effect of prone position in ards metanalysis icm jun 08 (nov-20-08).pdf

8/13/2019 Effect of Prone Position in ARDS metanalysis ICM Jun 08 (Nov-20-08).pdf

1/10

Intensive Care Med (2008) 34:10021011DOI 10.1007/s00134-008-1062-3 S Y S T E M A T I C R E V I E W

Fekri AbrougLamia Ouanes-BesbesSouheil Elatrous

Laurent Brochard

The effect of prone positioning in acute

respiratory distress syndrome or acute lung

injury: a meta-analysis. Areas of uncertaintyand recommendations for research

Received: 2 August 2007Accepted: 5 January 2008Published online: 19 March 2008 Springer-Verlag 2008

F. Abroug () L. Ouanes-BesbesCHU F. Bourguiba, Service de RanimationPolyvalente,5000 Monastir, Tunisiae-mail: [email protected]

S. ElatrousCHU T. Sfar, Service de RanimationMdicale,

Mahdia, Tunisia

L. BrochardHpital Henri Mondor, INSERM U 841,Service de Ranimation Mdicale, AP-HP,Crteil, France

Abstract Objective: To comparethe effects of ventilation in prone andin supine position in patients withacute lung injury/acute respiratorydistress syndrome (ALI/ARDS).

Design: Meta-analysis of randomisedcontrolled trials. Data sources:BioMedCentral, PubMed, CINAHL,and Embase (to November 2007), withadditional information from authors.Measurements and results: Fromselected randomised controlled trialscomparing positioning in ALI/ARDSwe extracted data concerning studydesign, disease severity, clinical out-comes, and adverse events. Five trialsincluding 1,372 patients met the inclu-sion criteria for mortality analysis; one

trial was added to assess the effectson acquisition of ventilator-associatedpneumonia (VAP). The included trialswere significantly underpoweredand enrolled patients with varyingseverity. Prone positioning durationand mechanical ventilation strategywere not standardised across studies.Using a fixed-effects model, we didnot find a significant effect of pronepositioning (proning) on mortality(odds ratio 0.97, 95% confidenceinterval 0.771.22). The PaO2/FiO2

ratio increased significantly more withproning (weighted means difference25 mmHg, p< 0.00001). Proningwas associated with a non-significant23% reduction in the odds of VAP(p= 0.09), and with no increasein major adverse airway complica-tions: OR 1.01, 95% CI 0.711.43.Length of intensive care unit staywas marginally and not significantlyincreased by proning. Conclusions:Prone position is not associated witha significant reduction in mortality

from ALI/ARDS despite a significantincrease in PaO2/FiO2, is safe, andtends to decrease VAP. Publishedstudies exhibit substantial clinicalheterogeneity, suggesting that anadequately sized study optimisingthe duration of proning and ventila-tion strategy is warranted to enabledefinitive conclusions to be drawn.

KeywordsARDS ALI Mechanicalventilation Prone positioning

Introduction

Acute respiratory distress syndrome (ARDS) and acutelung injury (ALI) remain associated with unacceptablyhigh mortality despite recent advances in supportivetreatment [1]. Several open-label studies showed that

ventilation in prone position improves arterial oxygena-tion with few untoward effects, in 58100% of paediatricand adult patients [211]. Prone position enhancesoxygenation via more even distribution of gravitationalgradient in pleural pressure [12, 13], better distributionof ventilation to the dorsal areas of the lungs [1416],


2/10

1003

and potentially lesser overdistension of airspace, reducingthereby the occurrence of ventilator-induced lung in-jury [1720]. It also improves lung mechanics and alveolarventilation [16].

In addition to a reduction in the intensity of ventilatorsupport (lower inspired oxygen concentration, lower

PEEP and mean airway pressure decreasing the ventilator-associated lung injury), resulting hence in facilitatedpatient recovery and earlier weaning from mechanicalventilation, beneficial physiologic effects of prone posi-tioning are expected to ensue in improved overall outcomeand especially in reduced mortality [21].

