo, Milano, italy; Bruno Dottore, Department of anesthesiology and intensive care, azienda ospedaliero-Universitaria “santa Maria della Misericordia”, Udine, italy; andrea Fanelli, Department of anesthesiology and intensive care, azienda ospedaliero-Universitaria Policlinico s. orsola-Malpighi, Bologna-italy; lavinia Fattorini, Department of anesthesiology and intensive care, asUr Marche av2 ospedale di Jesi, Jesi, italy; Daniela giubelli, Department of anesthesia and intensive care, azienda ospedaliera san Paolo, Milano, italy; Maurizia grazzini, U.o. di anestesia e rianimazione azienda ospedaliera Padova; stefano grossi, Department of regional anesthesia and Pain therapy, irccs Policlinico san Donato, san Donato Milanese, Milan, italy; letterio guglielmo, Department of anesthesia and intensive care, ospedale Buccheri la Ferla FBF, Palermo, italy; anna Maria locati, Day surgery Unit, cliniche Humanitas gavazzeni, Bergamo, italy; Maurizio Marchesini, Department of anesthesiology, intensive care and Pain therapy, University Hos-pital of Parma, Parma, italy; liliana Pasa, Department of anesthesiology, intensive care and emergency, azienda ospedaliero-Uni-versitaria “ospedali riuniti”, trieste, italy; Marzia Peron, Department of anesthesiology and intensive care, ospedale sant’andrea, vercelli, italy; giuseppina sabatino, Department of anesthesia and intensive care, ospedale Buccheri la Ferla FBF, Palermo, italy; alice somenzi, Department of regional anesthesia and Pain therapy, irccs Policlinico san Donato, san Donato Milanese, Milan, italy; gemma ternavasio cameroni, Department of anesthesiology and intensive care, ospedale sant’andrea, vercelli, italy; and-rea tognù, Department of anesthesiology and Postoperative intensive care, rizzoli orthopedic institute, Bologna, italy; Hanna van oven, Department of anesthesiology and Postoperative intensive care, rizzoli orthopedic institute, Bologna, italy; Michelangelo vitiello, Department of anesthesia and intensive care University campus Bio-Medico of rome, rome italy.Funding.—this work has been funded using research funds from istituto ortopedico rizzoli, Bologna, italy.Conflicts of interest.—The authors certify that there is no conflict of interest with any financial organization regarding the material discussed in the manuscript.Congresses.—Preliminary data were presented to the Xvii esra italian chapter Meeting (rome, 2011) and to the Xviii esra italian chapter Meeting (Milan, 2012).Acknowledgements.—all the authors would like to thank Marcus Dilallo — Department of anesthesiology, Hospital for special surgery, new York, nY — because of his precious work in language and grammar polishing.Article first published online: November 18, 2015. - Manuscript accepted: November 13, 2015. - Manuscript revised: October 23, 2015. - Manuscript received: July 23, 2015.


tions to avoid and/or minimize supine-induced gas exchange worsening due to excessive intra-abdominal pressure on dependent lung regions.1, 2 We have recently shown that prior to bariatric surgery (Bs), compared to healthy

current clinical guidelines recommend that critically ill obese patients remain upright

during perioperative and intensive care condi-


Postural effects on pulmonary gas exchange abnormalities in severe obesity

before and after bariatric surgeryeva riVas 1, 2, ebymar arisMeNDi 2, 3, alvar agUstÍ 2-4

concepción gistaU 3, 4, Peter D. WagNer 5, roberto roDrigUeZ-roisiN 2-4

1servei d’anestesiologia, Hospital clínic, Universitat de Barcelona, Barcelona, spain; 2institut d’investigacions Biomèdiques august Pi i sunyer (iDiBaPs) and Fundació clínic per a la recerca Biomédica (FcrB), Universitat de Barcelona, Barcelona, spain; 3ciBer enfermedades respiratorias (ciBeres), Barcelona, spain; 4servei de Pneumologia (institut clínic respiratori [icr]), Hospital clínic, Universitat de Barcelona, Barcelona, spain; 5Department of Medicine, University of california, san Diego (UcsD), ca, Usa*corresponding author: rodriguez-roisin r, servei de Pneumologia (institut clínic respiratori [icr]), Hospital clínic. Villarroel 170, 08036-Barcelona, spain. e-mail: [email protected]


lavoro: titolo breve: PostUral PUlMoNarY gas eXcHaNge eFFects iN oBesitYprimo autore: riVaspagine: 403-10citazione: Minerva anestesiol 2016;82:403-10

a B s t r a c tBacKgroUND: We hypothesized that in morbid obesity, pulmonary gas exchange abnormalities will worsen when supine and that bariatric surgery (Bs) will mitigate this effect.MetHoDs: gas exchange was investigated in 19 morbidly obese and 8 non-obese, age-matched control females, spontaneously breathing ambient air, both upright and supine, before and one year after Bs.resUlts: in control non-obese individuals, no postural changes in arterial blood gases (aBgs) were observed. While obese subjects had more altered Pao2, sao2 and aaPo2 values than controls (P<0.05 each) when upright, the values unexpectedly remained unchanged when supine. this was also the case in the subset of 6 normoxemic obese but the remaining 13 hypoxemic individuals actually improved aBgs when supine: Pao2 (by +2.7±1.3 mmHg, P=0.06), sao2 (by +1.5±0.6%), pH (by +0.01±0.01) and aaPo2 (by -3.4±1.4 mmHg); and cardiac output increased (by +0.4±0.2 l·min-1) (P<0.05 each). after Bs, Pao2 (from 75.5±2.4 to 89.4±2.4 mmHg), aaPo2 (from 27.0±2.0 to 15.4±2.1 mmHg) (P<0.05 each), and pulmonary gas exchange were improved compared to before Bs when upright, but aBgs worsened when supine (Pao2, by -4.6±1.7 mmHg; aaPo2, by +4.2±1.6 mmHg) (P<0.05 each).coNclUsioNs: Before Bs, aBgs are not altered in normoxemic obese subjects moving from upright to supine, even improving in those with hypoxemia when supine. after successful Bs, pulmonary gas exchange improved when upright in all subjects but aBgs deteriorated when supine. However, the important clinical observation is the lack of gas exchange deterioration when supine, which may have implications for critical care and anesthesia settings.(Cite this article as: rivas e, arismendi e, agustí a, gistau c, Wagner PD, rodriguez-roisin r. Postural effects on pul-rivas e, arismendi e, agustí a, gistau c, Wagner PD, rodriguez-roisin r. Postural effects on pul-monary gas exchange abnormalities in severe obesity before and after bariatric surgery. Minerva anestesiol 2016;82:403-10)Minerva anestesiol 2016;82:403-10)Key words: gases, blood - obesity - Pulmonary gas exchange - Bariatric surgery.

comment in p. 377


riVas PostUral PUlMoNarY gas eXcHaNge eFFects iN oBesitY


Measurements

Physiologic blood gases

arterial and mixed venous blood samples gases were analyzed for pH, Po2 and Pco2 (CIBA Corning 800. Medfield, MA, USA), and the alveolar-arterial Po2 gradient (aaPo2) was calculated using the measured exchange respiratory ratio as reported previously.3 oxy-gen uptake (V

.o 2) was calculated using stan-

dard formulae.3

Ventilatory and hemodynamic measure-ments

Minute ventilation was measured using a Wright spirometer (respirometer MK8, Boc-Healthcare. essex, UK). the electrocardio-gram, heart rate, mean arterial pressures and oxygen saturation through a pulseoximeter were continuously monitored (HP M 1001a-1006a &B, 1012a, 1020a, 1046a, 1166a; Hewlett-Packard, Waltham, Ma, Usa) to ensure safety conditions. an 18-gauge plastic cannula was inserted into the radial artery for monitoring systemic arterial pressure and for arterial blood gas sampling. in obese subjects with Pao2<80 mmHg (range, 55-79 mmHg; n, 13), a 7-French triple-lumen thermodilution balloon-tipped pulmonary artery catheter (ed-wards; Baxter Healthcare corporation, irvine, ca, Usa) was inserted under echography guidance to exclude the presence of pulmo-nary arterial hypertension. Pulmonary artery, capillary wedge and right atrial pressures were monitored and pulmonary (PVr) and system-ic vascular resistances (sVr) were calculated using standard formulae. accordingly, in hy-poxemic obese individuals, V

.a/Q

. distributions

were calculated using arterial, mixed venous and mixed expired inert gases concentrations and cardiac output (Q

.t) was determined by

thermodilution.3 in obese individuals without arterial hypoxemia (Pao2 ≥80 mmHg; range, 82-97 mmHg; n, 6), V

.a/Q

. distributions were

estimated without mixed venous sampling and cardiac output was measured by bio-impedance (PhysioFlow®, Manatec Biomedi-

subjects, morbidly obese individuals increased intrapulmonary shunt and ventilation-perfu-sion (V

.a/Q

.) imbalance inducing low arterial

Po2 (Pao2) and increased alveolar-arterial Po2 difference (aaPo2) when upright, whereas af-ter Bs overall gas exchange disturbances were substantially improved in relation to their own pre-operative data.3 Here, we hypothesized that prior to Bs recumbency would aggravate these pulmonary gas exchange abnormalities due to increased intrapulmonary shunt and further ventilation-perfusion (V

.a/Q

.) imbalance

other factors being equal and that Bs would improve them.

We tested this hypothesis in the same non-obese and obese individuals reported previous-ly,3 but separately, given the complexity of the study design. this prompted us to focus exclu-sively on the postural effects of pulmonary gas exchange before and after Bs.


Study population, study design and ethics

the principal characteristics and inclusion criteria of 8 control and 19 obese participants, all females, without major multi-morbidities, such as severe-to-moderate sleep apnea syn-drome, have been previously reported else-where.3 control subjects were normal weight, age-matched females who required thoracic computerized tomography (ct) scans for screening and follow- up of in situ cutaneous melanoma; otherwise, these subjects were completely healthy. We herein report the ef-fects of posture on gas exchange. all mea-surements in both control and obese subjects were always performed when upright with the legs down and when supine using a to-talcare® bed (Hillrom, Pluvigner, France) for 30 min each, in random order. all subjects refrained from any medication during the prior 24 h, before and after Bs (median, 51 weeks). all participants signed informed con-sent. the study was approved by the ethics committee of the Hospital clínic (Protocol 2008/4015).

PostUral PUlMoNarY gas eXcHaNge eFFects iN oBesitY riVas


Results

We enrolled 19 morbidly obese women (age, 51±[se]2 years; Body Mass index [BMi], 45±1 kg/m2), all but one never-smokers and 8 non-obese (BMi, 25±2 kg/m2), sex- and age-matched (50±3 years) never-smokers.

Findings before BS

arterial blood gases (aBgs), pH, oxygen saturation (sao2), inert gases and hemody-namic values in the upright and supine postures are set out in tables i, ii, respectively. Note that clinical and lung function tests, including aBgs, pulmonary and systemic hemodynam-ics and inert gas exchange measurements for all obese individuals when upright have been previously reported 3 but herein reproduced for convenience in both tables for compari-son with those measured when supine. aBgs in non-obese participants were within nor-mal limits (Pao2, 86.5±1.4 mmHg; Paco2, 37.1±1.2 mmHg; aaPo2, 18.4±1.4 mmHg; pH, 7.42±0.01; sao2, 99±1%) when upright with-out significant changes (PaO2, -0.2±2.7 mmHg; Paco2, +1.4±0.8 mmHg; aaPo2, -1.0±2.6 mmHg; pH, +0.01±0.01; sao2, +0.5±0.5%) when supine. similarly, in all obese individuals considered together, Pao2 (+0.6±1.6 mmHg), aaPo2 (-1.1±1.6 mmHg) and sao2 (+0.6±0.5%) did not deteriorate when supine. except for PcWP and sVr, systemic and pulmonary he-modynamic values and V

.a/Q

. descriptors were

not different from upright to supine (table ii). However, there were different postural aBgs effects in obese participants according to the presence or absence of arterial hypoxemia at enrollment (Figure 1). While no postural changes were observed in the 6 normoxemic obese subjects, the 13 hypoxemic individuals (table i, Figures 1, 2) improved aBgs from seated to supine. arterial pH (by +0.01±0.01), and sao2 (by +1.5±0.6%) increased and aaPo2 decreased (by -3.4±1.4 mmHg) (P<0.05 each) when supine, without significant changes in Pao2 (by +2.7±1.3 mmHg; P=0.06) and Paco2. except for a small increase in Q

.t (from

6.6±0.4 to 7.0±0.4, l·min-1) and in PcWP (by

cal. Paris, France).4 Paired inert gas studies were completed in all obese participants.

V.a/Q

. ratio distributions

Measurements of V.a/Q

. distributions were

estimated by the multiple inert gas elimina-tion technique (Miget) in the customary manner.5, 6 arterial and mixed venous blood samples and mixed expired inert gases were collected through a metallic heated box by duplicate. the dispersion of pulmonary distri-bution blood flow and that of alveolar ventila-tion distribution on a logarithmic scale (log sDQ and log sDV; upper normal limits, 0.60 and 0.65, respectively) 7 were calculated. Shunt and dead space were defined, respec-tively, as the fraction of blood flow perfusing units with V

.a/Q

. ratios <0.005 and the frac-

tion of alveolar ventilation distribution with V.a/Q

. ratios >100. all measurements were

determined under steady-state conditions, de-fined by stability (±5%) of both ventilatory and hemodynamic variables, and by the close agreement (±5%) between duplicate measure-ments of mixed expired and arterial oxygen and carbon dioxide.


results are presented as mean±seM. We used the unpaired student t-test to compare control and obese subjects before Bs. a gen-eralized linear mixed model was applied to all obese participants such that all their data were analyzed considering three interventions (Bs, inspired oxygen fraction, and posture). in this model each individual was considered as a random factor to assess the effects induced by each intervention and their potential in-teractions. Given the presence of significant consistent interactions among them, we then focus the analysis on the postural effects while breathing ambient air, before and after Bs, using paired student t-test. corrections were made for multiple comparisons of interven-tions in obese individuals. a P<0.05 value was considered significant at all instances.



table i.—�Gas exchange findings at upright and supine in obese individuals before and after bariatric surgery.Before surgery after surgery

all obese individuals (N.=19)

Upright supine P* Upright supine P‡

Pao2, mmHg 75.5±2.4 † 76.1±2.6 Ns 89.4±2.4 † 84.8±3.1 0.02Paco2, mmHg 39.0±1.0 39.7±1.0 Ns 39.8±1.2 39.7±1.3 NsaaPo2, mmHg 27.0±2.0 † 25.9±2.0 Ns 15.4±2.1 † 19.6±2.1 0.02pH 7.41±0.01 7.41±0.01 Ns 7.41±0.01 7.42±0.01 0.01sao2, % 96±1 † 96±1 Ns 99±1 † 99±1 Ns

Normoxemic obese individuals (N.=6)


Pao2, mmHg 88.4±2.7 † 84.3±5.7 Ns 100.1±3.8 † 97.2±5.8 NsPaco2, mmHg 35.8±1.7 35.9±1.6 Ns 36.3±2.0 34.6±2.3 NsaaPo2, mmHg 17.9±2.2 † 22.1±4.3 Ns 9.6±2.7 † 13.9±2.9 NspH 7.45±0.01 7.43±0.01 Ns 7.43±0.01 7.44±0.01 Nssao2, % 99±1 98±1 Ns 100±0 100±0 Ns

Hypoxemic obese individuals (N.=13)


Pao2, mmHg 69.5±1.9 † 72.3±2.3 0.06 84.4±1.9 † 79.1±2.4 0.01Paco2, mmHg 40.5±1.1 41.5±0.9 Ns 41.4±1.2 42.1±1.1 NsaaPo2, mmHg 31.2±1.9 † 27.8±2.0 0.03 18.1±2.5 † 22.2±2.4 0.07pH 7.39±0.01 7.41±0.01 0.04 7.40±0.01 7.41±0.01 0.01sao2, % 94±1 † 96±1 0.03 99±1 † 98±1 NsValues are expressed as mean±seM. aaPo2: alveolar to arterial Po2 difference; * denotes P-values for comparisons between upright and supine before surgery; † denotes P<0.05 for comparisons at upright between pre- and postoperative conditions; ‡ denotes P-values for com-parisons between upright and supine after surgery.

table ii.—�Pulmonary gas exchange, ventilatory, hemodynamic and metabolic findings at upright and supine in all obese (N.=19) individuals before and after bariatric surgery.

obese individuals

BeforeP*

afterP‡

Upright supine Upright supine

Pvo2, mmHg 38±1 39±1 Ns 39±1 39±1 Nsshunt, %Qt 4.3±1.1 † 3.8±0.6 Ns 1.9±0.4 † 2.0±0.3 Nslog sDQ 0.83±0.06 0.82±0.05 Ns 0.71±0.06 0.82±0.05 Nslog sDV 0.69±0.04 † 0.73±0.05 Ns 0.87±0.07 † 0.72±0.05 0.03Dead space, %Va 23.8±3.5 24.1±2.6 Ns 24.8±1.6 23.9±1.1 NsVe, l·min-1 6.7±0.5 † 6.5±0.5 Ns 5.9±0.5 † 5.4±0.4 NsQt, l·min-1 6.6±0.3 † 6.9±0.4 Ns 5.2±0.2 † 5.5±0.3 Nsci, l·min-1·m2 2.6±0.1 † 2.7±0.1 Ns 2.1±0.1 † 2.2±0.1 NsPsa, mmHg 100±2 † 98±3 Ns 91±3 † 87±3 NsPaP, mmHg 19±1 † 22±2 Ns 13±1 † 17±1 <0.001PcWP, mmHg 8±1 † 11±1 0.04 6±1 † 10±1 <0.001sVr, dyn·s·cm-5 1197±78 1079±71 0.01 1295±84 1236±100 NsPVr, dyn·s·cm-5 132±16 † 122±14 Ns 99±13 † 108±14 NsVo2, ml·min-1 251±14 † 254±17 Ns 191±11 † 182±10 NsValues are expressed as mean±seM; Pvo2: mixed venous Po2; shunt: unventilated units (V

.a/Q

.) ratios <0.005, expressed as % of Q

.t); log

SDQ: dispersion of blood flow distribution and Log SDV, dispersion of ventilation distribution (both dimensionless); dead space: alveolar units with V

.a/Q

. ratios >100, expressed as % of alveolar volume (V

.a); V

.e: minute ventilation; Q

.t: cardiac output; ci: cardiac index; Psa: mean

systemic arterial pressure; PaP: mean pulmonary artery pressure; PcWP: pulmonary capillary wedge pressure; sVr: systemic vascular resist-ance; PVr: pulmonary vascular resistance; V

. o2: oxygen consumption; * denotes P values for comparisons between upright and supine before

surgery; † denotes P<0.05 for comparisons at upright between pre- and postoperative conditions; ‡ denotes P values for comparisons between upright and supine after surgery.

·

··

·

·



creased Pao2 and decreased aaPo2 (P<0.05 each) post-Bs in the upright posture3 (table i, Figure 1). likewise, all abnormal V

.a/Q

. de-

scriptors but the alveolar ventilation disper-sion (log sDV) improved and all hemody-namic outcomes (systemic arterial pressure, Qt, PaP, PcWP, and PVr) and Vo2 decreased (P<0.05 each) after Bs as previously reported (table ii).3 However, aBgs worsened in the supine position: Pao2 decreased (by -4.6±1.7 mmHg), and aaPo2 (by +4.2±1.6 mmHg) and pH (by +0.01±0.01) increased (P<0.05 each) without changes in Paco2 and sao2. While no postural effects were observed in normoxemic obese subjects, hypoxemic obese individuals decreased Pao2 (by -5.3±1.7 mmHg; P=0.01) and increased pH (by +0.02±0.01; P=0.01) when supine without changes in aaPo2 (by +4.2±2.1 mmHg; P=0.07), Paco2 and sao2. Moreover, except for a decrease in log sDV (improvement), intrinsically related to increas-es in both PaP and PcWP (by +4±1 mmHg each) (P<0.05 each), no other changes in V

.a/Q

.

descriptors (Figure 2) and systemic and pul-monary hemodynamics were observed.

Discussion

this study shows that, in a selected popu-lation of morbidly obese female candidates for Bs without major multi-morbidities, pul-monary gas exchange abnormalities includ-ing aBgs remain essentially unchanged when moved from upright to supine postures as was also observed in control subjects. Unexpect-edly, hypoxemic obese individuals actually improved arterial oxygenation in the supine position. after Bs, all obese participants con-sidered together considerably improved pul-monary gas exchange when upright (compared to pre-Bs), but then aBgs deteriorated alone when supine. all pre- and post-operative aBg findings were small and associated with few hemodynamic and V

.a/Q

. changes.

Previous studies

Our findings in normoxemic obese individu-als are at variance with those shown by Fare-

+3±1 mmHg) associated with a decrease in sVr (by -118±40 dyn·s·cm-5) (P<0.05 each), no other effects were observed on pulmonary (V

.a/Q

. ratio distributions) (Figure 2) and/or oth-

er non-pulmonary (minute ventilation and/or V.

o2) factors governing pulmonary gas exchange and on systemic and pulmonary hemodynam-ics. Moreover, in hypoxemic obese individu-als, postural-induced changes in Q

.t and mixed

venous Po2 were linearly correlated (r, 0.62; P<0.03). the median residual sum of squares, the descriptor of the quality of Miget results, was 2.0, below the expected value of 5.4, indi-cating high quality of inert gas data.5

Findings one year after BS

Bs was successful in all obese subjects, as shown by a significant BMI reduction (31 kg/m2) (P<0.001)8. overall, obese individuals in-

Figure 1.—individual arterial Po2 (Pao2) and alveolar-arte-rial Po2 difference (aaPo2) values at upright (U) and supine (s) postures in obese individuals with normal (open symbols) and low Pao2 (closed symbols), before (circles) and after (squares) bariatric surgery. solid bars represent mean values. the 3 hypoxemic individuals who did not improve Pao2 be-fore Bs did not show any distinct features from the remainder 10 hypoxemic subjects who did increase Pao2 when supine.



participants would worsen in parallel to the expected deterioration of shunt and pulmonary vascular disturbances from upright to supine postures before Bs, pulmonary gas exchange abnormalities in our obese subset considered as a single group remained unchanged as well as in obese subjects with normal Pao2 from upright to supine. By contrast, in hypoxemic obese individuals, arterial oxygenation im-proved when supine so that it worsened when upright, a phenomenon characteristically seen, although of slightly more magnitude, in patients with hepatopulmonary syndrome (HPs).11, 12 Nevertheless, the underlying mech-anisms differ. in HPs, upright-induced arte-rial deoxygenation or orthodeoxia is caused by further V

.a/Q

. imbalance without changes

in non-pulmonary (i.e., minute ventilation,

brother et al.9 who investigated normoxemic obese smokers of both sexes, before and after starvation, and showed Pao2 decreases when supine before and after weight loss. Vaughan et al.10 observed no effects on aBgs in mild hypoxemic obese young nonsmoking females, before and at the third day after conventional surgery. Because these two studies differed con-siderably in design and inclusion criteria, direct comparisons with our results are difficult. Dis-crepancies can be also likely explained by our inclusion criteria, which excluded the coexis-tence of moderate to severe multi-morbidities.

Interpretation of findings

contrary to our original hypothesis, i.e. pul-monary gas exchange disturbances in obese

Figure 2.—Distributions of alveolar ventilation (V.a) (open circles) and pulmonary blood flow (Q

.) (solid circles) plotted

versus V.a/Q

. ratio from a representative hypoxemic obese female (age, 55 yrs) when upright (left panels) and when supine

(right panels), before (BMi, 38 kg/m2) and after Bs (BMi, 30 kg/m2). Note that, at each time point, both V.a/Q

. distributions

were always broadly unimodal. Note also that before Bs (top), Pao2 increased and aaPo2 decreased (i.e., arterial oxygena-tion improved) without changes in intrapulmonary shunt and V

.a/Q

. imbalance when supine. all inert dead space values were

reduced. after Bs (bottom), Pao2 increased and aaPo2 decreased (i.e., arterial oxygenation improved) while all but log sDV improved when upright; by contrast, Pao2 decreased and aaPo2 increased (i.e., arterial oxygenation worsened) while no other changes in V

.a/Q

. descriptors were observed when supine (for further explanation, see text).

aaPo2: alveolar-arterial Po2 difference; DS: dead space; Log SDQ: dispersion of blood flow distribution; Log SDV: disper-sion of alveolar ventilation distribution; V

.a/Q

.: ventilation-perfusion.



in all obese and in pre-operatively hypoxemic subjects when supine. although these supine-in-duced deleterious effects on arterial oxygenation remain ultimately unsettled, we have to consider that the log sDV improvement was the result of considerable pulmonary vascular changes within the context of a post-operative remnant of mild obesity.3, 16 Whether the current findings can improve or worsen potential aBg changes during sleep, anesthesia, surgery and critical care medicine settings remains unknown so that further investigations need to be conducted.

Strengths and limitations

there were strengths and shortcomings in our study. First, this is the first study that so far has addressed the effects of postural changes on pulmonary gas exchange before and after Bs through a comprehensive approach, in-cluding measurements of systemic and pul-monary hemodynamics and inert gas studies. second, the same participants were studied before and one year after Bs, thus providing a robust comparative insight into the difficult interplay of pulmonary and non-pulmonary factors governing pulmonary gas exchange.13 However, there were some limitations. one was the small number of patients due to the stringent complexity of our study design. an-other shortcoming was the exclusive inclusion of females given the fact that males have worse gas exchange than females due to the different regional distribution of adipose tissue. accord-ingly, our current data cannot be necessarily extrapolated to very severe obese males or to those who associated severe multi-morbidities.

Conclusions

Before Bs, in normoxemic obese subjects arterial oxygenation is not altered but in hy-poxemic individuals is paradoxically improved when supine. after successful Bs, pulmonary gas exchange considerably improved while ar-terial oxygenation increased in all obese and in pre-operatively hypoxemic individuals when upright but decreased when supine. Postural-induced aBg changes are small and of limited

cardiac output and/or oxygen uptake) deter-minants of gas exchange,13 likely produced by an abnormal pulmonary vasculature with more heterogeneous gravitational blood flow redis-tribution to dependent lung areas.11, 12 By con-trast, in hypoxemic morbidly obese subjects, orthodeoxia is induced by decreased cardiac output without ensuing gravitational negative changes in shunt and V

.a/Q

. mismatching, likely

aggravated by the coexistence of a pulmonary vasculature absent of hypoxic pulmonary va-soconstriction3. should increased cardiac out-put in the hypoxemic morbidly obese when supine be accompanied by higher increases in intrapulmonary shunt, it is most likely that the net effect would have offset the observed im-provement in arterial oxygenation.

alternatively, the absence of aBg chang-es in all obese subjects when supine can be related to the combination of small postural-induced lung volume changes 14, 15 along with the absence of noticeable Q

.t effects. in a re-

cent study in morbidly obese individuals, lung volumes were more restricted than in healthy individuals along with important reductions in expiratory reserve volume and in end-tidal functional residual capacity (Frc) when up-right, but similar when supine.14 in another study investigating supine changes in obese and lean subjects, total lung capacity (tlc) and its subdivisions, it was shown that despite the increased extra-pulmonary mass load in obese subjects, further falls in tlc and Frc were negligible when supine.15 likely, this was not the case in the subset of hypoxemic obese subjects in whom the increase in Q

.t

played a positive influential effect on PaO2 through mixed venous Po2. it remains un-known though what can be the duration of this supine-induced Pao2 improvement in the clinical setting of the real-world of morbidly obese individuals. it may be plausible that af-ter a few hours at supine, the increase in Q

.t is

limited so that the presence of upright-induced arterial deoxygenation is not patent any more.

in parallel to the substantial post-operative weight loss,3 all gas exchange indices but log sDV in all obese individuals improved when upright. However, after Bs, Pao2 decreased



3. rivas e, arismendi e, agustí a, sanchez M, Delgado s, gistau c, et al. Ventilation-perfusion distribution ab-normalities in morbidly obese subjects before and after bariatric surgery. chest 2015;147:1127-34.

4. Brown cV, Martin MJ, shoemaker Wc, Wo cc, chan l, azarow K, et al. the effect of obesity on bioimpedance cardiac index. am J surg 2005;189:547-50; discussion 550-1.

5. roca J, Wagner PD. contribution of multiple inert gas elimination technique to pulmonary medicine. 1. Prin-ciples and information content of the multiple inert gas elimination technique. thorax 1994;49:815-24.

6. Polverino e, gómez FP, Manrique H, soler N, roca J, Barberà Ja, et al. gas exchange response to short-acting beta2-agonists in chronic obstructive pulmonary dis-ease severe exacerbations. am J respir crit care Med 2007;176:350-5.

7. cardus J, Burgos F, Diaz o, roca J, Barbera Ja, Mar-rades rM, et al. increase in pulmonary ventilation-per-fusion inequality with age in healthy individuals. am J respir crit care Med 1997;156:648-53.

8. colquitt J, Picot J, loveman e, clegg a. surgery for obesity (review). cochrane Database syst rev 2009;cD003641.

9. Farebrother MJ, McHardy gJ, Munro JF. relation be-tween pulmonary gas exchange and closing volume be-fore and after substantial weight loss in obese subjects. Br Med J 1974;3:391-3.

10. Vaughan rW, Wise l. Postoperative arterial blood gas measurement in obese patients: effect of position on gas exchange. ann surg 1975;182:705-9.

11. gómez FP, Martínez-Pallí g, Barberà J a, roca J, Navasa M, rodríguez-roisin r. gas exchange mechanism of orthodeoxia in hepatopulmonary syndrome. Hepatology 2004;40:660-6.

12. rodriguez-roisin r, Krowka MJ. Hepatopulmonary syn-drome--a liver-induced lung vascular disorder. N engl J Med 2008;358:2378-87.

13. Rodriguez-Roisin R. Nonpulmonary influences on gas exchange. compr Physiol 2014;4:1455-94.

14. steier J, lunt a, Hart N, Polkey Mi, Moxham J. obser-vational study of the effect of obesity on lung volumes. thorax 2014;69:752-9.

15. Watson ra, Pride NB. Postural changes in lung volumes and respiratory resistance in subjects with obesity. J appl Physiol 2005;98:512-7.

16. arismendi e, rivas e, agustí a, ríos J, Barreiro e, Vidal J, et al. The systemic inflammome of severe obe-sity before and after bariatric surgery. Plos one 2014;9: e107859.

clinical relevance. However, the finding that a short period of time (up to 30 min) when su-pine does not induce arterial deoxygenation prior to surgery can be of interest to anesthesi-ologists for their daily clinical practice. obese individuals are usually placed on supine to induce anesthesia, intubation and instrumenta- and instrumenta-tion before any surgical intervention.

Key messages

— Prior to bariatric surgery, arterial oxygenation in a cohort of obese females without serious multi-morbidities is not altered but in hypoxemic individuals im-proved when supine.

— Preoperative supine-induced arterial oxygenation improvement is associated to an increase in cardiac output without ensu-ing changes in shunt and ventilation-perfu-sion mismatching.

— after successful bariatric surgery, arterial oxygenation significantly increased in all obese and in pre-operative hypox-emic individuals when upright but worsen when supine.

References

1. lewandowski K, lewandowski M. intensive care in the obese. Best Pract res anaesthesiol 2011;25:95-108.

2. shashaty Mgs, stapleton rD. Physiologic and Manage-ment implications of obesity in critical illness. ann am thorac soc 2014;11:1286-97.

Authors’ contributions.—eva rivas, ebymar arismendi, concepción gistau and roberto rodriguez-roisin conducted the experi-mental work and were involved in the acquisition and analysis of the data. roberto rodriguez-roisin, Peter D. Wagner, alvar agustí and eva rivas contributed to the conception, design, and data interpretation of the study. eva rivas, ebymar arismendi, concepción gistau and roberto rodriguez-roisin were involved in the planning and coordination of the study. eva rivas, alvar agustí, Peter D. Wagner and roberto rodriguez-roisin contributed to the writing of the article and/or had substantial involvement in its critical revision before submission.Funding.—the study was funded by the Fondo de investigación sanitaria (Fis) Pi 080311, ciBeres, the generalitat de catalunya (2014sgr661), and by a grant-in-aid by almirall.Congresses.—the work has been presented in the american thoracic society congress 2011, Denver (Usa), in european respira-the work has been presented in the american thoracic society congress 2011, Denver (Usa), in european respira-tory society (ers) congress 2012, Vienna (austria) and in euroanaesthesia 2015, Berlin (germany).Conflicts of interest.—The authors certify that there is no conflict of interest with any financial organization regarding the material discussed in the manuscript.Article first published online: June 9, 2015. - Manuscript accepted: June 4, 2015. - Manuscript revised: May 14, 2015. - Manuscript received: March 3, 2015.


abilities despite receiving multimodal analge-sic therapy.6

according to Bogduk,7 neck and cervical radicular pain are distinct entities. the iasP defines radicular pain as “pain arising in a limb or caused by ectopic activation of nociceptive afferent fibers in a spinal nerve or its roots or other neuropathic mechanisms. structural le-sions can directly compromise the dorsal root ganglion, or indirectly compromise the spinal nerve and its roots by causing ischemia or in-flammation of the axons”. The pain is typically described as knife-like in quality and radi-

Neck and shoulder pain (i.e., cervicobrachi-al pain, [cBP]) is a surprisingly common

spine disorder. in a general adult population in-volving 8356 subjects, swedish investigators reported that 43% suffered from neck pain.1 a recent review by cohen from the Mayo clinic in the Usa reported an annual prevalence rate of CBP exceeding 30%.2 Persistent cBP has been reported in almost 50% of individuals suffering from neck pain,3-5 and 14% of cBP patients report severe pain with physical dis-


treatment of chronic cervicobrachial pain with periradicular injection of meloxicam

lucia aUriNi 1, 2, Battista BorgHi 2, 3 *, Paul F. WHite 2, 4, 5, andrea togNÙ 6,Barbara rossi 2, greta FiNi 2, Pierfrancesco FUsco 7,

Massimiliano Mosca 8, raffaele BorgHi 3, 8

1Department of Medical and surgical sciences, University of Bologna, Bologna, italy; 2research Unit of anesthesia and Pain therapy, rizzoli orthopedic institute, Bologna, italy; 3Department of Biomedical and Neuromotor sciences, University of Bologna, Bologna, italy; 4Departments of anesthesiology, cedars sinai Medical center, los angeles, ca, Usa; 5White Mountain institute, the sea ranch, ca, Usa; 6Department of anaesthesia and Postoperative intensive care, istituto ortopedico rizzoli, Bologna, italy; 7Department of anesthesia and intensive care Unit, san salvatore academic Hospital of l’aquila, l’aquila, italy; 81st orthopedic clinic, rizzoli orthopedic institute, Bologna, italy*Corresponding author: Battista Borghi, Rizzoli Orthopedic Institute, via Pupilli 1, 40136 Bologna, Italy. E-mail: [email protected]


lavoro: titolo breve: MeloXicaM aND cHroNic cerVicoBracHial PaiNprimo autore: BorgHipagine: 411-8citazione: Minerva Anestesiol 2016;82:411-8

a B s t r a c tBacKgroUND: cervicobrachial pain (cBP) is often resistant to conventional oral analgesics. We hypothesized that the periradicular injection of meloxicam would produce a significant reduction in their intractable CBP. The secondary objective was to assess the impact of the treatment on functional recovery.MetHoDs: 48 patients with persistent cBP (>3 months of duration) despite multimodal analgesic therapy received 1-3 periradicular injections of meloxicam, 5-20 mg, at the dermatomal level(s) corresponding to their pain symptoms. Pain level (0=none to 10=severe), rescue analgesics, and functional activity were recorded at baseline and for 90d after the last injection. the injection was repeated if the pain score remained >3 or paresthesia persisted.RESULTS: The mean pain score was reduced from a baseline of 8.9 (±1) to 1.7 (±2.2) at 90 days after the last meloxicam injection. Following meloxicam treatment(s), only 13% of the patients required oral analgesic rescue medication. All patients increased their functional activity level.coNclUsioNs: Cervical periradicular injection of meloxicam reduced CBP by 81% at 90-day follow-up and also improved functional recovery.(Cite this article as: aurini l, Borghi B, White PF, tognù a, rossi B, Fini g, et al. treatment of chronic cervicobrachial pain with periradicular injection of meloxicam. Minerva Anestesiol 2016;82:411-8)Key words: Meloxicam - Spinal injections - Ultrasonography - Brachial plexus neuritis - Chronic pain.

Comment in p. 380.

Minerva Anestesiologica 2016 April;82(4):411-8© 2015 EDIZIONI MINERVA MEDICAthe online version of this article is located at http://www.minervamedica.it

BorgHi MeloXicaM aND cHroNic cerVicoBracHial PaiN


without any significant improvement in their cBP symptoms. thirteen patients (27%) had a herniated cervical disc diagnosed by magnetic resonance imaging (Mri) while in the remain-ing 35 (73%) no structural cause for the cBP symptoms (negative Mri). all patients were initially evaluated to determine their baseline Nrs pain score and presence of any paresthe-sias or other neurological symptoms. a careful physical examination was performed by pain specialist to detect any neurologic impairment. Using a binary (yes/no) questionnaire, the patient’s physical activity was assessed (e.g., ability to work, perform sporting activity, re-quirements for assistance in activities of daily living, and quality of sleep [insomnia]). Pa-tients did not receive any other analgesic treat-ments during the 90-day study period except for rescue analgesic medications.

after the dermatomal level(s) involved in the CBP symptoms was identified (C5-c7), periradicular meloxicam was administered with the patient lying in the lateral position. an ultrasound (Us) transducer was applied transversely to obtain a short-axis view (Fig-ure 1).12, 13 a 22-gauge blunt-tip needle was introduced under real-time Us guidance from posterior to anterior with in-plane technique externally to the foraminal opening.14 When injecting the nerve roots, it was important not to inject near the origin as the dura coats the nerve root at its origin and the injection can in-advertently traverse into the intrathecal space. the periradicular approach is not the same as transforaminal injections as the needle point is ~1.5 cm away from the foramina. this injec-tion approach is more similar to an iM injec-tion in close proximity to the inflamed nerve root.

if only one cervical level was involved, the patient received meloxicam 10 mg (1 mL). However, if the pain involved >1 dermatomal level; the patient received 5 mg (0.5ml) at each level to a maximum total dose of 20 mg (2 mL) per treatment session. after performing the periradicular meloxicam injection(s), the NRS scores were reassessed at 1, 5, 10, 20, 40, 60 min and subsequently at 1, 5, 7, 15, 30 and 90 days via telephone follow-up by a research as-

ates in a dermatomal distribution 8 that is not necessarily the same as neuropathic pain. a study from the Mayo clinic found an annual incidence of cervical radicular pain of 83 per 100,000,9 with c7 involvement in 45-60% of cases. Numerous analgesic modalities have been recommended for managing cBP;10 how-ever, rigorous scientific evidence supporting their efficacy is often lacking.

