American Journal of Emergency Medicine 35 (2017) 1420–1425

Contents lists available at ScienceDirect

American Journal of Emergency Medicine

j ourna l homepage: www.e lsev ie r .com/ locate /a jem

Original Contribution

A randomized comparison of three chest compression techniques andassociated hemodynamic effect during infant CPR: A randomizedmanikin study

Jacek Smereka, MD, PhD a, Lukasz Szarpak, EMT-P, DPH, PhDb,⁎, Antonio Rodríguez-Núñez, MD, PhD, Prof. c,Jerzy R. Ladny, MD, PhD, Prof. d, Steve Leung, MDe, Kurt Ruetzler, MD, PhD, Prof. e,f

a Department of Emergency Medical Service, Wroclaw Medical University, Wroclaw, Polandb Department of Emergency Medicine, Medical University of Warsaw, Warsaw, Polandc Paediatric Emergency and Critical Care Division, Clinical University Hospital, University of Santiago de Compostela, Santiago de Compostela, Institute of Research of Santiago [IDIS] and SAMIDNetwork, Spaind Department of Emergency Medicine and Disaster, Medical University Bialystok, Bialystok, Polande Department of Outcomes Research, Anesthesiology Institute, Cleveland Clinic, Cleveland, USAf Department of General Anesthesiology, Anesthesiology Institute, Cleveland Clinic, Cleveland, USA

⁎ Corresponding author at: Department of EmergencyWarsaw, 4 Lindleya Str., 02-005 Warsaw, Poland.

E-mail address: lukasz.szarpak@gmail.com (L. Szarpak

http://dx.doi.org/10.1016/j.ajem.2017.04.0240735-6757/© 2017 Elsevier Inc. All rights reserved.

a b s t r a c t

a r t i c l e i n f o

Article history:Received 13 January 2017Received in revised form 8 April 2017Accepted 13 April 2017

Introduction: Pediatric cardiac arrest is an uncommonbut critical life-threatening event requiring effective cardiopul-monary resuscitation. High-quality cardio-pulmonary resuscitation (CPR) is essential, but is poorly performed, evenby highly skilled healthcare providers. The recently described two-thumb chest compression technique (nTTT) con-sists of the two thumbs directed at the angle of 90° to the chestwhile having thefingersfist-clenched. This techniquemight facilitate adequate chest-compression depth, chest-compression rate and rate of full chest-pressure relief.Methods: 42 paramedics from the national EmergencyMedical Service of Poland performed three single-rescuer CPRsessions for 10minutes each. Each session was randomly assigned to the conventional two-thumb (TTHT), the con-ventional two-finger (TFT) or the nTTT. The manikin used for this study was connected with an arterial blood pres-sure measurement device and blood measurements were documented on a 10-seconds cycle.Results: The nTTT provided significant higher systolic (82 vs. 30 vs. 41mmHg). A statistically significant differencewasnoticed between nTTT and TFT (pb .001), nTTT and TTHT (p b 0.001), TFT and TTHT (p=0.003). Themedian diastolicpreassure using nTTTwas 16mmHg comparedwith 9mmHg for TFT (p b 0.001), and 9.5mmHg for TTHT (p b 0.001).Mean arterial pressure using distinctmethods varied and amounted to 40 vs. 22. vs. 26mmHg (nTTT vs. TFT vs. TTHT,respectively). A statistically significant differencewas noticed between nTTT and TFT (p b 0.001), nTTT and TTEHT (p b

0.001), and TFT and TTHT (p b 0.001). The highestmedian pulse pressurewas obtained by the nTTT 67.5mmHg. Pulsepressurewas 31.5mmHg in theTTHT and24mmHg in theTFT. Thedifference betweenTFT andTTHT (p=0.025), TFTand nTTT (p b 0.001), as well as between TTHT and nTTT (p b 0.001) were statistically significant.Conclusions: The new nTTT technique generated higher arterial blood pressures compared to established chest com-pression techniques using an infant manikin model, suggesting a more effective chest compression. Our results haveimportant clinical implications as nTTT was simple to perform and could be widely taught to both healthcare profes-sionals and bystanders. Whether this technique translates to improved outcomes over existing techniques needs fur-ther animal studies and subsequent human trials.