Two large randomised controlled trials (RCTs) andtwo moderately sized RCTs comparing prone pos-ition ventilation with supine ventilation in ARDS/ALIpatients failed to show an improved survival rate in suchpatients [2225]. As a consequence of the accumulationof published negative trials, recent surveys have recordeda statistically significant decrease in the use of prone

positioning at any time from 13% of ARDS patients in1998 to 7% in 2004 [26]. This phenomenon has beenrecorded in the same intensive care units (ICUs) that weresurveyed in 1998. It seems that the lack of evidence of anefficacy of prone positioning on a patient-centred outcome(mortality) has been interpreted as evidence of the absenceof clinically relevant effect of prone positioning. Thiswas reinforced by the lack of information on the trade-offbetween benefits and risks of prone positioning [increasein PaO2/FiO2 and reduced rate of ventilator-associatedpneumonia (VAP) on the one hand, and increased work-load, accidental tracheal tube displacement, and pressuresores on the other hand].

However, some of these RCTs were stopped beforethe termination of patients inclusion, yielding studiessubstantially underpowered to demonstrate potentialbenefits of prone ventilation [22, 24, 25]. Additionalmethodological limitations accrue from inclusion ofpatients with different ARDS severity, patients in the latestages of ARDS, absence of ventilation guidelines, andthe use of high tidal volumes. Moreover, these studieshighlighted the lack of general agreement on the durationof prone position, both on a 24-h basis and throughout thecourse of ARDS.

Of interest, the most recent RCT on prone positioningshowed in a multivariate analysis that prone ventilationwas an independent factor associated with improved sur-vival in patients with severe ARDS [25]. This study, whichwas prematurely stopped because of a slow inclusion rate,had the peculiar design of targeting a prolonged prone pos-ition (20 h per day).

Since none of the published RCTs was sufficientlyhighly powered to enable definitive conclusions, weundertook this meta-analysis to systematically reviewthe published randomised trials assessing the effect ofprone positioning on mortality of patients with ARDS/ALI.

Methods

Search strategy

Pertinent studies were independently searched in PubMed,EMBASE, CINAHL, and BioMedCentral (updated30 November 2007), by two trained investigators (F.A.,

L.O.-B.) using the following MeSH and keyword terms:acute respiratory distress syndrome, acute lung injury,acute respiratory failure, and prone position ventila-tion. The literature search included both adult and paedi-atric populations. No language restriction was imposed.

Study selection

Titles, abstracts, and citations were independently assessedby both reviewers to assess the potential relevance for fullreview. From the full text, both reviewers independentlyassessed studies for inclusion based on the criteria for

population, intervention, study design, and outcomes. In-cluded studies met the following criteria: (1) All studyparticipants, whether adults or children, had a clinical diag-nosis of acute respiratory failure (ARF), acute respiratorydistress syndrome or acute lung injury. ARDS and ALIwere respectively defined by the radiographic evidenceof bilateral pulmonary infiltrates, the absence of clinicalevidence of left atrial hypertension, and a PaO2/FiO2ratio of 300 or less (characteristic of ALI), or 200 or less(characteristic of ARDS). (2) Intervention: conventionalventilation in supine position compared with ventilationin prone position whatever its duration, on a 24-h basis,and during the ICU stay. (3) Design: prospective RCTs.

(4) Outcomes: mortality rate, whether the ICU mortalityor 28-day mortality, was the main outcome of this meta-analysis. It had to be clearly reported in the manuscript.Secondary outcomes corresponded to the following: theeffect of prone positioning on PaO2/FiO2 ratio. Thisusually corresponded to the average change of PaO2/FiO2ratio during the duration of the technique implementation.We also analysed the reported rate of VAP whatever itsdiagnosis method. Analysis also encompassed the inci-dence of procedure-related major airway complications.These were defined by the occurrence of accidental extu-bation, selective intubation, or accidental displacement oftracheal tube. We also compared the length of ICU stay.