We recently reported 11 complete pain relief and daily-life activities (including work and sports) resuming in patients with chronic low back pain (lBP) following periradicular injec-tion of meloxicam who had failed to respond to conventional multimodal analgesic therapy. therefore, we conducted a similar case study in cBP patients who failed to respond to multi-modal analgesic therapy. We hypothesized that periradicular injections of meloxicam would reduce CBP scores by >30% from their base-line pain level and lead to improved function activity levels.


after obtaining institutional review board approval for periradicular injection of meloxi-cam and informed consent from the patients, 48 patients (age: 60±13.6 years) with CBP symptoms (mean duration of 4±6.5 months and a range of 3 months to 30 years) were enrolled between 2011 and 2014 in Bologna, italy. inclusion criteria included age >18 years, cBP symptoms for more than 3 months despite receiving multimodal oral analgesic drug ther-apy, a baseline Numeric rating scale (Nrs) pain score >3 on an 11-point NRS with 0=none to 10=severe [worse pain imaginable], and no contraindication to receiving a non-steroidal anti-inflammatory drug (NSAID). Exclusion criteria included known allergy to meloxicam, unstable cardiorespiratory or neurologic dis-orders, or a cutaneous infection at the needle insertion site.

all cBP patients in this case series had been receiving multimodal oral analgesic therapies (e.g., NsaiDs, steroids, paracetamol, oral opioids and gabapentinoids), epidural or peri-facet injections and regular physical therapy

MeloXicaM aND cHroNic cerVicoBracHial PaiN BorgHi


were summarized as mean ± SD and categori-cal variables as frequencies and percentages. Data for patients receiving 1, 2, or 3 injec-tions were compared using one-way aNoVa, followed by a Bonferroni correction, and for categorical variables were assessed using χ² test. repeated-measures general linear models were used to analyze the changes in the levels of pain over time. analyses were performed using IBM SPSS Statistics v.20.

Results

the demographic characteristics and data regarding pain symptoms for the study popula-tion are summarized in table i. the baseline pain score was 9±1 (mean±SD) and the mean duration of CBP symptoms was 4±6.5 years (range: 3 months to 30 years). Data regard-ing the treatment and the physical disability responses are summarized in table ii. the dif-ferent subgroups were based on the changes in their functional status after receiving perira-dicular injection of meloxicam.

twenty-one patients (44%) required only one meloxicam injection, 18 patients (37%) required two injections, and 9 patients (19%)

sistant not involved in administering the peri-radicular meloxicam treatment(s). Changes in the patient physical activity level and quality of sleep were also assessed at each of the as-sessment intervals using a standardized binary response type of questionnaire. the primary endpoint for the study was to achieve a 30% or greater reduction of their Nrs pain score or a Nrs <3. the secondary endpoints included an improvement in their level of physical activity and overall quality of sleep.


an a-priori power calculation determine that a group size of 48 would be adequate to determine a 30% decrease in NRS pain score from the baseline value at the end of the 3 months assessment period assuming a power of 80% and an effect size of 0.37. The calcula-tion was performed using G*Power 3.1.9.2, for a paired-sample, one-tailed t-test examining the difference between two dependent means (matched pairs). given the actual observed de-crease in the cervical pain score from baseline to 90 days (mean 7.25±2.16), the study has a post-hoc power of 0.97. Continuous variables

Figure 1.—the hyperechoic structure represents the rudimental anterior and posterior tubercles (i.e., the “2-humped camel” sign of c4-c6 transverse processes) with the hypoechoic round-to-oval nerve root in between.scM: sternocleidomastoid muscle; ca: carotid artery; ta: anterior tubercle; tP: posterior tubercle.



were reported in 45 of the 48 patients (94%). The meloxicam injections failed to provide significant pain relief in 4 patients (8%) and 2 patients withdrew from the study prior to completing the 90-day follow-up. No patient experienced side effects from the NSAID medication or complications related to the periradicular injections.

Prior to the meloxicam treatment(s), 90% of the patients were receiving non-steroidal anti-inflammatory drugs (NSAIDs), 94% glucocor-ticoids, 85% paracetamol 56% oral opioids, 17% gabapentanoids, 25% received epidural steroids and/or local anesthetics (la), 15% received peri-facet injections, and 65% were administered physical therapy. after receiving the meloxicam injection(s), 42 patients (87%) required no chronic analgesics and the remain-ing 6 patients (13%) required only occasional oral NsaiDs. all patients who completed the study reported significant improvements with respect to their level of physical and functional activity at the 90d follow-up assessment (Ta-ble ii).

the cBP patients population consisted of 15 retired/non-workers and 33 actively em-ployed, and 19 were previously involved in sports-related activities. Due to their cBP, all patients were no longer able to carry out their normal daily activities and suffered from sleep impairment (i.e., chronic insomnia). three pa-tients also needed assistance with the activities of daily living. After receiving the meloxicam treatment(s), all ‘working-age’ patients were able to resume their work-related activities and/or sports-related activities without expe-riencing any ‘rebound’ pain symptoms during the 90-day follow-up period.

Discussion

chronic cBP is an important cause of dis-ability and contributes to loss of functional status.6 Many different pharmacologic and non-pharmacologic therapies have been tried with limited success. currently, there are no evidence-based treatments for chronic radicu-lar pain.15, 16 interestingly, the FDa recently issued a statement that they no longer support

received three injections due to persistent pain. comparisons of the characteristics of patients who received 1, 2, or 3 injections did not re-veal any demographic differences except for a lower BMi in the 3 injections group (aNo-Va F=7.11, P=0.002; post-hoc tests 3 vs. 1, p=0.022, 3 vs. 2, P<0.01). The patients were revaluated for a 2nd or 3rd injection, at 7-30 days intervals and if they continued to report a NRS≥3 and/or paresthesia at one week after the 1st or 2nd injection, the initial treatment was repeated. Overall, pain declined significantly for up to 60 minutes (overall within-subject effect F=90.5, P<0.001) after the first meloxi-cam injection (Figure 2); however, the onset was faster in males after adjusting for age and BMi (Figure 3).

the overall post-treatment Nrs score de-creased significantly over time (overall with-in-subject effect F=10.2, P<0.001) during the 90 days study period, with a mean pain score of 1.7±2.2 at 90 days after the last meloxicam injection (range from 1.2 to 2.5) (Figure 4). complete remission of their cBP symptoms

Table I.—�Demographic characteristics, specific dermatomes levels involved, and duration of CBP symptoms for the 48 CBP patients receiving periradicular injections of meloxicam.

study sample (N.=48)

age (years)Mean±SDrange

60±13.635-88

BMi (kg/m2)Mean±SDrange

26.7±4.419-39

gender (male) 14 (29%)N. of involved levels (c3-c7) 1 (23.48%)

2 (19.40%3 (4.8%)4 (1.2%)5 (1.2%)

Duration of cBP (years)Mean±SDrange

4±6.53 mo-30 yrs

Baseline pain score (0-10) 8.9±1.0Patients who suffered from insomnia 48 (100%)Patients who had to discontinue working activity 33 (69%)

Patients with impaired sport activity 19 (40%)Patients who needed daily assistance 3 (6%)CBP: cervicobrachial pain; BMI: Body Mass Index.



Table II.—�Pain characteristics, meloxicam dosage, number of treatments required, and the early and late (90-day) outcomes following for each of the 48 patient enrolled in this preliminary study.

Patients (N.)

age (years)

Duration of pain (years) Weight (kg) Dermatomal

level(s) treatedBaseline

NRS (0-10)Meloxicam dose (mg) N. treatments NRS at 90 days

post-treatment

Patients experiencing improvement in quality of sleep due to reduced neck pain4 70 20 65 c6-c7 8 5+5 2 05 88 12 75 c5-c6-c7 10 5+5+5+5+5 2 17 68 10 79 c7 7 10 1 011 84 4 68 c6 10 5+5 2 024 86 1 60 c6 10 10 3 231 67 1 74 c6 7 10 1 041 71 0.4 54 c7 9 10 3 443 82 0.3 80 c5 8 10 1 3

Patients able to return to work and experiencing improvement in quality of sleep due to reduced neck pain1 62 30 84 c5-c6-c7 10 5+5+5 2 22 57 23 80 c6-c7 9 5+5 2 03 56 20 80 c6 8 10 1 16 54 10 63 c7 9 10 1 09 54 5 54 c6-c7 10 5+5 3 010 53 5 63 c6-c7 10 5+5 2 112 54 4 78 c6-c7 8 5+5 1 014 64 4 60 c6 5 10 1 020 53 2 65 c7 9 10 2 023 50 1.5 77 c6 10 5+5 2 025 40 1 84 c6-c7 10 5+5 1 332 52 0.9 72 c6 9 10 2 133 49 0.9 117 c6 9 5+5 2 134 49 0.7 52 c6-c7 8 5+5 1 235 64 0.7 70 c6-c7 9 5+5 1 236 36 0.7 46 c6-c7 8 5+5 3 338 63 0.5 68 c6-c7 9 5+5 2 1

40 60 0.5 60 c6-c7 9 5+5 1 145 52 0.25 70 c6 9 5+5 1 1

Patients able to resume sport activities and experiencing improvement in quality of sleep due to reduced neck pain15 69 3 70 c6-c7 10 5+5 1 316 76 3 83 c6 9 5+5 1 219 66 2 111 c5-c6 9 5+5 2 121 77 1.9 90 c7 10 10 2 229 85 1 75.5 c6 8 10 1 0

Patients able to return to work, to resume sport activities and experiencing improvement in quality of sleep due to reduced neck pain

17 41 2.5 75 c6 7 10 1 122 63 1.5 83 c6 8 10 2 026 42 1 65 c5-c6-c7 10 5+5+5 3 228 54 1 125 c6-c7 3 5+5 2 142 53 0.4 82 c7 10 10 3 344 60 0.3 102 c6 9 10 1 146 60 0.25 50 c6-c7 8 5+5 3 147 43 0.25 55 c6 9 10 2 048 58 0.25 76 c7 10 10 1 0

Patients no longer requiring assistance with activities of daily living and able to return to work, to resume sport activities and experiencing improvement in quality of sleep due to reduced neck pain

18 35 2 63 c5-c6-c7 9 5+5+5 2 227 44 1 64 c5-c6-c7 10 5+5+5 3 237 53 0.7 60 c5-c6 10 5+5 2 1

Patients who withdrew informed consent8 52 7 74 c6-c7 10 5+5 1 930 72 1 52 c6 10 10 1 10

Patients not recovered13 84 4 78 c6 8 10 2 639 53 0.5 65 c7 10 10 3 7

Nrs: Numeric rating scale for cervicobrachial pain (cBP).



pain of uncertain origin include NsaiDs and paracetamol. Patients who do not respond to these commonly prescribed non-opioid anal-gesics are often administered oral opioids and/or centrally-active myo-relaxants. However, Peloso et al.18 found limited, if any, scientific evidence supporting the use of myo-relaxants and NsaiDs. some authors have reported on the use of iM injection of la for chronic me-chanical neck disorders and the iV injection of methylprednisolone for acute whiplash,19 as well as acupuncture,20 in the management of CBP. Even IM botulinum toxin A has been tried for cBP but in a placebo-controlled study was no better than saline in improving cervical pain-related disability.15 a review of 359 invasive and non-invasive interventions for neck pain concluded that manual and exer-cise interventions, laser therapy and acupunc-ture were more effective than no treatment; however, none of these non-pharmacologic therapies were clearly superior to any other with respect to either short- or long-term out-comes.21

Despite the FDa warning, both epidural 22 and periradicular injection of steroids com-bined with la are still commonly used for treating cervical neck pain. Karppinen et al.23 observed that periradicular injection of

the common practice of injecting epidural ste-roids and/or las for treating cervical neck pain.17

The first-line analgesic drugs for cervical

Figure 2.—Mean values and 95% confidence intervals of Nrs pain scores. in patients receiving 1 injection (21 pa-tients, 43.8%, black line) pain declined significantly over time (overall within-subject effect F=76.5, P<0.001) and de-creased significantly up to 5 days and stabilized afterwards. in patients receiving 2 injections (18 patients, 37.5%, dis-continuous line) pain declined significantly over time up to 40 minutes and then increased until the second inject when pain declined linearly (overall within-subject effect F=41.7, P<0.001) significantly up to 7 days after. In patients receiv-ing 3 injections of meloxicam (9 patients, 18.8%, grey line) pain declined significantly over time up to 20 minutes and then gradually increased. a decline was subsequently ob-served from the second injection (overall within-subject ef-fect F=31.98, P<0.001). All values were significantly differ-ent from the pre-treatment baseline values (P<0.05).

Figure 3.—the time-course of pain reduction differed sig-nificantly by gender. The onset of pain relief was significant-ly faster in males compared with females after adjusting for age and BMI (P<0.05). A significant gap between males and females was already observed at 1 minute.

Figure 4.—in the overall sample the Nrs score decreased from 9 (at baseline) to 2.4 at 7 days after the last effective treatment (73%) and to 1.7 at 90 days (81%). The NRS change from baseline to 90 days in patients treated with ei-ther 1, 2, or 3 injections of meloxicam was 78%, 89% and 72%, respectively. “End of tx” indicates 7 days after the last effective treatment.



previously (and unsuccessfully) treated cBP using standard multimodal analgesic thera-pies. Importantly, over 90% of the patients in this preliminary study achieved significant relief from their cBP symptoms and also im-provement in their functional activity. these preliminary results strongly support the use-fulness of this technique in patients suffer-ing from chronic cBP which fail to respond to standard multimodal analgesic therapy. However, larger-scale multi-center, double-blinded, placebo-controlled clinical trials are clearly needed in patients with neck and upper extremity radicular pain to confirm the safety and efficacy of periradicular infiltration with meloxicam on short and long-term clinical outcomes.

Conclusions

We conclude that periradicular injection of meloxicam appears to be a novel and highly beneficial therapy for patients with intractable cBP who failed to respond to standard multi-modal analgesic therapies. these results sug-gest that this approach could be used also as early treatment.

Key messages

— these preliminary data demonstrat-ed that ultrasound-guided cervical perira-dicular meloxicam injections are highly ef-fective in treating chronic cBP in patients who failed to respond to standard multi-modal analgesic drug regimens and also early treatment.

— the periradicular injection technique described in this report can be safely per-formed in the ambulatory setting when ul-trasound guidance is used.

— larger-scale prospective, random-ized, double-blind, placebo (or active comparator)-controlled trials are clearly needed to confirm the short and long-term benefits of periradicular meloxicam injec-tions in patients suffering from intractable cBP.

steroids for sciatica produced a transient im-provement in pain symptoms, but, at 3 and 6 months follow-up; however, patients often experienced a return of their pain symptoms. the addition of morphine to a steroid-la combination,24 provide no significant benefit but did increase opioid-related side effects. Despite questionable evidence supporting the clinical use of cervical epidural steroid-la in-jections 25 and the potential for serious com-plications,9 they are still widely used through-out the world.15

in patients with a herniated thoracic disc, stetkarova et al.26 reported that periradicular injection of a steroid-la combination pro-vided significant pain relief. Narozny et al.27 concluded that la nerve root blocks were ef-fective in ‘minor’ mono-radiculopathy. More recently, Borghi et al.10 demonstrated that periradicular injection of meloxicam pro-duced a sustained decrease in chronic lBP and was associated with a significant im-provement in quality of life even in patients who had been previously been treated with periradicular steroids. the periradicular injec-tion technique we employed is not the same as a trans-foraminal injection because the use of Us guidance allows the operator to inject near the nerve after it has passed through the foramina.

Meloxicam (Mobic®, Boehringer ingel-heim) produces a marked anti-inflammatory and analgesic effect.28-30 as a result of its pref-erential COX-2 inhibition, meloxicam is bet-ter tolerated with respect to gastrointestinal and anti-platelet side effects than the classical NsaiDs.31 The use of NSAIDs like meloxi-cam are an important component of all mul-timodal analgesic regimen.32 However, these data suggest that periradicular injection of meloxicam offers significant advantages over standard oral and parenteral administration of this NsaiD.

Although these preliminary findings with periradicular injection of meloxicam are im-pressive, the lack of a placebo (control) group and the failure to ‘blind’ the patients are ob-vious deficiencies in the study design. Nev-ertheless, the enrolled patients had all been



recovery in workers. third edition. elk grove Village, il: american college of occupational and environmen-tal Medicine (ACOEM); 2011:1-332.

16. chou r. Pharmacological management of low back pain. Drugs 2010;70:387-402.

17. agency for Healthcare research and Quality. National guidelines clearinghouse. FDa note: cervical and thorac-ic spine disorders [internet]. available from: www.guide-line.gov/content.aspx?id=35207 [cited 2016 Mar 18].

18. Peloso P, gross a, Haines t, trinh K, goldsmith cH, Burnie s. cervical overview group. Medicinal and injec-tion therapies for mechanical neck disorders. cochrane Database Syst Rev 2007;4:CD000319.

19. Peloso P, Gross A, Haines T, Trinh K, Goldsmith CH, aker P; cervical overview group. Medicinal and injec-tion therapies for mechanical neck disorders. cochrane Database Syst Rev 2005;(2):CD000319.

20. Trinh KV, Graham N, Gross AR, Goldsmith CH, Wang e, cameron iD, et al. cervical overview group. acu-puncture for neck disorders. cochrane Database syst rev 2006 19;(3):CD004870.

21. Hurwitz el, carragee eJ, Van Der Velde g, carroll lJ, Nordin M, guzman J, et al. treatment of neck pain: noninvasive interventions. results of the Bone and Joint Decade 2000-2010 Task Force on Neck Pain and its associated Disorders. J Manipulative Physiol ther 2009;32(2 Suppl):S176-93.

22. stout a. epidural steroid injections for low back pain. Phys Med Rehabil Clin N Am 2010;21:825-34.

23. Karppinen J, Malmivaara a, Kurunlahti M, Kyllönen e, Pienimäki t, Nieminen P, et al. Periradicular infiltration for sciatica. Spine 2001;26:1059-67.

24. castagnera l, Maurette P, Pointillart V, Vital JM, erny P, sénégas J. long-term results of cervical epidural steroid injection with and without morphine in chronic cervical radicular pain. Pain 1994;58:239-43.

25. Manchikanti l, singh V, Derby r, Helm s, trescot aM, staats Ps, et al. review of occupational medicine prac-tice guidelines for interventional pain management and potential implications. Pain Physician 2008;11:271-89.

26. Stetkarova I, Chrobok J, Ehler E, Kofler M. Segmental abdominal wall paresis caused by lateral low thoracic disc herniation. Spine 2007;32:E635-9.

27. Narozny M, Zanetti M, Boos N. Therapeutic efficacy of selective nerve root blocks in the treatment of lumbar radicular leg pain. Swiss Med 2001;131:75-80.

28. Ogino K, Saito K, Osugi T, Satoh H. Meloxicam (Mo-bic): a review of its pharmacological and clinical profile. Nippon Yakurigaku Zasshi 2002;120:391-7.

29. Euller-Ziegler L. Meloxicam: a review of its pharmacoki-netics, efficacy and tolerability following intramuscular administration. Inflamm Res 2001;50(Suppl 1):S5-9.

30. Fleischmann R. Meloxicam. Expert Opin Pharmacother 2002;3:1501-12.

31. ghozlan Pr, Bernhardt M, Vélicitat P, Bluhmki e. tol-erability of multiple administration of intramuscular meloxicam: a comparison with intramuscular piroxicam in patients with rheumatoid arthritis or osteoarthritis. Br J Rheumatol 1996;35(Suppl 1):51-5.

32. White PF. Multimodal analgesia: its role in preventing postoperative pain. Curr Opin Invest 2008;9:76-82.

References

1. guez M, Hildingsson c, Nilsson M, toolanen g. the prevalence of neck pain. a population-based study from northern Sweden. Acta Orthop Scand 2002;73:455-9.

2. cohen sP. epidemiology, diagnosis, and treatment of neck pain. Mayo Clin Proc 2015;90:284-99.

3. enthoven P, skargren e, oberg B. clinical course in pa-tients seeking primary care for back or neck pain: a pro-spective 5-year follow-up of outcome and health care con-sumption with subgroup analysis. Spine 2004;1:2458-65.

4. Manchikanti l, Boswell MV, singh V, Benyamin rM, Fellows B, abdi s, et al.; asiPP-iPM. comprehensive evidence-based guidelines for interventional techniques in the management of chronic spinal pain. Pain Physician 2009;12:699-802.

5. Hogg-Johnson s, van der Velde g, carroll lJ, Holm lW, cassidy JD, guzman J, et al. the burden and determi-nants of neck pain in the general population: results of the Bone and Joint Decade 2000-2010 Task Force on Neck Pain and its associated Disorders. J Manipulative Physiol Ther 2009;32:46-60.

6. côté P, van der Velde g, cassidy JD, carroll lJ, Hogg-Johnson s, Holm lW, et al. the burden and determi-nants of neck pain in workers: results of the Bone and Joint Decade 2000-2010 Task Force on Neck Pain and its associated Disorders. J Manipulative Physiol ther 2009;32:S70-86.

7. Bogduk N. the anatomy and pathophysiology of neck pain. Phys Med Rehabil Clin N Am 2011;22:367-82.

8. salt e, Wright c, Kelly s, Dean a. a systematic literature review on the effectiveness of non-invasive therapy for cervicobrachial pain. Man Ther 2011;16:53-65.

9. Radhakrishnan K, Litchy WJ, O’Fallon WM, Kurland LT. epidemiology of cervical radiculopathy. a population-based study from Rochester, Minnesota, 1976 through 1990. Brain 1994;117:325-35.

10. Diwan S, Manchikanti L, Benyamin RM, Bryce DA, geffert s, Hameed H, et al. effectiveness of cervical epi-dural injections in the management of chronic neck and upper extremity pain. Pain Physician 2012;15:E405-34.

11. Borghi B, aurini l, White PF, Mordenti a, lolli F, Borghi r, et al. Long-lasting beneficial effects of periradicular injection of meloxicam for treating chronic low back pain and sciatica. Minerva Anestesiol 2013;79:370-8.

12. Narouze sN, Vydyanathan a, Kapural l, sessler Di, Me-khail N. Ultrasound-guided cervical selective nerve root block: a fluoroscopy-controlled feasibility study. Reg An-esth Pain Med 2009;34:343-8.

13. Martinoli c, Bianchi s, santacroce e, Pugliese F, graif M, Derchi LE. Brachial plexus sonography: a tech-nique for assessing the root level. aJr am J roentgenol 2002;179:699-702.

14. Narouze s, Peng PWH. Ultrasound-guided interventional procedures in pain medicine: a review of anatomy, sono-anatomy, and procedures – Part II: axial structures. Reg Anesth Pain Med 2010;35:386-96.

15. Hegmann Kt. cervical and thoracic spine disorders. oc-cupational medicine practice guidelines. evaluation and management of common health problems and functional

Funding.—research support was provided by the rizzoli orthopedic istitute, Bologna, italy, and by the White Mountain insitute, the sea ranch, ca, Usa.Conflicts of interest.—The authors certify that there is no conflict of interest with any financial organization regarding the material discussed in the manuscript.Acknowledgements.—the authors wish to thank Dr. Maria specchia for providing a M-turbo Ultrasound system sonosite echography.Article first published online: September 3, 2015. - Manuscript accepted: August 14, 2015. - Manuscript revised: July 29, 2015. - Manuscript received: April 7, 2015.


nary edema (Pe) is among the most common causes of weaning failure and, more specifical-ly, of failure to successfully complete a spon-taneous breathing trial (sBt).4 the well-docu-mented underlying pathophysiology 5 indicates a high likelihood of reversibility with appro-priate treatment.4 the diagnosis of Pe, howev-er, can be challenging. the reference standard diagnostic tool is right heart catheterization,

Weaning from endotracheal mechanical ventilation (MV) can raise many chal-

lenges, which may affect the outcomes of patients in the intensive care unit (icU). the weaning process may contribute 40% to 50% of the total MV duration. Many factors can re-sult in weaning failure, which occurs in about 30% of patients.1-3

left ventricle (lV) dysfunction with pulmo-


Diagnostic accuracy of hemoconcentration for pulmonary edema as the cause of weaning failure

Véronique Pottier 1 *, Xavier Valette 2, amélie segUiN 2, romain MassoN 2, Jean-Jacques ParieNti 3, Bertrand saUNeUF 2, Damien DU cHeYroN 2,4, Nicolas terZi 2, 5

1Department of anesthesia and surgical intensive care, caen University Hospital, France; 2Department of Medical intensive care, caen University Hospital, France; 3Department of clinical research, caen University Hospital, France; 4University of caen Basse-Normandie, U2rM-ea4655, caen, France; 5University of caen Basse-Normandie, iNserM U1075, caen, France*corresponding author: Department of anesthesia and surgical intensive care, caen University Hospital (6th floor), avenue Côte de Nacre, 14033 caen cedex, France. e-mail: [email protected]


lavoro: titolo breve: HeMocoNceNtratioN For DetectiNg Pe DUriNg WeaNiNg FailUreprimo autore: Pottierpagine: 419-28citazione: Minerva anestesiol 2016;82:419-28

a B s t r a c tBacKgroUND: our objective was to assess the diagnostic accuracy of hemoconcentration for cardiogenic pulmonary edema (PE) as the cause of weaning failure, using left ventricular filling pressure elevation assessed by transthoracic echocardiography as the reference standard.METHODS: This prospective observational study included 41 patients who failed their first spontaneous breathing trial of weaning from mechanical ventilation. they were divided into two groups, with and without Pe by echocardiographic criteria. Hemoconcentration and other hemodynamic parameters were compared between the groups.resUlts: the group (N.=21) with Pe by echocardiographic criteria had a higher frequency of failure of the second spontaneous breathing trial (P=0.03) and a longer total weaning time (P=0.02) compared to the other group. the receiver-operating characteristics curve indicated that changes in plasma protein or hemoglobin concentration from initiation to completion of the second spontaneous breathing trial did not predict Pe as the cause of failure (areas under the receiver-operating characteristics curve, 0.47±0.09 and 0.51±0.09, respectively). the only factor predicting failure due to Pe was a positive fluid balance from intensive care unit admission to study inclusion (P=0.01). The increase in mean arterial blood pressure seemed suggestive of weaning failure due to cardiac causes.coNclUsioNs: Compared to echocardiographic criteria for left ventricular filling pressure elevation, hemoconcentra-tion assessed based on plasma protein and hemoglobin levels did not help to diagnose cardiogenic Pe as the cause of weaning failure.(Cite this article as: Pottier V, Valette X, seguin a, Masson r, Parienti J, sauneuf B, et al. Diagnostic accuracy of hemocon-centration for pulmonary edema as the cause of weaning failure. Minerva anestesiol 2016;82:419-28)Key words: Ventilator weaning - Blood proteins - Pulmonary edema.


Pottier HeMocoNceNtratioN For DetectiNg Pe DUriNg WeaNiNg FailUre


Nord-Ouest III) classified the research project as a study of standard care, approved the pro-tocol (a10-D13-Vol.10), and waived the re-quirement for informed consent. each patient or next of kin was given a study information sheet at study inclusion.

Study protocol

the patients were included at initiation of a second sBt. Before performing the second sBt, we recorded the following: heart rate (Hr); mean arterial blood pressure (MaP); respiratory rate (rr); plasma protein, hemo-globin, and BNP; arterial blood gas levels in a blood sample drawn from an arterial cath-eter; an electrocardiogram; and TTE findings. the lV ejection fraction (lVeF) was as-sessed using the biplane simpson’s method. Pulsed-wave Doppler was performed to re-cord transmitral flow for measurement of the early E and late A mitral inflows, and the E/A ratio was computed. tissue Doppler imag-ing of the lateral mitral annulus on the api-cal four-chamber view allowed peak velocity measurement (ea) and determination of the e/ea ratio.

a second set of measurements was per-formed at the end of the sBt, i.e., just before extubation in patients who successfully com-pleted the 2-hour trial and just before recon-nection to the ventilator in the other patients. tte (sonos 7500, Philips Medical systems, andover, Ma, Usa; then cX 50 from the same manufacturer starting in December 2011) was carried out by physicians having over two years of experience with Doppler echocardiog-raphy in the icU. all measurements were re-viewed by a single person, who was blinded to patient data. Plasma protein was assayed using the synchron system (Beckman coulter, Vil-lepinte, France) and hemoglobin using sysmex Xe-2100 (sysmex corporation, Kobe, Japan).

the patients were divided into two groups, with and without Pe by echocardiographic cri-teria (Pe+ and Pe- groups). in keeping with previous data,6 we defined LV dysfunction as e/a>0.95 and e/ea>8.5, and we took lV dys-function to indicate Pe.

although the findings may prove difficult to in-terpret during marked inspiratory efforts with large intrathoracic pressure fluctuations or in patients with dyspnea. studies evaluating non-invasive diagnostic investigations showed that transthoracic echocardiography (tte) was ac-curate for detecting LV filling pressure eleva-tion, a sign of lV dysfunction with Pe.6 Fur-thermore, laboratory findings such as serum brain natriuretic peptide (BNP) elevation 4, 7 and plasma protein elevation as a marker for hemoconcentration 8 performed well versus right heart catheterization.

acute cardiogenic Pe is related to an in-crease in LV filling pressure and, consequently, in pulmonary capillary hydrostatic pressure, leading to the extravasation of hypo-oncotic fluid from the pulmonary capillaries to the lung interstitium and alveoli. This fluid is char-acterized by a low protein level, usually lower than half the protein level in the plasma. in acute cardiogenic Pe, the semipermeable cap-illary membrane is intact. When the amount of extravasated fluid exceeds 3 L, hydrostatic PE may result in hemoconcentration that mani-fests as increases in the plasma protein and he-moglobin concentrations.9-11

the aim of this study was to evaluate the di-agnostic accuracy of hemoconcentration dur-ing an sBt for detecting Pe as the cause of sBt failure. our reference standard was echo-cardiographic evidence of LV filling pressure elevation during the sBt.


From august 2010 to March 2012, we in-cluded consecutive patients having failed a 2-hour sBt on a t-tube after at least 48 hours of MV. exclusion criteria were preexisting neuromuscular disease, tracheostomy, mitral valve disease, and chronic or acute (onset during the weaning trial) atrial fibrillation. Both mitral valve disease and atrial fibril-lation hinder the interpretation of echocar-diography findings.6 Weaning was attempted when patients met previously published cri-teria.12

the appropriate ethics committee (cPP

HeMocoNceNtratioN For DetectiNg Pe DUriNg WeaNiNg FailUre Pottier


was non-normal and they were therefore com-pared using the nonparametric Mann-Whitney U test. Qualitative data were described as n (%) and compared using Fisher’s exact test. the two groups were compared using stu-dent’s t-test for independent samples. the plasma protein difference between patients with and without Pe was 5 g/l (+7 g/l and +2 g/l, respectively) in the seminal work by anguel et al. 8 to have 80% power to detect a standardized plasma protein difference of 5 g/l with alpha set at 5%, 16 patients were re-quired in each group.

P values <0.05 were considered significant. receiver-operating characteristics (roc) curves were plotted for differences between the Pe+ and Pe- groups regarding changes in plasma protein and hemoglobin concentrations from sBt initiation to completion. statistical analyses were performed using sigmastat 3.5 and sigmaplot 10.0 (systat software inc., chi-cago, il, Usa), and sas 9.4 software (sas institute, cary, Nc, Usa).

Results

Study patients

Figure 1 is the flow chart of weaning out-comes. During the study period (august 2010-March 2012), 834 patients in our medi-

Data collection

We collected the following baseline patient characteristics: age, sex, Body Mass index (BMi), comorbidities such as heart disease or respiratory failure, reason for icU admission, and illness severity at admission assessed us-ing the Simplified Acute Physiology Score (saPs ii) 13 (table i). We also collected the following data on endotracheal MV: MV du-ration before inclusion, total MV duration, du-ration of sedation, duration of weaning from MV (time from the first SBT to extubation), and duration of the second sBt. changes from icU admission to study inclusion were recorded for body weight and plasma protein and albumin concentrations. We collected the daily urinary output, body temperature, and use of renal replacement therapy at inclusion; lVeF at icU admission or weaning initia-tion, and at inclusion; outcome of the second sBt; and use of the following drugs: anti-biotics, cardiovascular drugs (angiotensin-converting enzyme inhibitors, beta-blockers, and calcium-channel inhibitors), and diuretics (table ii).


Quantitative data were described as mean±standard deviation; their distribution

Table I.—�Baseline characteristics of the patients with and without pulmonary edema.Pe+

(n=21)Pe-

(n=20) P value

age (years) 61±14 60±11 0.72Male gender 19 (91%) 13 (65%) 0.07comorbidities

Heart disease 9 (43%) 6 (30%) 0.52respiratory failure 8 (38%) 6 (30%) 0.74

reason for icU admissionrespiratory distress 8 (38%) 10 (50%) Nashock 4 (19%) 7 (35%) Nacardiac arrest 5 (24%) 1 (5%) Nacoma 3 (14%) 2 (10%) Naacute kidney injury 1 (5%) 0 Na

BMi 28±9 30±7 0.14saPs ii 49±16 50±15 0.89Death 4 (19%) 1 (5%) 0.34Quantitative data are mean±standard deviation and categorical data are n (%).PE: pulmonary edema; ICU: intensive care unit; BMI: Body Mass Index; SAPS II: Simplified Acute Physiology Score version II;13 Na: not applicable.



4 patients had either ventilator-associated pneumonia (N.=2) or critical-illness neu-romuscular abnormalities (N.=2) and were finally extubated after a mean of 1±3 addi-tional days.

Comparison of patients with and without PE: failure of the second spontaneous breathing trial

the two groups showed no statistically significant differences at baseline (Table I). table ii displays the clinical characteristics during the icU stay and compares the two groups regarding variables collected during the second SBT. SBT failure was signifi-cantly more common in the Pe+ group (57% vs. 20%, P=0.03), which had a longer total weaning process and longer time from the second sBt to extubation. Moreover, patients with Pe during the second sBt had a larger change in body weight from icU admission to study inclusion (4±6 vs. -2±9, P=0.01) and a shorter duration of the second sBt (79±42 minutes vs. 111±17 minutes, P=0.02) com-pared to those without Pe diagnosed based on tte criteria.

cal icU required MV. among them, 103 (12%) met noninclusion criteria and 214 (26%) died before weaning was attempted. of the remaining 517 patients, 476 (92%) were extubated after successfully completing the first SBT. This left 41 patients, who were included in the study at the second sBt. the main reason for MV was acute respiratory failure.

Weaning outcomes

of the 41 patients, 21 (51%) had Pe as as-sessed using tte during the second sBt (Pe+ group) and 20 did not (Pe- group). of the 21 Pe+ patients, 9 were extubated after success-fully completing the second sBt and survived. the other 12 Pe+ patients (12/41, 29%) failed the second SBT and were finally extubated after a mean of 2±4 additional days. in these 12 patients, investigations showed no cause of weaning failure other than Pe. of the 12 pa-tients, 4 died in the icU, at least 48 hours after extubation.

of the 20 Pe- patients, 16 (80%) were extubated after successfully completing the second sBt and survived. the remaining

Table II.—�Clinical characteristics in patients with and without pulmonary edema.Pe+

(n=21)Pe-

(n = 20) P value

lVeF at icU admission (%) 44±15 48±11 0.61MV duration before inclusion (days) 10±5 10±7 0.57total MV duration (days) 13±7 11±8 0.34Duration of sedation (days) 5±5 5±4 0.90Duration of weaning from MV (days) a 6±5 5±4 0.21Weight change, icU admission to inclusion (Kg) 4±6 -2±9 0.01Plasma protein level change, icU admission to inclusion (g/l) -1±8 3±12 0.44Plasma albumin level change, icU admission to inclusion (g/l) -5±6 -2±8 0.22lVeF at inclusion (%) 51±14 56±13 0.60Urine output at inclusion (l/day) 2.3±1.6 2.3±1.3 0.66Body temperature at inclusion (°c) 37.4±0.7 37.5±0.5 0.76antibiotic therapy at inclusion 13 (62%) 12 (60%) 1Diuretic therapy at inclusion 7 (33%) 5 (25%) 0.73cardiovascular drug therapy at inclusion 16 (76%) 12 (60%) 0.33rrt at inclusion 2 (10%) 0 0.49Duration of second sBt (min) 79±42 111±17 0.02Failure of second sBt 12 (57%) 4 (20%) 0.03Days from second sBt failure to extubation 2±4 1±3 0.02Quantitative data are mean±standard deviation and categorical data are n (%).aTime from the first spontaneous breathing trial to extubation.Pe: pulmonary edema; lVeF: left ventricular ejection fraction; icU: intensive care unit; MV mechanical ventilation; rrt: renal replacement therapy; sBt: spontaneous breathing trial.



Figure 1.—Flow chart of patients and weaning outcomes.



and hemoglobin were not significantly differ-ent. only the MaP increase at sBt comple-tion was significantly greater in the PE+ group (Figure 2).

table iii reports the changes in clinical, lab-oratory, and Doppler echocardiography vari-ables during the second sBt in the Pe+ and Pe- groups. the changes in plasma protein

Figure 2.—changes in plasma protein and brain natriuretic peptide concentrations, e/ea ratio, and mean arterial pressure during the second spontaneous breathing trial in the groups with and without pulmonary edema.

Table III.—�Clinical, laboratory, and echocardiographic changes induced by the second spontaneous breathing trial in predefined groups.