© 2017 Elsevier Inc. All rights reserved.

Keywords:Infant CPRChest compressionHemodynamics

1. Introduction

Pediatric cardiac arrest is an uncommon but critical life-threateningevent requiring effective cardiopulmonary resuscitation (CPR). About16,000 pediatric cardiac arrests occurs in the United States annually

Medicine, Medical University of

).

[1]. Only 8% of the patients survive to hospital discharge [2] and ofthese, up to two-thirds have neurological sequelae [3]. Majority of pedi-atric cardiac arrest was below age of two and had poorer chance of sur-vival versus older children [3,4].

Quality of the CPR directly affects hemodynamics, survival and neu-rological outcomes following cardiac arrest and is therefore of centralimportance [5]. Nevertheless infant CPR is poorly performed, evenamong highly skilled healthcare providers [5,6,7,8,9]. Chest compres-sions are crucial in generating circulation to vital organs in infants

1421J. Smereka et al. / American Journal of Emergency Medicine 35 (2017) 1420–1425

who are highly susceptible to ischemia due to their high metabolism.Current guidelines for infant CPR suggest the so called 2-finger tech-nique (TFT) for lone rescuer and the 2-thumb–encircling hands(TTHT) technique for two rescuers for providing chest compressions[10,11]. Several studies demonstrated, that TTHT generates higherblood and coronary perfusion pressures compared to TFT [12,13,14],but is not recommended for lone rescuers due to potential difficulty inalternating between compression and ventilation.

Our study group recently investigated a new two-thumb technique(nTTT) in an infant CPR setting. The initial results are quite impressive,as this new technique might facilitate adequate chest-compressiondepth, chest-compression rate and rate of full chest pressure relief andwas superior compared to the two established techniques TFT andTTHT. The nTTT consists of two thumbs directed at the angle of 90° tothe chest while closing the fingers of both hands in a fist (Fig. 1).

Fig. 1. Chest compression techniques used in study: (A) the new two thumb chestcompression technique; (B) two finger technique; (C) two thumb technique.

We believe we have found a better alternative to existing standardmethods of chest compression during simulated CPR in infants. Higherarterial blood pressures have been associated with improved survivaland is a measure of quality of chest compression [15,16,17]. Thereforeour aimwas to quantitatively compare the quality of chest-compressionof nTTT versus the current standard techniques (TFT, TTHT) as definedby systolic-, diastolic-, mean arterial- and pulse-pressures. Our hypoth-esis is that nTTT generates higher arterial blood pressures compared tothe established TFT and TTHT in an infant manikin model.

2. Methods

2.1. Simulation of the scenario

This trial is a continuation of research conducted by the authors toimprove the quality of chest compressions in the newborn [18,19]. Tosimulate the scenario of infant CPR, an ALS Baby trainer manikin(Laerdal Medical, Stavanger, Norway) simulating a 3-month-old infant.The manikin was connected with a fixed-volume arterial system at-tached to a monitor (Draeger Infinity® Delta; Draegerwerk AG & Co.KGaA, Luebeck, Germany) via an arterial pressure transducer (EdwardLifesciences: TruWave Disposable Pressure Transducer; Irvine, CA,USA). The arterial circuit composed of a 50-mL bag of normal saline so-lution (air removed) attached to themanikin chest plate and connectedto the transducer with a 20-gauge intravenous catheter and tubing. Themanikinwas placed on a high adjustable hospital stretcher. The bedwasleveled to the iliac crest of each rescuer for standardization. Themanikinwas previously intubated and ventilation was performed using a resus-citator bag.

2.2. Study population

The participants for this study were recruited from the Staff of theNational Emergency Medical Service (EMS) teams of Poland betweenJuly and October 2016. After written informed consent was obtained,forty-two paramedics from the EMS volunteered to participate in thisstudy. All participants areworking full time in the EMS and are routinelyinvolved in the management and treatment of critically ill patients, in-cluding pediatrics in the out-of-hospital setting.