Excluded from the meta-analysis were: (1) non-controlled studies; (2) studies that examined only thephysiologic effects of prone positioning. These studiesincluded either ARDS/ALI patients or patients with acuteexacerbation of COPD.

Data abstraction and study characteristics

We extracted study design (including patient selection andrandomisation), population, prone position duration on


3/10

1004

a 24-h basis, ventilatory strategy (whether a pulmonaryprotective strategy was used or not), and duration offollow-up. We also made a post-hoc power analysiscorresponding to the power of the analysed study to detectthe relative risk of mortality actually observed in thatexperiment. The primary end-point of our analysis was

to determine the effect of prone ventilation on the inci-dence of ICU or 28-day mortality. Secondary end-pointsincluded effects on oxygenation during the acute phase ofillness assessed by the PaO2/FiO2 change, the incidenceof VAP and that of major airway adverse effects in relationwith prone position. We also analysed the duration of ICUstay. We attempted to contact authors of included trials torequest additional data if necessary.

Internal validity assessment

The methodological quality of each trial was evaluated

using the 5-point scale (0 = worst and 5 = best) as de-scribed by Jadad et al. [27]. This instrument assesses theadequacy of randomisation, blinding, and the handling ofwithdrawals and dropouts.

Data analysis and synthesis

Our primary outcome was mortality in the ICU or at28 days. Secondary end-points included respective effectsof prone and supine position on oxygenation duringthe acute phase of illness (ranging from 4 to 10 days),the incidence of VAP, and that of major airway adverse

complications (extubation, selective intubation) andminor complications (pressure sores) in relation to proneposition. We also compared the duration of ICU stay.

Binary outcomes from individual studies were ana-lysed according to the MantelHaenszel model to computeindividual odds ratios (ORs) with pertinent 95% confi-dence intervals (CIs), and a pooled summary effectestimate was calculated by means of a fixed-effects model.Weighted mean differences (WMDs) and 95% CI werecomputed for continuous variables. Statistical hetero-geneity and inconsistency were measured by using Coch-ran Q tests and I2, respectively [28]. Since the majorityof published studies were negative, the risk of publicationbias was assessed by using visual inspection of funnel plot.Statistical significance is set at the two-tailed 0.05 levelfor hypothesis testing and 0.10 for heterogeneity testing.I2 values around 25%, 50%, and 75% were consideredto represent low, moderate, and severe statistical incon-sistency, respectively. Unadjusted P values are reportedthroughout. The relationship between baseline PaO2/FiO2ratio (a surrogate for patients disease severity) or theduration of prone positioning (a surrogate for treatmentintensity) on one hand, and the odds of mortality on theother hand, were evaluated using Spearman correlation. To

investigate the hypothesis that higher effect might be dis-closed in more severely ill patients, we also sought for a re-lation between the relative risk reduction recorded in eachstudy and the mortality rate in the corresponding control(supine) group. The meta-analysis was conducted usingRevMan 4.2.10. This study was performed in compliance

with The Cochrane Collaboration and the Quality ofReporting of Meta-Analyses (QUOROM) guidelines [29].

Results

Study characteristics

Figure 1 shows a flow chart of studies assessed andexcluded at various stages of the review. Finally, a total ofsix prospective RCTs were selected. One of these studiesexamined the effect of prone positioning on the prevention

Fig. 1 Flow chart of the meta-analysis


4/10

1005

of lung worsening in comatose patients [30]. It contributedonly to the analysis of the effect of prone positioning onthe occurrence of VAP. The remaining five studies in-cluded hypoxaemic patients with ARF with ARDS or ALI.One multicentre study included paediatric patients [24].Four studies included only adult patients (one including

only severe ARDS patients [25], two including patientssuffering either from ALI or ARDS [22, 31], and oneincluding all patients suffering from ARF [23]). Althoughinclusion criteria corresponded not only to ALI/ARDSpatients in the latter study, analysis concerned all includedpatients without stratification according to ARF aetiology.Following personal contact, the authors reported thatmortality rates were fairly similar across ARF aetiologies.One of the studies included patients with post-traumaticALI/ARDS [31]. Table 1 gives details of the studiescharacteristics. There was total agreement between thetwo independent reviewers on inclusion of studies and theJadad study quality grading.