Pe+ (N.=21) Pe- (N.=20)*P value

Before sBt end of sBt change Before sBt end of sBt change

e/a ratio 1.2±0.3 1.2±0.3 0.1±0.2 0.9±0.3 0.9±0.3 0.1±0.2 0.51e/ea ratio 9.4±2.5 11.7±2.4 2.3±2.2 7.5±2.7 7.5±2.6 0.03±1.4 0.0003Plasma protein level (g/l) 52±7 54±6 2±3 53±9 56±9 2±2 0.95Hemoglobin level (g/dl) 9.5±1.4 9.8±1.3 0.3±0.5 10.4±2.3 10.6±2.3 0.2±0.4 0.92pH 7.45±0.03 7.44±0.06 -0.01±0.04 7.47±0.03 7.46±0.03 -0.02±0.02 0.56Pao2 (kPa) 11±2 10±3 -2±2 12±3 10±4 -1±2 0.99Paco2 (kPa) 5±1 5±1 0.3±0.6 5±1 5±1 0.2±0.4 0.92BNP (pg/ml) 1062±1582 1182±1788 120±233 565±759 589±803 24±95 0.12Hr 87±16 96±18 9±9 100±15 105±17 5±12 0.22MaP (mmHg) 88±8 95±9 8±9 89±13 90±14 1±8 0.04rr 22±6 28±7 6±5 21±7 27±8 4±5 0.29Quantitative data are mean±standard.Pe: pulmonary edema; sBt: spontaneous breathing trial; lVeF: left ventricular ejection fraction; BNP: brain natriuretic peptide; Hr: heart rate; MaP: mean arterial blood pressure; rr: respiratory rate.*student’s t-test for independent samples comparing mean change during sBt in the Pe+ and Pe- groups.



center study.6 However, tte is operator-dependent and exhibits limited interobserver and intraobserver reproducibility in icU pa-tients 14, 15. Finally, the previous study 8 re-quired two failed sBts, instead of one in our study, and the two studies may have differ-ences in terms of weaning criteria, sBt tim-ing, and case-mix.

The same team confirmed their results by showing a correlation between blood volume contraction and an increase in extravascular lung water. all patients were also assessed by right heart catheterization.16 Nevertheless, in another study,17 none of the tested diagnosis tools, including plasma protein and hemoglo-bin concentrations, helped to predict the out-come of the sBt.

Four findings from our study deserve spe-cial attention. First, positive fluid balance, de-fined as weight gain from ICU admission to inclusion (initiation of the second sBt), was the only predictor of sBt failure, in keeping with earlier work.7, 18, 19 Thus, aggressive fluid and sodium restriction after the acute dis-ease phase might improve sBt success rates and decrease MV duration in patients having failed a first SBT. Second, in our patients with Pe and sBt failure, time from sBt initiation to ventilator reconnection was longer than the 30 minutes recommended at an international consensus conference held in april 2005 and was significantly different between the groups with and without Pe.12 a study of e/a and e/ea to diagnose increased PaoP as a cause of weaning failure showed a similar differ-ence between the groups with and without Pe (35±22 min vs. 50±10 min, respectively), although the time from sBt initiation to re-connection was far shorter than in our study (about half the time in the Pe group).6 thus, although weaning failure due to lV dysfunc-tion occurs early during the sBt, a 30-minute test time may be too short to detect weaning failure of cardiac origin.6, 9 in addition to sBt duration, parameters such as fluid balance should be taken into account before decid-ing to extubate, and the appropriateness of treatments such as diuretics, vasodilators, and postextubation noninvasive ventilation must

ROC curve analysis of diagnostic accuracy

the area under the roc curves for changes in plasma protein and hemoglobin concentra-tions from sBt initiation to completion as pre-dictors of Pe were 0.47±0.09 and 0.51±0.09, respectively. thus, neither variable showed any usefulness for diagnosing Pe as the cause of sBt failure.

Discussion

Our main finding is that failure of a second sBt due to Pe, diagnosed based on tte evi-dence of LV filling pressure elevation, was not associated with hemoconcentration, defined as increases in plasma protein and hemoglobin levels.

We are aware of a single previous study evaluating changes in plasma protein con-centration during an sBt.8 of 46 patients who had failed two previous sBts, 24 ex-perienced PE defined as a pulmonary artery occlusion pressure (PaoP) increase to more than 18 mmHg as determined by right heart catheterization before and after a third sBt. in this subgroup of 24 patients, the plasma protein concentration increased by 11% (range, 3%-25%), and a greater than 6% in-crease had 87% sensitivity and 95% specific-ity for Pe. However, this study did not as-sess the value of a plasma protein increase in indicating sBt failure: although all pa-tients with PaoP>18 mmHg at sBt comple-tion failed the trial, 47% of those with lower PaoP values also failed the trial. the authors suggested that the change in plasma protein concentration between starting and ending the sBt might serve as an alternative to right heart catheterization for assessing weaning-induced Pe. the discrepancies between our findings and theirs may be related to several factors. PAOP is extremely difficult to inter-pret in patients undergoing sBts, most nota-bly those with dyspnea and large pleural and intrathoracic pressure swings during inspira-tory efforts. We therefore used tte, which proved accurate versus pulmonary artery catheter pressure measurements in a single-



selected patients (i.e., with coPD). thus, our findings may not apply to other populations. Zapata et al. 7 reported that BNP elevation during sBt predicted weaning failure due to heart failure, whereas ouanes-Besbes et al. 17 found that Nt-proBNP was of no help in pre-dicting the sBt outcome. our results further indicate that BNP and Nt-proBNP fail to pre-dict weaning failure due to cardiac causes. Given our findings and the conflicting data in the literature, caution is in order when using these biomarkers.

several limitations of our study must be ad-dressed. this is a single-center study with a very low frequency of failure of the first SBT. therefore, the general applicability of our re-sults may be limited. second, we detected Pe using tte instead of the reference standard, i.e., right heart catheterization. echocardiogra-phy can be used to estimate the end-diastolic lV pressure as a means of diagnosing wean-ing failure due to cardiac causes.14, 15 We used a combination of two echocardiography vari-ables, e/a>0.95 and e/ea>8.5, obtained at the end of the weaning trial. this choice deserves discussion. Both variables are easily assessed on a single echocardiography window. thus, they are of special interest in cases of weaning failure that require a rapid evaluation.6 Howev-er, their assessment is operator-dependent and their variability can make their interpretation challenging. assuming that e/ea<8 indicates normal LV filling pressures and E/Ea>15 high LV filling pressures then there is a gray zone in the 8-15 range that requires the use of other variables to detect a filling pressure increase, such as the e-wave deceleration time (eDt) measured by transmitral Doppler.21 these variables were first widely validated in cardi-ology patients who were breathing spontane-ously (i.e., were not receiving endotracheal ventilation) and only subsequently suggested as relevant for the echocardiographic diagno-sis of Pe. thus, the tte variables used in our study, although validated in previous work, may be highly sensitive but less specific for Pe,6 as some patients may have elevated lV filling pressures with no hydrostatic pressure increase within the pulmonary arteries.11 these

be discussed. third, the MaP increase at sBt completion seemed associated with weaning failure of cardiac origin, as in previous stud-ies.6, 8, 9, 14 there is no clear evidence in the literature to support this possibility, which, however, is consistent with the well-described pathophysiology of weaning failure due to cardiac causes,5 as well as with the effects of controlled MV on the cardiovascular system and the impact of physiological changes on spontaneous breathing. controlled MV pro-duces a positive intrathoracic pressure, there-by decreasing cardiac flow by diminishing the systemic venous return and also decreasing cardiac work by reducing the lV afterload. in contrast, switching from positive-pressure ventilation to spontaneous ventilation is as-sociated with an increase in venous return and a negative intrathoracic pressure, which increase the lV afterload and myocardial oxygen consumption. consequently, sBt in-creases cardiac work and may induce cardio-genic Pe via PaoP elevation.5 lV afterload is mainly represented by MaP. thus, an MaP increase during sBt suggests weaning failure of cardiac origin. Fourth, although baseline BNP was elevated in the group that developed Pe, the discriminating power of this sign was low (0.66±0.09). Baseline BNP eleva-tion probably correlates with weight gain and, therefore, with higher filling pressures before any sBt. indeed, there is evidence that many factors such as advanced age, sepsis, fluid management, mechanical ventilation, female gender, diuretics, angiotensin-converting en-zyme inhibitors, beta-blockers, renal dysfunc-tion, and pulmonary arterial hypertension are confounders that can interfere with BNP ho-meostasis, in the absence of cardiac dysfunc-tion. in addition, published data on BNP are conflicting. Mekontso-Dessap et al. 4 reported that the baseline plasma BNP level before the first weaning attempt independently predicted weaning failure and correlated with weaning duration. grasso et al. 20 found that plasma Nt-proBNP levels showed significantly greater increases in patients with weaning failure due to cardiac causes. the main limitation of our study is the relatively small number of highly



References

1. esteban a, Frutos F, tobin MJ, alia i, solsona JF, Valver-du i, et al. a comparison of four methods of weaning pa-tients from mechanical ventilation. spanish lung Failure collaborative group. N engl J Med 1995;332:345-50.

2. ely eW, Baker aM, Dunagan DP, Burke Hl, smith ac, Kelly Pt, Jet al. effect on the duration of mechanical ventilation of identifying patients capable of breathing spontaneously. N engl J Med 1996;335:1864-9.

3. esteban a, anzueto a, Frutos F, alia i, Brochard l, stewart te, et al. characteristics and outcomes in adult patients receiving mechanical ventilation: a 28-day inter-national study. Jama. 2002;287:345-55.

4. Mekontso-Dessap a, de Prost N, girou e, Braconnier F, lemaire F, Brun-Buisson c, et al. B-type natriuretic pep-tide and weaning from mechanical ventilation. intensive care Med. 2006;32:1529-36.

5. lemaire F, teboul Jl, cinotti l, giotto g, abrouk F, steg g, et al. acute left ventricular dysfunction during unsuc-cessful weaning from mechanical ventilation. anesthesi-ology 1988;69:171-9.

6. lamia B, Maizel J, ochagavia a, chemla D, osman D, richard c, et al. echocardiographic diagnosis of pulmonary artery occlusion pressure elevation during weaning from mechanical ventilation. crit care Med 2009;37:1696-701.

7. Zapata l, Vera P, roglan a, gich i, ordonez-llanos J, Betbese aJ. B-type natriuretic peptides for prediction and diagnosis of weaning failure from cardiac origin. inten-sive care Med 2011;37:477-85.

8. anguel N, Monnet X, osman D, castelain V, richard c, teboul Jl. increase in plasma protein concentration for diagnosing weaning-induced pulmonary oedema. inten-sive care Med 2008;34:1231-8.

9. teboul Jl, Monnet X, richard c. Weaning failure of car-diac origin: recent advances. crit care 2010;14:211.

10. Ware lB, Matthay Ma. clinical practice. acute pulmo-nary edema. N engl J Med 2005;353:2788-96.

11. Figueras J, Weil MH. increases in plasma oncotic pres-sure during acute cardiogenic pulmonary edema. circula-tion 1977;55:195-9.

12. Boles JM, Bion J, connors a, Herridge M, Marsh B, Melot c, et al. Weaning from mechanical ventilation. eur respir J 2007;29:1033-56.

13. Le Gall JR, Lemeshow S, Saulnier F. A new Simplified acute Physiology score (saPs ii) based on a european/North american multicenter study. Jama 1993;270:2957-63.

14. caille V, amiel JB, charron c, Belliard g, Vieillard-Bar-on a, Vignon P. echocardiography: a help in the weaning process. crit care 2010;14:r120.

15. Jardin F, Vieillard-Baron a. Weaning failure from cardio-vascular origin. intensive care Med 2006;32:937; author reply 8.

16. Dres M, teboul Jl, anguel N, guerin l, richard c, Monnet X. extravascular lung water, B-type natriuretic peptide, and blood volume contraction enable diagnosis of weaning-induced pulmonary edema. crit care Med 2014;42:1882-9.

17. ouanes-Besbes l, Dachraoui F, ouanes i, Bouneb r, Jalloul F, Dlala M, et al. Nt-proBNP levels at spontane-ous breathing trial help in the prediction of post-extuba-tion respiratory distress. intensive care Med 2012;38: 788-95.

18. Upadya a, tilluckdharry l, Muralidharan V, amoateng-adjepong Y, Manthous ca. Fluid balance and weaning outcomes. intensive care Med 2005;31:1643-7.

19. Frutos-Vivar F, Ferguson ND, esteban a, epstein sK, arabi Y, apezteguia c, et al. risk factors for extubation

considerations invite questions about the va-lidity of echocardiography for diagnosing Pe in the icU, where examination conditions are far from optimal. Finally, our choices regard-ing the timing of the sBt and of the second measurements during the sBt may explain a number of discrepancies with other studies. Moreover, we did not evaluate the multiple confounding factors that may have affected the development of sBt failure. in particular, the impact of weight gain in our study may have been overestimated.

Conclusions

our study does not support the diagnostic usefulness of plasma protein or hemoglobin concentration monitoring during sBts to de-tect Pe, diagnosed based on validated echocar-diographic criteria, as the cause of sBt failure. other studies are needed to identify the best echocardiographic criteria for heart failure in patients receiving MV and to assess their in-terobserver and intraobserver variability in a range of tte conditions. However, our data suggest that close attention to fluid balance control during the weaning process might im-prove the frequency of successful sBts. Fi-nally, an MaP increase during sBt suggests weaning failure of cardiac origin.

Key messages

— Neither plasma protein nor hemoglo-bin levels help to detect pulmonary edema, diagnosed based on validated echocardio-graphic criteria, as the reason for failing a second spontaneous breathing trial.

— the only factor predicting failure of spontaneous breathing trial failure due to pulmonary edema was a positive fluid balance from intensive care unit admis-sion to initiation of the second spontaneous breathing trial.

— a larger increase in mean arterial pressure during the second spontaneous breathing trial suggested weaning failure due to pulmonary edema.



chronic obstructive pulmonary disease. crit care Med 2007;35:96-105.

21. Nagueh sF, appleton cP, gillebert tc, Marino PN, oh JK, smiseth oa, et al. recommendations for the evalua-tion of left ventricular diastolic function by echocardiog-raphy. eur J echocardiogr 2009;10:165-93.

failure in patients following a successful spontaneous breathing trial. chest 2006;130:1664-71.

20. grasso s, leone a, De Michele M, anaclerio r, ca-farelli a, ancona g, et al. Use of N-terminal pro-brain natriuretic peptide to detect acute cardiac dysfunction during weaning failure in difficult-to-wean patients with

Funding.—the study received institutional support only.Conflicts of interest.—The authors certify that there is no conflict of interest with any financial organization regarding the material discussed in the manuscript.Congresses.—this work was presented in part at the annual congress of the Société de Réanimation de Langue Française (srlF) held in January 2013, in Paris.Acknowledgements.—We thank the intensive care unit physicians, nursing staff, and secretaries for contributing to the study; as well as a Wolfe for revising the manuscript.Article first published online: September 16, 2015. - Manuscript accepted: September 10, 2015. - Manuscript revised: September 8, 2015. - Manuscript received: January 8, 2015.


pulmonary resuscitation (cPr) during transport is challenging as recent manikin 6 and human studies 7 have demonstrated that the quality of manual chest-compression deteriorates during transport and prolonged cPr attempts amongst others due to increasing fatigue of the emer-

High quality chest-compressions and re-duced hands-off intervals are important de-

terminants for survival with good neurological outcome after cardiac arrest.1-5 However, cardio-


Mechanical lUcas resuscitation is effective, reduces physical workload and improves mental performance of helicopter teams

gregor reHatscHeK 1, 2, Marcus MUeNcH 1, ingmar scHeNK 3, Wienand DittricH 4, Jens-christian scHeWe 2, cornelius DirK 5, rudolf HeriNg 1, 2 *

1Department of anesthesia, intensive care, emergency and Pain Medicine, Kreiskrankenhaus Mechernich gmbH, Mechernich, germany; 2Department of anesthesia and intensive care Medicine, University Hospital Bonn, germa-ny; 3Forel Klinik, competence center for addiction Medicine, ellikon an der thur, switzerland; 4Kci competence center for interdisciplinary economics and Behavioral economics, Frankfurt am Main, germany; 5Department of oral technology, University of Bonn, germany*corresponding author: rudolf Hering, Department of anesthesia, intensive care, emergency and Pain Medicine, Kreiskrankenhaus Mechernich gmbH, st.-elisabeth str. 2-6, 53894 Mechernich, germany. e-mail-address: [email protected]


lavoro: titolo breve: MecHaNical VersUs MaNUal resUscitatioN iN HelicoPtersprimo autore: reHatscHeKpagine: 429-37citazione: Minerva anestesiol 2016;82:429-37

a B s t r a c tBacKgroUND: Physical and mental workload during cardiopulmonary resuscitation (cPr) is challenging under ex-treme working conditions. We hypothesized that the mechanical chest-compression device lund University cardiac assist system (lUcas) increases the effectiveness of cPr, decreases the physical workload and improves the mental performance of the emergency medical service (EMS) staff during simulated emergency helicopter flights.METHODS: During simulated helicopter flights, 12 EMS teams performed manual or LUCAS-CPR on a manikin at random order. Compression depth, rate, overall time of compressions, application of drugs and defibrillation were re-corded to test the quality of cPr. Heart rate monitoring of eMs members was used as a surrogate of physical workload. Cognitive performance was evaluated shortly after each flight by a questionnaire and a memory test about medical and extraneous items presented to the teams during the flights.resUlts: overall times of chest-compressions were similar, compression rate (101.7±9.6/min) was lower and compres-sions were deeper (3.9±0.2cm) with lUcas as compared to manual cPr (113.3±19.3/min and 3.7±0.4cm) (P<0.01, respectively). Heart rates of the eMs staff were increased after manual as compared to mechanical cPr (100.1±21.0 vs. 80.4±11.3, P<0.01). results of the questionnaire (93.6±6.9% vs. 87.0±7.3% correct answers, P<0.01) and memory test (22.4±15.4% vs. 11.3±7.5%, P<0.02) were significantly better after LUCAS resuscitation. Dosing of drugs, application intervals and rate of correct handling of drugs and defibrillation were not different between LUCAS or manual CPR.CONCLUSIONS: During simulated helicopter flights LUCAS-CPR improved the efficacy of chest-compressions, was physically less demanding and provided enhanced cognitive performance of the eMs team as compared to manual cPr.(Cite this article as: rehatschek g, Muench M, schenk i, Dittrich W, schewe Jc, Dirk c, et al. Mechanical lU-cas resuscitation is effective, reduces physical workload and improves mental performance of helicopter teams. Minerva anestesiol 2016;82:429-37)Key words: cardiopulmonary resuscitation - chest wall oscillation - air ambulances - Workload - task performance and analysis.

comment in p. 383.


reHatscHeK MecHaNical VERSUS MaNUal resUscitatioN iN HelicoPters


North rhine, Duesseldorf, germany. twelve advanced life support-certified EMS teams consisting of an emergency physician and a paramedic underwent two simulated helicop-ter flights of 20 minutes duration, respectively. During each flight the teams had to manage two cardiac arrest scenarios using either manu-al CPR or LUCAS. The sequence of the flights with manual or mechanical cPr was randomly assigned by the sealed envelopes technique im-mediately before the first flight of each team.

Materials and measurements

the emergency helicopter simulator chris-toph sim® (helicopter emergency medical ser-vice academy, allgemeiner Deutscher auto-mobil club, st. augustin, germany) which is a replica in proportion to the actual dimensions of an eurocopter ec 135 was used. an intu-bated manikin (ambuMan megacode®, ambu gmbH, Bad Nauheim, germany) was placed on a stretcher locked in the front position of the helicopter during the whole study (Figure 1).

gency medical service (eMs) staff. Moreover, performing manual chest-compressions on a patient in a moving vehicle may be dangerous for the eMs staff itself.8, 9 in contrast, mechani-cal chest-compression devices have been shown to be applicable in ground based ambulances as well as in helicopters 6, 7, 10, 11 and may maintain good quality cPr during patient transport and increase the safety of the staff. However, little is known about potentially different physical and mental demands by performing cPr with manual or mechanical chest-compression. es-pecially under extreme working conditions, such as transports with ongoing cPr, physi-cally demanding work that is performed con-currently with important cognitive demands may impact mental workload of eMs members by impairing mental processing or decreasing performance resulting in incorrect handling of drugs, ventilation and defibrillation.

We hypothesized that in an emergency heli-copter mechanical cPr would be more effec-tive than manual cPr, and reduce physical de-mands and mental workload resulting in a lower heart rate 12, 13 and better cognitive performance of the eMs team. to test this hypothesis eMs teams performed two different cPr scenarios on a manikin during simulated helicopter flights with and without the lund University cardiac assist system lUcas (Jolife inc., lund, swe-den) which is a battery-driven device providing mechanical active compression-decompression cPr. to determine the effectiveness and qual-ity of cPr the chest-compression rate and depth, the ratio of correct compressions and overall compression times as well as dosing and application intervals of drugs and defibrillation were measured. Heart rates of the medical crew were recorded continuously and mental work-load and performance were estimated via cog-nitive tasks after the simulated flight episodes.


Design and participants

approval of this prospective, randomized, cross-over manikin study was waived by the ethics committee of the Medical association of

Figure 1.—Fully instrumented manikin with attached lU-cas within helicopter simulator.

MecHaNical VERSUS MaNUal resUscitatioN iN HelicoPters reHatscHeK


simulated with good flight conditions at calm weather without turbulences. the medical crew heard the typical propeller and engine noises of a helicopter flight and were given different relevant information about the pa-tient, the target hospital and the flight, and further usual communication pattern through headphones (supplementary table i, online content only). In addition, five packages each consisting of five extraneous words such as everyday items were presented at irregular in-tervals (supplementary table ii, online con-supplementary table ii, online con-tent only). three minutes after take-off the first cardiac arrest due to ventricular fibrilla-tion occurred. in order to simulate real-life de-cision making, it remained at the discretion of the physician and paramedic who started with chest-compressions and who administered drugs and shocks, and whether and how often they rotated. During lUcas-cPr it also re-mained at the discretion of the eMs team to begin with manual chest compressions imme-diately after recognition of cardiac arrest or to delay chest-compressions until lUcas was attached. return of spontaneous circulation was simulated both ten minutes after the first cardiac arrest and again another four minutes after the second cardiac arrest was simulated. after a total time of 14 minutes of cardiac ar-rest and a total flight time of 20 minutes the test was stopped.

Questionnaire and Memory Test

Ten minutes after each flight the physician and paramedic had to answer a questionnaire about the medical items (supplementary table iii, online content only) and to perform a mem-) and to perform a mem-ory test for the added extraneous non-medical items which were presented during the flight.


results are expressed as mean±standard deviation (sD). the data was assumed to be normal distributed due to the central limit theorem. student’s paired t-test was used to compare parametric and Fisher´s exact test to compare categorical data of the two resuscita-

the manikin was mechanically ventilated (oxy-log3000, Draeger Medical gmbH, luebeck, germany) and connected to an integrated de-fibrillator/monitoring device (Corpuls3, G. stemple gmbH, Kaufering, germany) with defibrillation patches attached to the thorax. During simulated flights with LUCAS the back-board of the device was already placed under the manikin from the beginning of the scenario while the device itself was not allowed to be at-tached to the patient until recognition of cardiac arrest. compression depth, compression rate on a beat to beat basis, the time between recogni-tion of cardiac arrest and the first compression, and the overall time without chest-compression were recorded. since the compression depth of lUcas was set according to the 2005 aHa guidelines,14 both manual and mechanical com-pressions with a depth of at least 40 mm were judged as correct.

all team members were equipped with an ambulatory digital Holter electrocardi-ography monitor (custoflash 510, custom med, ottobrunn, germany) for continuous heart rate recording. overall performance of cPr, dosing, application times and intervals of drugs and external defibrillations were analysed by video recording. all data were stored on an attached laptop-computer and extracted to excel XP (Microsoft, Usa) and sPss (iBM sPss statistics 19) for further analysis.

Protocol

all participants were blinded to the scenari-os and measurements. since no participant had worked with lUcas before, each team was allowed to exercise with the device without a time limit outside the helicopter. all partici-pants were told that they would perceive medi-cal relevant and irrelevant information during the flight and that a questionnaire had to be answered and a memory test to be performed after each flight. The interval between the two flights was 60 minutes to allow the teams to recover from the first flight.

in both scenarios, transfers to a tertiary hospital for coronary revascularization were



results of the questionnaire about the pre-sented relevant medical information shortly after the flight simulation phase were bet-ter after lUcas-cPr (93.6±6.9% cor-rect answers) compared with manual cPr (87.0±7.3% correct answers, P<0.01) (Figure 3). With the more challenging free irrelevant word test, differences were also significant: after lUcas-cPr, participants remem-bered 22.4±15.4% of the items compared to 11.3±7.5 % after manual cPr (P<0.02) (Fig-ure 3). this tendency of both tests was similar for each participant.

total doses of adrenaline (epinephrine) and amiodarone, the time of first application of adrenaline and the rate of failed or incor-rect application of adrenaline, amiodarone and defibrillation were not different during CPR with manual or mechanical chest compression (table iii).

tion scenarios. randomness of the sequence of the cPr-order of the twelve teams was tested and verified by the runs-test (supplementary table iV, online content only). Differences were considered to be statistically significant if P<0.05.

Results

all teams fully completed the simulated he-licopter flights and used LUCAS correctly as judged by video analysis.

the time intervals from recognition of car-diac arrests to first compression as well as overall compression times, and the ratio of compression time to total time of cardiac ar-rest were similar with manual and lUcas-cPr (table i). With lUcas the compression rate was significantly lower (101.7±9.6/min), and compressions were significantly deeper (3.9±0.2cm) as compared to manual cPr (113.3±19.3/min and 3.7±0.4cm, P<0.01, re-spectively). analysis of every single chest-compression showed that the recommended depth was achieved in 92.4±19.4% of all lUcas compressions and in 56.0±40.4% of all manual compressions (P<0.01). the ratio of recorded to recommended compres-sion depths rapidly improved during ongoing LUCAS-CPR but was significantly lower and did not improve over time during manual cPr (Figure 2).

Heart rates of the EMS teams were signifi-cantly increased immediately after manual as compared to mechanical chest compression (P<0.01) (table ii).

Figure 2.—rate of correct compression depth during manu-al and LUCAS-CPR. Differences were significant at all time points, (P<0.01, student’s t-test).

Table I.—� Data of cardiopulmonary resuscitation (CPR) with manual and LUCAS-CPR performed by 12 emergency medical service teams.

Manual cPr* lUcas cPr* P

compression rate (b/min) 113.3±19.9 101.7±9.6 <0.01compression depth (cm) 3.7±0.4 3.9±0.2 <0.01rate of correct compression depth (%) 56.0±40.4 92.4±19.4 <0.01Time interval from recognition of first cardiac arrest to first compression (s) 73.1±33.8 85.6±40.5 n.s.Time interval from recognition of second cardiac arrest to first following compression (s) 91.5±56.6 92.2±48.7 n.s.total compression time during cardiac arrest (s) 659.1±78.8 710.1±56.2 n.s.ratio of compression time to total time of cardiac arrest (%) 68.7±8.2 74.0±5.9 n.s.Values are mean±standard deviation.*tested on a randomized basis, student´s t-test for dependent samples.



compared with manual cPr during simulated helicopter flights. We found that LUCAS-cPr was more effective, physically less de-manding and provided enhanced cognitive performance as compared to manual cPr. However, application of drugs and defibrilla-tion were not affected by the method of chest-compression.

in patients with out-of-hospital cardiac arrest two large randomized trials, liNc15 and ParaMeDic,16 recently have failed to demonstrate an improved outcome after lU-cas-cPr as compared to manual cPr. in these trials patients with the whole spectrum of cardiac arrests except a few causes such as traumatic cardiac arrests were included. However, problems of ongoing cPr during transports, especially helicopter transport, were not specifically addressed. Therefore, the authors of the ParaMeDic trial em-phasized the accepted role mechanical com-pression devices will continue to have when

Discussion

the aim of our study was to investigate the effectiveness of lUcas-cPr and its im-pact on the physical workload and cognitive performance of the medical emergency team

Table II.—�Heart rates (HR) of all members of the 12 emergency medical service teams.Manual cPr* lUcas cPr* P

HR at the start of the simulated flight (b/min) 78.0±13.0 82.2±12.4 n.s.HR after 2 minutes of simulated flight (b/min) 83.3±15.4 84.0±14.6 n.s.Hr at time of recognizing cardiac arrest (b/min) 86.9±19.5 87.8±16.3 n.s.Hr at time of rosc (b/min) 100.1±21.0 80.4±11.3 <0.01Hr 2 minutes after rosc (b/min) 91.1±20.4 83.1±12.0 n.s.cPr: cardiopulmonary resuscitation; rosc: return of spontaneous circulation.Values are mean±standard deviation.*tested on a randomized basis, student´s t-test for dependent samples.

Table III.—�Dosing and application intervals of drugs and defibrillation with manual and LUCAS-cardiopulmonary resuscitation (CPR) used by 12 emergency medical service teams.

Manual cPr* lUcas cPr* P

Time interval from recognition of first cardiac arrest to first defibrillation (s)

100.0±78.2 90.0±40.5 n.s.§

Time interval from recognition of second cardiac arrest to first following defibrillation (s)

65.5±82.5 32.7±72.8 n.s.§

Rate of failed or incorrect interval of defibrillation (%) 24.2±20.9 23.6±18.9 n.s.§total adrenaline dose (mg) 3.8±1.5 3.6±1.1 n.s.§Time interval from recognition of first cardiac arrest to first

adrenaline application (s)260.0±126.0 270.0±110.0 n.s.§

rate of failed or incorrect intervals of adrenaline dosing (%) 32.0±31.9 38.5±20.5 n.s.§total amiodarone dose (mg) 287.5±186 312.5±135.1 n.s.§rate of correct amiodarone dosing in all teams (%) 8.3 41.7 n.s.¥

Values are mean±standard deviation.*the sequence of manual and lUcas cPr both performed by 12 emergency medical service teams was randomly assigned.§student’s t-test for dependent samples; ¥χ2 test.

Figure 3.—rate of correct results in a questionnaire and a memory test after manual and lUcas-cPr. Differences were significant (*P<0.02, **P<0.01, Student’s t-test).



variable which in some cases may have been caused by an initially not optimal positioning of the device on the sternum. therefore, we propose to apply not only the backboard of lUcas but the complete device preventive-ly to high-risk patients and check its´ correct position before the flight. Since a cardiac as-sist device like LUCAS may only be benefi-cial when it is used as quickly as possible and as long as necessary our observations clearly emphasize the importance of a device related training program even for experienced eMs teams. similarly, the authors of the Para-MEDIC Trial found it essential that sufficient resources are made available to support ini-tial and regular refresher training and ongo-ing quality assurance when implementing mechanical devices into real world eMs sys-tems.16

our study only included transport time within the helicopter, and simulated calm weather. However, in real transport scenar-ios manual cPr may have to be performed even during loading or unloading of the pa-tient to and off the vehicle which has been shown to be ineffective.21 During turbulent flight conditions manual CPR may further deteriorate. therefore, cPr with a mechani-cal compression device should be even more effective under those actual helicopter rescue simulations as nicely confirmed by Putzer 10 et al. in their manikin study where mechani-cal chest-compression performed better not only within the helicopter but also during transport of the manikin from the scene to the helicopter and from the helicopter to the emergency room.

the present study seems unique in combin-ing both the narrow cPr conditions as well as the physical and mental workload condi-tions (e.g. intensive movements, high vigi-lance, safety concerns, and constant decision making) of emergency-trained physicians and paramedics in a relatively long-term fully re-alistic flight simulation including emergency-related and additional mental task-specific extraneous cockpit-communications and noise parameters related to mental overload and performance. thus, we tried to adopt a more

manual cPr is impracticable or involves an increased risk (e.g., in a moving ambu-lance).16 in fact, manual cPr during patient transport is challenging and the quality of prolonged cPr has been shown to deterio-rate during any transport mainly due to the extreme working conditions.6, 7, 17 in previ-ous manikin and human studies mechanical chest-compression devices have been shown to be applicable in ground based ambulance systems and in emergency helicopters with acceptable results.6, 7, 10, 11, 18 similarly, we also could demonstrate that lUcas-cPr was applicable and more effective than manual cPr in an emergency helicopter and did not deteriorate but even improved over the entire time of resuscitation. While in our study the time intervals from recogni-tion of cardiac arrests to the first chest-com-pressions as well as the overall compression times and the ratio of compression time to total time of cardiac arrest were similar dur-ing cPr with and without lUcas, chest-compressions were actually performed in only 68.7±8.2% and 74.0±5.9% of the de-manded time during manual and mechanical chest-compression, respectively. these re-sults are similar to the ratios of failed com-pression times during simulation courses of in-hospital scenarios,19 on scene or in am-bulance vehicles.17, 20, 21 analyses of video recordings of our cPr scenarios revealed that pre- and post-shock pauses to reassess cardiac rhythm could have been minimised and resumption of chest-compression could have been achieved faster as recommended by recent guidelines.5 Moreover, in our study mechanical chest-compression was frequent-ly interrupted for defibrillations by some of the eMs teams even during lUcas-cPr although continuing mechanical chest-com-pression during defibrillation is supposed to be an advantage of mechanical over manual compression. Despite in our study all eMs teams were able to attach lUcas easily and quickly to the backboard of the device which was already placed under the manikin the rate of correct chest compressions during the first minutes of LUCAS-CPR was quite



load and cognitive performance of health care professionals.22 obviously, decreasing physi-cal demands by using lUcas improved cog-nitive capacity of the eMs members. Howev-er, this result had no impact on the adherence of the teams to resuscitation guidelines. thus, irrespective of cPr with or without the me-chanical compression device our results re-vealed substantial deficits even in experienced eMs teams as concerns correct or recom-mended handling of drugs and defibrillation in a highly volatile resuscitation scenario within a helicopter which is in accordance with re-cent data obtained from simulation courses for physicians and nurses in hospitals.19 there-fore, continuous training and surveillance of basic elements of advanced cardiac life sup-port is of paramount importance even if a so-phisticated device such as lUcas is imple-mented to an eMs. While instituting written checklists may help to minimize distraction and errors of omission simulation training is another promising method to improve the clin-ical performance of eMs teams.19 However, as Bressan et al. recently pointed out those courses should focus both on technical and non-technical skills to produce a specific and measurable outcome in terms of situational awareness and clinical performance.23

Limitations

as with all simulation studies the results of the present study cannot be extrapolated to real life-scenarios of patients with cardiac arrest in a flying helicopter. All participants of our study were experienced emergency physicians and paramedics but none of them had previously worked with lUcas. therefore, it is possible that getting more experienced with lUcas might increase not only cognitive capacity but also improve overall cPr performance even in a helicopter. However, we tried to simulate real-life events with regular eMs teams and did not intend to demonstrate that specialized eMs teams being intensively trained in the handling of lUcas perform better with than without the device.

complementary or, in psychological terms, a more ecological approach. in fact, in our study not only the flight duration and medi-cal scenario but also the noise and busy com-munication pattern of emergency helicopter flights have been simulated in order to match the test and the real-life event as much as pos-sible. in particular, the noise level and the ongoing-cockpit communication might affect the mental workload of the eMs team. such extraneous factors to the mental workload of the eMs team in stress-related situations seem often neglected and not specifically addressed. During our simulated flights heart rates of the eMs staff were markedly increased during manual cPr. in addition to the physically ex-hausting effects of manual cPr, the combina-tion of an increased physical workload with extreme working conditions within the con-fined space of a helicopter, rapid movements of the eMs members while performing man-ual cPr, substantially impaired visual and acustic signals from the patient and monitor-ing devices in such a noisy environment, and finally potential personal safety concerns due to the unfastened seat-belts may have contrib-uted to the increase in heart rate most likely reflecting an enhanced sympathetic tone of the medical crew.

if physical and extraneous mental load are high, it is clear that performance may be hampered, but having these two forms of load set low does not guarantee that performance and outcome will succeed. on the one hand some technological advancements may have increased the requirement for many health professionals to execute cognitive tasks con-currently with physical activity, on the other hand technological advancements such as the device tested in the present study may lower the physical and probably also the mental workload of emergency teams, and thereby may improve their cognitive performance. in the present study mental performance as tested by memory tests shortly after the simulation phase was significantly better after the physi-cally less demanding mechanical cPr which may provide further evidence to the close rela-tionship between physical load, mental work-



sion rates and outcomes from cardiac arrest. circulation 2012;125:3004-12.

3. Zuercher M, Hilwig rW, ranger-Moore J, Nysaether J, Nadkarni VM, Berg MD, et al. leaning during chest compressions impairs cardiac output and left ventricular myocardial blood flow in piglet cardiac arrest. Crit Care Med 2010;38:1141-6.

4. christenson J, andrusiek D, everson-stewart s, Kuden-chuk P, Hostler D, Powell J, et al. chest compression fraction determines survival in patients with out-of-hos-pital ventricular fibrillation. Circulation 2009;120:1241-7.

5. Berg ra, Hemphill r, abella Bs, aufderheide Pt, cave tM, Hazinski MF, et al. Part 5: adult basic life support: 2010 american Heart association guidelines for cardiopulmonary resuscitation and emergency cardiovascular care. circulation 2010;122 (suppl 3):s685-s705.

6. gässler H, Ventzke MM, lampl l, Helm M. transport with ongoing resuscitation: a comparison between manu-al and mechanical compression. eMJ 2013;30:589-92.

7. schewe Jc, Heister U, Hoeft a, Krep H. emergency phy-sician and autoPulse--a good duo in preclinical emergen-cy services? anaesthesist 2008;57:582-8.

8. stapleton er. comparing cPr during ambulance trans-port: manual vs mechanical methods. JeMs 1991;16:63-4.

9. thomas sH, stone cK, Bryan-Berge D. the ability to perform closed chest compressions in helicopters. am J emerg Med 1994;12:296-8.

10. Putzer g, Braun P, Zimmermann a, Pedross F, strapaz-zon g, Brugger H, et al. lucas compared to manual cardiopulmonary resuscitation is more effective during helicopter rescue - a prospective, randomized, cross-over manikin study. am J emerg Med 2013;31:384-9.

11. omori K, sato s, sumi Y, inoue J, okamoto K, Uzura M, et al. The analysis of efficacy for AutoPulse system in flying helicopter. Resuscitation 2013;84:1045-50.