A prospective, randomized, crossover trial designwas chosen for thisstudy.

This studywas performed in accordancewith the Consolidated Stan-dards of Reporting Trials (CONSORT) statement and has been approvedby the Institutional Review Board of the Polish Society of Disaster Med-icine (Approval: IRB N20.07.2016) [20].

2.3. Compression techniques

All paramedics participated in a 30 min lasting theoretical trainingsession, covering the relevant aspects of pediatric advanced life support(PALS) according to the current CPR guidelines.

After the theoretical session, three different chest compressionstechniques were introduced, explained in detail and demonstrated tothe paramedics by highly experienced senior paramedics. These threechest compression techniques included:

a) Two finger technique TFT: the pediatric thorax is compressed withthe tips of two fingers and is recommended for lone rescuer duringinfant CPR by international CPR guidelines [10,11].

b) Two thumb technique TTHT: the two thumbs of the rescuer areplaced over the lower third of the sternum, with the fingersencircling the torso and supporting the back. This technique is rec-ommended for two rescuers during infant CPR by internationalCPR guidelines [10,11].

c) ‘new two-thumb technique’ (nTTT): this technique consists in using

1422 J. Smereka et al. / American Journal of Emergency Medicine 35 (2017) 1420–1425

two thumbs directed at the angle of 90° to the chest while closingthe fingers of both hands in a fist (Fig. 1).

All 42 paramedics were randomly assigned to one out of threegroups (TFT, TTHT, nTTT) using the ResearchRandomized software(www.randomizer.org), resulting in 14 paramedics in each group.

All paramedics were asked to perform infant CPR for 10 min with achest compression: ventilation ratio of 15:2 according to internationalCPR guidelines. Chest compressions were performed according to therespective group assignments. After completion of the initial CPR set-ting, the paramedics had a break for 2 days. Afterwards the paramedicswere asked to perform a second CPR settingwith an randomly assignedchest compression technique. After another rest for two days, the para-medics performed the third CPR setting with the third and remainingchest compression technique. Randomization flow chart is presentedin Fig. 2.

All paramedics were blinded to the arterial pressure tracing duringall complete study setting.

2.4. Power calculation

Based on data acquired by a previous unpublished pilot study, thefollowing assumptions were made to calculate the number of para-medics to be included: we proposed an alpha risk of 0.05, and a betarisk of 0.2. The quality CPR score (QCPRmeasured by Laerdal QCPR feed-back device) in pilot data amounted to 21% vs. 24% vs. 30% in the TFT,TTHT and nTTT, respectively. Using the t-test, paired, two-sided, 40 par-ticipants were required.

Fig. 2. Randomizat

2.5. Measurements

The bloodmeasurement systemwas calibrated before each CPR set-ting and blood pressure including systolic blood pressure (SBP), diastol-ic blood pressure (DBP), mean arterial pressure (MAP), were recordedin mmHg.

Each CPR setting was recorded with a camera (Panasonic HC-VX980EP-K). Blood pressure parameters (SBP, DBP, MAP) in 10-s inter-vals were retrospectively added to the data-form. Pulse pressures (PP)were calculated (SBP-DBP) in 10-s intervals and recorded in mm Hg.

2.6. Statistical analysis

The Statistica statistical package version 12.0 forWindowswas used(StatSoft, Tulsa, OK, USA). All results are shown as absolute numbers(percentages),means and standard deviation (±SD)ormedians and in-terquartile range (IQR). Normal distribution was confirmed by the Kol-mogorov-Smirnov test.We use two-way repeated-measures analysis ofvariance (RMANOVA) using a generalized linear model. The validity ofthe F-statistic used in RMANOVA was examined by performingMauchly's test of sphericity. If sphericity could not assumed, we useda stepwise approach to adjusting the degrees of freedom in the F-test.The conservative Greenhouse-Geisser adjustment criteria were used.The p-value of b0.05 was considered significant.

3. Results

Forty-two paramedics were participating in this study (27 male and15 female; mean age 30 ± 5 years, mean experience 6 ± 4 years).

ion flow chart.