Data for 1,372 patients were available for analysisof the primary outcome, namely mortality; 713 patientswere ventilated in prone position, and 659 were ventilatedin supine position. All eligible reports were describedas RCTs. They were published between 2001 and 2006.In the most recently published studies, patients wereventilated using a lung-protective ventilation strategycorresponding to a limited tidal volume (68 ml/kg ofbody weight) and a plateau pressure not exceeding 35 cmH2O [24, 25, 31]. The remaining two studies did not usesuch a ventilator strategy [22, 23]. In these five studies,duration of ventilation in prone position varied from 7 hto 18 h a day (mean duration per day 12 5 h). It lasted

between 4 days and the whole duration of hospitalisation.Three studies allowed crossover from one study armto another [22, 23, 25] . This usually corresponded topatients in the supine position who were subsequentlyventilated in prone position because of the persistence ofsevere hypoxaemia. In these studies, analysis was madeon intention-to-treat basis. The main outcome evaluated inthese studies was mortality reduction (either ICU mortal-ity, or 28-day mortality) by ventilation in prone positionin comparison with supine ventilation. Three studies wereprematurely stopped because of a slow inclusion rate [22,24, 25]. Calculation of studies power showed that allpositive studies were underpowered (< 10%40%) todetect a statistical significance of the observed mortalityreduction by prone positioning (Table 1).

Quantitative data synthesis

Effects on mortality

When we pooled all studies, ventilation in prone positionwas associated with a non-significant 3% reduction of theodds of mortality [249 of 713 patients (34.9%) in the prone T

able1

Characteristicsoftrialsincludedinthemeta-analysis

Trial

Numberof

PaO2/FiO2

Patients

Patients

Durationprone

Crossover

Jadad

Primary

P

remature

Study

patients

(mmHg)

prone(n)

supine(n)

(hours/day)

allowed

score

outcome

stop

power

Gattinoni2001

304AR

DS/ALI

127

152

152

7

1.8

Yes

3

28-daymortality

Y

es

49). Case-mix might alsopertain to the age of ARDS. In Mancebo et al.s study pa-tients were included within 48 h of meeting inclusion cri-teria, while in Gattinonis study inclusion of a substantial

sample of the population occurred even late in the courseof ARDS [22, 25].

Another explanation stems from the lack of standardi-sation of the prone procedure and that of ventilation strat-egy. Indeed, duration of prone positioning varied from 7 hto 17 h on a 24-h basis in the five studies included in the

meta-analysis. The study by Mancebo et al. [25], which ap-plied the longest prone duration (17 h/24 h) and found po-tential benefit from this procedure in the multivariate ana-lysis, suggests that prone positioning should be applied forthe longest time (more than 20 h) during the 24-h period.Along with this inference, it has been shown that the timecourse of alveolar recruitment during prone position is notconsistent and in fact differs markedly from one patientto another [38]. In some patients, the plateau of completealveolar recruitment may not be reached even after 8 h ofprone positioning [38]. Accordingly, it has been suggestedthat the duration of prone positioning should be tailored onan individual basis by repeated measurements of alveolar

recruitment. Future studies evaluating prone position ven-tilation in ARDS should probably apply a prone positionduration closer to that of Mancebo et al. [25] than to thatof Gattinonis study [22].

Standardisation of the ventilation strategy is alsoto consider. Three of the five studies included in thismeta-analysis were conducted prior to the generalisationof protective lung ventilation in ARDS [22, 23, 31].Protective lung strategy could indeed be synergistic withthe effects of prone position.