12. Kamal aa, Dammak M, caillard JF, couzinet M, Paris c, ragazzini i. relative cardiac cost and physical, men-tal and psychological work load among a group of post-operative care personnel. int arch occup environ Health 1991;63:353-8.

13. sammer g. Heart period variability and respiratory changes associated with physical and mental load: non-linear analysis. Ergonomics 1998;41:746-55.

14. anonymous. Part 4: adult Basic life support: 2005 american Heart association guidelines for cardiopul-monary resuscitation and emergency cardiovascular care. circulation 2005;112:V-19-iV-34.

15. rubertsson s, lindgren e, smekal D, oestlund o, silf-verstolpe J, lichtveld ra, et al. Mechanical chest com-Mechanical chest com-pressions and simultaneous defibrillation vs conventional cardiopulmonary resuscitation in out-of-hospital cardiac arrest: the liNc randomized trial. JaMa 2014;311:53-61.

16. Perkins gD, lall r, Quinn t, Deakin cD, cooke MD, Horton J, et al. Mechanical versus manual chest compres-sion for out-of-hospital cardiac arrest (ParaMeDic): a pragmatic cluster randomised controlled trial. lancet 2015;385:947-55.

17. odegaard s, olasveengen t, steen Pa, Kramer-Johansen J. the effect of transport on quality of cardiopulmonary resuscitation in out-of-hospital cardiac arrest. resuscita-tion 2009;80:843-8.

18. Havel c, schreiber W, riedmuller e, Haugk M, richling N, trimmel H, et al. Quality of closed chest compres-Quality of closed chest compres-sions in ambulance vehicles, flying helicopters and at the scene. resuscitation 2007;73:264-70.

19. Mayer V, schulz cM, Kreuzer M, Wagner KJ, schnei-der g, Kochs eF. cardiopulmonary resuscitation perfor-cardiopulmonary resuscitation perfor-

Conclusions

Mechanical chest-compression in an emer-gency helicopter using lUcas was more effective, physically less demanding and en-hanced cognitive performance of the eMs staff as compared to standard manual cPr. although we could not demonstrate improved decision making as concerns application of drugs and defibrillation the present findings can be taken as encouragement for the use of mechanical cPr devices in highly volatile res-cue situations within helicopters. Before a me-chanical compression device is implemented in an emergency helicopter service training of the eMs teams should focus not only on the handling of the device itself but most impor-tantly on the integration of the device into the resuscitation algorithm.

Key messages

— lUcas-cPr is more effective, physically less demanding and provides enhanced cognitive performance as com-pared to manual cPr during simulated he-licopter emergency flights.

— Despite the cognitive capacity of the eMs members improved with lUcas-cPr that did not automatically result in a better adherence of the teams to resuscita-tion guidelines.

— Before implementing lUcas in emergency helicopters we recommend to establish simulation courses which should focus not only on the technical skills re-quired for the handling of lUcas but also on the integration of the device into the re-suscitation algorithm to be performed.

References

1. Vadeboncoeur t, stolz U, Panchal a, silver a, Venuti M, tobin J, et al. chest compression depth and sur-vival in out-of-hospital cardiac arrest. resuscitation 2014;85:182-8.

2. idris aH, guff ey D, aufderheide tP, Brown s, Morrison lJ, Nichols P, et al. relationship between chest compres-



svensson l, Fellows B, et al. Quality of cardiopulmonary resuscitation during out-of-hospital cardiac arrest. JaMa 2005;293:299-304.

22. DiDomenico a, Nussbaum Ma. effects of differ-ent physical workload parameters on mental workload and performance. int Journal of industrial ergonomics 2011;41:255-60.

23. Bressan F, cabrini l. simulation and non-technical skills: the way is correct but more accurate researches are mandatory. Minerva anestesiol 2012;78:1306.

mance during simulator-based trainings: a comparative retrospective analysis of adherence to 2005 and 2010 guidelines. Minerva anestesiol 2013;79:264-73.

20. Krarup NH, terkelsen cJ, Johnsen sP, clemmensen P, olivecrona gK, Hansen tM, et al. Quality of cardiop-ulmonary resuscitation in out-of-hospital cardiac arrest is hampered by interruptions in chest compressions--a nationwide prospective feasibility study. resuscitation 2011;82:263-9.

21. Wik l, Kramer-Johansen J, Myklebust H, sørebø H,

Authors’ contributions.—gregor rehatschek and Marcus Muench contributed equally to the paper.Conflicts of interest.—The authors certify that there is no conflict of interest with any financial organization regarding the material discussed in the manuscript.Acknowledgments.—We thank all physicians and paramedics who participated in the study, M. Monnig (allgemeiner Deutscher automobil club) for providing the simulation center, s. stegherr for providing the manikin, and M. strohm for his invaluable or-ganizational support.Article first published online: November 17, 2015. - Manuscript accepted: November 13, 2015. - Manuscript revised: October 28, 2015. - Manuscript received: June 21, 2015.For supplementary materials, please see the online version of this article.


side effect profile.1 anaesthetists are reluctant to prescribe premedication (with benzodiaz-epines) in a day-case setting as patients could be too somnolent postoperatively. Consequently, this may prolong their time to discharge which should be avoided, especially, in day-case sur-gery. However, a Cochrane Review could not support this hypothesis,2 although they did not focus specifically on benzodiazepine premedi-cation. Furthermore, withholding premedica-tion may not be justified as almost half of the

Introduction

Benzodiazepines are among the most pre-scribed drugs used for premedication.1 in a

clinical setting, anaesthetists frequently admin-ister benzodiazepines preoperatively as they have unique properties like anxiolysis — one of the main goals of premedication — calming ef-fects and anterograde amnesia as a favourable

R E V I E W

effectiveness of benzodiazepine premedication on recovery in day-case surgery:

a systematic review with meta-analysisHerjan MIJDERWIJK *, Stefan VAN BEEK, Hugo J. DUIVENVOORDEN, Robert J. STOLKER

Department of Anaesthesiology, Erasmus University Medical Centre, Rotterdam, The Netherlands*Corresponding author: Herjan Mijderwijk, Erasmus University Medical Centre, room HS-203, PO Box 2040, Rotterdam, The Neth-erlands. E-mail: [email protected]


lavoro: titolo breve: EFFECTIVENESS OF BENZODIAZEPINE PREMEDICATIONprimo autore: MIJDERWIJKpagine: 438-64citazione: Minerva anestesiol 2016;82:438-64

a B s t r a c tINTRODUCTION: Benzodiazepines are frequently used as a premedication. In day-case surgery, anesthetists are re-luctant to administer benzodiazepines preoperatively for reasons of delayed recovery. However, premedication with benzodiazepines might be beneficial regarding postoperative somatic symptoms/complaints (i.e. time to recovery and postoperative side effects) and psychological phenomena.EVIDENCE ACQUISITION: A systematic review with meta-analysis was performed using all important search engines. Study methodological quality was assessed using risk of bias tables. Mean differences (MD) and odds ratios (OR) were used for continuous data (time to recovery and psychological phenomena) and categorical data (postoperative somatic symptoms) respectively. Random effects modelling was applied. Nineteen studies were included. Overall time to re-covery was significantly delayed in patients receiving benzodiazepines (MD 1.75; 95% CI 0.82 to 2.69) although time to discharge was not significantly affected. Postoperative side effects were significantly reduced in patients receiving benzodiazepines (OR 0.47; 95% CI 0.36 to 0.63). Regarding psychological outcome, only anxiety could be statistically analyzed showing no statistical difference (MD 1.47; 95% CI -1.01 to 3.96).EVIDENCE SYNTHESIS: Although overall time to recovery was significantly prolonged by benzodiazepine premedica-tion, withholding premedication in day-case surgery patients is not justified for such reason, as time to discharge was not negatively affected. Furthermore, benzodiazepines show to have beneficial effects on postoperative side effects.CONCLUSIONS: For a firm conclusion regarding psychological phenomena, more research is needed. Anaesthetists should take into account this new evidence when they apply their premedication regime in day-case surgery.(Cite this article as: Mijderwijk H, Van Beek S, Duivenvoorden HJ, Stolker RJ. Effectiveness of benzodiazepine premedica-Mijderwijk H, Van Beek S, Duivenvoorden HJ, Stolker RJ. Effectiveness of benzodiazepine premedica-effectiveness of benzodiazepine premedica-tion on recovery in day-case surgery: a systematic review with meta-analysis. Minerva Anestesiol 2016;82:438-64)Key words: Benzodiazepines - Meta-analysis as topic - Preoperative care.

Minerva anestesiologica 2016 april;82(4):438-64© 2015 EDIZIONI MINERVA MEDICAThe online version of this article is located at http://www.minervamedica.it

Comment in p. 386.

EFFECTIVENESS OF BENZODIAZEPINE PREMEDICATION MIJDERWIJK

Vol. 82 - No. 4 Minerva anestesiologica 439

cochrane central register of controlled tri-als, PubMed Publisher, and Google Scholar, updated until January 2014. There were no language or publication date restrictions. Main key words for the search queries included day surgery, postoperative psychological aspect, somatic symptoms/complaints, and premedica-tion. The full search is presented in Appendix I.

Study selection

all randomized controlled trials including human adult patients undergoing day-case surgery were included when they really test-ed premedication. We defined “premedica-tion tested” as any medication given prior to induction of anaesthesia that was continued neither during the surgical procedure nor post-operatively. In addition, the intervention had to be tailored to the premedication itself. We excluded studies where no postoperative out-comes were assessed. Original articles written in English, German or French that included an abstract were maintained in the set in order to reduce language bias. Next, we excluded tri-als based on clinical anaesthetic and surgical selection criteria. Anesthetic exclusion crite-ria concerned not undergoing general or re-gional anaesthesia. Surgical exclusion criteria included not undergoing day-case surgery or undergoing abortion, dental surgery or oph-thalmology surgery. Also, trials testing non-benzodiazepines as a premedication were ex-cluded. Finally, non-randomised studies and studies with no placebo control (i.e. method-ological criteria) were excluded.

Data extraction and management

Three authors (HM, SVB, RJS) independ-ently analyzed studies for inclusion in the analysis. Two authors (HM, SVB) independ-ently assessed all included studies with respect to their quality. Data was extracted using a pre-set collection sheet. Authors were not blinded for information regarding the identified studies (e.g. journal, author, institution, date of pub-lication), as it was previously shown that not blinding did not influence the results of meta-

patients in day-case surgery request something to relieve their stress and anxiety.3

In day-case surgery, patient’s somatic symp-toms became of minor interest as perioperative morbidity and mortality are extremely low.4 Therefore, day-case surgery patients place high-er priority on psychological phenomena rather than physical recovery in the postoperative peri-od. Along with this, literature shows more atten-tion for psychological aspects of perioperative care.5, 6 research evaluating benzodiazepine administration in day-case surgery is focussing on psychological phenomena including anxiety, fatigue, aggression and depressive moods.7

Preoperative benzodiazepine administration is mostly evaluated preoperatively. However, the reluctance of anaesthetists is based on po-tential postoperative concerns. Therefore, to determine whether benzodiazepine premedi-cation in day-case surgery is appropriate, thor-ough research focussing on the postoperative period is needed.

to evaluate the effectiveness of benzodi-azepine administration in day-case surgery, we conducted a systematic review with me-ta-analysis of randomised trials focussing on postoperative somatic symptoms/complaints and psychological phenomena.

this systematic review and meta-analysis tested three related hypotheses:

1. in adult day-case surgery, benzodiaz-epines as a premedication do elongate time to recovery from general or regional anaesthesia;

2. benzodiazepines as a premedication do beneficially affect postoperative somatic symp-toms/complaints;

3. benzodiazepines as a premedication do reduce postoperative psychological sequelae.

Evidence acquisition

the Preferred Reporting Items for System-atic Review and Meta-Analyses guidelines were adhered to.8

Literature search

Literature search was performed in Embase, Medline OvidSP, ISI Web of Science, Scopus,

MIJDERWIJK EFFECTIVENESS OF BENZODIAZEPINE PREMEDICATION


sion 5.3 from the Cochrane Collaboration for analysis. Meta-regression was done in SPSS version 20.0 (IBM, NY, Armonk). Statistical significance was fixed at p < 0.05 (two-tailed).

Evidence synthesis

Study selection and characteristics

The search was performed on 31 January 2014. After excluding duplicates (N.=508), 586 studies were screened on the base of title/abstract. Based on the selection criteria, clini-cal criteria and studies that did use benzodi-azepines as a premedication we excluded 497 articles. Fifty full articles were reviewed ac-cordingly. However, three articles could not be retrieved. We excluded 31 studies based on methodological and clinical criteria. Ulti-mately, 19 Articles were eligible for systematic review/meta-analysis (Figure 1).7, 13-30 table i shows the study characteristics. The risk of bias of the included studies is shown in Appendix II. The risk of bias graph showing each risk of bias as percentages across all included studies is shown in Appendix III. Appendix IV shows the risk of bias scoring of the individual studies.

Somatic symptoms/complaints: time to recovery

twelve studies were included in me-ta-analysis with 1445 patients altogeth-er.13, 15, 16, 19, 21-25, 27, 29, 30 applying random effects model resulted into tau2=2.78, chi2=69.73, df=26, P<0.01, I2=63%. Overall time to recovery was significantly delayed by benzodiazepines with 1.75 minutes (95% CI 0.82-2.69) (Figure 2A).

Time to eye opening was significantly de-layed with 1.47 minutes by benzodiazepines (95% CI 0.51-2.42), but we could not find statistical significant differences regarding time to first correct response (P=0.06), time to early recovery (P=0.24) and time to discharge (P=0.39), Figure 2A. Sensitivity analyses did not provide new insights. None of the covari-ates could explain the heterogeneity among the studies for time to eye opening, early recovery and discharge. However, time to first correct re-

analysis.9 Disagreement among authors was resolved by consensus.

included studies were reviewed for data on any of the following outcomes:

1. Somatic symptoms/complaints: time to recovery, time to eye opening, time to first cor-rect response (TCR), time to early recovery (i.e. discharge form the recovery room) and time to discharge; postoperative side effects, including, nausea, vomiting/emesis, dizziness, pain, headache and miscellaneous (coughing and double vision).

2. Psychological outcomes: anxiety, de-pression, fatigue and aggression.

all outcomes were eligible for assessment when they were measured up to the first post-operative day.

Quality assessment

The Cochrane Handbook of Systematic reviews of interventions was used to evalu-ate the risk of bias of each included study.10 Selection-, performance-, detection-, attrition-, reporting- and other biases are assessed in this risk of bias tool.


categorical outcome data were evaluated by odds ratios (or) while continuous outcome data were evaluated using mean differences (MD). Random effects model was used for each outcome. Meta-regression was performed in order to evaluate heterogeneity among stud-ies and included the following covariates: year of publication, quality assessment and the journal’s impact factor. Sensitivity analyses were performed in order to evaluate whether eliminating influential studies affect the results. Studies that exceed Cook’s distance 1/n were found influential. Funnel plots were used to graphically examining small study effects as an asymmetrical funnel shape may indicate publi-cation bias.11 inconsistency among studies was evaluated by means of the I2-statistic. An I2-sta-tistic of 25%, 50% and 75% were respectively defined as low, moderate and high inconsisten-cy.12 We used Review Manager (RevMan) ver-



and raybould et al. could not be subjected to meta-analyses but did report on time to recovery.17, 24, 29, 31 Beechey et al. found no difference in the time to awaken from anaes-thesia.17, 31 Hargreaves found that awakening

sponse (TCR) increased significantly in studies with a higher methodological quality (Appen-dix V). All funnel plots of these studies were not considered asymmetrical (Appendix VI).

In addition, Beechey et al., Hargreaves

Figure 1.—Flow of information.



Table I.—�Characteristics of included studies.reference Methods Participants intervention control outcomes notes

abdul-latif Ms et al. (2001) 13

randomised placebo controlled double-blind study

50 female patients undergoing day case breast surgery, aged 18-70 years

7.5 mg oral midazolam

Placebo somatic symptoms/complaints (time to recovery)

source of funding not stated

ahmed n et al. (1995) 14

randomised placebo controlled double-blinded study

50 mixed patients undergoing day-case surgery, aged 20-60 years

7.5 mg oral midazolam

Placebo somatic symptoms/complaints (postoperative side effects); Psychological (anxiety)


Bailie r et al. (1989) 15

randomised controlled trial

65 female patients undergoing day-case surgery, aged 16-75 years

20 mg oral temazepam

identical placebo capsule

somatic symptoms/complaints (time to recovery)

Financially supported by Cognitive Drug research

Bauer KP et al. (2004) 16

Prospective randomised placebo-controlled study


0.04 mg/kg intravenous midazolam

intravenous saline

somatic symptoms/complaints (time to recovery; postoperative side effects); Psychological (anxiety)


Beechey APG et al. (1981) 17


60 mixed patients undergoing elective minor surgery as day cases, aged 18-70 years

10 mg oral temazepam

identical placebo capsule



Berendes e et al. (1996) 18

randomised placebo-controlled double-blinded study

85 female patients scheduled for breast biopsy

7.5 mg oral midazolam; 20 mg oral clorazepate dipotassium

Placebo Psychological (anxiety, depression)


De Witte JL et al. (2002) 19

randomised placebo controlled double blinded study


0.5 mg oral alprazolam; 7.5 mg oral midazolam

oral placebo somatic symptoms/complaints (time to recovery; postoperative side effects)

Financially supported by NIH Grant GM 58273, the Joseph Drown Foundation, and the commonwealth of Kentucky Research Challenge Trust Fund

Duggan M et al. (2002) 20



0.1 mg/kg oral diazepam, 60 min preoperatively; 0.1 mg/kg oral diazepam, 90 min preoperatively

Placebo Psychological (anxiety)


Forrest P et al. (1987) 21

randomised placebo controlled study

120 mixed patients undergoing day-case surgery; aged 20-60 years

0.25 mg oral triazolam; 15 mg oral midazolam; 10 mg oral diazepam

oral placebo somatic symptoms/complaints (time to recovery)

sources of funding not stated

Fredman B et al. (1999) 22


90 patients undergoing brief procedures, aged 65-81 years

0.5 mg intravenous midazolam; 2 mg intravenous midazolam

Equal volume intravenous saline



Greenwood BK et al. (1983) 23



20 mg oral temazepam; 30 mg oral oxazepam



(to be continued)



Table I.—�Characteristics of included studies.reference Methods Participants intervention control outcomes notes

Hargreaves J et al. (1988) 24

Double-blinded study


15 mg oral midazolam; 20 mg oral temazepam

identical placebo


Roche Products supplied the double-blind randomized premedications

Kain ZN et al. (2000) 25



5 mg intramuscular midazolam

intramuscular saline

somatic symptoms/complaints (time to recovery; postoperative side effects), Pyschological (anxiety)

Financially partly supported by a grant from the national institutes of Helat, Bethesda, Maryland, Roche Pharmaceuticals, Nutly, New Jersey, and the Patrick and Catherine Weldon Donaghue Medical Research Foundation, hartford, CT (Dr. Kain)

loach a et al. (1975) 26



1 mg oral lorazepam

Placebo somatic symptoms/complaints (postoperative side effects)


Mijderwijk H et al. (2013) 7


398 mixed patients undergoing day-case surgery, aged at least 18 years

1 to 1.5 mg/mL intravenous lorazepam

1 to 1.5 mL intravenous saline

Psychological (anxiety, depression, fatigue, aggression)

Financial support was provided by the Department of Anaesthesiology, Erasmus University Medical centre

Oxorn DC et al. (1997) 27


60 female patients undergoing day-case surgery, aged >19 years

30 ug/kg intravenous midazolam

0.03 mL/kg intravenous saline

somatic symptoms/complaints (time to recovery); Psychological (anxiety, depression, aggression)

Financial support was provided by a grant from roche Pharmaceuticals and the First International anesthesia research Society Frontiers in anesthesia award

Raeder JC et al. (1987) 28


193 female patients undergoing day-case surgery

0.1 mg/kg intramuscular midazolam; 0.8 to 1 mL intramuscular Mo-scop (i.e. morphine 10 mg/mL and scopolamine 0.4 mg/mL)

0.8 to 1.0 mL intramuscu-lar saline

somatic symptoms/complaints (postoperative side effects)


Raybould D et al. (1987) 29


60 mixed patients undergoing day-case surgery; aged 16-65 years

7.5 mg oral midazolam; 15 mg oral midazolam



shafer a et al. (1989) 30



5 mg (1 mL) intramuscular midazolam; 1 mg (2 ml) intravenous oxymorphone; 100 ug (2 ml)

1 ml intra-muscular saline; 2 ml intravenous saline

somatic symptoms/complaints (time to recovery; postoperative side effects)


from anaesthesia was significantly longer in the midazolam group compared to placebo and temazepam groups.24 raybould et al. showed that the group receiving a benzodiazepine did not show significantly longer recovery times.29

Somatic symptoms/complaints: postoperative side effects

seven studies were included in meta-analysis with 1530 patients altogether.14, 16, 17, 23, 26, 28, 30 re-

(continues).



al. could not be subjected to meta-analysis but did report on anxiety.14, 16, 18-20, 22 ahmed et al. found no significant difference in patient’s anx-iety levels.14 Bauer et al. did not found statis-tical significant difference in patient’s anxiety levels in PACU and at discharge from PACU.16 Berendes et al. found that clorazepate dipotas-sium had significant lower anxiety scores com-pared to placebo.18 No significant difference was found between midazolam and placebo. De Witte et al. found that all patients in the mi-dazolam group reported a sufficient quality of anxiety reduction; 2 patients in the alprazolam group reported insufficient anxiety reduction and 1 patient did not know; 7 patients in the pla-cebo group reported insufficient anxiety reduc-tion and 3 patient did not know, which was sta-tistically significant among groups.19 Duggan et al. found no significant difference in anxiety scores (vas and state part of the state-trait Anxiety Inventory [STAI-State]) at discharge.20 Fredman et al. found that anxiety scores were unaffected during PACU admission.22

The studies by Kain et al. and Mijderwijk et al. measured anxiety beyond the first postop-erative day.7, 25 Kain et al. found a significant greater reduction in anxiety in the benzodiaz-epine group compared with placebo from 2-30 days after surgery.25 Mijderwijk et al., on the seventh day after surgery, found significant greater reduction in anxiety measured by means of the Trait part of the State-Trait Anxiety In-ventory (STAI-Trait) and by means of the Hos-pital Anxiety and Depression Scale (HADS) in the placebo group, although no significant result was found regarding STAI-state.7

None of the covariates enabled explaining heterogeneity among the studies (Appendix V). Sensitivity analyses did not provide new insights. The funnel plot was not considered asymmetrical (Appendix VIII).

Only the studies by Mijderwijk et al. and Oxorn et al. have reported about depression, fatigue and aggression.7, 27 Oxorn et al. have reported results up to first postoperative day.27 They found no significant differences on de-pression and aggression while Mijderwijk et al. measured these outcomes beyond the first postoperative day.7 Although no significant

sults of random effects model yielded tau2=0.00, chi2=19.25, df=24, P=0.74, I2=0%. Overall postoperative side effects occurred significantly less in patients treated with benzodiazepines (OR 0.47, 95% CI 0.36 – 0.63) (Figure 2B, C).

Nausea (OR 0.34, 95% CI 0.21-0.55) and headache (OR 0.44, 95% CI 0.25-0.78) oc-curred significantly less in the patients treated with benzodiazepines. However, we could not find statistical significant differences regard-ing vomiting (P=0.08), dizziness (P=0.68) and the miscellaneous group (P=0.21). Categori-cal data regarding pain showed no statistical significant difference (P=0.86) as well as pain scored on a continuous scale (P=0.55).

In addition, de Witte et al., Hargreaves and Kain et al. could not be subjected to meta-analyze but did report postoperative side ef-fects.19, 24, 25 De Witte et al. found no statis-tical difference in the incidence of nausea or vomiting and other side effects including diz-ziness and headache.19 Hargreaves found no statistical difference in the incidence of minor side effects.24 However, nausea was found in 8 patients receiving temazepam, which was statistically significant when compared to the placebo group. Kain et al. found no signifi-cant difference regarding a postoperative pain score (Visual Analogue Score [VAS] >30) on discharge form the Post Anaesthesia Care Unit (PACU).25 Furthermore, undefined adverse ef-fects were not significantly different in PACU.

sensitivity analysis and meta-regression did not provide new insights (Appendix V). Fun-nel plot for dizziness was asymmetrical sug-gesting publication bias; the other funnel plots were considered symmetrical (Appendix VII).

Psychological outcomes

A total of 4 studies assessing anxiety were included in meta-analysis with 653 pa-tients.7, 25, 27, 30 random effects model yielded tau2=3.96, Chi2=15.33, df=3, P<0.01, I2=80%. Anxiety was not significantly affected (mean difference 1.47, 95% CI -1.01-3.96) (Figure 2D).

ahmed et al., Bauer et al., Berendes et al., de Witte et al., Duggan et al. and Fredman et



further detail, benzodiazepine premedication clearly reduce the risk of postoperative head-ache (56%) and nausea (66%), and show a ten-dency to reduce the risk of vomiting. The risk of postoperative dizziness, pain and miscellane-ous side effects was not statistically significant-ly affected by benzodiazepines. The articles not meeting the standards for this meta-analysis are considered to have no influence.19, 24, 25

After discharge from day-case surgery, head-ache, nausea and vomiting emerge with inci-dence rates of 27%, 17-21% and 5.7-8% respec-tively.32, 33 Benzodiazepine premedication could be a favourable way to reduce these postopera-tive side effects, and, consequently, enhancing the quality of patients’ somatic recovery. We are fully aware that with regards to postoperative somatic symptoms/complaints, more factors could emerge in the postoperative period, such as sore throat and back pain.32, 33 Unfortunately, due to lack of data we could not test for these factors with regards to benzodiazepine premedi-cation. This could be subject of further research.

PsychologIcal Phenomena

We were able to perform meta-analysis for anxiety, showing no statistical significant dif-ferences between benzodiazepine or placebo groups. With regards to the other psychologi-cal outcomes (i.e. depression, aggression and fatigue), meta-analysis could not be performed at all as they did not meet the eligibility criteria with regards to the time span. Therefore, the ef-fects of premedication with benzodiazepines on psychological phenomena remain inconclusive. given the shift towards psychological outcome in day-case surgery,5-7 we need more research on this topic. This should also be studied beyond the first postoperative day as recommend by oth-ers.32 Furthermore, we need more research on in-terventions in the preoperative period that could beneficiallyaffect the postoperative period.34

meThodologIcal sTrengThs and weakness-es

The rationale for our extensive literature search was to get a solid understanding of the

differences were found on depression, fatigue and trait aggression on the seventh day after surgery, yet they found a statistically signifi-cant result regarding state aggression.

Discussion

PrIncIPal fIndIng

The principal finding of this systematic re-view with meta-analysis is that overall ben-zodiazepines did unfavorably affect time to recovery, did reduce the incidence of postop-erative side effects but they did not statistically significantly affect psychological outcomes. These findings will be discussed in further de-tail below.

TIme To recovery

The overall test of significance showed that time to recovery is significantly prolonged in patients administered benzodiazepines preop-eratively. However, only time to eye opening is significantly prolonged by 1.47 minutes in the benzodiazepine group. Time to early recovery and time to discharge were not affected by ben-zodiazepines. Although not statistically signifi-cant, benzodiazepines tend to prolong TCR. The articles that could not be subjected to meta-anal-ysis are considered to have no influence.17, 24, 29

Considering all this, we agree with (some) anesthetists that recovery time is prolonged by benzodiazepines but only at the first stage of recovery. Time to early recovery and time to discharge are clearly not affected by benzo-diazepine premedication which is in line with a previous review.2 Therefore, withholding benzodiazepine premedication for reasons of delayed discharge time seems not justified in day-case surgery.

PosToPeraTIve sIde effecTs

The overall test of significance showed that premedication with benzodiazepines sig-nificantly reduced the incidence of postopera-tive side effects with 53%. When looking into



a

D

c

nomena emerging in the postoperative period. By doing so, we minimized the risk of selection bias too. As a consequence, we had to evalu-

effects of benzodiazepine premedication in day-case surgery patients regarding somatic symptoms/complaints and psychological phe-

Figure 2.—A) Forest plot for time to recovery; B, C) Forest plot for postoperative side effects. Regarding Raeder JC et al., postoperative N. was determined proportionally and data regarding “nausea at home” was ignored as this data is likely cor-related with “nausea at recovery”; D) Forest plot for psychological phenomena (anxiety).13-30



B

minimize language bias by including articles not only written in english but also in german and France, when appropriate. This strategy

ate many studies for eligibility. Selection bias was also minimized by not excluding studies based on their quality assessment. We tried to



to these within study variation, in between study heterogeneity is possibly caused by in-tra- and/or interjudgement bias. Physicians, nurses or investigators may have alternately assessed time to recovery in patients, which is likely to induce bias.

the moderate inconsistency regarding diz-ziness can be caused by the vague definition of dizziness interpreted by patients. However, it was previously shown that heterogeneity might also be due to publication bias 36 (i.e. small studies with expressive results are likely to be published) and in the case of dizziness publication bias has emerged.

Heterogeneity can also be present due to conceptual concerns, especially in case of psychological outcomes. For example, anxi-ety itself can be considerably heterogeneous and reasons for heterogeneity can be difficult to clarify.37 Furthermore, heterogeneity among studies may always be due to change.11

In this study, we have focussed on a homog-enous group of drugs used for premedication. As a consequence, we were able to perform meta-analysis and could draw conclusions from our results accordingly. Such a statisti-cal synthesis was previously not feasible due to too many different premedication drugs.2 a total of eight different benzodiazepines, three different administration routes and differ-ent times of administration were nonetheless present in our meta-analysis. Unfortunately, it was methodologically statistically not fea-sible to evaluate this possible heterogeneity in our meta-regression. However, administra-tion of benzodiazepines is characterized by high efficacy despite differences in route of administration and pharmacological proper-ties.

Conclusions

this systematic review with meta-analysis provides new evidence for beneficial effects of premedication with benzodiazepines in day-case surgery. Benzodiazepine premedication does prolong time to recovery but only at the first stage of postoperative recovery – time to discharge is not affected. Furthermore, ben-

seems justified as we included an article writ-ten in German.18 However, potential language bias could not be ruled out. We were able to check the possible influence of the year of pub-lication on the outcomes, as we did not have year of publication restrictions in our literature search. We used a pre-set standardized form for data-extraction and management. Risk of bias tables were used to evaluate the quality of each study, which was done by two authors individually. Discrepancies were resolved by consensus. Meta-regression was performed on each outcome individually. Meta-regression was not always possible as numerical data was not always clearly provided or not provided. To deal with this, we systematically described the results of the outcomes of these studies.

random effects model was performed for each meta-analysis when heterogeneity was present. For consistency’s sake, we also per-formed random effects model when heteroge-neity was statistically not significant. Another justification for this analysis strategy is that judgements for heterogeneity in meta-analysis can be misleading when the number of includ-ed patients is insufficient.35 thus we possibly could have under-estimated heterogeneity and therefore we have applied random effects model even when heterogeneity was statisti-cally insignificant.

the fact that heterogeneity is predominantly present in time to recovery and psychologi-cal phenomena while heterogeneity is nearly present in postoperative side effects suggests that the assessment sources may be differ-ent. With regards to time to recovery, in the majority of the studies it is unclear who ac-tually measured time to recovery. However, in the study by De Witte et al. study nurses observed the patients in the recovery room.19 Likely, although not specifically stated, they assessed time to early recovery. Accordingly, interjudgement unreliability bias may have emerged and this could be the reason for the wide confidence intervals of this particular study. On the other hand, intrajudgement unre-liability may have played a role when, for ex-ample, the pre-set definitions for specific time to recovery were not clear. Furthermore, next



13. Abdul-Latif MS, Putland AJ, McCluskey A, Meadows DP, Remington SA. Oral midazolam premedication for day case breast surgery, a randomised prospective double-blind placebo-controlled study. Anaesthesia 2001;56:990-4.

14. Ahmed N, Khan FA. Evaluation of oral midazolam as pre-medication in day care surgery in adult Pakistani pa-tients. J Pak Med Assoc 1995;45:239-41.

15. Bailie R, Christmas L, Price N, Restall J, Simpson P, Wesnes K. Effects of temazepam premedication on cog-nitive recovery following alfentanil-propofol anaesthe-sia. Br J Anaesth 1989;63:68-75.

16. Bauer KP, Dom PM, Ramirez AM, O’Flaherty JE. Preop-erative intravenous midazolam: benefits beyond anxioly-sis. J Clin Anesth 2004;16:177-83.

17. Beechey AP, Eltringham RJ, Studd C. Temazepam as pre-medication in day surgery. Anaesthesia 1981;36:10-5.

18. Berendes E, Scherer R, Rotthove K, Prien T. Anxiolyse, Sedierung und Streßreduktion nach oraler Prämedika-tion mit Midazolam bei erwachsenen ein vergleich mit Dikaliumclorazepat bzw. Plazebo. Der Anaesthesist 1996;45:506-11.

19. De Witte JL, Alegret C, Sessler DI, Cammu G. Preop-erative alprazolam reduces anxiety in ambulatory surgery patients: a comparison with oral midazolam. Anesth An-alg 2002;95:1601-6.

20. Duggan M, Dowd N, O’Mara D, Harmon D, Tormey W, Cunningham AJ. Benzodiazepine premedication may at-tenuate the stress response in daycase anesthesia: a pilot study. Can J Anaesth 2002;49:932-5.

21. Forrest P, Galletly DC, Yee P. Placebo controlled com-parison of midazolam, triazolam and diazepam as oral premedicants for outpatient anaesthesia. Anaesth Inten-sive Care 1987;15:296-304.

22. Fredman B, Lahav M, Zohar E, Golod M, Paruta I, Jedeikin R. The effect of midazolam premedication on mental and psychomotor recovery in geriatric patients undergoing brief surgical procedures. Anesth Analg 1999;89:1161-6.

23. Greenwood BK, Bradshaw EG. Preoperative medication for day-case surgery. A comparison between oxazepam and temazepam. Br J Anaesth 1983;55:933-7.

24. Hargreaves J. Benzodiazepine premedication in minor day-case surgery: comparison of oral midazolam and temazepam with placebo. Br J Anaesth 1988;61:611-6.

25. Kain ZN, Sevarino F, Pincus S, Alexander GM, Wang SM, Ayoub C, et al. Attenuation of the preoperative stress response with midazolam: effects on postoperative out-comes. Anesthesiology 2000;93:141-7.

26. Loach A, Fisher A. Lorazepam as a premedicant for day-case surgery: an assessment. Anaesthesia 1975;30:545-9.

27. Oxorn DC, Ferris LE, Harrington E, Orser BA. The ef-fects of midazolam on propofol-induced anesthesia: pro-pofol dose requirements, mood profiles, and periopera-tive dreams. Anesth Analg 1997;85:553-9.

28. Raeder JC, Breivik H. Premedication with midazolam in out-patient general anaesthesia. A comparison with morphine-scopolamine and placebo. Acta Anaesthesiol Scand 1987;31:509-14.

29. Raybould D, Bradshaw EG. Premedication for day case surgery. A study of oral midazolam. Anaesthesia 1987;42:591-5.

30. Shafer A, White PF, Urquhart ML, Doze VA. Outpatient premedication: use of midazolam and opioid analgesics. Anesthesiology 1989;71:495-501.

31. Olkkola KT, Ahonen J. Midazolam and other benzodi-azepines. Handb Exp Pharmacol 2008;335-60.

32. Wu CL, Berenholtz SM, Pronovost PJ, Fleisher LA. Sys-tematic review and analysis of postdischarge symptoms af-ter outpatient surgery. Anesthesiology 2002;96:994-1003.

33. Mattila K, Toivonen J, Janhunen L, Rosenberg PH,

zodiazepines seem to reduce the incidence of postoperative side effects with 53%. However, effects on psychological outcomes remain in-conclusive. It is recommended that future stud-ies should also focus on other postoperative side effects, and on psychological phenomena.

Key messages

— Withholding benzodiazepine pre-medication for reasons of delayed discharge time is not justified in day-case surgery.

— Premedication with benzodiazepines reduces postoperative side effects with 53%.

— In day-case surgery, more research on postoperative psychological sequelae is needed.

References

1. Barash PG, Cullen BF, Stoelting RK, Cahalan M, Stock MC. Clinical Anesthesia. Lippincott Williams & Wilkins; 2011.

2. Walker KJ, Smith AF. Premedication for anxiety in adult day surgery. The Cochrane Database Syst Rev 2009;4:CD002192.

3. van den Berg AA. Towards needleless induction of anaes-thesia. Anaesthesia 2003;58:806-7.

4. Mathis MR, Naughton NN, Shanks AM, Freundlich RE, Pannucci CJ, Chu Y, et al. Patient selection for day case-eligible surgery: identifying those at high risk for major complications. Anesthesiology 2013;119:1310-21.

5. Bellani ML. Psychological aspects in day-case surgery. Int J Surg 2008;6:S44-6.

6. Mitchell M. General anaesthesia and day-case patient anxiety. J Adv Nurs 2010;66:1059-71.

7. Mijderwijk H, van Beek S, Klimek M, Duivenvoorden HJ, Grüne F, Stolker RJ. Lorazepam does not improve the quality of recovery in day-case surgery patients: a randomised placebo-controlled clinical trial. Eur J An-aesthesiol 2013;30:743-51.

8. Moher D, Liberati A, Tetzlaff J, Altman DG, PRISMA Group. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med 2009;6:e1000097.

9. Berlin JA. Does blinding of readers affect the results of meta-analyses? University of Pennsylvania Meta-analysis Blinding Study Group. Lancet 1997;350:185- 6.

10. Higgins JPT, Green S (editors). Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 [up-dated March 2011]. The Cochrane Collaboration, 2011. Available from www.cochrane-handbook.org.

11. Egger M, Smith GD, Altman D. Systematic Reviews in Health Care. BMJ Books; 2001.

12. Higgins JPT, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. Br Med J 2003;327:557-60.



timates and Their 95% Confidence Intervals in Large Meta-Analyses. PLoS ONE 2012;7:e39471.

36. Easterbrook PJ, Berlin JA, Gopalan R, Matthews DR. Publication bias in clinical research. Lancet 1991;337:867-72.

37. Rosenthal R, DiMatteo MR. Meta-analysis: recent devel-opments in quantitative methods for literature reviews. Annu Rev Psychol 2001;52:59-82.