Table 1Meadian values with interuartile range [IQR] for all participants

00:10 1 min 2 min 3 min 4 min 5 min 6 min 7 min 8 min 9 min 10 min

TFT SBP 44 [40–47] 46 [39–53] 40,5 [32–56] 32 [29–45] 42 [26–48] 42 [28–50] 34 [26–44] 38 [31–41] 32,5 [23–49] 29,5 [29–35] 30 [24–34]TTHT SBP 45 [38–55] 50 [40–59] 52 [47–61] 53 [50–55] 50 [43–56] 47,5 [43–53] 42,5 [38–45] 43 [42–46] 42,5 [39–45] 42 [39–48] 41 [37–45]nTTT SBP 78,5 [73–92] 85 [81–89] 83,5 [81–86] 85 [78–87] 86 [85–90] 80,5 [80–83] 85 [78–87] 82,5 [71–92] 88,5 [77–90] 83 [82–87] 84 [82–90]TFT DBP 12,5 [11–14] 12,5 [10–14] 11,5 [11−12] 10,5 [9–13] 10 [10−11] 9 [8–9] 9 [8–12] 10 [9–11] 8.5 [7–10] 9 [7–13] 8.5 [7–10]TTHT DBP 19 [17–21] 14,5 [10–16] 12 [7–14] 11.5 [9–18] 11 [9–12] 10 [7–11] 7.5 [7–13] 7 [6–12] 9.5 [8–11] 7 [7–11] 8.5 [6–12]nTTT DBP 16 [16–17] 16 [16–16] 16 [15–17] 17.5 [17–18] 17 [15–20] 16.5 [14–18] 16.5 [13–19] 15 [14–19] 16.5 [14–17] 16.5 [15–19] 17 [15–18]TFT MAP 24.5 [20–27] 24.5 [21–29] 23.5 [21–26] 20 [18–23] 21 [19–24] 22 [18–24] 20 [15–22] 21.5 [18–22] 20.5 [14–21] 18.5 [17–20] 17.5 [15–22]THTT MAP 31.5 [28–33] 30.5 [24–31] 27 [27–28] 29 [28–32] 26 [25–29] 26.5 [24–28] 24 [22–26] 24.5 [23–26] 24 [24–25] 22.5 [21–25] 23.5 [20–27]nTTT MAP 37.5 [36–41] 40 [38–41] 39.5 [38–41] 40 [38–42] 41.5 [39–42] 39 [39–40] 41 [40–42] 40.5 [39–44] 42 [38–42] 40.5 [40–42] 41 [39–43]TFT PP 33.5 [28–36] 32.5 [27–43] 28 [20–45] 21.5 [19–35] 32 [16–33] 29 [20–41] 24.5 [17–36] 25.5 [20−32] 21.5 [15–41] 23 [16–28] 19.5 [15–28]THTT PP 25 [17–40] 35.5 [32–43] 39.5 [33–48] 42 [37–43] 40 [34–45] 37 [36–44] 33.5 [31–38] 35.5 [31–37] 32 [31–40] 34.5 [31–37] 31.5 [28–37]nTTT PP 62.5 [56–69] 68.5 [61–73] 67.5 [64–71] 67.5 [65–70] 67 [65–77] 64.5 [62–67] 67.5 [60–74] 68 [53–73] 70.5 [60–75] 67 [59–69] 68 [66–72]

SBP – systolic blood pressure; DBP – diastolic blood pressure; MAP – mean arterial pressure; PP – pulse pressure.

1423J. Smereka et al. / American Journal of Emergency Medicine 35 (2017) 1420–1425

The SBP, DBP, MAP and PP, and interquartile ranges for each chestcompression technique are presented in Table 1.

The median SBP value using the distinct compression techniquesvaried and amounted to be 30 [IQR; 24–34] mmHg for TFT, 41 [IQR;37–45] mmHg for TTHT, and 84 [IQR; 82–90] mmHg for nTTT (Fig. 3).The difference in median SBP between TFT and TTHT (p = 0.003), TFTand nTTT (p b 0.001), as well as between TTHT and nTTT (p b 0.001)was statistically significant.