Compliance with assigned treatment is another con-cern raised by open-label studies assessing ventilatorytechniques such as prone positioning. Indeed, Gattinoni

et al. planned to use prone position for at least 6 h perday [22]. However, 27% of the patients allocated to pronepositioning were ventilated prone for fewer hours thanexpected. A similar proportion of patients (25%) ran-domised to prone ventilation in the study by Guerin et al.were actually ventilated in prone position for a shortertime than the scheduled 8 h [23]. In addition, a largenumber of patients assigned in both studies to supineventilation actually crossed over to prone ventilationbecause of a worsening in arterial oxygenation. Therefore,ascertainment bias, inherent to every unblinded trial, andlimited compliance with the scheduled prone positionsessions might have seriously compromised the accuracyof reported results of RCTs on prone position ventilationin ARDS/ALI patients.

Obviously, the involvement of the aforementionedlimitations in published studies would make the drawingof definitive conclusions a matter of great difficulty. Giventhe difficulties in properly applying the results of RCTs inthe real world, individualised care should definitively notbe abandoned based on negative RCTs [39]. Despite thelack of strong evidence for the efficacy of prone positionventilation in reducing mortality in ARDS patients, andthe fact that prone positioning has relatively lost ground,


9/10

1010

this technique remains in more than only exceptional usein most ICUs, and our study shows no sign of any harmfuleffect. From a pragmatic standpoint, physicians prefer todeal with a patient who is well oxygenated and requiringthe lowest FiO2. Meanwhile, the optimal randomised

controlled study devised to definitively answer the ques-tion of whether prone position ventilation should be usedin ARDS is still to be conducted. Several studies arecurrently at different stages (ClinicalTrials.gov identifier:NCT00159939 and NCT00527813) [40, 41].

References

1. The Acute Respiratory Distress Syn-drome Network (2000) Ventilationwith lower tidal volumes as comparedwith traditional tidal volumes for acutelung injury and the acute respiratorydistress syndrome. N Engl J Med342:13011308

2. Guerin C, Badet M, Rosselli S,Heyer L, Sab JM, Langevin B, Philit F,Fournier G, Robert D (1999) Effects ofprone position on alveolar recruitmentand oxygenation in acute lung injury.Intensive Care Med 25:12221230

3. Mure M, Lindahl SG (2001) Proneposition improves gas exchange but how? Acta Anaesthesiol Scand45:150159

4. Lee DL, Chiang HT, Lin SL, Ger LP,Kun MH, Huang YC (2002) Prone-position ventilation induces sustainedimprovement in oxygenation in pa-tients with acute respiratory distresssyndrome who have a large shunt. CritCare Med 30:14461452

5. McAuley DF, Giles S, Fichter H,Perkins GD, Gao F (2002) What is theoptimal duration of ventilation in theprone position in acute lung injury andacute respiratory distress syndrome?

Intensive Care Med 28:4144186. Gainnier M, Michelet P, Thirion X,

Arnal JM, Sainty JM, Papazian L(2003) Prone position and positiveend-expiratory pressure in acute res-piratory distress syndrome. Crit CareMed 31:27192726

7. Mentzelopoulos SD, Roussos C, Za-kynthinos SG (2005) Prone positionreduces lung stress and strain in severeacute respiratory distress syndrome.Eur Respir J 25:534544

8. Papazian L, Gainnier M, Marin V, Do-nati S, Arnal JM, Demory D, Roch A,Forel JM, Bongrand P, Bregeon F,Sainty JM (2005) Comparison of

prone positioning and high-frequencyoscillatory ventilation in patients withacute respiratory distress syndrome.Crit Care Med 33:21622171

9. Vieillard-Baron A, Rabiller A,Chergui K, Peyrouset O, Page B,Beauchet A, Jardin F (2005) Proneposition improves mechanics andalveolar ventilation in acute respiratorydistress syndrome. Intensive Care Med31:220226

10. Mure M, Martling CR, Lindahl SG(1997) Dramatic effect on oxygenationin patients with severe acute lung insuf-ficiency treated in the prone position.Crit Care Med 25:15391544

11. Curley MA, Thompson JE, Arnold JH(2000) The effects of early and re-peated prone positioning in pediatricpatients with acute lung injury. Chest118:156163