Hynynen M. Postdischarge Symptoms After Ambulatory Surgery: First-Week Incidence, Intensity, and Risk Fac-tors. Anesth Analg 2005;101:1643-50.

34. Roden E, Walder B. Preoperative evaluation and prepara-tion by anaesthesiologists only, please! Eur J Anaesthe-siol 2013;30:731-3.

35. Thorlund K, Imberger G, Johnston BC, Walsh M, Awad T, Thabane L et al. Evolution of Heterogeneity (I2) Es-

Authors’ contributions.—Herjan Mijderwijk, Robert J. Stolker and Hugo J. Duivenvoorden conception and design; Herjan Mijder-Herjan Mijderwijk, Robert J. Stolker and Hugo J. Duivenvoorden conception and design; Herjan Mijder-, Robert J. Stolker and Hugo J. Duivenvoorden conception and design; Herjan Mijder-Robert J. Stolker and Hugo J. Duivenvoorden conception and design; Herjan Mijder- and Hugo J. Duivenvoorden conception and design; Herjan Mijder-Hugo J. Duivenvoorden conception and design; Herjan Mijder- conception and design; Herjan Mijder-Herjan Mijder-wijk, Stefan Van Beek, and Robert J. Stolker: data collection and extraction. Herjan Mijderwijk, Stefan Van Beek, Hugo J. Dui- Stefan Van Beek, and Robert J. Stolker: data collection and extraction. Herjan Mijderwijk, Stefan Van Beek, Hugo J. Dui-Stefan Van Beek, and Robert J. Stolker: data collection and extraction. Herjan Mijderwijk, Stefan Van Beek, Hugo J. Dui- and Robert J. Stolker: data collection and extraction. Herjan Mijderwijk, Stefan Van Beek, Hugo J. Dui-Robert J. Stolker: data collection and extraction. Herjan Mijderwijk, Stefan Van Beek, Hugo J. Dui-: data collection and extraction. Herjan Mijderwijk, Stefan Van Beek, Hugo J. Dui-Herjan Mijderwijk, Stefan Van Beek, Hugo J. Dui-, Stefan Van Beek, Hugo J. Dui-Stefan Van Beek, Hugo J. Dui-, Hugo J. Dui-Hugo J. Dui-venvoorden and Robert J. Stolker: data analysis. Herjan Mijderwijk, Stefan Van Beek and Robert J. Stolker: interpretation of data; drafting of the manuscript and revising it critically for important intellectual content; final approval of the manuscript submitted.Funding.—This work was supported by the Department of Anaesthesiology, Erasmus University Medical Centre, Rotterdam, The Netherlands.Acknowledgements.—We thank Wichor M. Bramer, information specialist at Erasmus MC, for performing the literature search.Conflicts of interest.—The authors certify that there is no conflict of interest with any financial organization regarding the material discussed in the manuscript.Article first published online: September 29, 2015. - Manuscript accepted: September 25, 2015. - Manuscript revised: September 8, 2015. - Manuscript received: April 28, 2015.

Appendix I.—Complete literature search, 31 January 2014.

embase 213 210Medline OvidSP 301 157Web-of-science 60 26scopus 354 119cochrane 66 6PubMed publisher 0 0google scholar 100 68total 1094 586Duplicates removed: 508

Embase('ambulatory surgery'/de OR (((ambul* OR day OR daycare OR daycase OR outpatient* OR office*) NEAR/3 (surg* OR operati*))):ab,ti) AND ('psychological aspect'/de OR psychology/exp OR emotion/exp OR depression/exp OR fatigue/de OR exhaustion/de OR stress/exp OR 'adaptive behavior'/de OR 'surgical stress'/de OR (psycholog* OR emotion* OR anxiet* OR fatigue OR exhaust* OR depress* OR perception* OR (somatic NEAR/3 (symptom* OR complain*)) OR ((mental OR preoperat* OR postoperat* OR perioperat* OR operative* OR surg*) NEXT/1 stress) OR (adapt* NEAR/3 behavio*) OR coping):ab,ti) AND (adult/exp OR 'middle aged'/de OR aged/exp OR (adult* OR aged):ab,ti) AND (premedication/de OR benzodiazepine/de OR 'anesthetic agent'/exp OR 'antidepressant agent'/exp OR tranquilizer/exp OR (premedicat* OR pretreatment* OR (pre NEXT/1 (medicat* OR treatment*)) OR preanesthe* OR preanaesthe* OR anaesthetic* OR anesthetic* OR ((anxiet* OR antianxiet* OR ataract*) NEAR/3 (agent* OR drug*)) OR benzodiazepine* OR anxiolytic* OR tranquill* OR antidepress*):ab,ti) AND ((random* OR factorial* OR crossover* OR (cross NEXT/1 over*) OR placebo* OR ((doubl* OR singl*) NEXT/1 blind*) OR assign* OR allocat* OR volunteer*):ab,ti OR 'crossover procedure'/de OR 'double-blind procedure'/de OR 'random-ized controlled trial'/de OR 'single-blind procedure'/de) NOT ([animals]/lim NOT [humans]/lim)

Medline OvidSP("Ambulatory Surgical Procedures"/ OR (((ambul* OR day OR daycare OR daycase OR outpatient* OR office*) ADJ3 (surg* OR operati*))).ab,ti.) AND (exp psychology/ OR psychology.xs. OR exp emotions/ OR depression/ OR exp fatigue/ OR "Stress, Psychological"/ OR "Adaptation, Psychological"/ OR "surgical stress"/ OR (psycholog* OR emotion* OR anxiet* OR fatigue OR exhaust* OR depress* OR perception* OR (somatic ADJ3 (symptom* OR complain*)) OR ((mental OR preoperat* OR postoperat* OR perioperat* OR operative* OR surg*) ADJ stress) OR (adapt* ADJ3 behavio*) OR coping).ab,ti.) AND (exp adult/ OR (adult* OR aged).ab,ti.) AND (premedication/ OR



"Preanesthetic Medication"/ OR exp benzodiazepines/ OR exp "anesthetics"/ OR exp "antidepressive agents"/ OR exp "Tranquilizing Agents"/ OR (premedicat* OR pretreatment* OR (pre ADJ (medicat* OR treatment*)) OR pre-anesthe* OR preanaesthe* OR premedicat* OR pretreatment* OR preanesthe* OR preanaesthe* OR anaesthetic* OR anesthetic* OR ((anxiet* OR antianxiet* OR ataract*) ADJ3 (agent* OR drug*)) OR benzodiazepine* OR anxiolytic* OR tranquill* OR antidepress*).ab,ti.) AND (Clinical Trial.pt. OR randomized.ab,ti. OR placebo.ab,ti. OR dt.fs. OR randomly.ab,ti. OR trial.ab,ti. OR groups.ab,ti. NOT (Animals/ NOT Humans/))

Cochrane((((ambul* OR day OR daycare OR daycase OR outpatient* OR office*) NEAR/3 (surg* OR operati*))):ab,ti) AND ((psycholog* OR emotion* OR anxiet* OR fatigue OR exhaust* OR depress* OR perception* OR (somatic NEAR/3 (symptom* OR complain*)) OR ((mental OR preoperat* OR postoperat* OR perioperat* OR operative* OR surg*) NEXT/1 stress) OR (adapt* NEAR/3 behavio*) OR coping):ab,ti) AND ((adult* OR aged):ab,ti) AND ((premedicat* OR pretreatment* OR (pre NEXT/1 (medicat* OR treatment*)) OR preanesthe* OR preanaesthe* OR anaesthetic* OR anesthetic* OR ((anxiet* OR antianxiet* OR ataract*) NEAR/3 (agent* OR drug*)) OR ben-zodiazepine* OR anxiolytic* OR tranquill* OR antidepress*):ab,ti)

Web-of-scienceTS=(((((ambul* OR day OR daycare OR daycase OR outpatient* OR office*) NEAR/3 (surg* OR operati*)))) AND ((psycholog* OR emotion* OR anxiet* OR fatigue OR exhaust* OR depress* OR perception* OR (somatic NEAR/3 (symptom* OR complain*)) OR ((mental OR preoperat* OR postoperat* OR perioperat* OR operative* OR surg*) NEXT/1 stress) OR (adapt* NEAR/3 behavio*) OR coping)) AND ((adult* OR aged)) AND ((pre-medicat* OR pretreatment* OR (pre NEXT/1 (medicat* OR treatment*)) OR preanesthe* OR preanaesthe* OR anaesthetic* OR anesthetic* OR ((anxiet* OR antianxiet* OR ataract*) NEAR/3 (agent* OR drug*)) OR benzodi-azepine* OR anxiolytic* OR tranquill* OR antidepress*)) AND (random* OR factorial* OR crossover* OR (cross NEAR/1 over*) OR placebo* OR ((doubl* OR singl*) NEAR/1 blind*) OR assign* OR allocat* OR volunteer*))

ScopusTITLE-ABS-KEY(((((ambul* OR day OR daycare OR daycase OR outpatient* OR office*) W/3 (surg* OR opera-ti*)))) AND ((psycholog* OR emotion* OR anxiet* OR fatigue OR exhaust* OR depress* OR perception* OR (so-matic W/3 (symptom* OR complain*)) OR ((mental OR preoperat* OR postoperat* OR perioperat* OR operative* OR surg*) PRE/1 stress) OR (adapt* W/3 behavio*) OR coping)) AND ((adult* OR aged)) AND ((premedicat* OR pretreatment* OR (pre PRE/1 (medicat* OR treatment*)) OR preanesthe* OR preanaesthe* OR anaesthetic* OR anesthetic* OR ((anxiet* OR antianxiet* OR ataract*) W/3 (agent* OR drug*)) OR benzodiazepine* OR anx-iolytic* OR tranquill* OR antidepress*)) AND (random* OR factorial* OR crossover* OR (cross W/1 over*) OR placebo* OR ((doubl* OR singl*) W/1 blind*) OR assign* OR allocat* OR volunteer*))

PubMed publisher((((ambul*[tiab] OR day[tiab] OR daycare[tiab] OR daycase[tiab] OR outpatient*[tiab] OR office*[tiab]) AND (surger*[tiab] OR surgic*[tiab] OR operati*[tiab])))) AND ((psycholog*[tiab] OR emotion*[tiab] OR anxiet*[tiab] OR fatigue[tiab] OR exhaust*[tiab] OR depress*[tiab] OR perception*[tiab] OR (somatic symptom*[tiab] OR somatic complain*[tiab] OR ((mental stress[tiab] OR preoperative stress[tiab] OR postoperative stress[tiab] OR perioperative stress[tiab] OR operative stress[tiab] OR surgical stress[tiab])) OR (adaptive behavio*[tiab]) OR cop-ing)) AND ((adult*[tiab] OR aged[tiab])) AND ((premedicat*[tiab] OR pretreatment*[tiab] OR pre medicat*[tiab] OR pre treatment*[tiab] OR preanesthe*[tiab] OR preanaesthe*[tiab] OR anaesthetic*[tiab] OR anesthetic*[tiab] OR ((anxiet*[tiab] OR antianxiet*[tiab] OR ataract*[tiab]) AND (agent*[tiab] OR drug[tiab] OR drugs[tiab])) OR benzodiazepine*[tiab] OR anxiolytic*[tiab] OR tranquill*[tiab] OR antidepress*[tiab]) AND (random*[tiab] OR factorial*[tiab] OR crossover*[tiab] OR cross over*[tiab] OR placebo*[tiab] OR double blind*[tiab] OR single blind*[tiab] OR assign*[tiab] OR allocat*[tiab] OR volunteer*[tiab])) AND publisher[sb]

Google Scholar"ambulatory|day|daycare surgery|operation" psychological|psychology|emotions|depression|fatigue|exhaustion|stress|coping|anxiety premedication|benzodiazepine|tranquilizer|anxiolytic|antidepressant random|randomized|randomised|factorial|trial|placebo|group



Appendix II.—Risk of bias table – authors’ judgement

reference Random sequence generation (selection bias)

allocation concealment (selection bias)

Blinding of participants and personnel (performance bias)

Blinding of outcome assessment (detection bias)

incomplete outcome data (attrition bias)

selective reporting (reporting bias) other bias

abdul-latif Ms et al.13 Unclear risk Unclear risk Low risk Unclear risk Low risk Low risk Unclear riskSupport for judgement Comment: not reported

whether random sequence generation was performed.

Comment: not reported whether allocation concealment was achieved.

Comment: Anaesthetists were blinded as well as the patients.

Comment: not reported whether time to recovery was blindly assessed.

Comment: table 2 shows that all patients were assessed for time to recovery.

Comment: all measures with respect to time to recovery were presented.

Comment: no other sources of bias identified

ahmed n et al.14 Unclear risk Unclear risk Low risk Low risk High risk Low risk High riskSupport for judgement Quote: “..patients..were

randomly allocated to receive either Midazolam 7.5 mg or a placebo..”

Comment: insufficient information about the sequence generation process.


Quote: “The study was done in a double blind manner.”

Quote: “The anaesthetist involved in recording observations was unaware of the patient grouping.”

Comment: 9 patients are excluded from the analyses. Reasons for the exclusions are not given. Furthermore, it is likely that these missings all belong to the placebo group, and, consequently, imbalancedness is likely to emerge.

Comment: all outcomes are reported as dictated by the study protocol as written in the Methods section.

Comment: the groups are imbalanced regarding sex: the male/female ratio for the midazolam group equals 2/3, whereas the ratio for the placebo group equals 15/1.

Bailie r et al.15 High risk Low risk Low risk Unclear risk High risk Low risk Unclear riskSupport for judgement Quote: “The hospital

pharmacist, who was not involved in the study, had previously allocated them to one of two groups.” However: “Because of a delay in obtaining matched placebo capsules, the first 26 patients of 65 patients were allocated to group 1.”

Quote: “..was unknown to the researcher undertaking the cognitive assessments and to those conducting initial evaluation of data.”

Quote: “All testing and initial evaluation of data was conducted double-blind.”

Comment: not reported if time to recovery was blindly assessed.

Quote: “Six patients in the placebo group were withdrawn. Four patients in the temazepam group were withdrawn.

Comment: Imbalancedness emerged: >20% withdrawn from the placebo group versus 10% withdrawn. Furthermore, investigators themselves have withdrawn the patients.


Comment: no other sources of bias identified.

Bauer KP et al.16 Unclear risk Low risk Low risk Low risk High risk Low risk Unclear riskSupport for judgement Comment: not reported


Quote: “Study syringes were prepared by the pharmacy. Patients, anesthesiologists, and investigators were blinded to the contents of each syringe until the study was completed.”

Quote: “Patients, anesthesiologists, and investigators were blinded to the contents of each syringe until the study was completed.”


Quote: “..a total of 118 patients signed a consent form. Of that number, 13 patient were withdrawn from data analysis for protocol violations, and 17 patients were withdrawn from data analysis because of missing data.”

Comment: no reasons for missing data provided.



Beechey APG et al.17 Unclear risk Unclear risk Low risk Unclear risk Low risk Low risk Unclear riskSupport for judgement Quote: “The patients, who

gave informed consent, were randomly allocated into two groups..”


Comment: not reported how allocation concealment was achieved.

Quote: “A double-blind trial was therefore undertaken..” “..placebo capsules of identical appearance were used..”

Comment: not reported whether blinding of outcome assessment was achieved.

Comment: no missing outcome data.



Berendes e et al.18 Low risk Low risk Low risk Unclear risk Low risk Low risk Unclear riskSupport for judgement Quote: “Die Randomisierung

in 3 Gruppen erfolgte anhand einer Tabelle mit Zufallszahlen.”

Quote: “Zwei Stunden präoperative erhielten die Patientinnen in einem verschlossenen un numerierten Umschlag entweder..Midazolam,..Dikaliumclorazepat oder ein Plazebopräparat..”

Quote: “Die Studie wurde als prospektive randomisierte Doppelblindstudie durchgeführt.”

Comment: unclear if blinding of outcome assessment was achieved.












abdul-latif Ms et al.13 Unclear risk Unclear risk Low risk Unclear risk Low risk Low risk Unclear riskSupport for judgement Comment: not reported



Comment: Anaesthetists were blinded as well as the patients.

Comment: not reported whether time to recovery was blindly assessed.

Comment: table 2 shows that all patients were assessed for time to recovery.

Comment: all measures with respect to time to recovery were presented.

Comment: no other sources of bias identified

ahmed n et al.14 Unclear risk Unclear risk Low risk Low risk High risk Low risk High riskSupport for judgement Quote: “..patients..were

randomly allocated to receive either Midazolam 7.5 mg or a placebo..”



Quote: “The study was done in a double blind manner.”

Quote: “The anaesthetist involved in recording observations was unaware of the patient grouping.”

Comment: 9 patients are excluded from the analyses. Reasons for the exclusions are not given. Furthermore, it is likely that these missings all belong to the placebo group, and, consequently, imbalancedness is likely to emerge.


Comment: the groups are imbalanced regarding sex: the male/female ratio for the midazolam group equals 2/3, whereas the ratio for the placebo group equals 15/1.

Bailie r et al.15 High risk Low risk Low risk Unclear risk High risk Low risk Unclear riskSupport for judgement Quote: “The hospital

pharmacist, who was not involved in the study, had previously allocated them to one of two groups.” However: “Because of a delay in obtaining matched placebo capsules, the first 26 patients of 65 patients were allocated to group 1.”

Quote: “..was unknown to the researcher undertaking the cognitive assessments and to those conducting initial evaluation of data.”

Quote: “All testing and initial evaluation of data was conducted double-blind.”

Comment: not reported if time to recovery was blindly assessed.

Quote: “Six patients in the placebo group were withdrawn. Four patients in the temazepam group were withdrawn.

Comment: Imbalancedness emerged: >20% withdrawn from the placebo group versus 10% withdrawn. Furthermore, investigators themselves have withdrawn the patients.



Bauer KP et al.16 Unclear risk Low risk Low risk Low risk High risk Low risk Unclear riskSupport for judgement Comment: not reported


Quote: “Study syringes were prepared by the pharmacy. Patients, anesthesiologists, and investigators were blinded to the contents of each syringe until the study was completed.”



Quote: “..a total of 118 patients signed a consent form. Of that number, 13 patient were withdrawn from data analysis for protocol violations, and 17 patients were withdrawn from data analysis because of missing data.”

Comment: no reasons for missing data provided.



Beechey APG et al.17 Unclear risk Unclear risk Low risk Unclear risk Low risk Low risk Unclear riskSupport for judgement Quote: “The patients, who

gave informed consent, were randomly allocated into two groups..”



Quote: “A double-blind trial was therefore undertaken..” “..placebo capsules of identical appearance were used..”





Berendes e et al.18 Low risk Low risk Low risk Unclear risk Low risk Low risk Unclear riskSupport for judgement Quote: “Die Randomisierung

in 3 Gruppen erfolgte anhand einer Tabelle mit Zufallszahlen.”

Quote: “Zwei Stunden präoperative erhielten die Patientinnen in einem verschlossenen un numerierten Umschlag entweder..Midazolam,..Dikaliumclorazepat oder ein Plazebopräparat..”

Quote: “Die Studie wurde als prospektive randomisierte Doppelblindstudie durchgeführt.”

Comment: unclear if blinding of outcome assessment was achieved.












De Witte JL et al.19 Unclear risk Unclear risk Low risk Unclear risk Low risk Low risk Unclear riskSupport for judgement Quote: “..patients were

randomly assigned to receive..”



Quote: “..outpatients participated in a double-blinded study..” “..the commercially available drug tablets were placed in opaque capsules filled with an inactive powder.”





Duggan M et al.20 Low risk Low risk Low risk Unclear risk Low risk Low risk Unclear riskSupport for judgement Quote: “Randomization was

performed using a random numbers table.”

Quote: “Randomization was performed using a random numbers table by the research division, pharmacy, Beuamont Hospital.”

Comment: central allocation.

Quote: “We conducted a..double-blind, randomized study..” “..Group III received a placebo.”


Quote: “One patient was excluded as she required admission to the hospital overnight..”

Comment: this missing data was judged to have no clinical impact.



Forrest P et al.21 Unclear risk Unclear risk Low risk Unclear risk Low risk Low risk Unclear riskSupport for judgement Quote: “..patients were

randomly allocated to receive..one of four oral premedicants from coded envelopes..”



Quote: “Patients were randomly allocated to receive, in a double-blind manner..”


Comment: no missing outcome data relative to time to recovery. However, some missing data in other parts of the study; unclear if this leads to risk of bias.



Fredman B et al.22 Low risk Unclear risk Low risk Low risk Low risk Low risk Unclear riskSupport for judgement Quote: “computer-generated

randomization table”.Comment: not reported how

allocation concealment was achieved.

Quote: “90 geriatric patients were enrolled in to this..double-blinded study.”

Quote: “..a “blinded” coinvestigator continuously monitored the patient’s..” “PACU staff was unaware of patient enrolment.”

Quote: “Because of the need to perform an open transvesical prostatectomy, one patient in Group 0.5mg was excluded.”

Comment: this missing is judged to have no clinical impact.



Greenwood BK et al.23 Unclear risk Unclear risk Low risk Unclear risk Low risk Low risk Unclear riskSupport for judgement Quote: “..the patients were

randomly allocated to one of the three groups.”



Quote: “A double-blind, between-patient trial was designed..” “All patients received a similar soft gelatin capsule..”

Comment: not reported whether blinding outcome assessment was achieved.




Hargreaves J et al.24 Unclear risk Unclear risk Low risk Unclear risk Low risk Low risk Unclear riskSupport for judgement Quote: “The author

wishes to thank Rocke Products for supplying the double-blind randomized premedication..”

Comment: not reported whether random sequence generation was achieved.


Quote: “Ninety patients..were allocated to three groups in a double-blind study” “This comprised an active and a dummy preparation for the two study groups and tow dummy preparation for the placebo group.





Appendix II.—Risk of bias table – authors’ judgement (continues).









De Witte JL et al.19 Unclear risk Unclear risk Low risk Unclear risk Low risk Low risk Unclear riskSupport for judgement Quote: “..patients were

randomly assigned to receive..”



Quote: “..outpatients participated in a double-blinded study..” “..the commercially available drug tablets were placed in opaque capsules filled with an inactive powder.”





Duggan M et al.20 Low risk Low risk Low risk Unclear risk Low risk Low risk Unclear riskSupport for judgement Quote: “Randomization was

performed using a random numbers table.”

Quote: “Randomization was performed using a random numbers table by the research division, pharmacy, Beuamont Hospital.”


Quote: “We conducted a..double-blind, randomized study..” “..Group III received a placebo.”


Quote: “One patient was excluded as she required admission to the hospital overnight..”

Comment: this missing data was judged to have no clinical impact.



Forrest P et al.21 Unclear risk Unclear risk Low risk Unclear risk Low risk Low risk Unclear riskSupport for judgement Quote: “..patients were

randomly allocated to receive..one of four oral premedicants from coded envelopes..”



Quote: “Patients were randomly allocated to receive, in a double-blind manner..”


Comment: no missing outcome data relative to time to recovery. However, some missing data in other parts of the study; unclear if this leads to risk of bias.



Fredman B et al.22 Low risk Unclear risk Low risk Low risk Low risk Low risk Unclear riskSupport for judgement Quote: “computer-generated

randomization table”.Comment: not reported how

allocation concealment was achieved.

Quote: “90 geriatric patients were enrolled in to this..double-blinded study.”

Quote: “..a “blinded” coinvestigator continuously monitored the patient’s..” “PACU staff was unaware of patient enrolment.”

Quote: “Because of the need to perform an open transvesical prostatectomy, one patient in Group 0.5mg was excluded.”

Comment: this missing is judged to have no clinical impact.



Greenwood BK et al.23 Unclear risk Unclear risk Low risk Unclear risk Low risk Low risk Unclear riskSupport for judgement Quote: “..the patients were

randomly allocated to one of the three groups.”



Quote: “A double-blind, between-patient trial was designed..” “All patients received a similar soft gelatin capsule..”





Hargreaves J et al.24 Unclear risk Unclear risk Low risk Unclear risk Low risk Low risk Unclear riskSupport for judgement Quote: “The author

wishes to thank Rocke Products for supplying the double-blind randomized premedication..”

Comment: not reported whether random sequence generation was achieved.


Quote: “Ninety patients..were allocated to three groups in a double-blind study” “This comprised an active and a dummy preparation for the two study groups and tow dummy preparation for the placebo group.














Kain ZN et al.25 Low risk Low risk Low risk Low risk High risk Low risk Unclear riskSupport for judgement Quote: “Randomization was

performed according to a computer-generated list created from a random numbers table.”

Quote: “Blinding an randomization were handled by Yale-New Haven Hospital’s investigational pharmacy,..”


Quote: “Blinding an randomization were handled by Yale-New Haven Hospital’s investigational pharmacy, an no other individuals (e.g. anesthesiologists, surgeons, investigators) were informed of the particular treatment group of which a particular subject was assigned.” “Subjects received the intervention, midazolam or saline from coded syringes..”

Comment: Likely that patients were blinded too.

Quote: “..no other individuals (e.g. anesthesiologists, surgeons, investigators) were informed of the particular treatment group of which a particular subject was assigned.”

Quote: “Six subjects were excluded from the final sample because of noncompliance of the anesthesia staff to the study protocol. These subjects were excluded on the day of surgery, and no data were obtained regarding their postoperative course.”

Comment: unclear to which treatment group these patients belong; unbalancedness likely. Furthermore, no specific reasons for exclusion provided.



loach a et al.26 Low risk High risk Low risk Unclear risk Low risk Low risk Unclear riskSupport for judgement Quote: “The allocation

was made on the basis of numbers from random tables..”

Quote: “The allocation was made on the basis of numbers from random tables, and the code held by the Ward Sister.”

Comment: Although the code was held by the Ward Sister, it is unclear who enrolled the patients in the study. Furthermore, it is likely that no appropriate safeguards are taken.

Quote: “A randomised double blind trial was carried out..” “..patients then received two tablets which were either placebo or lorazepam..”


Comment: no missing outcome data on the outcome of interest for this review. From two patients, however, blood samples could not be obtained and these patients were missing on some other outcomes.



Mijderwijk H et al.7 Low risk Low risk Low risk Low risk Low risk Low risk Unclear riskSupport for judgement Quote: “Randomisation

was done by a computer-generated table.”

Quote: “Randomisation was done by a computer-generated table, and patients were assigned subsequent numbers upon inclusion. Nurses who were not further involved in the care of these patients prepared the study medication according to the randomisation table.”


Quote: “The study was double blinded; the researchers, patients and all healthcare professionals involved in patient care were blinded to the treatment allocation.” “Blinding was achieved by preparation of the transparent fluids in identical syringes.”

Comment: as ‘blinding of participants and personnel.’

Quote: “An Intention-to-treat analysis was applied.” “14 patients from the lorazepam group and six patients from the NaCl 0.9% group were lost to follow-up for at least one of the measurement points. This difference was not significant.”

Comment: reasons for lost-to -follow up clearly described; no imbalancedness detected.



Oxorn DC et al.27 Unclear risk Low risk Low risk Low risk Low risk Low risk Unclear riskSupport for judgement Quote: “On enrolment,

patients were randomly allocated..”


Quote: “Sixty sealed envelopes were prepared.”

Quote: “A syringe labelled study drug was prepared on the day of surgery by an anesthesiologist not involved with the conduct of the anesthetic. The anesthesiologist administering anesthesia was blinded to the patients group.”

Quote: “The patient’s responses were assessed by the same nurse who was blind tot the patient’s treatment group.”

Comment: no missing outcome data for our outcomes of interest. Furthermore, 4 missing data detected which was balanced over the treatment groups.











Raeder JC et al.28 Unclear risk Unclear risk Low risk Low risk Low risk Low risk Unclear riskSupport for judgement Quote: “The patients were

then randomly allocated to one of the three groups”



Quote: “The patients did not know which premedicant they received. Mo-Scop or placebo were given double-blind from coded ampoules.”

Comment: the outcome relevant for this review was based on patient’s self-assessment. However, outcomes reported that were not relevant for this review were assessed by health care professionals. It remains unclear if those were blinded.

Comment: 186 of 193 patients did not complete postoperative questionnaire. Reasons for these missing data are not provided. However, these missings are judged to have no clinical impact.



Raybould D et al.29 Unclear risk Unclear risk Low risk Low risk Low risk Low risk Unclear riskSupport for judgement Quote: “..they were allocated

at random to one of three groups.”



Quote: “A double-blind, between-patient trial was designed..” “All patients received a gelatin capsule with up to 20ml of water..”

Quote: “The observer was trained, was the same throughout and was unaware of the medication given.”

Quote: “Several patients had to be withdrawn from the study because of problems in the timing of operation or because more major surgery was indicated.”

Comment: 7 patients had to be withdrawn, which was judged to satisfactory balanced over the treatment groups (placebo versus active). This was judged to have no clinical impact.



shafer a et al.30 Low risk Low risk Low risk Low risk Low risk Low risk Unclear riskSupport for judgement Quote: “Patients were

randomly assigned using a computer-generated random number list..”

Quote: “Only the pharmacist who prepared the study-drug vials knew to which group a patient was assigned.”


Quote: “Only the pharmacist who prepared the study-drug vials knew to which group a patient was assigned.” “..all study drugs were administered in a double-blinded manner.”

Quote: “A blinded observer noted the incidence of coughing,..and assessed overall difficulties during the induction and maintenance phases of anesthesia.”

Comment: outcome relative to this review are assessed by patient’s self-assessment.

Quote: “Overall, an 89% response rate was obtained on the follow-up questionnaires.”

Comment: Although no reasons for these missing are provided, these missings are judged to have no clinical impact.






Appendix III.—Summary risk of bias graph.

Appendix IV.—Individual risk of bias scores.



Appendix V.—Meta-regression.

covariate b95% CI

lower bound Upper bound

time to recoverytime to eye opening

Year of publication -0.11 -0.28 -0.06Methodological quality 0.64 -0.08 1.36impact factor -0.19 -1.16 0.77

tcrYear of publication -0.19 -0.75 0.37Methodological quality 1.68 0.53 2.83impact factor 0.46 -3.30 4.22

time to early recoveryYear of publication 0.37 -1.81 2.55Methodological quality -1.27 -9.37 6.83impact factor -2.56 -12.66 7.54

time to dischargeYear of publication 0.28 -0.85 1.42Methodological quality 0.19 -11.91 12.29impact factor -0.18 -3.06 2.70

Postoperative side effectsHeadache

Year of publication 0.08 -0.31 0.47Methodological quality -0.86 -2.61 0.89impact factor -0.50 -1.12 0.11

nauseaYear of publication -0.00 -0.05 0.05Methodological quality -0.17 -0.58 0.25impact factor 0.20 -0.02 0.42

vomitingYear of publication 0.01 -0.08 0.09Methodological quality 0.32 -0.15 0.80impact factor -0.17 -0.82 0.49

DizzinessYear of publication -0.16 -0.49 0.16Methodological quality 0.20 -1.31 1.71impact factor -0.47 -2.28 1.33

PainYear of publication$ na na naMethodological quality$ na na naimpact factor$ na na na

MiscellaneousYear of publication na na naMethodological quality na na naimpact factor na na na

Psychological outcomeAnxiety

Year of publication 0.12 -0.26 0.50Methodological quality -1.01 -6.33 -4.32impact factor -0.20 -3.57 3.18

B=regression coefficient; CI=confidence interval; TCR=time to first correct response; Na=not applicable.



Appendix VI.—Funnel plots for time to recovery.

Appendix VIII.—Funnel plots for time to recovery.

Time to eye opening.

Time to early recovery.

Headache.

Time to first correct response.

Time to discharge.

Nausea.



Vomiting.

Pain.

Miscellaneous.

Dizzines.

Pain, continuous.



Appendix VIII.—Funnel plot for phenomena.

Anxiety.


suffering and fear of one’s own death. Many icU procedures are invasive and may cause considerable distress. Furthermore, frightening hallucinations and paranoid delusions are com-mon.1, 2 research suggests that the cumulative stress patients experience in icU can affect their

the intensive care unit (icU) is an envi-ronment that induces significant stress in

many patients. stressors include pain, hunger, thirst, sleep deprivation, inability to commu-nicate, loss of control, loss of dignity, an un-natural environment, witnessing other people’s

R E V I E W

Non-pharmacological interventions to reduce icU-related psychological distress:

a systematic reviewDorothy M. WaDe 1*, Zoe MooN 2, sula s. WiNDgasseN 3,

anthony M. HarrisoN 4, lucy Morris 5, John a. WeiNMaN 6

1critical care Unit, University college Hospital, london, UK; 2institute of Psychiatry, Psychology and Neuroscience, King’s college london, london, UK; 3Department of Psychological Medicine, King’s college london, london, UK; 4institute of Psychiatry, Psychology and Neuroscience, King’s college london, london, UK; 5King’s college london school of Medicine, london, UK; 6institute of Pharmaceutical science, King’s college london, london, UK*corresponding author: Dorothy Wade, critical care, University college Hospital, 235 euston rd, london NW1 2BU. e-mail: [email protected]


lavoro: titolo breve: NoN-PHarMacological iNterVeNtioNs to reDUce icU-re-lateD PsYcHological Distressprimo autore: WaDepagine: 465-78citazione: Minerva anestesiol 2016;82:465-78

a B s t r a c tiNtroDUctioN: Patients frequently suffer stress in intensive care units (icUs) and many develop serious psychologi-cal morbidity after discharge. Little is known about the nature and efficacy of interventions to reduce ICU-related dis-tress. there is growing evidence that administering sedative drugs can be harmful. therefore we carried out a systematic review of non-pharmacological interventions to reduce icU-related distress.eViDeNce acQUisitioN: a systematic search was conducted using Medline, embase, Psychinfo, cinahl and the Web of science. included studies evaluated the effect of non-pharmacological interventions to reduce icU stress. study populations were adults in mixed or general icUs. outcomes were stress or psychological distress in or after the icU, using self-report or physiological measures. No meta-analysis was possible due to heterogeneity, therefore studies were arranged according to intervention type, and outcomes examined together with risk of bias criteria.eViDeNce sYNtHesis: twenty-three studies were eligible, including 15 randomi�ed controlled trials. Non-pharma-: twenty-three studies were eligible, including 15 randomi�ed controlled trials. Non-pharma-cological interventions included music therapy (11 studies), mind-body practices (5) and psychological interventions (7). 12 studies showed a beneficial effect. However only three of the 12 had a low risk of bias, and many studies in the review were under-powered to detect an effect. only 5 studies measured a medium/long term psychological outcome such as PtsD or depression at 2-12 months.coNclUsioNs: evidence indicates that non-pharmacological approaches to reducing icU distress, in particular psy-chological interventions, may be beneficial. The evidence base would be strengthened by the implementation of fully-powered studies using robust designs, that measure longer-term outcomes.(Cite this article as: Wade DM, Moon Z, Windgassen ss, Harrison aM, Morris l, Weinman Ja. Non-pharmacological inter-Wade DM, Moon Z, Windgassen ss, Harrison aM, Morris l, Weinman Ja. Non-pharmacological inter-ventions to reduce icU-related psychological distress: a systematic review. Minerva anestesiol 2016;82:465-78)Key words: intensive care units - Psychotherapy - Music therapy.


WaDe NoN-PHarMacological iNterVeNtioNs to reDUce icU-relateD PsYcHological Distress


tions to reduce stress and psychological dis-tress carried out during or following an icU admission. any non-pharmacological inter-ventions were included, providing that the intervention aim was to reduce stress, anxiety or psychological distress in patients who were expected to survive the icU. any study design was accepted if its aim was to evaluate an in-tervention. end-of-life interventions were ex-cluded.

Types of parTicipanTs

the study populations were adults who had been admitted to mixed or general icUs. stud-ies only including specific diseases were ex-cluded to ensure the results could be extrapo-lated to a general intensive care population. studies of fewer than 20 participants were excluded, as they are extremely likely to have bias.

Types of ouTcome measures

studies were included if they measured stress or psychological distress in or following an icU admission using a validated method. studies could also use a physiological measure of stress such as heart rate or blood pressure. sleep was also included as sleep problems can be seen as a proxy measure of stress.

Exclusion criteria

studies were excluded if they were not in the english language, if the full text was not available or if they were published prior to 2000 considering that the nature of intensive care provision has changed significantly over time. only papers published in peer reviewed journals were included.

Search strategy

a systematic search was conducted in March 2015 using the following databases: Medline, embase and Psycinfo, cinahl and the Web of science. as an example, the search terms for the Medline database search are shown in

long-term psychological well-being and recov-ery.3, 4 systematic reviews show that after icU, the prevalence of post-traumatic stress disorder (PtsD) ranges from 19-27% 5, 6 and the median prevalence of depression is around 28%.7

Poor psychological outcomes are associated with intrusive memories, delusional memo-ries, stressful experiences and disturbed mood in icU.1, 3, 4 Furthermore, acute stress in the icU has been found to be a strong risk factor for later adverse psychological outcomes.4, 8, 9 High levels of anxiety, trauma and depression are likely to significantly impair an individu-al’s physical and functional recovery, as well as causing a significant reduction in quality of life and well-being.10, 11 therefore there is considerable clinical interest in identifying interventions that could reduce psychological distress in- or post-icU and to improve the psychological well-being of survivors.

in practice, stressed icU patients have been treated with sedative drugs such as dia�epam and other ben�odia�epines. However the admin-istration of these drugs to icU patients has in-creasingly been associated with adverse effects such as delirium and hallucinations in the icU, and flashbacks following the ICU.4, 12 However, little is known about the efficacy of non-phar-macological interventions to reduce distress in icU and long-term psychological morbidity af-ter icU, and no systematic reviews have been conducted to assess this. this paper aims to systematically review the evidence from stud-ies published in the last 15 years reporting the effects of non-pharmacological interventions aimed at reducing stress and psychological dis-tress in patients who are expected to survive the icU. interventions for patients at the end of life were not included as the aim and content of these interventions would be different from interven-tions designed to reduce stress in survivors.

Evidence acquisitionCriteria for study selection

Types of sTudies

included studies were designed to examine the effect of non-pharmacological interven-

NoN-PHarMacological iNterVeNtioNs to reDUce icU-relateD PsYcHological Distress WaDe


trol group; intervention details, outcome mea-sures and results. the data extraction of each paper was conducted independently by one of three researchers (sW, ZM, aH).