Use of nTTT allowed the highest SBP among all compression tech-niques studied. This proportion was the highest both at initiation ofCPR and at 10 min of CPR (Fig. 4).

The observed median DBP was 9 [IQR; 9–12.5] mmHg in the TTHT,9.5 [IQR; 9–10.5] mmHg in the TFT, and 16 [IQR; 15–19] mmHg in thenTTT. ThemedianDBP observed for the nTTTwas statistically significanthigher compared to the TFT (p b 0.001) and to TTHT (p b 0.001), where-as difference between the TFT and the TTHT did not differ significantly(p = 0.429) (Fig. 5).

The median MAP was 22 [IQR; 18.5–23] mmHg for TFT, 26 [IQR;25.5–26.5] mmHg in TTHT, and 40 [IQR; 40–40.5] mmHg in the nTTT(Fig. 6). Differences between TFT and TTHT (p b 0.001), TFT and nTTT(p b 0.001) and between TTHT and nTTT (p b 0.001; Fig. 6) were statis-tically significant.

The highest median pulse pressure (PP) was obtained by the nTTT67.5 [IQR; 63–70] mmHg. Pulse pressure was 31.5 [IQR; 28–37]mmHg in the TTHT and 24 [IQR; 19–39] mmHg in the TFT. The differ-ence between TFT and TTHT (p = 0.025), TFT and nTTT (p b 0.001), aswell as between TTHT and nTTT (p b 0.001)were statistically significant(fig. 7).

92% of all participants preferred the nTTT, and 8% preferred TTHT. Allparticipants declared fingers pain during TFT and pain in the ball of the

Fig. 3.Median systolic blood pressure during the CPR.

thumb during TTHT. These ailments were not observed in the nTTTgroup.

4. Discussion

We recently demonstrated that our novel two-thumb chest com-pression technique (nTTT) for infant CPR was feasible and met the rec-ommended target in areas of chest compression death, compressionrate, and chest recoil. Regardless of cardiac arrest etiology, arterialblood pressure is the ultimate metric for quality of chest compression.In the current study, nTTT had significantly higher SBP, DBP, MAP andPP when compared with the current recommended chest compressiontechniques for infant CPR. Additionally, nTTT consistently achieved thehighest pressures at all-time points during CPR.

Our results are consistent with the findings by Na et al., which eval-uated a comparable technique called “vertical two-thumb technique”and shown to be superior to the TTHT in 40-week old infant CPR [21].However, this study was limited by use of a new born infant manikinand also lack arterial blood pressure measurements, making extrapola-tion to a clinical situation difficult. Nonetheless, we further confirmedthe insufficiency of the current TTHT technique.

TTHTwas regarded as themost effective compression technique be-cause of its ability to generate significantly greater compression depthand blood pressure than TFT [22]. Ourfinding is novel and important be-cause we showed that the nTTT outperformed TTHT in all arterial bloodpressures. Coronary perfusion pressure (CPP) is a predictor of return ofspontaneous circulation [23]. While we could not measure CPP in ourstudy, we can speculate that nTTT provides higher CPP versus TTHTdue to significantly higher DBP in the nTTT group (nTTT 16 versusTTHT 9 mmHg). A CPP of minimum of 15 mmHg is necessary for returnof spontaneous circulation [16]. Surprisingly, TTHT does not meet thethreshold of 15 mmHg. In the initial 1 min of CPR, TTHT had a medianDBP of 14.5 mmHg but consistently decreased in DBP over the courseof 10 min of CPR, indicating suboptimal CPR quality. In contrast, nTTTmaintained DBP of 16–17mmHg throughout the entire duration of CPR.

Fig. 4. Median SBP over the complete CPR setting.

Fig. 5. The median DBP. Fig. 7. Median Pulse Pressure.

1424 J. Smereka et al. / American Journal of Emergency Medicine 35 (2017) 1420–1425

If we take CPP throughout the entire duration of CPR into account,the total dose (area under the curve) of coronary perfusion over10 min, reflects a more serious deficiency of TTHT in meeting adequatecoronary perfusion compared to sufficient nTTT.