12. Mutoh T, Guest RJ, Lamm WJ, Al-bert RK (1992) Prone position alters theeffect of volume overload on regional

pleural pressures and improves hypox-emia in pigs in vivo. Am Rev RespirDis 146:300306

13. Wiener CM, Kirk W, Albert RK (1990)Prone position reverses gravitationaldistribution of perfusion in dog lungswith oleic acid-induced injury. J ApplPhysiol 68:13861392

14. Lamm WJ, Graham MM, Albert RK(1994) Mechanism by which the proneposition improves oxygenation in acutelung injury. Am J Respir Crit Care Med150:184193

15. Pappert D, Rossaint R, Slama K,Gruning T, Falke KJ (1994) Influenceof positioning on ventilationper-

fusion relationships in severe adultrespiratory distress syndrome. Chest106:15111516

16. Richter T, Bellani G, Scott Harris R,Vidal Melo MF, Winkler T, Venegas JG,Musch G (2005) Effect of prone pos-ition on regional shunt, aeration, andperfusion in experimental acute lunginjury. Am J Respir Crit Care Med172:480487

17. Broccard A, Shapiro RS, Schmitz LL,Adams AB, Nahum A, Marini JJ (2000)Prone positioning attenuates and redis-tributes ventilator-induced lung injuryin dogs. Crit Care Med 28:295303

18. Gattinoni L, Carlesso E, Cadringher P,

Valenza F, Vagginelli F, Chiumello D(2003) Physical and biological triggersof ventilator-induced lung injury andits prevention. Eur Respir J Suppl47:15s25s

19. Gattinoni L, Pesenti A (2005) Theconcept of baby lung. Intensive CareMed 31:776784

20. Mure M, Domino KB, Lindahl SG,Hlastala MP, Altemeier WA,Glenny RW (2000) Regionalventilation-perfusion distribution ismore uniform in the prone position.J Appl Physiol 88:10761083

21. Reignier J (2005) Prone position: canwe move from better oxygenationto better survival? Crit Care Med33:453455

22. Gattinoni L, Tognoni G, Pesenti A,Taccone P, Mascheroni D, Labarta V,Malacrida R, Di Giulio P, Fumagalli R,

Pelosi P, Brazzi L, Latini R (2001)Effect of prone positioning on the sur-vival of patients with acute respiratoryfailure. N Engl J Med 345:568573

23. Guerin C, Gaillard S, Lemasson S,Ayzac L, Girard R, Beuret P, Palmier B,Le QV, Sirodot M, Rosselli S, Cadier-gue V, Sainty JM, Barbe P, Com-bourieu E, Debatty D, Rouffineau J,Ezingeard E, Millet O, Guelon D,Rodriguez L, Martin O, Renault A,Sibille JP, Kaidomar M (2004) Effectsof systematic prone positioning inhypoxemic acute respiratory failure:a randomized controlled trial. Jama292:23792387

24. Curley MA, Hibberd PL, Fineman LD,Wypij D, Shih MC, Thompson JE,Grant MJ, Barr FE, Cvijanovich NZ,Sorce L, Luckett PM, Matthay MA,Arnold JH (2005) Effect of prone pos-itioning on clinical outcomes in childrenwith acute lung injury: a randomizedcontrolled trial. Jama 294:229237

25. Mancebo J, Fernandez R, Blanch L,Rialp G, Gordo F, Ferrer M, Ro-driguez F, Garro P, Ricart P, Vall-verdu I, Gich I, Castano J, Saura P,Dominguez G, Bonet A, Albert RK(2006) A multicenter trial of prolongedprone ventilation in severe acute re-spiratory distress syndrome. Am J

Respir Crit Care Med 173:1233123926. Esteban A, Ferguson ND, Meade MO,

Frutos-Vivar F, Apezteguia C,Brochard L, Raymondos K, Nin N,Hurtado J, Tomicic V, Gonzalez M,Elizalde J, Nightingale P, Abroug F,Pelosi P, Arabi Y, Moreno R, Jibaja M,DEmpaire G, Sandi F (2007) Evolutionof mechanical ventilation in response toclinical research. Am J Respir Crit CareMed:200706200893OC