Evidence synthesis

No meta-analysis could be carried out due to the heterogeneity of interventions, outcomes, length of follow-up and statistical methods used and the lack of consistency in reporting results. therefore studies were arranged ac-cording to intervention type, and outcomes examined together with risk of bias criteria (tables i-iii).

as seen in Figure 1, the original search re-trieved 2639 references. after screening of titles, abstracts and full texts, and manually searching reference lists, 23 papers were in-cluded in the review.14-36

Characteristics of the studies

summaries of the characteristics of the in-cluded studies are shown in tables. these in-cluded 11 music interventions (table i), 5 mind-body interventions (table ii), and seven psychological interventions (table iii). of these 23 studies, 15 were rcts (including a pilot rct) and two were randomised cross-over designs. the remaining papers were: one comparative controlled study, one comparative time series, two pre-post-studies (3 periods); one historical control study, and one second-ary analysis of a sub-group from an rct. in total, data from 2,135 patients from 44 icUs were included in the review. twelve studies included only ventilated patients. six of the 23 studies were multi-centre. of the 23 studies, two were conducted in iran, two in turkey, one in canada, three in the Us, two in taiwan, two in china and the rest in europe (Netherlands, sweden, UK, France, italy, or multi-centre).

Population characteristics

all patients were adults in a mixed or gen-eral icU. the percentage of males in the study ranged from 34% to 79%. the range of aver-

supplementary table i (online content only). the search terms were adapted slightly for the remaining databases. results from the search strategy were managed using endnote and all duplicates were removed. one author (aH) se-quentially reviewed the titles and abstracts to determine which studies were eligible for in-clusion in the review. any papers not meeting the eligibility criteria were removed. the full texts of the remaining papers were then read by two authors (aH, ZM or sW) to determine whether they were eligible. the reference lists of the included papers were read to identify any other potentially relevant studies.

Assessment of risk of bias in included studies

each included study was assessed for the risk of bias. this assessment was conducted using relevant criteria, as advised in the Pris-Ma guidelines.13 the criteria were random-i�ation, similarity between groups at baseline, outcome measurement, statistical analysis and sample si�e. studies were assigned a score of low risk (2), unclear risk (1) or high risk (0) for each aspect, with a total possible score of 10 (supplementary table ii, online content only). studies scoring 0-6 overall were deemed to have a high risk of bias, studies scoring 7-8 had an unclear risk of bias and only studies scoring 9 or 10 had a low risk of bias. every paper was assessed for risk of bias by two re-searchers (from sW, ZM, JW or DW). agree-ment was reached between researchers on all studies’ risk of bias.

Data extraction

a standardi�ed data extraction table was drawn up and validated by all researchers. Once the final set of papers had been deter-mined, data from each paper were extracted into the standardised data table. Data were ex-tracted on the study’s aim; design; participants recruited, included and followed up; recruit-ment process; setting; inclusion and exclusion criteria; length of stay in icU, psychological history, gender, ethnicity, socio-economic sta-tus, icU clinical characteristics, details of con-



Table i.—�Characteristics and results of studies investigating music / nature sound interventions.

studyrisk of bias

score (10 lowest risk; 0 highest)

DesignN of Ps re-cruited (N. in statistical

analysis)inclusion criteria/setting exclusion criteria Patient characteristics intervention

outcome of inter-est – anxiety/stress measure

results(outcome of interest)

other outcomes (Measures)

results(other outcomes)

saadatmand et al. 2012 14

9 rct(intervention vs.

control groups)

60 (60) Mechanically ventilated patients in icU, iran.

e.g. glasgow coma score 8 or below; mentally incompetent and unable to communicate, psychiatric, neurological illnesses

34% maleMean age 44Mean length of stay:

7.6 days

90 minutes listening to nature sounds in rest period

(controls rested in silence)

anxiety using Faces anxiety scale, score range 1-5, every 30 minutes during intervention period and 30 minutes after intervention

Difference between anxiety scores of the two groups (P<0.001). odds of higher anxiety score 4.5 times more in control group

agitation measured using richmond agitation sedation scale every 30 minutes during intervention and 30 minutes after intervention

Difference between agitation scores of the two groups (P<0.001). odds of higher agitation scores 11.24 times more in control group

Beaulieu-Boire et al. 2013 15

9 randomi�ed crossover study

(intervention vs. control –groups)

55 (49) Ventilated patients in mixed icU, canada.

ventilation > 3d

Deafness, pregnancy 65% maleMean age 62Mean length of stay:

11.5 daysacute Physiology

and chronic Health evaluation (aPacHe ii) 26

60 minutes slow-tempomusic morning and evening

arterial blood pressure (mmHg) at end of intervention period/60 minutes

No differences in arterial blood pressure

Blood cortisol (nmol/l) at end of intervention period

Unclear if cortisol differed between groups, but it decreased during music intervention (815 vs. 727 nmol/l)

su et al. 2012 16 8 rct(intervention vs.

control groups)

28 (28) all patients in mixed medical icU, taiwan, aPacHe <25. icU stay >1 day

alcoholism, hearing impairment, physical restraint

61% maleMean age 62aPacHe ii 19

45 minutes of “sedating piano

music “ at bedtime

sleep quality using VsH sleep scale the following morning (score range =0-1500mm based on 15 visual analogue scales

sleep scale scores higher in intervention group: 545 vs 497, P<0.05. (Higher scores = better sleep quality)

“relaxation indices” - Heart rate (Hr), respiratory rate (rr) and mean arterial pressure (MaP) measured every five minutes during the intervention

intervention group had lower Hr, rr and MaP at varying time points, suggesting benefit for intervention group

Korhan et al. 2011 17


control groups)

60 (60) Patients ventilated in mixed icU. turkey. icU stay> 1 day

e.g. Psychiatric or neurological illnesses, taken neuromuscular blocker or antihypertensive drug, inadequate hearing


8.3 days

60-minute music intervention (not specified when)

respiratory rate at 30, 60 and 90 minutes (0 minutes = start of intervention period)

Significant difference in rr between groups (P<0.05, rr lower in intervention group)

1. change in rr over time (90 mins)

2. change in Hr over time (90 mins)

1. reduction in rr and Hr over time in intervention group

Wong et al. 2001 18


(intervention and control groups)

20 (20) Patients ventilated in mixed icU. china. icU stay >1 day

e.g. Not mentally competent, hearing problems, unable to communicate, not undergoing mechanical ventilation


6.1 days

30-minutes listening to chinese or Western music, or 30-minute rest (all Ps had both, crossing over)

reduction in anxiety (state-trait anxiety inventory: stai, score range 20-80) from baseline to 30 minutes

greater reduction in anxiety in intervention than control conditions

(mean difference:11, P<0.05)

Differences in rr (number of respirations in 1 minute) from baseline to 30 minutes

No differences found.

Han et al. 2010 19 8 rct(Music vs.

headphone vs. usual care groups)

137 (137) Ventilated patients mixed icU.

chinaicU stay >1 day

Hearing impairment, skull injury patients on cMV or cPaP


3.5 days

one 30- minute session listening to choice of relaxing music (classical or chinese)

change in state anxiety(using stai, scores 20-80)

between baseline and 30 minutes

Significant difference in anxiety reduction across 3 groups (P<0.01). Mean reductions: music 10.7, headphone 3.3 and control 0.8

Mean difference between groups for

1. Hr2. rr

Differences in Hr (P<0.05) and rr (P<0.01).

intervention group had greater reduction in Hr and rr over music period.

chlan 2013 20 7 rct(Music vs.

headphones vs. usual care)

373 (373) Patients ventilated in 12 general icUs. Usa. icU stay >1 day

incapacity, unresponsive, delirious, aggressive ventilation


7 daysaPacHe iii 63.6

self-initiated music whenever desired while ventilated (max 30 days). guided by music therapist

anxiety assessed daily with the 100mm Vas-a. 0=not anxious at all 100=most anxious ever

anxiety scores consistently lower during study period, by 19.5mm on Vas-a, in intervention group cf controls (P<0.01)

sedation exposure, operationalised as:

1. Daily sedative drug intensity score

2. sedative dose frequency

greater decrease over time in sedation intensity score and sedation frequency in music group (P<0.05, P<0.05)

Dijkstra2010 21

7 Pilot rct (intervention vs. control groups)

20 (20) 3 icUs in a university teaching hospital, Netherlands

ramsay scoreof 1, “too sluggish”, non-

responsive to any touch or auditory stimulus


27.3 daysaPacHe ii 22.8

three 30 minute music sessions over 2 days (patient or family choice)

Decrease in systolic blood pressure (sBP) between baseline and 30 minutes

No significant difference in decrease of sBP between groups

Difference in decrease of rr and Hr between groups

No significant differences identified between groups

lee et al. 2005 22 7 rct(intervention vs.

control groups)

64 (64) Patients ventilated in general icU. china. icU stay >not stated

Haemodynamically unstable participants


2.5 days

single 30-minute music intervention (chinese or Western)

anxiety (stai) at 30 minutes

No significant reduction in anxiety

Difference in rr at 30 minutes

No significant difference in RR at 30 minutes between groups

chlan & engeland 2013 23

7 Descriptive design (secondary data analysis of sub-group from an

rct with 3 groups – music vs. headphones vs. usual care)

65 (65) 12 ICUs in five hospitals in Usa.

Patients receiving aggressive ventilator support, were unstable hemodynamically, unresponsive, receiving chronic ventilator support prior to hospitali�ation, or had a documented mental health problem


7.9 daysaPacHe iii 57.2

Patients listen to preferred music when desired – while on ventilator (30-day maximum)

Urinary free cortisol – 24 hour urine collections but timing of collections not clear in relation to intervention

treatment showed no decreases in cortisol over time or when compared to control groups

N/a N/a

almerud & Petersson 2003 24

4 comparative study (intervention vs. control groups)

20 (20) Patients ventilated in general icU. sweden. icU stay >1 day

severe psychiatric condition, severe depression or learning difficulties


11.5 days

30 minute music intervention at bedtime

rr at 30 minutes No differences between groups

Hr at 30 minutes No differences between groups



Table i.—�Characteristics and results of studies investigating music / nature sound interventions.

studyrisk of bias


DesignN of Ps re-cruited (N. in statistical

analysis)inclusion criteria/setting exclusion criteria Patient characteristics intervention

outcome of inter-est – anxiety/stress measure


other outcomes (Measures)


saadatmand et al. 2012 14


control groups)

60 (60) Mechanically ventilated patients in icU, iran.

e.g. glasgow coma score 8 or below; mentally incompetent and unable to communicate, psychiatric, neurological illnesses


7.6 days

90 minutes listening to nature sounds in rest period

(controls rested in silence)

anxiety using Faces anxiety scale, score range 1-5, every 30 minutes during intervention period and 30 minutes after intervention

Difference between anxiety scores of the two groups (P<0.001). odds of higher anxiety score 4.5 times more in control group

agitation measured using richmond agitation sedation scale every 30 minutes during intervention and 30 minutes after intervention

Difference between agitation scores of the two groups (P<0.001). odds of higher agitation scores 11.24 times more in control group

Beaulieu-Boire et al. 2013 15


(intervention vs. control –groups)

55 (49) Ventilated patients in mixed icU, canada.

ventilation > 3d

Deafness, pregnancy 65% maleMean age 62Mean length of stay:

11.5 daysacute Physiology

and chronic Health evaluation (aPacHe ii) 26

60 minutes slow-tempomusic morning and evening

arterial blood pressure (mmHg) at end of intervention period/60 minutes

No differences in arterial blood pressure

Blood cortisol (nmol/l) at end of intervention period

Unclear if cortisol differed between groups, but it decreased during music intervention (815 vs. 727 nmol/l)

su et al. 2012 16 8 rct(intervention vs.

control groups)

28 (28) all patients in mixed medical icU, taiwan, aPacHe <25. icU stay >1 day

alcoholism, hearing impairment, physical restraint

61% maleMean age 62aPacHe ii 19

45 minutes of “sedating piano

music “ at bedtime

sleep quality using VsH sleep scale the following morning (score range =0-1500mm based on 15 visual analogue scales

sleep scale scores higher in intervention group: 545 vs 497, P<0.05. (Higher scores = better sleep quality)

“relaxation indices” - Heart rate (Hr), respiratory rate (rr) and mean arterial pressure (MaP) measured every five minutes during the intervention

intervention group had lower Hr, rr and MaP at varying time points, suggesting benefit for intervention group

Korhan et al. 2011 17


control groups)

60 (60) Patients ventilated in mixed icU. turkey. icU stay> 1 day

e.g. Psychiatric or neurological illnesses, taken neuromuscular blocker or antihypertensive drug, inadequate hearing


8.3 days

60-minute music intervention (not specified when)

respiratory rate at 30, 60 and 90 minutes (0 minutes = start of intervention period)

Significant difference in rr between groups (P<0.05, rr lower in intervention group)

1. change in rr over time (90 mins)

2. change in Hr over time (90 mins)

1. reduction in rr and Hr over time in intervention group

Wong et al. 2001 18


(intervention and control groups)

20 (20) Patients ventilated in mixed icU. china. icU stay >1 day

e.g. Not mentally competent, hearing problems, unable to communicate, not undergoing mechanical ventilation


6.1 days

30-minutes listening to chinese or Western music, or 30-minute rest (all Ps had both, crossing over)

reduction in anxiety (state-trait anxiety inventory: stai, score range 20-80) from baseline to 30 minutes

greater reduction in anxiety in intervention than control conditions

(mean difference:11, P<0.05)

Differences in rr (number of respirations in 1 minute) from baseline to 30 minutes

No differences found.

Han et al. 2010 19 8 rct(Music vs.

headphone vs. usual care groups)

137 (137) Ventilated patients mixed icU.

chinaicU stay >1 day

Hearing impairment, skull injury patients on cMV or cPaP


3.5 days

one 30- minute session listening to choice of relaxing music (classical or chinese)

change in state anxiety(using stai, scores 20-80)

between baseline and 30 minutes

Significant difference in anxiety reduction across 3 groups (P<0.01). Mean reductions: music 10.7, headphone 3.3 and control 0.8

Mean difference between groups for

1. Hr2. rr

Differences in Hr (P<0.05) and rr (P<0.01).

intervention group had greater reduction in Hr and rr over music period.

chlan 2013 20 7 rct(Music vs.

headphones vs. usual care)

373 (373) Patients ventilated in 12 general icUs. Usa. icU stay >1 day

incapacity, unresponsive, delirious, aggressive ventilation


7 daysaPacHe iii 63.6

self-initiated music whenever desired while ventilated (max 30 days). guided by music therapist

anxiety assessed daily with the 100mm Vas-a. 0=not anxious at all 100=most anxious ever

anxiety scores consistently lower during study period, by 19.5mm on Vas-a, in intervention group cf controls (P<0.01)

sedation exposure, operationalised as:

1. Daily sedative drug intensity score

2. sedative dose frequency

greater decrease over time in sedation intensity score and sedation frequency in music group (P<0.05, P<0.05)

Dijkstra2010 21

7 Pilot rct (intervention vs. control groups)

20 (20) 3 icUs in a university teaching hospital, Netherlands

ramsay scoreof 1, “too sluggish”, non-

responsive to any touch or auditory stimulus


27.3 daysaPacHe ii 22.8

three 30 minute music sessions over 2 days (patient or family choice)

Decrease in systolic blood pressure (sBP) between baseline and 30 minutes

No significant difference in decrease of sBP between groups

Difference in decrease of rr and Hr between groups

No significant differences identified between groups

lee et al. 2005 22 7 rct(intervention vs.

control groups)

64 (64) Patients ventilated in general icU. china. icU stay >not stated

Haemodynamically unstable participants


2.5 days

single 30-minute music intervention (chinese or Western)

anxiety (stai) at 30 minutes

No significant reduction in anxiety

Difference in rr at 30 minutes

No significant difference in RR at 30 minutes between groups

chlan & engeland 2013 23

7 Descriptive design (secondary data analysis of sub-group from an

rct with 3 groups – music vs. headphones vs. usual care)

65 (65) 12 ICUs in five hospitals in Usa.

Patients receiving aggressive ventilator support, were unstable hemodynamically, unresponsive, receiving chronic ventilator support prior to hospitali�ation, or had a documented mental health problem


7.9 daysaPacHe iii 57.2

Patients listen to preferred music when desired – while on ventilator (30-day maximum)

Urinary free cortisol – 24 hour urine collections but timing of collections not clear in relation to intervention

treatment showed no decreases in cortisol over time or when compared to control groups

N/a N/a

almerud & Petersson 2003 24

4 comparative study (intervention vs. control groups)

20 (20) Patients ventilated in general icU. sweden. icU stay >1 day

severe psychiatric condition, severe depression or learning difficulties


11.5 days

30 minute music intervention at bedtime

rr at 30 minutes No differences between groups

Hr at 30 minutes No differences between groups



nature sounds. the mind-body interventions (table ii) included tactile touch,26 valerian acu-pressure,28 reflexology 27 and massage,25, 28 de-livered in the icU by nurses, family or others.

the psychology interventions (table iii) in-cluded one clinical psychology intervention; three icU patient diary interventions;30, 32, 33 one relaxation and guided imagery interven-tion delivered by nurses;34 one intervention for nurses to facilitate family participation in psy-chological care;36 and one post-icU self-help rehabilitation programme including advice about psychological self-care.31 the clinical psychology study involved a range of inter-ventions, including psycho-education, coun-selling and stress management, delivered by three clinical psychologists and other trained staff members. the diary studies involved an icU diary being kept for the duration of the pa-tient’s stay by healthcare professionals or fam-

age ages of patients was 44-71. Nine studies included a measure of illness severity such as the aPacHe (acute Physiology and chron-ic Health evaluation) ii score. the average length of stay ranged from 2.5-27.3 days.

Interventions

music and mind-body Therapies

in most studies (table i), music was played to patients through headphones while they were mechanically ventilated in the intensive care Unit. control groups mostly used noise-cancelling headphones without music at the same time. Music was played for 30-90 min-utes for one or more sessions per day, or at the patient’s discretion. the majority of the inter-ventions used calming, slow-tempo classical or other traditional music, and one study used

Table ii.—�Characteristics and results of studies investigating mind / body interventions.

studyrisk of bias


Design

N. of Ps recruited

(N. in statistical analysis)

inclusion criteria/setting exclusion criteria Patient characteristics intervention

outcome of interest – anxiety/

stress measureresults

(outcome of interest)other outcomes

(Measures)results

(other outcomes)

Vahedian-a�imi et al. 2014 25

9 rct (intervention vs. control groups)

90 (90) Patients in general icU, tehran, > 10 days hospitalised, gcs 7 - 12

contraindication for changes of body position or body massage

51% malegcs 8

60 minute massage delivered by trained family member

sBP at one hour post intervention

SBP significantly lower in intervention (122) than control (128) group.

DBP (diastolic blood pressure) and Hr at one hour

No differences between groups in DPB and Hr

Henricson et al. 2008 26


control groups)

45 (44) all patients, general icU, sweden. icU stay >24 hours

acute neurological disease, head injury, psychosis.


5 daysaPacHe 21

Tactile touch (effleurage) with soft background music for one hour per day during rest period for 5 days

anxiety (Faces anxiety scale, score 1-5) on days 1-5 immediately post-intervention

No significant differences in anxiety between groups on day 1, 3, 4, 5. Day 2: post-intervention anxiety was significantly lower in intervention group (P<0.05)

Daily doses of sedatives (mida�olam and propofol) observed and reported in m/l per day

No significant differences found in sedation requirement between groups on any day

Korhan et al. 2014 27 7 rct (intervention vs. control groups)

60 (60) Ventilated patients in icU, turkey. gc s>9

Psychiatric/ neurological illness

53% maleMean age 51

30 minutes of reflexology twice daily for five days

change in sBP from 0 to 30 minutes

Significant difference between groups in change in mean sBP from 0 to 30 mins, on all days (sBP decreased in intervention group)

change in DBP and Hr from 0 to 30 minutes

Significant difference between groups in change in DBP and Hr from 0 to 30 mins, on all days (sBP decreased in intervention group)

chen et al. 2012 28 7 rct(intervention vs.

control groups)

86 (85) stable patients in icU, taiwan, aPs <15

Hand/foot amputees, bilateral paralysis, sedative users, sleeping pills > month

76% maleMean age 71aPs 12.1

Valerian acupressure (applying acupressure to points on wrist and foot with valerian oil) on 2nd night post-enrolment

sleep hours measured by actigraphy on night after intervention

intervention group had more hours of sleep (7.8 vs. 7.1, P<0.01)

Minutes spent awake.sleep quality (stanford

sleepiness scale, sss). Highest level of wakefulness =1 on the sss. Highest level of sleepiness – 7

intervention group had fewer minutes awake (14.2 vs. 54.6, P<0.01) and lower sss scores (2.5 vs. 3.4 P<0.01)

Hayes & cox 1999 29

4 comparison of pre-, during and post- intervention

25 (25) Patients in general icU, UK

Head trauma, raised intracranial pressure

52% maleMean age 54

5 minute foot massage by trained researcher (range: 1-10 sessions per patient)

Hr (bpm) during intervention compared to pre- and post- intervention

Hr lower during intervention (94.7 bpm) than pre- (97.3, P<0.01) or post- (96.3, P<0.05)

arterial blood pressure (mmHg) and respirations (breaths per minute) during intervention compared to pre- or post- intervention

arterial BP lower during intervention than pre- (83.6 vs. 85.5, P<0.01)

respirations lower during intervention (21.3) and post- (21) than pre- (23.1) (P<0.01)



Table ii.—�Characteristics and results of studies investigating mind / body interventions.

studyrisk of bias


Design

N. of Ps recruited

(N. in statistical analysis)

inclusion criteria/setting exclusion criteria Patient characteristics intervention

outcome of interest – anxiety/

stress measureresults

(outcome of interest)other outcomes

(Measures)results

(other outcomes)

Vahedian-a�imi et al. 2014 25

9 rct (intervention vs. control groups)

90 (90) Patients in general icU, tehran, > 10 days hospitalised, gcs 7 - 12

contraindication for changes of body position or body massage

51% malegcs 8

60 minute massage delivered by trained family member

sBP at one hour post intervention

SBP significantly lower in intervention (122) than control (128) group.

DBP (diastolic blood pressure) and Hr at one hour

No differences between groups in DPB and Hr

Henricson et al. 2008 26


control groups)

45 (44) all patients, general icU, sweden. icU stay >24 hours

acute neurological disease, head injury, psychosis.


5 daysaPacHe 21

Tactile touch (effleurage) with soft background music for one hour per day during rest period for 5 days

anxiety (Faces anxiety scale, score 1-5) on days 1-5 immediately post-intervention

No significant differences in anxiety between groups on day 1, 3, 4, 5. Day 2: post-intervention anxiety was significantly lower in intervention group (P<0.05)

Daily doses of sedatives (mida�olam and propofol) observed and reported in m/l per day

No significant differences found in sedation requirement between groups on any day

Korhan et al. 2014 27 7 rct (intervention vs. control groups)

60 (60) Ventilated patients in icU, turkey. gc s>9

Psychiatric/ neurological illness

53% maleMean age 51

30 minutes of reflexology twice daily for five days

change in sBP from 0 to 30 minutes

Significant difference between groups in change in mean sBP from 0 to 30 mins, on all days (sBP decreased in intervention group)

change in DBP and Hr from 0 to 30 minutes

Significant difference between groups in change in DBP and Hr from 0 to 30 mins, on all days (sBP decreased in intervention group)

chen et al. 2012 28 7 rct(intervention vs.

control groups)

86 (85) stable patients in icU, taiwan, aPs <15

Hand/foot amputees, bilateral paralysis, sedative users, sleeping pills > month

76% maleMean age 71aPs 12.1

Valerian acupressure (applying acupressure to points on wrist and foot with valerian oil) on 2nd night post-enrolment

sleep hours measured by actigraphy on night after intervention

intervention group had more hours of sleep (7.8 vs. 7.1, P<0.01)

Minutes spent awake.sleep quality (stanford

sleepiness scale, sss). Highest level of wakefulness =1 on the sss. Highest level of sleepiness – 7

intervention group had fewer minutes awake (14.2 vs. 54.6, P<0.01) and lower sss scores (2.5 vs. 3.4 P<0.01)

Hayes & cox 1999 29

4 comparison of pre-, during and post- intervention

25 (25) Patients in general icU, UK

Head trauma, raised intracranial pressure

52% maleMean age 54

5 minute foot massage by trained researcher (range: 1-10 sessions per patient)

Hr (bpm) during intervention compared to pre- and post- intervention

Hr lower during intervention (94.7 bpm) than pre- (97.3, P<0.01) or post- (96.3, P<0.05)

arterial blood pressure (mmHg) and respirations (breaths per minute) during intervention compared to pre- or post- intervention

arterial BP lower during intervention than pre- (83.6 vs. 85.5, P<0.01)

respirations lower during intervention (21.3) and post- (21) than pre- (23.1) (P<0.01)

Figure 1.—Flow-chart showing identification of records; screening of titles, abstracts and full-text articles; and inclusion of final studies.



Outcome measures

Five music studies used a scale measure of anxiety as their primary outcome and one16 measured sleep quality using a sleep scale (see table i for details of measures). The other five music studies used physiological measures

ily members. the diary was given to the patient either at discharge or at one month post-dis-charge. three of the psychology interventions were clearly delivered within the icU,33, 34, 36 the rehabilitation intervention was delivered post-icU,31 while the diary method included both in-icU and post-icU components.

Table iii.—�Characteristics and results of studies investigating psychological interventions.

study

risk of bias

score (10 lowest risk; 0

highest)

DesignN of Ps

recruited (n in statistical

analysis)

inclusion criteria/setting exclusion criteria Patient

characteristics intervention outcome of interest – anxiety/stress measure


other outcomes (measures)


Jones et al. 2010 30 10 rct(intervention vs.

control groups)

352 (333) Ventilated patients,6 european mixed

icUs.icU stay > 72 hours,

Ventilation > 24 hours

Pre-existing psychosis or PtsD.


13 daysaPacHe ii 19

icU diary kept during stay, given to patients one month post-discharge

Prevalence of PtsD at 3 months (using Post traumatic stress Diagnostic scale, PDs)

PtsD lower in intervention group (5% vs.13 %, P<0.05)

PtsD-related symptoms (Ptss-14 screening tool) at 3 months

No differences in Ptss-14 scores between groups

Jones et al. 2003 31 9 rctcontrol (routine

icU follow-up) vs. intervention (routine follow-up plus rehabilitation manual)

126 (102) Patients who had previously been ventilated in icU in three hospitals in the UK.

icU <48 hours, burn injury, neurological patients, pre-existing psychotic illness.


14 daysaPacHe ii17

in addition to routine follow-up, intervention patients used 6-week self-help rehab manual of psychological, and physical advice

Depression (Hospital anxiety and Depression scale, HaDs) at 8 weeks

No significant differences between groups in anxiety or depression

PtsD (impact of events scale, ies) at 8 weeks

No difference in PtsD between groups

garrouste-orgeas et al. 2012 32

7 intervention period between two control periods

143 (52) all patients in general icU, France icU stay >4 days

Dementia, not fluent in French


18 dayssaPs ii 41.7

icU diary, started on 4th day of admission, given to patients on discharge from icU

PtsD related symptoms (ies-revised) at 12 months

lower PtsD scores for diary (21) compared to pre-diary (35) and post-diary periods (30, P<0.05)

anxiety and depression at 3 months postdischarge (HaDs)

No differences in anxiety or depression at 3 months

Peris et al. 2011 33 7 Historical control 376 (209) Ventilated trauma patients, mixed icU, italy. icU stay >72 hours

Psychotic illnesses, previous critical illness, psychiatric medication, drug use or addiction


19 dayssaPs ii 41.8

clinical psychology intervention delivered for duration of icU stay

PtsD(ies-r) 12-months

postdischarge

lower rate of PtsD in intervention group (21% vs. 57%, P<0.01)

anxiety and depression (HaDs) at 12 months

No differences in depression or anxiety

richardson 2003 34 7 rct(intervention vs.

control groups)

36 (29) Medical, surgical and coronary patients in 3 mixed icUs, Utah

E.g. dementia, psychosis, neurological impairment, severe hypotension

47% maleMean age 58

Nurse-delivered relaxation and guided imagery sessions on two consecutive evenings

sleep quality on mornings following intervention sessions

(VsH sleep scale – score range 0-800mm based on 8 visual analogue scales)

No significant difference in sleep quality scores between groups

N/a N/a

Knowles & tarrier 2009 35


control groups)

36 (36) Patients in general icU, UK. icU stay >48 hours

suicide attempt, psychological symptoms, dementia

58% maleMean length of stay:

9 daysaPacHe ii 15

icU diary, given to patients one month postdischarge

anxiety (HaDs) at 2 months postdischarge

reduction in anxiety scores for intervention group (mean change 1.9, P<0.05). No reduction in control group. But between-group difference unclear

Depression (HaDs) at 2 months postdischarge

reduction in depression scores for intervention group (mean change=2.5, P<0.01). No reduction in control group. But between-group difference unclear

Black2010 36

6 comparative time series

170 (138) general icU in an inner city public hospital Northern ireland

Patients with family members who were physically unable to participate in the intervention due to being unable to visit, patients with no living family, patients with a terminal diagnosis.

Mean length of stay: 13 days

Nurse-facilitated family participation in psychological care during icU stay

Hr No differences in Hr between groups

BP No significant differences in BP



ately after the intervention period, in some cases over a number of days. Five of the seven psychological interventions (diaries, clinical psychology, self-help rehabilitation) measured psychological distress at 2-12 months using validated measures for PtsD, depression or anxiety (table iii). the family participation

of stress such as heart rate, respiratory rate, urinary free cortisol23 or blood pressure. the mind-body interventions (table ii) measured blood pressure, heart rate, hours of sleep and anxiety as primary indicators of icU stress. outcomes in music and mind-body studies were generally measured during or immedi-

Table iii.—�Characteristics and results of studies investigating psychological interventions.

study

risk of bias

score (10 lowest risk; 0

highest)

DesignN of Ps

recruited (n in statistical

analysis)

inclusion criteria/setting exclusion criteria Patient

characteristics intervention outcome of interest – anxiety/stress measure


other outcomes (measures)


Jones et al. 2010 30 10 rct(intervention vs.

control groups)

352 (333) Ventilated patients,6 european mixed

icUs.icU stay > 72 hours,

Ventilation > 24 hours

Pre-existing psychosis or PtsD.


13 daysaPacHe ii 19

icU diary kept during stay, given to patients one month post-discharge

Prevalence of PtsD at 3 months (using Post traumatic stress Diagnostic scale, PDs)

PtsD lower in intervention group (5% vs.13 %, P<0.05)

PtsD-related symptoms (Ptss-14 screening tool) at 3 months

No differences in Ptss-14 scores between groups

Jones et al. 2003 31 9 rctcontrol (routine

icU follow-up) vs. intervention (routine follow-up plus rehabilitation manual)

126 (102) Patients who had previously been ventilated in icU in three hospitals in the UK.

icU <48 hours, burn injury, neurological patients, pre-existing psychotic illness.


14 daysaPacHe ii17

in addition to routine follow-up, intervention patients used 6-week self-help rehab manual of psychological, and physical advice

Depression (Hospital anxiety and Depression scale, HaDs) at 8 weeks

No significant differences between groups in anxiety or depression

PtsD (impact of events scale, ies) at 8 weeks

No difference in PtsD between groups

garrouste-orgeas et al. 2012 32

7 intervention period between two control periods

143 (52) all patients in general icU, France icU stay >4 days

Dementia, not fluent in French


18 dayssaPs ii 41.7

icU diary, started on 4th day of admission, given to patients on discharge from icU

PtsD related symptoms (ies-revised) at 12 months

lower PtsD scores for diary (21) compared to pre-diary (35) and post-diary periods (30, P<0.05)

anxiety and depression at 3 months postdischarge (HaDs)

No differences in anxiety or depression at 3 months

Peris et al. 2011 33 7 Historical control 376 (209) Ventilated trauma patients, mixed icU, italy. icU stay >72 hours

Psychotic illnesses, previous critical illness, psychiatric medication, drug use or addiction


19 dayssaPs ii 41.8

clinical psychology intervention delivered for duration of icU stay

PtsD(ies-r) 12-months

postdischarge

lower rate of PtsD in intervention group (21% vs. 57%, P<0.01)

anxiety and depression (HaDs) at 12 months

No differences in depression or anxiety

richardson 2003 34 7 rct(intervention vs.

control groups)

36 (29) Medical, surgical and coronary patients in 3 mixed icUs, Utah

E.g. dementia, psychosis, neurological impairment, severe hypotension

47% maleMean age 58

Nurse-delivered relaxation and guided imagery sessions on two consecutive evenings

sleep quality on mornings following intervention sessions

(VsH sleep scale – score range 0-800mm based on 8 visual analogue scales)

No significant difference in sleep quality scores between groups

N/a N/a

Knowles & tarrier 2009 35


control groups)

36 (36) Patients in general icU, UK. icU stay >48 hours

suicide attempt, psychological symptoms, dementia

58% maleMean length of stay:

9 daysaPacHe ii 15

icU diary, given to patients one month postdischarge

anxiety (HaDs) at 2 months postdischarge

reduction in anxiety scores for intervention group (mean change 1.9, P<0.05). No reduction in control group. But between-group difference unclear

Depression (HaDs) at 2 months postdischarge

reduction in depression scores for intervention group (mean change=2.5, P<0.01). No reduction in control group. But between-group difference unclear

Black2010 36

6 comparative time series

170 (138) general icU in an inner city public hospital Northern ireland

Patients with family members who were physically unable to participate in the intervention due to being unable to visit, patients with no living family, patients with a terminal diagnosis.

Mean length of stay: 13 days

Nurse-facilitated family participation in psychological care during icU stay

Hr No differences in Hr between groups

BP No significant differences in BP



effecT of psychological inTervenTions

three of the seven psychological inter-ventions found a significant difference in the primary outcome. of these one was low risk and two were unclear risk. the low-risk study was an rct of a patient diary intervention.30 there was a reduction in PtsD prevalence at 3 months in the intervention group, who re-ceived a diary kept during their icU stay one month post-discharge. another diary inter-vention 32 and the clinical psychology inter-vention 33 were associated with reductions in longer-term psychological distress.

Discussion

in this systematic review, 23 studies were identified that evaluated non-pharmacological interventions to reduce distress or psychologi-cal distress in icU patients. of these, 12 stud-ies showed a clear beneficial effect on primary outcomes including physiological stress indi-cators or self-report measures of psychological distress.

Music/nature sounds was the most com-mon intervention, with eleven studies. there were also five mind-body studies and seven psychological studies. Both music and mind-body studies were characterised by short-term measures of stress including physiological in-dicators and anxiety scales, while more of the psychological studies measured longer-term outcomes using validated questionnaires for PtsD, depression or anxiety.

there was mixed evidence for the music therapy group, with six of 11 studies show-ing an effect on stress. the risk of bias was low in only one of these studies.14 a limitation of the music studies was the lack of longer-term follow up. However, focus on short-term outcomes did reflect the aim of such studies, which was to reduce anxiety during uncom-fortable medical procedures. some of these studies had small sample si�es. another po-tential limitation was the heavy reliance on physiological measures of stress, as the auto-nomic stress response has been found to vary across individuals.37 The group of five mind-

study used Hr to indicate stress, while the re-laxation/guided imagery study measured sleep quality with a sleep scale.34

Risk of bias assessment

Overall there were five studies with a low risk of bias (scoring 9 or 10: supplementary table iii, online content only). Four studies had a high risk of bias (scoring 0-6), while in 14 studies the risk of bias was unclear (scores of 7 or 8). the music grouping included 2 low risk, 8 unclear risk and one high risk studies. the mind-body grouping included one low-risk, 3 unclear risk and one high-risk studies. Finally the psychology grouping included two low-risk, 3 unclear risk and two high-risk stud-ies.

Effect of interventions

twelve of the 23 studies showed a clear ef-fect of the intervention on the outcome of in-terest. Of these studies showing a benefit, three were low-risk, eight were unclear risk and one was a high-risk study.

effecT of music inTervenTions

six out of 11 studies found that music ther-apy reduced stress. one of these studies was low-risk, the others were unclear risk. the low-risk study 14 showed that anxiety and agi-tation were reduced in ventilated patients lis-tening to nature sounds for 90 minutes, during the intervention and at 30 minutes after it.

effecT of mind-body inTervenTions

Improvement was found in four of the five mind-body intervention studies. of these stud-ies, one was low-risk, three were unclear risk and one was high risk. the low risk study 25 showed a reduction in systolic blood pressure after one hour, in icU patients who received a 60-minute massage from a trained family member. Reflexology, valerian acupressure and foot massage were also associated with reduced stress.



it is thought that diaries could lessen the impact of delusional memories, which are associated with post-icU PtsD, by strengthening factual memories (thought to be protective against PtsD). However diaries have been critiqued for a lack of sound theoretical basis underlying their proposed mechanism of action. there is also a great deal of variation in the practice of how diaries are compiled and when they are given to patients. Finally their acceptability to patients is not known.39

the single study in the review that evalu-ated the provision of psychotherapeutic sup-port from psychologists in the icU 33 had an unclear risk of bias. it covered a wide range of activities including stress management, coun-selling, psycho-education, and coping strate-gies aimed at anxiety, depression, discomfort, hopelessness and helplessness. consequently there was no standardised procedure, making interpretation difficult, and constraining scien-tific replication. However, the study resulted in a very large significant reduction in PTSD at 12 months, suggesting that further exploration of the effectiveness of psychological interven-tions for critical care patients would be worth-while. this would require careful consideration of key risk factors for long-term post-icU dis-tress such as mood disturbance, acute stress, early intrusive memories, hallucinations and delusions,1, 3, 4 to guide the selection of psy-chotherapeutic techniques that could have an impact on these specific symptoms.

the present review is limited by a number of factors. First, the variety of outcome mea-sures used meant that performance of a meta-analysis was not possible. in this review, scale measures of anxiety and sleep quality, as well as physiological stress markers were included to reflect ICU stress. However, future reviews may benefit from limiting the outcome mea-sures included, to provide more generali�able findings, aiding interpretation and comparison. the risk of bias assessment could have been improved by using the recently published, validated cochrane risk of bias assessment tool.13 However several of its key criteria (e.g. blinding) were not relevant to these studies. these limitations could have resulted in an

body interventions were found to be generally effective in the short term for reducing stress. However only one of these studies (family-delivered massage) 25 was low risk for bias, sample si�es were fairly small and all studies focused on short-term measures.

three of the group of seven psychological studies proved effective (two diary interven-tions and a clinical psychology service) in re-ducing longer-term psychological outcomes, but only one was low-risk.30 evidence is stron-gest for this third group taking into account study designs used, length of follow-up, mea-surements methods employed and strength of effects (table iii).