While there is no consensus on the adequateMAP in infant CPR, it isundisputed that higher MAPs will better perfuse vital organs, such asthe brain. The poorest performer in terms of MAP is TFT, followed byTTHT. Once again nTTT achieved significant higher median MAP at allpoints compared to both established chest compression techniques.

Effective chest compression are needed in a consistent manner tomaximize chance of successful resuscitation [24]. In our study nTTTwas able to deliver consistently high arterial blood pressures through-out the 10 min of CPR, whereas, THT and TTHT had much variableblood pressures. Furthermore, with regards to MAP, both THT andTTHT had a time-dependent decline. This can be attributed rescuer fa-tigue which will reduce CPR quality [25]. THT has been sited to causehand fatigue [22]. Similarly, TTHT relies on the circumferential compres-sion of the thorax with both hands. Hand grip strength is highly influ-enced by factors such as age, gender, height and weight [26]. Rescuerswithweak grip strength or small handsmight potentially cause hand fa-tigue and again, this might affect CPR quality. In a study by Pitcher et al.,where right hand grip strengthwas assessed in healthymales, the forceexerted by hand-grip declined rapidly after three minutes of isometricgrip exercises [27]. Our novel nTTT aligns rescuers' thumbs with armsso that upper body weight and strength can be directed to the infant,similar to adult CPR. We hypothesize that by use of multiple musclessuch as pectoralis major, latissiumus dorsi, erector spinae, and rectusabdominis, nTTT's reliance on more and bigger muscles can potentially

Fig. 6.Median MAP.

generate greater compressive strength and reduce fatigue due to spreadof the work load [28].

Our post-hoc questionnaire to our participants indicated, that that avastmajority (92%) preferred the nTTT technique. The fact that nTTTdidnot cause finger or thumb pain, which is common in TFT and TTHT, sug-gested it leads to less hand fatigue andmight be overall classified easierand less exhausting.

Our study has notable strengths arising from its novelty and design.Previous studies on CPR techniques used surrogates of arterial bloodpressure or are short in duration [22,29]. The 10 min CPR session inour study more closely simulates a real-life CPR scenario. We used elec-tronically recorded blood pressures which allowed us to characterizemultiple arterial blood pressure measurements in 10 s to 10 s detail.Our large sample size of 42 paramedics and cross-over design allowedus to achieve a proper statistical power.

Therewere several limitations to our study. Firstly, we utilized an in-fant manikin which cannot simulate a real cardiac arrest infant. LaerdalManikin have been widely used in CPR studies [21,30]. Use of manikinalso allows us to achieve statistically power via a cross-over design. Fu-ture studiesmight include the use porcinemodels that enables us to di-rectly measure CPP and cerebral perfusion pressures with intra-arterialcatheters, simulate real-life cardiac arrest and observe outcomes, suchas ROSC and survival rates.

5. Conclusions

In conclusion, the new nTTT technique generated higher arterialblood pressures compared to established chest compression techniquesusing an infant manikin model, suggesting a more effective chest com-pression. Our results have important clinical implications as nTTT wassimple to perform and could bewidely taught to bothhealthcare profes-sionals and bystanders. Whether this technique translates to improvedoutcomes over existing techniques needs further animal studies andsubsequent human trials.

References

[1] Topjian AA, Berg RA, Nadkarni VM. Pediatric cardiopulmonary resuscitation: ad-vances in science, techniques, and outcomes. Pediatrics 2008;122(5):1086–98.http://dx.doi.org/10.1542/peds.2007-3313.

[2] JayaramN,McNally B, Tang F, Chan PS. Survival after out-of-hospital cardiac arrest inchildren. J Am Heart Assoc 2015;4(10) e002122 10.1161/JAHA.115.002122.

[3] Young KD, Gausche-Hill M, McClung CD, Lewis RJ. A prospective, population-basedstudy of the epidemiology and outcome of out-of-hospital pediatric cardiopulmo-nary arrest. Pediatrics 2004;114(1):157–64.