10/10

1011

27. Jadad AR, Moore RA, Carroll D, Jenk-inson C, Reynolds DJ, Gavaghan DJ,McQuay HJ (1996) Assessing the qual-ity of reports of randomized clinicaltrials: is blinding necessary? ControlClin Trials 17:112

28. Higgins JP, Thompson SG, Deeks JJ,Altman DG (2003) Measuring in-

consistency in meta-analyses. BMJ327:557560

29. Moher D, Cook DJ, Eastwood S,Olkin I, Rennie D, Stroup DF (1999)Improving the quality of reports ofmeta-analyses of randomised controlledtrials: the QUOROM statement. Qualityof Reporting of Meta-analyses. Lancet354:18961900

30. Beuret P, Carton MJ, Nourdine K,Kaaki M, Tramoni G, Ducreux JC(2002) Prone position as preventionof lung injury in comatose patients:a prospective, randomized, controlledstudy. Intensive Care Med 28:564569

31. Voggenreiter G, Aufmkolk M,

Stiletto RJ, Baacke MG, Waydhas C,Ose C, Bock E, Gotzen L, Obertacke U,Nast-Kolb D (2005) Prone positioningimproves oxygenation in post-traumaticlung injury a prospective randomizedtrial. J Trauma 59:333341; discussion341333

32. Lenfant C (2003) Shattuck lecture:clinical research to clinical practice lost in translation? N Engl J Med349:868874

33. Wilson K, Mills EJ, McGowan J,Guyatt G (2002) Teaching evidence-based complementary and alternativemedicine. 5. Interpreting the results of

a study on therapy and applying themto a patient. J Altern Complement Med8:867873

34. Malbouisson LM, Busch CJ, Puybas-set L, Lu Q, Cluzel P, Rouby JJ (2000)Role of the heart in the loss of aerationcharacterizing lower lobes in acuterespiratory distress syndrome. CT ScanARDS Study Group. Am J Respir CritCare Med 161:20052012

35. Takala J, Ruokonen E, Webster NR,Nielsen MS, Zandstra DF, Vun-delinckx G, Hinds CJ (1999) Increasedmortality associated with growth hor-mone treatment in critically ill adults.N Engl J Med 341:785792

36. Yusuf S, Sleight P, Pogue J, Bosch J,Davies R, Dagenais G (2000) Effectsof an angiotensin-converting-enzymeinhibitor, ramipril, on cardiovascu-lar events in high-risk patients. TheHeart Outcomes Prevention EvaluationStudy Investigators. N Engl J Med342:145153

37. Montgomery AB, Stager MA, Car-rico CJ, Hudson LD (1985) Causesof mortality in patients with the adultrespiratory distress syndrome. Am RevRespir Dis 132:485489

38. Reutershan J, Schmitt A, Dietz K,Unertl K, Fretschner R (2006) Alveolarrecruitment during prone position: time

matters. Clin Sci (Lond) 110:65566339. Hayward RA, Kent DM, Vijan S,

Hofer TP (2005) Reporting clinicaltrial results to inform providers, payers,and consumers. Health Aff (Millwood)24:15711581

40. Mentzelopoulos SD, Roussos C, Za-kynthinos SG (2007) Prone positionin early and severe acute respiratorydistress syndrome: a design for a defini-tive randomized controlled trial. AnesthAnalg 104:466468

41. Curley MA, Arnold JH, Thompson JE,Fackler JC, Grant MJ, Fineman LD,Cvijanovich N, Barr FE, Molitor-Kirsch S, Steinhorn DM, Matthay MA,

Hibberd PL (2006) Clinical trialdesigneffect of prone positioningon clinical outcomes in infants andchildren with acute respiratory distresssyndrome. J Crit Care 21:2332;discussion 3227

effect of prone position in ards metanalysis icm jun 08 (nov-20-08).pdf

Documents