What conclusions can be drawn from this systematic review for the future? caution is needed because the strength of evidence for the effect of non-pharmacological interven-tions to reduce icU stress is unclear at present, based on the findings of this review. Neverthe-less, elements of these interventions appear promising. studies suggest that music, nature sounds, various forms of massage and reflex-ology could all be employed to reduce acute stress among icU patients, including mechani-cally ventilated patients. as acute stress is an important precursor of long-term psychologi-cal morbidity such as PtsD, these short-term practices could also have an impact on long-term outcome. they might certainly be popu-lar with patients and families, given a trend to using more complementary or mind-body ther-apies, with less reliance on drugs.38 the risk of harm from such therapies is not generally high, but protocols and training for staff are needed as some critically ill patients would not be able to receive certain complementary therapies.

Potentially these elements could be deliv-ered within complex interventions that include psychological techniques to bring about lon-ger-term improvement for patients with stress in the icU. the strongest evidence for psycho-logical interventions to date is for the use of icU patient diaries. the idea of diaries is that they could reduce patients’ risk of developing PtsD (a disorder characterised by intrusive memories of a trauma) by filling in memory gaps, which are common among icU patients.



reduce patient stress in or following an icU admission have an unclear or high risk of bias, and predominantly centre around short- to medium-term outcomes. the present review highlights the need for larger studies in this area, utilising more robust study designs such as rcts, and measuring longer-term outcomes with validated measures of psychological dis-tress. Future research could include the evalu-ation of complex interventions that include elements of music, mind-body work or psy-chological support, to reduce in-icU or post-icU psychological distress.

Key messages

— as critical care patients frequently suffer acute and long-term stress that may be exacerbated by the effects of sedative drugs, a review of non-pharmacological interventions to reduce icU-related stress is timely.

— This review identified 23 studies, including 15 randomi�ed controlled trials, evaluating three types of non- pharmaco-logical intervention to reduce icU stress: Music or nature sound therapy; mind/body practices and psychological interventions.

— of the 23 studies, more than half (12) showed that a non-pharmacological inter-vention reduced short-term or long-term stress in critical care patients. However the overall pattern of evidence was inconsis-tent and study designs were weak.

— given early cautiously promising re-sults, robust studies such as rcts should evaluate non-pharmacological interven-tions to reduce stress in ICU, specifically psychological interventions, as the stron-gest of the three groups of interventions reviewed here.

References

1. Jones c, Backman c, capu��o M, Flaatten H, rylander C, Griffiths RD. Precipitants of post-traumatic stress dis-order following intensive care: a hypothesis generating study of diversity in care. intens care Med 2007;33:978-85.

over-estimation of the strength of the evidence. However this was avoided by using a stringent cut-off for identifying low-risk studies (risk of bias scores of 9 or 10).

Many questions remain unanswered from this review. It is not possible to say definitively whether there is a clinically important effect of non-pharmacological interventions to reduce icU-related stress, given the heterogeneity of both methods and results across studies. Fur-thermore, it remains unclear whether non-phar-macological interventions, if effective, might be of more benefit to certain sub-sets of pa-tients, such as those with high baseline anxiety or a predisposition to psychological morbidity due to family background, previous trauma or prior psychological conditions. Future studies might determine whether non-pharmacological interventions are more effective when patients are screened for acute stress or anxiety.

it is also not known how non-pharmacolog-ical interventions compare to “pharmacologi-cal” interventions, such as changing sedation practice. Hypothesised associations between drugs, delirium and post-icU psychologi-cal outcomes have not yet been established by robust intervention studies. one small trial 40 found that post-traumatic stress symp-toms were reduced in a group of patients who received daily sedation interruption. However, the study (of only 32 patients) found no sig-nificant difference in PTSD diagnosis rate. Furthermore, a study 41 of lightly sedated com-pared to heavily sedated patients found no sub-sequent difference in PtsD, anxiety or depres-sion in lightly sedated patients.

Conclusions

evidence from this review indicates that non-pharmacological approaches to reducing ICU-related distress may be beneficial to pa-tients. these could include elements of music therapy, mind-body practices or psychological support. However this review also highlights the need for improved conceptualisation of psychological interventions to be used in icU. Furthermore the majority of studies investi-gating non-pharmacological interventions to



19. Han l, li JP, sit JW, chung l, Jiao ZY, Ma Wg. ef-fects of music intervention on physiological stress re-sponse and anxiety level of mechanically ventilated pa-tients in china: a randomised controlled trial. J clin Nurs 2010;19:978-87.

20. chlan ll, Weinert cr, Heiderscheit a, tracy MF, skaar DJ, guttormson Jl. effects of patient-directed music in-tervention on anxiety and sedative exposure in critically ill patients receiving mechanical ventilatory support: a randomi�ed clinical trial. JaMa 2013;309:2335-44.

21. Dijkstra BM, gamel c, ven der Bijl JJ, Bots Ml, Keseci-oglu J. the effects of music on physiological responses and sedation scores in sedated, mechanically ventilated patients. J clin Nurs 2010;19:1030-9.

22. lee oK, chung YF, chan MF, chan WM. Music and its effect on the physiological responses and anxiety levels of patients receiving mechanical ventilation:a pilot study. J clin Nurs 2005;14:609-20.

23. Chlan LL, Engeland WC, Savik K. Does music influence stress in mechanically ventilated patients? intens crit care Nurs 2013;29:121-7.

24. almerud s, Petersson K. Music therapy-a complemen-tary treatment for mechanically ventilated intensive care patients. intens crit care Nurs 2003;19:21-30.

25. Vahedian-a�imi a, ebadi a, Jafarabadi Ma, saadat s, ahmadi F. effect of massage therapy on vital signs and gcs scores of icU patients:a randomised controlled trial. trauma Mon 2014;19:e17031.

26. Henricson M, ersson a, Maatta s, segesten K, Berglund al. the outcome of tactile touch on stress parameters in intensive care:a randomi�ed controlled trial. comple-ment ther clin Pract 2008;14:244-54.

27. Korhan EA, Khorshid L, Uyar M. Reflexology:Its effects on physiological anxiety signs and sedation needs. Holist Nurs Pract 2014;28:6-23.

28. chen JH, chao YH, lU sF, shiung tF, chao YF. the effectiveness of valerian acupressure on the sleep of icU patients:a randomi�ed clinical trial. int J Nurs stud 2012;29:913-20.

29. Hayes J, Cox C. Immediate effects of a five minute foot massage on patients in critical care. intens crit care Nurs 1999;15:77-82.

30. Jones c, Backman c, capu��o M, egerod i, Flatten H, granja c, et al. intensive care diaries reduce new-onset post traumatic stress disorder following critical illness: a randomised, controlled trial. crit care 2010;14:r168.

31. Jones C, Skirrow P, Griffiths RD, Humphris GH, In-gleby s, eddleston J, et al. rehabilitation after critical illness: a randomised controlled trial. crit care Med 2003;31:2456-61.

32. garrouste-orgeas M, coquet i, Perier a, timsit JF, Pochard F, lancrin F, et al. impact of an intensive care unit diary on psychological distress in patients and rela-tives. crit care Med 2012;40:2033-40.

33. Peris a, Boni��oli M, io��elli D, Migliaccio Ml, Zagli g, Bacchereti a. early intra-intensive care unit psycho-logical intervention promotes recovery from post trau-matic stress disorders, anxiety and depression symptoms in critically ill patients. crit care 2011;15:r41.

34. richardson s. effects of relaxation and imagery on the sleep of critically ill patients. Dimens crit care Nurs 2003;22:182-90.

35. Knowles re, tarrier N. evaluation of the effect of pro-spective patient diaries on emotional well-being in in-tensive care unit survivors:a randomi�ed controlled trial. crit care Med 2009;37:184-91.

36. Black P, Boore Jr, Parahoo K. the effect of nurse-facili-tated family participation in the psychological care of the critically ill patient. JaN 2011;67:1091-101.

37. cacioppo Jt, Berntson gg, Malarkey WB, Kiecolt-gla-ser JK, sheridan JF, Poehlmann KM, et al. autonomic,

2. Wade DM, Brewin cr, Howell DcJ, White e, Mythen M, Weinman Ja. intrusive memories of hallucinations and delusions in traumati�ed intensive care patients: an interview study. Br J Health Psychol 2015;20:613-31.

3. samuelson Ka, lundberg D, Fridlund B. stressful mem-ories and psychological distress in adult mechanically ventilated intensive care patients – a 2 month follow up study. acta anaesthesiol scand 2007;51:671-8.

4. Wade DM, Howell Dc, Weinman Ja, Hardy rJ, Mythen Mg, Brewin cr, et al. investigating risk factors for psy-chological morbidity three months after intensive care: a prospective cohort study. crit care 2012;16:r192.

5. Davydow Ds, gifford JM, Desai sV, Needham, DM, Bi-envenu oJ. Posttraumatic stress disorder in general inten-sive care unit survivors: a systematic review. gen Hosp Psychiatry 2008;30:421-34.

6. Wade Ds, Hardy r, Howell D, Mythen M. identifying clinical and acute psychological risk factors for PtsD af-ter critical care: a systematic review. Minerva anestesiol 2013;79:944-63.

7. Davydow Ds, gifford JM, Desai sV, Bienvenu oJ, Needham DM. Depression in general intensive care unit survivors:a systematic review. intens care Med 2009;35:796-809.

8. Davydow Ds, Zat�ick D, Hough cl, Katon WJ. in hos-pital acute stress symptoms are associated with impair-ment in cognition 1 year after intensive care unit admis-sion. ann am thorac soc 2013;10:450-7.

9. Wilcox Me, Brummel Ne, archer K, ely eW, Jackson Jc, Hopkins ro. cognitive dysfunction in icU patients: risk factors, predictors, and rehabilitation interventions. crit care Med 2013;41(9 suppl. 1):s81-98.

10. Brooks r, Kerridge r, Hillman K, Bauman a, Daffurn, K. Quality of life outcomes after intensive care – com-parison with a community group. intens care Med 1997;23:581-6.

11. Papaoiunnou ai, Konstantinos B, tsikrika s, Karakon-taki F, Kastanakis e, Banya W, et al. the impact of de-pressive symptoms on recovery and outcome of hospital-ised coPD exacerbations. eur respir J 2013;41:815-23.

12. Pandharipande P, shintani a, Peterson J, Pun Bt, Wilkin-Pandharipande P, shintani a, Peterson J, Pun Bt, Wilkin-son gr, Dittus rs, et al. lora�epam is an independent risk factor for transitioning to delirium in intensive care unit patients. anesthesiology 2006;104:21-6.

13. Moher D, liberati a, tet�laff J, altman Dg. Preferred re-porting items for systematic reviews and meta-analyses: the PrisMa statement. J clin epidemiol 2009;62:1006-12.

14. saadatmand V, rejeh N, Heravi-Karimooi M, tadrisi sD, Zayeri F, Vaismoradi M, et al. effect of nature-based sounds’ intervention on agitation, anxiety, and stress in patients under mechanical ventilator support:a ran-domised controlled trial. int J Nurs stud 2013;50:895-904.

15. Beaulieu-Boire g, Bourque s, chagnon F, chouinard l, gallo-Payet N, lesur o. Music and biological stress dampening in mechanically ventilated patients at the in-tensive care unit ward-a prospective interventional rand-omi�ed crossover trial. J crit care 2013;28:442-50.

16. su cP, lai Hl, chang et, Yiin lM, Perng sJ, chen PW. a randomi�ed controlled trial of the effects of listening to non-commercial music on quality of nocturnal sleep and relaxation indices in patients in medical intensive care unit. J adv Nurs 2013;69:1377-89.

17. Korhan ea, Khorshid l, Uyar M. the effect of music therapy on physiological signs of anxiety in patients receiving mechanical ventilatory support. J clin Nurs 2011;20:1026-34.

18. Wong Hl, lope�-Nahas V, Molassiotis a. effects of mu-sic therapy on anxiety in ventilator-dependent patients. Heart lung 2001;30:376-87.



40. Kress JP, gehlbach B, lacy M, Pliskin N, Pohlman as, Hall JB. the long-term psychological effects of daily sedative interruption on critically ill patients. am J respir crit care Med 2003;168:1457-61.

41. treggiari M, romand J, Yane� D, Deem s, goldberg J, Hudson l, et al. randomised trial of light versus deep sedation on mental health after critical illness. crit care Med 2009;37:2527-34.

neuroendocrine, and immune responses to psychologi-cal stress: the reactivity hypothesis. ann NY acad sci 1998;840:664-73.

38. Kramlich D. introduction to complementary, alternative and traditional therapies. crit care Nurse 2014;34:50-6.

39. aitken lM, rattray J, Hull a, Kenardy Ja, le Broque r, Ullman aJ. the use of diaries in psychological recovery from intensive care. crit care 2013;17:253

Funding.—Dorothy M. Wade is funded by the National institute for Health research (NiHr) University college london Hospitals Biomedical research Unit.Acknowledgments.—the authors would like to thank Pippa Davie who manually searched studies included from the online search for further references. Dorothy M. Wade would like to thank Dr David Howell and Professor Monty Mythen for their support in carrying out psychological research in the critical care Department at University college Hospital, london.Conflicts of interest.—The authors certify that there is no conflict of interest with any financial organization regarding the material discussed in the manuscript.For supplementary materials, please see the online version of this article.Article first published online: October 27, 2015. - Manuscript accepted: October 22, 2015. - Manuscript revised: October 20, 2015. - Manuscript received: august 14, 2015.


studies are usually necessary. such studies have to overcome center and country variabil-ity in the approval process. in italy, the level of this variability is striking and italian research ethics committee (recs) often take opposite

emergency research in incompetent patients is a common but problematic issue. Due

to the large sample size required, multicenter

P O I N T O F V I E W

How to protect incompetent clinical research subjects involved in critical

care or emergency settingsNereo ZaMPeretti 1 *, Mariassunta PicciNNi 2, rinaldo BelloMo 3,

giuseppe citerio 4, 5, giovanni Mistraletti 6, 7, giuseppe gristiNa 8, alberto giaNNiNi 9

1servizio Qualità, sicurezza ed accreditamento, azienda Ulss n. 6, Vicenza, italy; 2Dipartimento di scienze Politiche, giuridiche e studi internazionali, Università degli studi di Padova, Padua, italy; 3Department of intensive care, austin & repatriation Medical center, Heidelberg, australia; 4Unità di terapia intensiva Neurochirurgica, ospedale san gerardo, Monza, italy; 5Dipartimento di scienze della salute, Università Milano Bicocca, Milan, italy; 6Unità di terapia intensiva, azienda ospedaliera san Paolo – Polo Universitario, Milan, italy; 7Dipartimento di Fisiopatologia Medico-chirurgica e dei trapianti, Università degli studi di Milano, Milan, italy; 8gruppo di studio per la Bioetica - società italiana di anestesia, analgesia rianimazione e terapia intensiva (siaarti), rome, italy; 9terapia intensiva Pediatrica, Fondazione irccs ca’ granda - ospedale Maggiore Policlinico, Milan, italy*corresponding author: Nereo Zamperetti, servizio Qualità, sicurezza ed accreditamento, azienda Ulss n. 6, Dipartimento di anestesia e rianimazione, ospedale san Bortolo, Via Rodolfi, 37, 36100 Vicenza, Italia. e-mail: [email protected]


lavoro: titolo breve: researcH sUBJects ProtectioN iN eMergeNcY sitUatioNsprimo autore: ZaMPerettipagine: 479-85citazione: Minerva Anestesiol 2016;82:479-85

a B s t r a c tclinical research is an essential component of medical activity, and this is also true in intensive care. adequate informa-tion and consent are universally considered necessary for the protection of research subjects. However, in emergency situations, the majority of critical patients are unable to consent and a valid legal representative is often unavailable. the situation is even more complex in italy, where the relevant legislation fails to specify how investigators should manage research in emergency or critical care setting when it involves incompetent patients who do not have an appointed legal representative. While special measures for the protection of incompetent subjects during emergency research are neces-sary, not allowing such research at all dooms critically ill patients to receive non-evidence-based treatments without the prospect of improvement. The recently-issued EU Regulation n. 536/2014 will probably help shed light on this situation. Indeed, it specifically addresses the issue of “research in emergency situations” and introduces detailed rules aimed at protecting patients while allowing research.in this article, we argue that obtaining informed consent during emergency research on incompetent subjects in unreal-istic, and that in most cases substituted judgment on the part of a proxy carries major flaws. Strict criteria in evaluating the risk-benefit ratio of proposed intervention and a careful evaluation of the trial by a local or national Research Ethics committee are perhaps the most practicable solution.(Cite this article as: Zamperetti N, Piccinni M, Bellomo r, citerio g, Mistraletti g, gristina g, et al. How to protect incompetent clinical research subjects involved in critical care or emergency settings. Minerva Anestesiol 2016;82:479-85)Key words: research - informed consent - intensive care units - emergency medicine.

Minerva Anestesiologica 2016 April;82(4):479-85© 2016 EDIZIONI MINERVA MEDICAthe online version of this article is located at http://www.minervamedica.it

comment in p. 389.

ZaMPeretti researcH sUBJects ProtectioN iN eMergeNcY sitUatioNs


dition, as might be the case in cardiac arrest research, there is no time or opportunity for consent to be given.

The specific issues of clinical research on incompetent patients in emergency settings had not been addressed in the eU legal frame-work until April 2014, leading to uncertainty on the legal feasibility of such research 2.

The present situation in Italy

The European Directive 2001/20/EC will still be in force at least until May 2016. In italy, such directive has been implemented by the Legislative Decree n. 211/2003,3 whose art. 3 states that only the involved subject can give valid consent for research. according to art. 5, the “legal representative” is the only legal substitute for the incompetent patient; yet, the Decree fails to specify who is to be intended as legal representative in emergen-cies.

according to italian law, in case of incom-petent adults, the legal representative is a legal guardian specifically appointed by the Court — a process which optimistically takes weeks. How then could be possible to involve incom-petent adults in emergency research when a legal guardian is not available, as indeed hap-pens in most cases?

a possible, extreme solution is to consider every form of research on the incompetent adults forbidden.4

Yet, this construct makes emergency clini-cal research in adult incompetent persons im-possible at all, unless there is a previously ap-pointed legal substitute (as in case of previous mental impairment). this deprives critically ill patients of evidence-based improvements in care and is clearly unethical.

an alternative solution is to consider that the protection measure of legal representation, as it has been developed in the italian legal order, does not cover emergency situations, in the absence of a previously appointed legal substitute.5 as a consequence, art. 5, lett. a) of the Legislative Decree n. 200/2007 should not apply to emergency situations. on the con-trary, according to art. 4 of the same Decree,6

decisions based only on the different interpre-tation of the current law, as we will discuss further on.

this raises important issues in relation to re-search in critical care or emergency settings, when involving temporarily incompetent pa-tients (e.g. comatose patients), or highly dis-tressed patients that cannot rapidly provide a truly informed consent. the aim of this paper is to present the current legal framework in ita-ly, compare it with the recent developments in european Union legislation and discuss such developing regulation.

Information and consent in emergency research

clinical research is an essential component of medical activity, even in emergency situ-ations. effective therapies mean better out-comes and reduced morbidity and mortality; these goals can be achieved only by means of research. Not allowing such studies stops progress in knowledge, condemning critically ill patients to receive inferior non-evidence based treatments, without any prospect of im-provement. critically ill patients should be protected from such undesirable lack of evi-dence.

in order to also protect subjects involved in research studies, adequate information and subsequent valid consent are mandatory both in the international context and in the euro-pean Union (eU) legal framework (see WMa Declaration of Helsinki, 2013, points 25-26 and Directive 2001/20/EC, articles 2.j and 3.2.b-d). informed consent may be granted either by the patient or by the legal represen-tative of the incompetent patient (see WMa Declaration of Helsinki, 2013, points 27-29 and EU Directive 2001/20/EC, articles 3, 4 and 5). these measures of protection work in non-critical settings (e.g. research on psychi-atric patients or on incompetent patients with degenerative conditions). However, in emer-gency situations, the vast majority of patients are incompetent and legal representative are absent:1 they cannot be adequately informed and are consequently unable to consent. in ad-

researcH sUBJects ProtectioN iN eMergeNcY sitUatioNs ZaMPeretti


28 May 2016 (see art. 99 in the Appendix). the recital (36) states:

“This Regulation should provide for clear rules concerning informed consent in emer-gency situations. (…) For such cases, inter-vention within an ongoing clinical trial, which has already been approved, may be pertinent. However, in certain emergency situations, it is not possible to obtain informed consent prior to the intervention. this regulation should therefore set clear rules whereby such patients may be enrolled in the clinical trial under very strict conditions”.

accordingly, the new regulation provides special protection measures at art. 10 and 35 (appendix) and it mostly seems in line with art. 30 of the WMA Declaration of Helsinki, 2013, setting four fundamental principles:

1. good clinical research must be promoted, even on incompetent subjects and in emergen-cy situations, in order to meet health needs and priorities of this group of patients;

2. clinical research in this setting requires special protection measures, including: a) a strict relationship between the clinical trial and the medical condition which causes the patient’s incompetence, b) the necessity to conduct the clinical trial in emergency situ-ations; c) the expectations of benefits and of only minimal risks or burdens for the perspec-tive subjects.

3. informed consent from the subject or from a legally designated representative should be sought as soon as possible (delayed consent).

4. the clinical trial should have been previ-ously reviewed by a rec.

Protecting incompetent patients through informed consent?

two kinds of measures are currently used in order to protect the subjects of clinical re-search.

The first kind of protective measures relate to the acceptability of the risk/benefit ratio of the study design. they may give answers to the need of reasonably balancing the promotion of scientifically sound and clinically relevant re-

the point 4.8.15 of the annex i to the Ministe-rial Decree 15.7.1997, which implemented the gcP cPMP/icH/135/1995 in italy, must be considered applicable:

“When prior consent of the subject is not possible, and the subject’s legally acceptable representative is not available, enrolment of the subject should require measures described in the protocol and/or elsewhere, with docu-mented approval/favorable opinion by the irB/iec, to protect the rights, safety and well-being of the subject and to ensure compliance with applicable regulatory requirements. the subject or the subject’s legally acceptable rep-resentative should be informed about the trial as soon as possible and consent to continue and other consent as appropriate (see 4.8.10) should be requested”.

this means that currently the destiny of a research protocol in emergencies depends on a case by case evaluation of the local rec. the rec has to assess if the special measures provided in the protocol are adequate for the protection of the rights, safety and well-being of the perspective subjects. the aim can be achieved through:

— measures concerning the “acceptability” of the risk/benefit ratio — involving the study design, limitation of inherent risks, necessity of a second expert opinion, stress on the inves-tigator responsibility, use of more strict inclu-sion criteria etc., or

— measures concerning the participation of the perspective subjects — involving the infor-mation and assent of the person even partially competent, the respect of any previously ex-pressed objection by the patient, a proxy con-sent, the delayed consent.

in the absence of other applicable regulatory requirements, this evaluation totally relies on the recs’ discretion.

Things are changing: the EU scenario

On April 2014 the European Parliament and the Council issued EU Regulation n. 536/2014 on clinical trials on medicinal products for hu-man use, and repealing Directive 2001/20/EC. the new regulation shall apply no earlier than



clinical trials may be conducted only where — among other conditions — there are scien-tific grounds for expecting that participation in the clinical trial (1) will produce a direct benefit for the subject concerned outweigh-ing the risks and burdens involved; or (2) […] will pose only minimal risk to, and will im-pose minimal burden on the subject concerned in comparison with the standard treatment of the minor’s condition. on the contrary, in article 35 (clinical trials in emergency situ-ations) the two standards of “direct clinically relevant benefit for the subject” and “minimal risk to, and minimal burden on, the subject in comparison with the standard treatment” are both mandatory for the subject’s inclusion. these two strict prerequisites are designed to protect the incompetent subject of emergency research, when he/she cannot decide (he/she is incompetent) and a guardian is not (yet) avail-able. the role of the rec here is crucial as it has the task to verify that the clinical trial de-sign really respects these requirements.

As a matter of fact, an efficient local REC can provide a much better evaluation of the planned intervention, by ensuring effective protection of research subjects and promoting good clinical research in emergency settings. all rec’s mem-bers should be experienced professionals who should be able to evaluate all the relevant issues related to protocol safety and have sufficient time for adequate discussion. No other people, and surely not the subject’s relatives pressed in an emergency situation, can make a better eval-uation. Yet, such a committee should work in close contact with the research clinicians.

at present, in italy, most recs do good pre-emptive work: they examine the different aspects of the protocol, the information sheets and the insurance issues. But after that, the task of optimal performance and protocol ad-herence is left to the clinicians.

our proposal is to trust clinicians: those who perform clinical research have specific legal and moral responsibilities. But, at the same time, we think that the strengthened criteria ver-ified by the RECs in evaluating the design of an emergency trial involving incompetent subjects should be sufficient for the inclusion (Figure 1).

search and the physical protection of incompe-tent patients involved in clinical research. the second kind of protection measures concern the involvement of the participants and the protec-tion of the identity of the research subjects and their rights to privacy and self-determination.

Both these kinds of measures have to be used, in line also with the eU regulation n. 536/2014.

Yet, in case of emergency research on in-competent subjects, a previously collected in-formed consent is simply unfeasible (by defi-nition, as the subjects are incompetent).

as for deferred consent, it is evident that con-sent can protect a patient only if given before an action is performed. Deferred consent can work only for the subsequent treatments and for the use of personal data. on the contrary, in prospective randomized protocols, after a potentially dangerous intervention (a drug has been administered, an operation has been per-formed), consent can have little space in prac-tice in order to protect the patient. collecting valid consent from next-of-kin is also problem-atic. in italy, though this practice has no full le-gal value,3, 6, 7 researchers often inform the rela-tives and take into account their reporting of the patient’s wishes. this practice is often not feasi-ble in emergencies due to several reasons; first, the time constraints, as a next-of-kin is usually unavailable in the therapeutic window frame-time;8 second, different family members could give different versions of the patient’s wishes and/or may fail to accurately report them;9 third, emotional stress can significantly bias the deci-sion of relatives in emergency situations.2, 10

Protecting incompetent patients through risk/benefit ratio

evaluation: the role of RECs

We believe that measures regarding the ac-ceptability of the risk/benefit ratio of the study design can protect the incompetent subjects of emergency trials much more than information and consent. Regulation EU n. 536/2014 sets suitable rules for this aim. in fact, article 31 (clinical trials on incapacitated subjects) and article 32 (clinical trials on minors) state that



Key messages — clinical research is an essential com-

ponent of medical activity; it has to always be guaranteed, both for individual and so-cial interests, even in emergency condi-tions.

— adequate information and consent are unfeasible in critically-ill patients and a legal representative is often not promptly available.

— special measures for the protection of incompetent patients have to be har-monized with the possibility to conduct research in emergencies, in order to reach evidence based treatments.

— the careful evaluation and approval of the trial by the competent ethics com-mittee appears as the best solution, togeth-er with surveillance during trial progress and the timely consent collection from pa-tient.

The problem of multicenter trials (and multiple RECs)

But again: what to do in multicenter trials, where many centers (and many recs) are in-volved? How can agreement be reached?

in a recent document,11 the italian National committee for Bioethics expressed hopes for a significant legislative change in Italy, which could enable multicenter emergency research on incompetent patients, provided that:

1. the clinical trial has been approved by an Ad-Hoc National rec composed by expert clinicians, lawyers, patients’ representatives, bioethicists;

2. if a patient has not previously refused to be involved in medical research, he/she has to be promptly enrolled;

3. deferred consent is used for subsequent treatments and for permission to use previous-ly gathered data;

4. publication of negative results is recom-mended.

the proposal of the National committee for Bioethics is open to challenge. the main problem is to assemble expert members in dif-ferent disciplines who are able to represent the different working realities (university and big city hospitals, little community hospi-tals in rural areas). Moreover, to evaluate all multicenter studies regarding emergency and involving temporarily unable patients, this “ad-hoc national REC” should have adequate number of planned meetings: a dire challenge, also from an economic point of view. another problem is that such a committee could be per-ceived as too far away and disconnected from researchers.

on the other hand, this solution could en-sure a more balanced equilibrium between the local interests (autonomy of local rec) and the community interests (the possibility to per-form phase iii studies).

We will wait and see how the scenario will evolve, hoping for a clearer set of rules that allow critically ill patients in emergency con-ditions or in intensive care unit to receive the best treatment, based on the best available re-search data.

Figure 1.—Flowchart for management of incompetent pa-tients eligible for emergency clinical trial.



6/11/2007 no. 200, art. 4 Gazzetta Ufficiale n.261 del 9-11-2007 - Supplemento Ordinario n. 228.

7. Point 4.8.15 of the annex i, Decreto del Ministero della salute 15.7.1997, recepimento delle linee guida dell’U-nione europea di buona pratica clinica per la esecuzione delle sperimentazioni cliniche dei medicinali. gazzetta Ufficiale Serie Generale n.191 del 18-8-1997 - Supple-mento ordinario n. 162.

8. Burns Ke, Zubrinich c, tan W, raptis s, Xiong W, smith o, et al. research recruitment practices and critically ill patients a multicenter, cross-sectional study (the consent Study). Am J Respir Crit Care Med 2013;187:1212-8.

9. ciroldi M, cariou a, adrie c, annane D, castelain V, cohen Y, et al. Famirea study group. ability of family members to predict patient’s consent to critical care re-search. Intensive Care Med 2007;33:807-13.

10. Azoulay E, Pochard F, Chevret S, Adrie C, Annane D, Bleichner g, et al. Half the family members of intensive care unit patients do not want to share in the decision-making process: a study in 78 French intensive care units. Crit Care Med 2004;32:1832-8.

11. Comitato Nazionale per la Bioetica (28 September 2012). la sperimentazione clinica in pazienti adulti o minori che non sono in grado di dare il consenso informato in situa-zioni di urgenza; 2012 [Internet]. available from: http://www.governo.it/bioetica/pdf/5La%20Sperimentazi-one%20clinica%20in%20pazienti%20adulti%20o%20minori%20che%20non%20sono%20in%20grado%20di%20dare%20il%20consenso%20informato%20in%20situazioni%20di%20urgenza.pdf [cited 2015, Feb 17].

References

1. Maas ai, Murray g, Henney H 3rd, Kassem N, legrand V, Mangelus M, et al. Efficacy and safety of dexanabinol in severe traumatic brain injury: results of a phase iii ran-domised, placebo-controlled, clinical trial. lancet Neurol 2006;5:38-45.

2. Kompanje eJo, Maas air, Menon DK, Kesecioglu J. Medical research in emergency research in the european Union member states: tensions between theory and prac-tice. Intensive Care Med 2014;40:496-503.

3. Attuazione della direttiva 2001/20/CE rela-tiva all’applicazione della buona pratica clinica nell’esecuzione delle sperimentazioni cliniche di medici-nali per uso clinico. Decreto legislativo 24/06/2003, no. 211, art. 3 Gazzetta Ufficiale n.184 del 9-8-2003 - Sup-plemento Ordinario n. 130.

4. azzoni g. la disciplina giuridica delle sperimentazioni cliniche su adulti incapaci di prestare il loro consenso in-formato. Anestesia Forum 2009;2:97-104.

5. Palermo Fabris E. La sperimentazione clinica: profili giu-ridici. in: lenti l, Palermo Fabris e, Zatti P, editors. trat-trat-tato di Biodiritto – i diritti in medicina. Milna: giuffrè Editore; 2011. p 682-6.

6. Attuazione della direttiva 2005/28/CE recante principi e linee guida dettagliate per la buona pratica clinica rela-tiva ai medicinali in fase di sperimentazione a uso umano, nonché requisiti per l’autorizzazione alla fabbricazione o importazione di tali medicinali. Decreto legislativo

Conflicts of interest.—The authors certify that there is no conflict of interest with any financial organization regarding the material discussed in the manuscript.Article first published online: July 8, 2015. - Manuscript accepted: July 6, 2015. - Manuscript revised: July 1, 2015. - Manuscript received: July 29, 2014.



medical condition because of which it is not pos-sible within the therapeutic window to obtain prior informed consent from the subject or from his or her legally designated representative and to supply prior information, and the clinical trial is of such a nature that it may be conducted ex-clusively in emergency situations;

(f) the clinical trial poses a minimal risk to, and im-poses a minimal burden on, the subject in com-parison with the standard treatment of the sub-ject’s condition.

3. Following an intervention pursuant to paragraph 1, informed consent in accordance with article 29 shall be sought to continue the participation of the subject in the clinical trial, and information on the clinical trial shall be given, in accordance with the following requirements: (a) regarding incapacitated subjects and minors, the informed consent shall be sought by the investigator from his or her legally designated representative without undue delay and the infor-mation referred to in article 29(2) shall be given as soon as possible to the subject and to his or her le-gally designated representative; (b) regarding other subjects, the informed consent shall be sought by the investigator without undue delay from the sub-ject or his or her legally designated representative, whichever is sooner and the information referred to in article 29(2) shall be given as soon as possible to the subject or his or her legally designated repre-sentative, whichever is sooner. For the purposes of point (b), where informed consent has been obtained from the legally designated representative, informed consent to continue the participation in the clinical trial shall be obtained from the subject as soon as he or she is capable of giving informed consent.

4. if the subject or, where applicable, his or her legally designated representative does not give consent, he or she shall be informed of the right to object to the use of data obtained from the clinical trial.

article 99 — entry into forcethis regulation shall enter into force on the twenti-

eth day following that of its publication in the Official Journal of the European Union. it shall apply as from six months after the publication of the notice referred to in article 82(3), but in any event no earlier than 28 May 2016.

EU Regulation n. 536/2014 on clinical trials on me-dicinal products for human use, repealing Directive 2001/20/EC

Article 10 — Specific considerations for vulnerable populations1. (…)2. Where the subjects are incapacitated subjects, specif-

ic consideration shall be given to the assessment of the application for authorisation of a clinical trial on the basis of expertise in the relevant disease and the patient population concerned or after taking advice on clinical, ethical and psychosocial questions in the field of the relevant disease and the patient popula-tion concerned.

article 35 — clinical trials in emergency situations1. By way of derogation from points (b) and (c) of ar-

ticle 28(1), from points (a) and (b) of article 31(1) and from points (a) and (b) of article 32(1), in-formed consent to participate in a clinical trial may be obtained, and information on the clinical trial may be given, after the decision to include the subject in the clinical trial, provided that this decision is taken at the time of the first intervention on the subject, in accordance with the protocol for that clinical trial” and that all of the following conditions are fulfilled:(a) due to the urgency of the situation, caused by a

sudden life-threatening or other sudden serious medical condition, the subject is unable to pro-vide prior informed consent and to receive prior information on the clinical trial;

(b) there are scientific grounds to expect that par-ticipation of the subject in the clinical trial will have the potential to produce a direct clinically relevant benefit for the subject resulting in a measurable health-related improvement alleviat-ing the suffering and/or improving the health of the subject, or in the diagnosis of its condition;

(c) it is not possible within the therapeutic window to supply all prior information to and obtain prior informed consent from his or her legally desig-nated representative;

(d) the investigator certifies that he or she is not aware of any objections to participate in the clin-ical trial previously expressed by the subject;

(e) the clinical trial relates directly to the subject’s

Appendix


“primary injury”) is generally considered irre-versible. the current therapeutic arsenal in the acute phase include, amongst all, admission to a specialized intensive care unit (icU), surgi-cal removal of space-occupying lesions, opti-mization of cerebral oxygenation and perfu-sion, control of increased intracranial pressure and prevention and treatment of other compli-cations (e.g. hyperthermia, seizures, sodium balance disorders, dysglycemia). guidelines have been published in order to standardize

traumatic brain injury (tBi) is a leading cause of death and disability among young

persons worldwide and the incidence is rising in the elderly population. the rehabilitation of tBi survivors usually takes years and many of them are left with permanent disability. the financial burden of TBI management to the so-ciety remains significant.1

the initial therapy of severe tBi patients is mainly aimed at the prevention of secondary brain injuries; the initial insult to the brain (or

J O U R N A L C L U B C R I T I Q U E

another failed attempt of neuroprotection: progesterone for moderate and severe traumatic brain injury

geert MeYFroiDt 1, Fabio s. taccone 2

1Department of intensive care Medicine, University Hospitals leuven, leuven, Belgium; 2Department of intensive care, Hopital erasme, Université libre de Bruxelles (UlB), Brussels, Belgium*corresponding author: geert Meyfroidt, Department of intensive care Medicine, University Hospitals leuven, Herestraat 49, 3000 leuven, Belgium. e-mail: [email protected]


lavoro: titolo breve: Progesterone For MoDerate anD severe tBiprimo autore: MeYFroiDtpagine: 486-491citazione: Minerva anestesiol 2016;82:486-491

a B s t r a c ttwo large phase-iii prospective, multicenter, controlled, double-blind, randomized clinical trials (the Protect iii study; the sYnaPse study) evaluated the effectiveness of an early administration of progesterone in patients with mod-erate to severe traumatic brain injury (tBi). in the Protect iii trial, patients were included if the admission glasgow coma scale (gcs) was within 4-12, whereas the sYnaPse trial only included patients with gcs 4-8. the total dose of progesterone was nearly similar in both studies and drug administration was initiated early after injury (within 4 hours for a total of 96 hours in Protect; within 8 hours for 120 hours in sYnaPse). in the Protect trial, primary outcome was 6-month favourable neurological outcome (defined using the Glasgow Outcome Scale), while in the SYNAPSE Trial it was the 6-month glasgow outcome scale (gos). secondary outcomes, in both studies, included 6-month mortality. the Protect trial was interrupted for futility after the second interim analysis (882 patients randomized out of the 1140 initially planned); the sYnaPse trial included 1195 patients. in Protect, the proportion of patients with favour-able outcome was similar between groups (51% for progesterone vs. 56% for placebo; rr 3.03 [95% ci 1.96-4.66]); in sYnaPse, no difference in gos between the progesterone and placebo group was found (or 0.96 [95% ci 0.77-1.18]). there was no difference in 6-month mortality or any of the other secondary outcomes between groups in the two trials. These studies demonstrated that early progesterone administration did not provide any benefit on the neurological recovery of tBi patients.(Cite this article as: Meyfroidt g, taccone Fs. another failed attempt of neuroprotection: progesterone for moderate and severe traumatic brain injury. Minerva anestesiol 2016;82:486-491)Key words: Brain injuries - Progesterone - neuroprotection - treatment outcome.