[4] Kuisma M, Suominen P, Korpela R. Paediatric out-of-hospital cardiacarrests—epidemiology and outcome. Resuscitation 1995;30(2):141–50.

[5] Cheng A, Brown LL, Duff JP, Davidson J, Overly F, Tofil NM, et al. International Net-work for Simulation-Based Pediatric Innovation, Research, & Education (INSPIRE)CPR Investigators. Improving cardiopulmonary resuscitation with a CPR feedback

1425J. Smereka et al. / American Journal of Emergency Medicine 35 (2017) 1420–1425

device and refresher simulations (CPR CARES Study): a randomized clinical trial.JAMA Pediatr 2015;169(2):137–44. http://dx.doi.org/10.1001/jamapediatrics.2014.2616.

[6] Smereka J, Szarpak L, Abelairas-Gomez C. Randomized comparison of two-thumb vs.two-finger chest compression during infant resuscitation performed by paramedics.Resuscitation 2016 106s: e-34 10.1016/j.resuscitation.2016.07.078.

[7] Abella BS, Sandbo N, Vassilatos P, Alvarado JP, O'Hearn N, Wigder HN, et al. Chestcompression rates during cardiopulmonary resuscitation are suboptimal: a prospec-tive study during in-hospital cardiac arrest. Circulation 2005;111(4):428–34.

[8] Szarpak Ł, Truszewski Z, Smereka J, Czyżewski Ł. Does the use of a chest compressionsystem in children improve the effectiveness of chest compressions? A randomisedcrossover simulation pilot study. Kardiol Pol 2016;74(12):1499–504. http://dx.doi.org/10.5603/KP.a2016.0107.

[9] Sutton RM, Wolfe H, Nishisaki A, Leffelman J, Niles D, Meaney PA, et al. Pushingharder, pushing faster, minimizing interruptions⋯ but falling short of 2010 cardio-pulmonary resuscitation targets during in-hospital pediatric and adolescent resusci-tation. Resuscitation 2013;84(12):1680–4. http://dx.doi.org/10.1016/j.resuscitation.2013.07.029.

[10] Berg MD, Schexnayder SM, Chameides L, Terry M, Donoghue A, Hickey RW, et al.American Heart Association. Pediatric basic life support: 2010 American Heart Asso-ciation Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascu-lar Care. Pediatrics 2010;126(5):e1345–60. http://dx.doi.org/10.1542/peds.2010-2972C.

[11] Maconochie IK, Bingham R, Eich C, López-Herce J, Rodríguez-Núñez A, Rajka T, et al.Paediatric life support section Collaborators. European Resuscitation Council Guide-lines for Resuscitation 2015: Section 6. Paediatric life support. Resuscitation 2015;95:223–48. http://dx.doi.org/10.1016/j.resuscitation.2015.07.028.

[12] Menegazzi JJ, Auble TE, Nicklas KA, Hosack GM, Rack L, Goode JS. Two-thumb versustwo-finger chest compression during CRP in a swine infant model of cardiac arrest.Ann Emerg Med 1993;22(2):240–3.

[13] Houri PK, Frank LR, Menegazzi JJ, Taylor R. A randomized, controlled trial of two-thumb vs two-finger chest compression in a swine infant model of cardiac arrest[see comment]. Prehosp Emerg Care 1997;1(2):65–7.

[14] Whitelaw CC, Slywka B, Goldsmith LJ. Comparison of a two-finger versus two-thumbmethod for chest compressions by healthcare providers in an infantmechan-ical model. Resuscitation 2000;43(3):213–6.

[15] Halperin HR, Lee K, Zviman M, Illindala U, Lardo A, Kolandaivelu A, et al. Outcomesfrom low versus high-flow cardiopulmonary resuscitation in a swinemodel of cardi-ac arrest. Am J Emerg Med 2010;28(2):195–202. http://dx.doi.org/10.1016/j.ajem.2009.10.006.

[16] Paradis NA, Martin GB, Rivers EP, Goetting MG, Appleton TJ, Feingold M, et al. Coro-nary perfusion pressure and the return of spontaneous circulation in human cardio-pulmonary resuscitation. JAMA 1990;263(8):1106–13.

[17] Sutton RM, French B, Nishisaki A, Niles DE, Maltese MR, Boyle L, et al. AmericanHeart Association cardiopulmonary resuscitation quality targets are associated

with improved arterial blood pressure during pediatric cardiac arrest. Resuscitation2013;84(2):168–72. http://dx.doi.org/10.1016/j.resuscitation.2012.08.335.

[18] Smereka J, Szarpak L, Smereka A, Leung S, Ruetzler K. Evaluation of new two-thumbchest compression technique for infant cardiopulmonary resuscitation performed bynovice physicians. A randomized, crossover, manikin trial. Am J Emerg Med 2017;35(4):604–9.

[19] Smereka J, Kasinski M, Smereka A, Ładny JR, Szarpak Ł. The quality of a newly devel-oped infant chest compression method applied by paramedics: a randomized cross-over manikin trial. Kardiol Pol 2017 Feb 2. http://dx.doi.org/10.5603/KP.a2017.0015.

[20] Schulz KF, Altman DG, Moher D. CONSORT Group. CONSORT 2010 statement: up-dated guidelines for reporting parallel group randomised trials. , 340BMJ; 2010c332. http://dx.doi.org/10.1136/bmj.c332.

[21] Na JU, Choi PC, Lee HJ, Shin DH, Han SK, Cho JH. A vertical two-thumb technique issuperior to the two-thumb encircling technique for infant cardiopulmonary resusci-tation. Acta Paediatr 2015;104(2):e70–5. http://dx.doi.org/10.1111/apa.12857.

[22] Dorfsman ML, Menegazzi JJ, Wadas RJ, Auble TE. Two-thumb vs. two-finger chestcompression in an infant model of prolonged cardiopulmonary resuscitation. AcadEmerg Med 2000;7(10):1077–82.

[23] Reynolds JC, Salcido DD, Menegazzi JJ. Coronary perfusion pressure and return ofspontaneous circulation after prolonged cardiac arrest. Prehosp Emerg Care 2010;14(1):78–84.

[24] Kramer-Johansen J, Myklebust H, Wik L, Fellows B, Svensson L, Sørebø H, et al. Qual-ity of out-of-hospital cardiopulmonary resuscitation with real time automated feed-back: a prospective interventional study. Resuscitation 2006;71(3):283–92.

[25] Ashton A,McCluskey A, Gwinnutt CL, Keenan AM. Effect of rescuer fatigue on perfor-mance of continuous external chest compressions over 3 min. Resuscitation 2002;55(2):151–5.

[26] Angst F, Drerup S, Werle S, Herren DB, Simmen BR, Goldhahn J. Prediction of gripand key pinch strength in 978 healthy subjects. BMC Musculoskelet Disord 2010;11:94. http://dx.doi.org/10.1186/1471-2474-11-94.

[27] Pitcher JB, Miles TS. Influence of muscle blood flow on fatigue during intermittenthuman hand-grip exercise and recovery. Clin Exp Pharmacol Physiol 1997;24(7):471–6.

[28] Tsou JY, Su FC, Tsao PC, Hong MY, Cheng SC, Chang HW, et al. Electromyography ac-tivity of selected trunk muscles during cardiopulmonary resuscitation. Am J EmergMed 2014;32(3):216–20. http://dx.doi.org/10.1016/j.ajem.2013.10.04.

[29] Haque IU, Udassi JP, Udassi S, Theriaque DW, Shuster JJ, Zaritsky AL. Chest compres-sion quality and rescuer fatigue with increased compression to ventilation ratio dur-ing single rescuer pediatric CPR. Resuscitation 2008;79(1):82–9. http://dx.doi.org/10.1016/j.resuscitation.2008.04.026.

[30] Udassi S, Udassi JP, Lamb MA, Theriaque DW, Shuster JJ, Zaritsky AL, et al. Two-thumb technique is superior to two-finger technique during lone rescuer infantmanikin CPR. Resuscitation 2010;81(6):712–7. http://dx.doi.org/10.1016/j.resusci-tation.2009.12.029.