Progesterone For MoDerate anD severe tBi MeYFroiDt


aquaporin 4 (which is a water channel playing an important role in cerebral water homeosta-sis). Progesterone also promotes neuronal sur-vival, by down-regulating several pro-apoptot-ic factors, while up-regulating anti-apoptotic factors and brain-derived neurotrophic factors. at the same time, progesterone acts on the inflammatory pathway by attenuating the ex-pression of pro-inflammatory cytokines (such as IL-1β, TNF-α, NF-κB, and IL-6), and by a modulatory effect on the complement system. In addition, other beneficial effects have been attributed to progesterone, such as the reduc-tion of oxidative stress and the attenuation of glutamate-mediated excito-toxicity (Figure 1).5, 9 two small phase ii trials 10, 11 showed that early administration of progesterone af-ter severe tBi was safe. in addition, patients treated with progesterone, had a lower mortal-ity at 30-days 10 or 6-months 11, and modestly improved functional neurological outcomes were observed in both trials.

guided by the results of these animal and clinical trials, the Protect (Progesterone for the treatment of traumatic Brain injury) iii 12 and the sYnaPse (study of the neuro-protective activity of Progesterone in severe traumatic Brain injuries) 13 studies have been set up and conducted in parallel to evaluate the

tBi management.2 as such, the neurological outcome and overall survival of tBi patients have gradually improved in the last decades, especially when aggressive therapy was pro-vided.3

importantly, in the hours, days and weeks after the initial injury, a complex cascade of molecular and cellular events is progressively initiated and intensified, which can eventually lead to additional tissue damage and cell death. none of the available supportive therapeutic interventions is able to specifically attenuate or inhibit these pathological pathways; however, the fact that tBi is not a single event occurring at the time of injury, but a continuous process, offers a window of opportunity for additional targeted therapies. Unfortunately, no pharma-cological intervention has been shown to posi-tively affect the neurological recovery of tBi patients;4 moreover, more than thirty clinical trials using different potential neuroprotective drugs have failed in this setting.5

Progesterone is an endogenous steroid hor-mone that is involved in the menstrual cycle, pregnancy and embryogenesis. However, it is not only synthesized by the ovaries and pla-centa, but also within the central nervous sys-tem, in particular by neurons and oligo-den-drocytes. Progesterone receptors are expressed in the fetal brain during gestation, where they are involved in normal brain development and myelination, but also in adults, where they have a potential role in neuro-regeneration.6 the fortuitous observation that pseudo-preg-nant female rats had improved functional out-comes and decreased edema after experimen-tal tBi 7 was the first hint towards a possible role for neurosteroids in brain protection. in the following years, several hundreds of ani-mal studies have reported on the neuroprotec-tive effects of progesterone in experimental tBi, stroke and spinal cord injury.8 Proges-terone has been shown to reduce brain edema, possibly by influencing the permeability of the blood-brain-barrier, through the up-regulation of the P-glycoprotein pump (which is an atP-dependent transport protein responsible for the very poor permeability of endothelial layers to many drugs) and decreasing the expression of

Figure 1.—Potential mechanisms of neuroprotection medi-ated by progesterone: improved integrity of the blood-brain barrier (BBB); reduced apoptosis and oxydative stress; mod-ulation of leucocytes activation and diapedesis; attenuation of neuro-inflammation; modulation of glutamate neurotrans-mission and excitotoxicity.

MeYFroiDt Progesterone For MoDerate anD severe tBi


errors encountered with other neuroprotectant agents, such as an insufficient statistical pow-er, the use of insensitive outcome measures or the inclusion of heterogeneous patients. all pre-clinical and phase ii human data ap-peared to be very promising, and were used to design both trials. efforts were made to avoid treatment variability across sites: in the Pro-tect iii trial, patients were monitored in real time and immediate feedback was provided in case of non-compliance with a standardized management protocol. robust outcomes were chosen and assessed early as well as after 6 months, in order to detect any possible long-term benefit. In view of the tremendous efforts and financial costs involved in these large tri-als, this failure is a huge disappointment.

When evaluating these studies, some par-ticular issues may be considered to explain the reasons for their negative findings. First, in both studies, TBI was classified according to the glasgow coma scale (gcs) on admission, a clinical assessment of the initial presentation,

effects of progesterone therapy on neurologi-cal outcome of tBi patients.

Discussion on the studies

the Protect iii trial 12 was premature-ly stopped for futility after two thirds of all planned patients were included. in the sYn-aPse trial 13, 1195 patients with severe tBi were finally included. Despite more than 2000 patients enrolled in these two trials, no benefit of progesterone over placebo with regards to neurological recovery or mortality was ob-served. The results of all five published ran-domized trials on progesterone therapy for tBi patients are summarized in table 1. 10-14

Unfortunately, the search for a potential neuroprotective drug for tBi has been once again unsuccessful. these two randomized controlled trials have been extremely well con-ducted and designed, according to all current methodological standards. Moreover, particu-lar attention had been paid to avoid the main

Table I.—�Summary of the randomized clinical trials evaluating the effects of progesterone on neurological outcome after traumatic brain. injury.

Protect 9 Xiao 10 shakeri 13 Protect iii 11 sYnaPse 12

settingPhase ii –

single centerUs

Phase ii – single center

china

Phase ii – single center

iran

Phase iii – multicenter

Us

Phase iii – multicenter

Multinational

number of patients (Pg/placebo)

100(77/23)

159(82/77)

76(38/38)

882(442/440)

1195(591/588)

tBi Moderate / severe

(gcs 4-12)

severe(GCS≤8)

severe(GCS≤8)

Moderate / severe(gcs 4-12)

severe(GCS≤8)

Primary outcome gos at 1 month gos at 3 months gos at 3 months gos-e at 6 months gos at 6 monthsMediane age (yrs) 36 31 34 35 35Median gcs on

admissionnr 6 6 nr 6

Favourable neurological outcome (Pg vs. placebo)

all patients – 25% vs. 16%

Moderate tBi– 21% vs. 27%severe tBi– 55% vs. 0%

3-month: 46% vs. 30%6-month: 57% vs. 42%

50% vs. 28% * all patients – 48% vs. 52%

Moderate tBi – 27% vs. 36%

Moderate to severe tBi

– 51% vs. 56%severe tBi

– 57% vs. 70%

50% vs. 50%

Mortality (Pg vs. placebo)

13% vs. 30% * 3-months: 18% vs. 32%* 31% vs. 45% 19% vs. 16% 22% vs. 22% **

*P<0.05; **mortality was not reported separately in the sYnaPse trial. Percentage refers to the proportion of patients who died or were severely disabled at 6 months (gos 4 and 5 combined).Pg: progesterone; tBi: traumatic brain injury; gos: glasgow outcome scale; gos-e: extended glasgow outcome scale; nr: not reported.



of thromboembolic events was significantly increased in patients treated with progester-one when compared to others (17% vs. 6%). a detailed investigation into the 33% of subjects who died from non-neurological causes, could lead to a better understating of unexpected side effects. Fifth, it is possible that the preliminary clinical trials on progesterone might not have been robust enough to justify these large stud-ies. the two phase ii trials were small and the observed benefit was only modest, observed at one month and not relevant with regard to functional neurological outcome, which was selected as the primary outcome in both Pro-tect iii and sYnaPse trials. this might have led to an over-optimistic estimation of the possible benefits for progesterone, which, in combination with a better than expected outcome in the placebo group, could have con-tributed to the failure of both phase iii trials. Finally, the poor understanding of the patho-physiology of TBI might explain the difficulty to translate laboratory findings to clinical prac-tice. the use of data from large international networks may understand to better describe the mechanisms of secondary brain injury in this setting and to a better standardization of patients’ care.16

in clinical practice, progesterone therapy cannot be recommended in moderate to se-vere tBi patients. However, the role of sex hormones on neuroprotection has not been completely elucidated and several unresolved issues deserve further clarification. There is an incomplete understanding of the inflam-matory and neuro-endocrine cascade leading to the increased serum and cerebral proges-terone levels that have been associated with poor outcome in patients with severe tBi.17 other stress hormone abnormalities, such as increased cortisol concentrations, accompany critical illness; their role in the amplification of additional neurosteroid production should be further investigated. other factors, such as changes in hormone transport and metabo-lism, and de novo synthesis of neurosteroids within the brain should be taken into account when investigating the role of sex steroids in neuroprotection for acute brain injury. More-

in order to estimate the severity of the underly-ing injury. However, tBi encompasses a wide and heterogeneous range of different types of injury that can all lead to the same non-spe-cific motor or pupillary symptoms, each with different mechanistic causes (e.g. subdural or epidural hematomas, contusions, edema, dif-fuse axonal injury, ischemia or inflammation). A “one-treatment-fits-all” approach is prob-ably not appropriate for these different types of brain damage. other parameters, including the presence of edema, of raised intracranial hypertension, of space-occupying lesions, may be considered in the future to further character-ize the subgroup of patients who may benefit the most from therapy. such a more individu-alized classification of TBI (based on biomark-ers or clinical/imaging findings) may reduce the heterogeneity of the studied population, al-though the price to pay may be a more limited generalizability of the final results. Second, other confounders, such as age and gender, should also be considered in the recruitment process. in the Protect study, there was no age limit to include patients (although most of them were around 35 years of age). it is dis-putable whether neuroprotective strategies are as effective in older tBi victims, because of their reduced capability of neuro-regeneration, or pre-existing decline in cognitive functions. Because female patients are still capable of producing valuable endogenous progesterone, it could have been useful to assess the initial levels of circulating neurosteroids hormones, in order to avoid the administration of supple-mental hormones to subjects with adequate levels. third, the outcome measures that are currently used, although robust and well vali-dated, might not be sensitive enough to detect all potential benefits. Although advanced neu-rocognitive testing is extremely time and ef-fort consuming, other large trials, for instance in critically ill children,15 have demonstrated that this is feasible, even at such a large scale, and that more subtle neurological benefits that matter to our patients can be picked up this way. Fourth, the administered therapy may have been associated with significant adverse events. in the Protect study, the occurrence

MeYFroiDt Progesterone For MoDerate anD severe tBi


Key messages

— early administration of progesterone offers no outcome benefit in moderate or severe tBi.

— an increased risk for thromboem-bolic events has been observed in patients treated with progesterone when compared to control patients in one study.

— a more coordinated and extensive early phase research might improve future trial design in tBi.

References

1. Hydera aa, Wunderlich ca, Puvanachandra P, guru-raj g, Kobusingye oc. the impact of traumatic brain injuries: a global perspective. neurorehabilitation 2007;22:341-53.

2. Brain trauma Foundation, american association of neu-rological surgeons (aans), congress of neurological surgeons (cns), aans/cns Joint section on neuro-trauma and critical care. guidelines for the manage-ment of severe traumatic brain injury. vi. indications for intracranial pressure monitoring. J neurotrauma 2007;24(suppl 1):s37-s44.

3. stein sc, georgoff P, Meghan s, Mizra K, sonnad ss. 150 years of treating severe traumatic brain injury: a sys-tematic review of progress in mortality. J neurotrauma 2010;27:1343-53.

4. stocchetti n, taccone Fs, citerio g, Pepe Pe, le roux PD, oddo M, Polderman KH, et al. neuroprotection in acute brain injury: an up-to-date review. crit care 2015;19:186.

5. loane DJ, Faden ai. neuroprotection for traumatic brain injury: translational challenges and emerging therapeutic strategies. trends Pharmacol sci 2010;31:596-604.

6. schumacher M, robel P, Baulieu ee. Development and regeneration of the nervous system: a role for neuroster-oids. Dev neurosci 1996;18:6-21.

7. attella MJ, nattinville a, stein Dg. Hormonal state af-fects recovery from frontal cortex lesions in adult female rats. Behav neural Biol 1987;48:352-67.

8. Wei J, Xiao g. the neuroprotective effects of progester-one on traumatic brain injury: current status and future prospects. acta Pharmacologica sinica 2013;34:1485-90.

9. Deutsch er, espinoza tr, atif F, Woodall e, Kaylor J, Wright DW. Progesterone’s role in neuroprotection, a re-view of the evidence. Brain res 2013;1530:82-105.

10. Wright DW, Kellermann al, Hertzberg vs, clark Pl, Frankel M, goldstein Fc, et al. Protect: a randomized clinical trial of progesterone for acute traumatic brain in-jury. ann amerg Med 2007;49:391-402.

11. Xiao g, Wei J, Yan W, Wang W, lu Z. improved out-comes from the administration of progesterone for pa-tients with acute severe traumatic brain injury: a rand-omized controlled trial. crit care 2008;12:r61.

12. Wright DW, Yeatts sD, silbergleit r, Palesh YY, Hertz-berg vs, Frankel M, et al. very early administration of progesterone for acute traumatic brain injury. n engl J Med 2014; 371:2457-66.

13. skolnick Be, Maas ai, narayan rK, gerritsen van der

over, some of the most important neuropro-tective properties of progesterone seem to be related to one of its metabolites, allopreg-nanolone.18 While progesterone effects are mediated by specific nuclear receptors, allo-pregnanolone acts as a potent positive modu-lator of γ-amino butyric acid type (GABA)-A receptors and of the mitochondrial permeabil-ity transition pore,19 suggesting that this com-pound could be a potential neuroprotectant for tBi. Finally, experimental studies in tBi have demonstrated that combining progester-one with magnesium or vitamin D could po-tentially be more effective than progesterone alone.20, 21

When exploring these potentially prom-ising research paths, it is important to learn from the experiences from Protect iii and sYnaPse trials. in the future, a more system-atic approach and evaluation of the preclinical data and phase ii clinical trials is necessary. initial studies should focus on identifying appropriate subgroups of patients, classified according to the pathophysiology of the in-jury, rather than the presentation; on defining relevant outcomes, as well as biomarkers to guide the clinical management; to optimize dosing and timing of potential neuroprotec-tant drugs. results from these studies should be pooled, critically evaluated and reported in an open and rigorous way. Findings should be replicated in more than one, and preferably several, phase ii trials. this more extensive and strategic early phase research will prob-ably lead to less compounds that will make it to phase II at a final stage. In addition, as small studies are only hypothesis-generating, clinical management of tBi patients should be validated by larger studies or at least meta-analysis including an adequate number of pa-tients, which will significantly reduce the risk of bias.22 Finally, innovative adaptive trial design might allow to better identifying those patients in which the drug has a higher prob-ability of benefit. These new approaches will hopefully result in the development of neu-roprotectants that are more specific but also more efficient for specific subgroups of TBI patients.



18. schumacher M, Mattern c, ghoumari a, oudinet JP, liere P, labombarda F, et al. revisiting the roles of pro-gesterone and allopregnanolone in the nervous system: resurgence of the progesterone receptors. Prog neurobiol 2014;113:6-39.

19. cooke Ps, nanjappa MK, Yang Z, Wang KK. therapeu-tic effects of progesterone and its metabolites in trau-matic brain injury may involve non-classical signaling mechanisms. Front neurosci 2013;7:108.

20. Uysal n, Baykara B, Kiray M, cetin F, aksu i, Dayi a, et al. combined treatment with progesterone and magne-sium sulfate positively affects traumatic brain injury in immature rats. turk neurosurg 2013;23:129-37.

21. aminmansour B, nikbakht H, ghorbani a, rezvani M, rahmani P, torkashvand M, et al. comparison of the administration of progesterone versus progesterone and vitamin D in improvement of outcomes in patients with traumatic brain injury: a randomized clinical trial with placebo group. adv Biomed res 2012;1:58.

22. Ioannidis JP. Why most published research findings are false. Plos Med 2005;2:e124-e124.

Hoop r, Macallister t, Ward JD, et al. a clinical trial of progesterone for severe traumatic brain injury. n engl J Med 2014;371:2467-76.

14. shakeri M, Boustani Mr, Pak n, Panahi F, salehpour F, Lotfinia I, et al. effect of progesterone administration on prognosis of patients with diffuse axonal injury due to severe head trauma. clin neurol neurosurg 2013;115:2019-22.

15. Mesotten D, gielen M, sterken c, claessens K, Hermans g, vlasselaers D, et al. neurocognitive development of children 4 years after critical illness and treatment with tight glucose control: a randomized controlled trial. JaMa 2012;308:1641-50.

16. tosetti P, Hicks rr, theriault e, Phillips a, Koroshetz W, Draghia-akli r. toward an international initia-tive for traumatic brain injury research. J neurotrauma 2013;30:1211-22.

17. santarsieri M, niyonkuru c, Mccullough eH, Dobos Ja, Dixon ce, Berga sl, et al. cerebrospinal Fluid cortisol and Progesterone Profiles and Outcomes Prognostica-tion after severe traumatic Brain injury. J neurotrauma 2014;31:699-712.

Conflicts of interest.—The authors certify that there is no conflict of interest with any financial organization regarding the material discussed in the manuscript.Article first published online: July 1, 2015. - Manuscript accepted: June 29, 2015. - Manuscript revised: June 19, 2015. - Manuscript received: May 5, 2015.


© 2015 eDiZioni Minerva MeDicathe online version of this article is located at http://www.minervamedica.itMinerva anestesiologica 2016 april;82 (4):492-3

spinal epidural abscess: stay focused,stay tuned! a clinical report with negative neurological outcome from the “italian registry of complications associated with regional anesthesia - ricalor”

Dear editor,

spinal-epidural abscess (sea) may occur with tho-racic epidural analgesia (tea). We inserted a 20-gauge epidural catheter at t7-t8 in a 57-years-old patient

(Whipple procedure), in aseptic conditions (surgical cap, face mask and sterile gloves; skin disinfection with 10% povidone-iodine and sterile draping). Her past medical history included hypertension (asa ii). a single-use elastomeric pump was prepared under sterile conditions and replaced on postoperative day (PoD) ii. On POD V, patients presented systemic inflammation characterized by pyrexia and leukocytosis with nega-tive blood cultures, attenuating (but not resolving) up to PoD viii; epidural was removed (documenting an erythematous swelling area at the entry point). on PoD viii mid-abdominal pain and distension with perihepat-ic fluid collections (negative for any bacterial growth) were reported and empirically treated with teicoplanin (400 mg/daily), with partial resolution of the inflam-matory status. on PoD Xii mid-back pain joined with progressive bilateral sensory loss (lower limbs, up to T4 dermatome) and motor deficit, leading to lower limb paralysis in a few hours; the Mri showed marked nar-rowing of the spinal canal and abnormal signal from the posterior epidural space (t6-t8). an emergency lami-nectomy of t6, t7, t8, and a partial laminectomy at t5

L E T T E R T O T H E E D I T O R


lavoro: titolo breve: primo autore: letter to tHe eDitorpagine: 492-3citazione: Minerva anestesiol 2016;82:492-3

Figure 1.—Mri cuts documenting the epidural abscess with enhanced signal; a) sagittal plane; B) sagittal view — particular at thoracic level with abscess; C) transverse view; D) sagittal plane — evidence of cavitations and air-fluid levels within the epidural abscess.

a D

B

c

letter to tHe eDitor


References

1. Manassero a, Ugues s, Bossolasco M, Meineri M, colet-ta g. Postoperative thoracic epidural analgesia; a prospec-tive audit of 1,515 patients. raPM sup 2014;39:e174.

2. allegri M, Bugada D, grossi P, Manassero a, Pinciroli rl, Zadra n, et al. italian registry of complications associated with regional anesthesia (ricalor). an incidence analysis from a prospective clinical survey. Minerva anestesiol 2016;82:392-402.

3. Horlocker tt, Wedel DJ. infectious complications of regional anesthesia. Best Pract res clin anaesthesiol 2008;22:451-75.

4. Krishnamahan P. spinal epidural abscess. curr infect Dis rep 2014;16:436-41.

5. royakkers aa, Willigers H, van der ven aJ, Wilmink J, Durieux M, van Kleef M. catheter-related epidural ab-scess. Don’t wait for neurological deficits. Acta Anaesth scand 2002;46:611-5.

alberto Manassero 1*, susanna UgUes 1, giuseppe coletta 1, Dario BUgaDa 2,3

1anesthesia, intensive care and Pain therapy Unit, s. croce e carle Hospital, cuneo,

italy; 2Department of surgical sciences, Parma University, Parma, italy; 3siMPar (study in

Multidisciplinary Pain research) group*corresponding author: alberto Manassero, anesthesia, inten-sive care and Pain therapy Unit, s. croce e carle Hospital, via Michele coppino 26, cuneo, italy. e-mail: [email protected] first published online: November 18, 2015. - Manuscript accepted: october 20, 2015. - Manuscript received: July 24, 2015.(Cite this article as: Manassero a, Ugues s, coletta g, Bugada D. stay tuned! a clinical report with negative neurological outcome from the “italian registry of complications associated with regional anesthesia - ricalor”. Minerva anestesiol 2016;82:492-3)

© 2016 eDiZioni Minerva MeDicathe online version of this article is located at http://www.minervamedica.itMinerva anestesiologica 2016 april:82(4):493-5

impact of axotracktM for ultrasound-guided central venous catheter insertion: a randomized controlled study conducted on inanimate manikin

Dear editor,

current guidelines recommend ultrasound guid-ance for central venous catheter insertion (Us-guided cvc insertion).1 However, all physicians do not use this method. An explanation could be the difficulty en-countered with an “in-plane” approach to follow needle progression because it has to be accurately placed in the

and t9 was performed (Figure 1). Methicillin-sensitive Staphylococcus Aureus was isolated, highly sensitive to teicoplanin, which was increased to 600 mg/daily to reach the therapeutic interval, and continued up to POD 30. Sensory deficit recovered within one month; a sphincters’ dysfunction persisted (urinary retention re-quiring intermittent self-catheterization) up to one year. Motor impairment was persisting at three years (patient unable to walk without crutches).

We are currently running a prospective audit on epidural in our institution,1 and this was the only sea on more than 1800 patients, as well as the only sea (on 4954 epidurals) documented in the “ricalor” study;2 otherwise, our experience shows that despite sea is infrequent, long-term negative outcomes are not. inoculation of bacteria along the needle track seems to be the most likely mechanism for epidural contamination.3 strict aseptic technique is a major is-sue both during catheter’s placement (hat, long sleeves surgical gowns, gloves and face-mask, large sterile drape, skin disinfection with chlorexidine), and man-agement (sterile infusions, closed delivery systems, minimal bag changes, filters); catheter’s insertion site should frequently be inspected and maintained sterile, and catheter should be removed after 96 hours in any case (prolonged catheterization is also a risk for sea), when exposed (due to dressing’s dislodgment) or if any sign of local infection occurs (reported in 70% of pa-tients with sea). likewise, infection may also origin by bacterial spread from an infectious focus through the blood stream.4

as most case reports do, we did not retrieve any ob-vious cause to epidural infection: since blood cultures and abdominal drainage were all negative, hematic spread from a distant focus is unlikely; we presume that the epidural space was directly contaminated by local bacteria. the strict aseptic procedure was followed during the procedure, but the prolonged catheteriza-tion (5 days) and pump’s replacement may account for infection. otherwise, the main issue was the delayed diagnosis; sea presentation can be insidious in na-ture: early signs can be unspecific,4 and misdiagnosis is common. in our case, delayed back pain, as well as the association of abdominal pain and abdominal fluid collections blamed abdominal complications for cre-ating systemic inflammation. Unfortunately, mild but persisting signs of systemic and skin inflammation did not raise the suspicion of sea, and nor dressing swabs nor catheter tip were cultured; furthermore, an antibi-otic therapy was started but resulted ineffective due to low serum concentrations, arguably allowing the pro-gression of the underlying abscess. This case confirms the importance of strict surveillance on patients devel-oping fever, as well as local and/or laboratory signs of infection5, without waiting for neurological deficits to occurr.5 early diagnosis and proper management are the mainstays for a positive outcome, promoting strict adherence to guidelines (asepsis, surveillance) since sea remains infrequent, but rapidly evolving and po-tentially disastrous.



tensive care medicine (22) - were included in the study. twenty two declared previous experience in Us-guided cvc insertion. any participant had previously used the tested device. the manikin (Blue Phantom ii, cae Healthcare st. louis, Mo) mimics right internal jugular and subclavian veins, and carotid and subclavian arter-ies. a manual pump produces arterial pulse. aspiration of blue fluid confirms venous puncture, whereas aspira-tion of red fluid rules in arterial puncture.

We used the M-turbo® device (sonosite, Bothewell, Ma, Usa) equipped with either a regular 7.5 MHz lin-ear probe (control approach) or a specific probe (experi-mental approach) designed to adapt the axotrack tM de-vice (soma access systems, greenville, sc, Usa). in clinical practice, using ultrasound guidance, the infra-clavicular approach to central venous catheterization is achieved by puncturing the axillary vein, whereas the supraclavicular approach may be obtained through in-ternal jugular or subclavian vein puncture. in the present study, the order in which punctures were performed on

field explored. This can be obtained by positioning the needle in the exact middle of the probe. if not achieved, no or only part of the needle is visualized and complica-tions occurred at a rate similar to what is observed with the landmark approach.2 Moreover, success rate at first attempt reported in studies is not 100% which is puz-zling if one assumes that the procedure is performed un-der visual control.3, 4 an observational study suggested usefulness of a device aimed at facilitate needle path vi-sualization: axotrack tM.5 the device is an ultrasound probe incorporating a needle guidance system that per-mits a real-time information on both direction and depth of the needle continuously during the whole procedure (Figure 1). We planned the present study to assess the impact of this device for Us-guided cvc insertion in a randomized placebo-controlled study conducted on inanimate manikin.

after signing an informed consent form, 37 physi-cians - median age 37 [28, 45], graduated (29) or being in their graduating process (8), in emergency (15) or in-

Figure 1.—Picture of the axotrack (a) and its ultrasound images (B).

a

B

Table I.—�Outcome measures in the classical approach versus experimental approach according to the puncture site.time between skin contact

and first puncturetime between

first puncture and success needle passes

internal jugular classic approach 9.5 [3-37] 8 [2-127] 1 [1-4]experimental approach 12.5 [2-81] a 8 [3-103] 1 [1-3]

axillary vein classic approach 12 [2-75] 21 [2-120] 1 [1-4]experimental approach 16 [3-60] 7.5 [2-37] b 1 [1-3] c

all values are expressed as median (range). times are given as seconds and needle passes as number of occurrences.aP=0.008 compared to classic approach; bP=0.0001 compared to classic approach; cP=0.02 compared to classic approach.



Fagon JY. improvement of internal jugular vein cannu-lation using an ultrasound-guided technique. intensive care Med 1997;23:916-9.

4. Milling tJ, Jr., rose J, Briggs WM, Birkhahn r, gaeta tJ, Bove JJ, et al. randomized, controlled clinical trial of point-of-care limited ultrasonography assistance of cen-tral venous cannulation: the third sonography outcomes assessment Program (soaP-3) trial. critical care Med 2005;33:1764-9.

5. Ferre rM, Mercier M. novel ultrasound guidance system for real-time central venous cannulation: safety and ef-ficacy. Western J Emerg Med 2014;15:536-40.

Authors’ contributions.—Mikael alves and eric Maury have substantial contributions in conception, and study design. Mi-kael alves, naike Bigé, claire Pichereau, Jean-luc Baudel, ar-naud Galbois, Hafid Ait-Oufella and Eric Maury have substan-tial contributions in acquisition of data, data analysis and writing of the manuscript. all the authors corrected and approved the final version of the manuscript.Acknowledgments.—the authors wish to sincerely thank aurélie Boudin (sonosite France) who made this study possible.Conflicts of interest.—The authors certify that there is no conflict of interest with any financial organization regarding the material discussed in the manuscript.Article first published online: January 5, 2016. - Manuscript ac-cepted: January 4, 2016. - Manuscript revised: December 16, 2015. - Manuscript received: august 10, 2015.(Cite this article as: alves M, Bigé n, Pichereau c, Baudel Jl, galbois a, oufella Ha, et al. impact of axotracktM for ultrasound-guided central venous catheter insertion: a rando-mized controlled study conducted on inanimate manikin. Miner-va anestesiol 2016;82:493-5)

© 2016 eDiZioni Minerva MeDicathe online version of this article is located at http://www.minervamedica.it Minerva anestesiol 2016;82:495-6

tetanus and tako-tsubo: is there a relationship?

Dear editor,

Tako-Tsubo Syndrome (TTS), also defined “Left ventricular apical Ballooning syndrome” because of the typical aspect of the left ventricle (lv) characterized by transient wall motion abnormalities involving the lv apex, is often precipitated by emotional or physical stress. clinically it mimics an acute coronary syndrome without obstructive coronary disease. etiopathogenesis of tts remains unclear but an important hypothesis is the excess of circulating catecholamines.1 some reports stated that autonomic dysfunction associated with med-ical illness, like tetanus, might conduce to the develop-ment of myocardial damage.2, 3 autonomic instability with labile blood pressure, myocardial dysfunction and sympathetic overactivity are common in tetanus.4 We present a case of tetanus and tts in order to reveal a

a simulator - supraclavicular (internal jugular) or infra-clavicular site (axillary vein), control or experimental - were randomized (using a table). the number of needle passes required before success and the times (seconds) between 1) skin contact and first needle pass, and 2) first needle pass and success were recorded. values are ex-pressed as median (range). the U Mann-Whitney test was used to compare quantitative values. Proportions were compared using the χ2 test or the Fisher exact test, as appropriate.

Finally, all participants performed their four cath-eter insertion and 148 approaches were therefore ana-lyzed. axotrack tM decreased time required for venous puncture and the number of needle passes only at the infraclavicular site (table i). this was observed what-ever previous experience in Us-guided cvc inser-tion, suggesting that it also facilitates insertion among the most skilled physicians. this could be due to the greater intrinsic difficulty encountered for CVC inser-tion at the infraclavicular site (size and depth of axillary vein compared to jugular vein, pleura proximity) which often delays the first skin contact with the larger probe used in the control group. conversely, at jugular site, the device increased the time between skin contact and first pass. This could be explained by chin proximity making axotrack tM sometimes difficult to handle in this area. Most of physicians approving axotrack tM use (60%) were naïve in the field of US-guiding CVC insertion (85% vs. 14%, P=0.001). these results sug-gest the potential interest of such devices for Us-guided cvc insertion.

Mikael alves 1*, naike BigÉ 2, claire PicHereaU 2, Jean-luc BaUDel 2,

arnaud galBois 2, Hafid Ait-OUFELLA 2, 3, eric MaUrY 2, 3

1intensive care Unit, centre Hospitalier Poissy saint-germain en laye, 10 rue du champ gaillard,

Poissy, France; 2Medical intensive care Unit, Hôpital saint-antoine, assistance Publique-Hôpitaux de

Paris, 184 rue du Faubourg saint-antoine, 75571 Paris cedex 12, France; 3UPMc, Paris, France

*corresponding author: Mikael alves, intensive care Unit, centre Hospitalier Poissy saint-germain en laye, 10 rue du champ gaillard, 78300 Poissy, France. e-mail: [email protected]

References

1. lamperti M, Bodenham ar, Pittiruti M, Blaivas M, au-goustides Jg, elbarbary M, et al. international evidence-based recommendations on ultrasound-guided vascular access. intensive care Med 2012;38:1105-17.

2. Blaivas M, adhikari s. an unseen danger: frequency of posterior vessel wall penetration by needles dur-ing attempts to place internal jugular vein central cath-eters using ultrasound guidance. critical care Med 2009;37:2345-9.

3. slama M, novara a, safavian a, ossart M, safar M,



gest that they may be catecholamine-induced rather than a direct effect on the myocardium of tetanus toxin.2 so the hypothesis of ventricular dysfunction as the result of a direct effect of the catecholamines on the myocardium is more credible. therefore we conclude that tetanus is a stressful event and it could represent a trigger for tts.

giulia BrUnelli *, simonetta tesoro, andrea trotta, nadia Dentini,

vito a. PeDUtoDepartment of surgical and Biomedical sciences,

institute of anesthesia, analgesia and critical care, University Hospital of Perugia, Perugia, italy

*corresponding author: giulia Brunelli, via del campanile 2, Foligno (Pg) caP 06034, italy. e-mail: [email protected]

References

1. Hurst rt, Prasad a, askew iii JW, sengupta PP, tajik aJ. takotsubo cardiomyopathy: a unique cardiomyopathy with variable ventricular morphology. J am coll cardiol img 2010;3:641-9.

2. rose ag. catecholamine-induced myocardial damage associated with phaeochromocytomas and tetanus. s afr Med J 1974;48:1285-9.

3. cho HJ, Kim HY, Han sH, Kim HJ, Moon Ys, oh J. takotsubo cardiomyopathy following cerebral infarction involving the insular cortex. J clin neurol 2010;6:152-5.

4. Henriques Filho gt, lacerda Hr, albuquerque a, Xi-menes ra. sympathetic overactivity and arrhythmias in tetanus: electrocardiographic analysis. rev inst Med trop sao Paulo 2007;49:17-22.

5. goodloe aH, evans JM, Middha s, Prasad a, olson tM. characterizing genetic variation of adrenergic signalling pathways in takotsubo (stress) cardiomyopathy exomes. eur J Heart Fail 2014;16:942-9.

Conflicts of interest.—The authors certify that there is no conflict of interest with any financial organization regarding the material discussed in the manuscript.Article first published online: January 28, 2016. - Manuscript accepted: January 26, 2016. - Manuscript revised: January 15, 2016. - Manuscript received: october 12, 2015.(Cite this article as: Brunelli g, tesoro s, trotta a, Dentini n, Peduto va. tetanus and tako-tsubo: is there a relationship? Minerva anestesiol 2016;82:495-6)

possible connection between the two conditions, repre-sented by autonomic dysfunction.

a 61 year-old woman was hospitalized because of cervical-chest pain and sweating. after worsening of pain, elevation of myocardial necrosis markers and ecg changes, she received therapy for myocardial ischemia. symptoms persisted and rigidity involving superior and inferior limbs appeared. We diagnosed generalized tetanus. she was treated with human tetanus immu-noglobulin and antibiotics. During hospitalization she presented severe hemodynamic instability. laboratory tests showed troponin levels by 2.82 ng/ml. ecg re-vealed sinus tachycardia, st-segment elevation and low voltage in v5-v6 leads. echocardiography showed total akinesis of the lv mid segments and of the apex with se-vere systolic function impairment. We considered the di-agnosis of tts. a gradual improvement of the patient’s clinical condition was confirmed by instrumental exams. at the discharge the patient was oriented and collabora-tive. Echocardiography confirmed the hypokinesis of the apex. a coronary angiography resulted normal.

Management of tetanus is essentially supportive, us-ing mechanical assisted ventilation, control of general spasticity and autonomic events. our patient had a se-vere form of tetanus characterized by generalized spas-ticity, impairment of respiratory function and autonomic events involving the cardio-vascular system (hypoten-sion alternating with hypertension). these autonomic manifestations could represent a connection between tetanus and tako-tsubo in our patient. sympathetic overactivity appears to play a very important role in the pathophysiology of tts. several studies found high lev-els of catecholamines in suffering patients.1 it is unclear why some patients are more susceptible than others, maybe genetic heterogeneity of the adrenergic recep-tors renders them more or less sensitive to adrenergic stimuli which may explain this variability.5 the reason why myocardial stunning localizes preferentially at the apex could be that it has a greater density of adrenergic receptors.5 our patient met all the Mayo clinic criteria for diagnosis of ttc 2 and she presented a global and multisegment ventricular dysfunction. the hypothesis of multi-vessel coronary spasm is unlikely. toxic myo-carditis may occur in tetanus and may lead to hypoten-sion, but the specific nature of myocardial lesions sug-

corrigenDUMin volume 81, issue no. 9 – March, page 1044, in the article entitled “an additional tip to facilitate glidescope

intubation”, the correct authors’ names are: turkstra t.P., rachinsky M., Batohi P.

Vol. 82 - No. 4 MINERVA ANESTESIOLOGICA 497

T O P 5 0 M I N E R V A A N E S T E S I O L O G I C A R E V I E W E R S

Anno: 2016Mese: AprilVolume: 82No: 4Rivista: MINERVA ANESTESIOLOGICACod Rivista: Minerva Anestesiol

Lavoro: titolo breve: primo autore: ACKNOWLEDGEMENTSpagine: 497

Most active reviewers between September 2015-February 2016

Surname Name Numberof revisions

Savoia Gennaro 10Biasucci Daniele Guerino 9Agrò Eugenio Felice 8Bertini Pietro 7Dalfino Lidia 7Camporesi Enrico Mario 7Brazzi Luca 7Montini Luca 6Peris Adriano 6Donadello Katia 6Deflandre Eric P. 6Cattano Davide 6Benes Jan 6Chelazzi Cosimo 6Alston Theodore 5Weiner Menachem 5Cata Juan P. 5El Tahan Mohamed 5Della Rocca Giorgio 5Guarracino Fabio 5Tritapepe Luigi 5Cinnella Gilda 5Zanella Alberto 5Bruder Nicolas 4Sorbello Massimiliano 4Cunha Francisco 4Ezzeldin Ibrahim Saleh 4Calzavacca Paolo 4Langer Thomas 4Cavallone Laura Francesca 4Camporota Luigi 4Faraoni David 4De Pascale Gennaro 4Caruselli Marco 4Szakmany Tamas 4Aceto Paola 4Boer Christa 4Max Martin 3Ting Chien-Kun 3Mauri Tommaso 3Mallat Jihad 3Sakai Tetsuro 3Gómez-Ríos Manuel Ángel 3Kopp Rüdger 3Somaini Marta 3De Cosmo Germano Amilcare 3Turnbull David 3Leone Marc 3Beilin Yaakov 3Abdel Raheem Mohamed Samy 3

minerva anestesiologica vol. 82 no. 4 pages 375 · pdf filemechanical resuscitation devices...

Documents