Changes of end-tidal carbon dioxide during cardiopulmonary resuscitation from ventricular fi brillation versus asphyxial cardiac arrest

Qing-ming Lin, Xiang-shao Fang, Li-li Zhou, Yue Fu, Jun Zhu, Zi-tong Huang

1Department of Emergency Medicine, Fujian Provincial Hospital, Fujian Medical University, Fuzhou 350001, China

2Institute of Cardiopulmonary Cerebral Resuscitation, Sun Yat-sen University, Guangzhou 510120, China

3Department of Emergency Medicine, Sun Yat-sen Memorial Hospital, Sun Yat-sen University, Guangzhou 510120, ChinaCorresponding Author:Zi-tong Huang, Email: huangztdoc@126.com

Changes of end-tidal carbon dioxide during cardiopulmonary resuscitation from ventricular fi brillation versus asphyxial cardiac arrest

Qing-ming Lin1, Xiang-shao Fang2,3, Li-li Zhou2,3, Yue Fu2,3, Jun Zhu2,3, Zi-tong Huang2,3

1Department of Emergency Medicine, Fujian Provincial Hospital, Fujian Medical University, Fuzhou 350001, China

2Institute of Cardiopulmonary Cerebral Resuscitation, Sun Yat-sen University, Guangzhou 510120, China

3Department of Emergency Medicine, Sun Yat-sen Memorial Hospital, Sun Yat-sen University, Guangzhou 510120, ChinaCorresponding Author:Zi-tong Huang, Email: huangztdoc@126.com

BACKGROUND:Partial pressure of end-tidal carbon dioxide (PETCO2) has been used to monitor the effectiveness of precordial compression (PC) and regarded as a prognostic value of outcomes in cardiopulmonary resuscitation (CPR). This study was to investigate changes of PETCO2during CPR in rats with ventricular fi brillation (VF) versus asphyxial cardiac arrest.

METHODS:Sixty-two male Sprague-Dawley (SD) rats were randomly divided into an asphyxial group (n=32) and a VF group (n=30). PETCO2was measured during CPR from a 6-minute period of VF or asphyxial cardiac arrest.

RESULTS:The initial values of PETCO2immediately after PC in the VF group were signi fi cantly lower than those in the asphyxial group (12.8±4.87 mmHg vs. 49.2±8.13 mmHg, P=0.000). In the VF group, the values of PETCO2after 6 minutes of PC were significantly higher in rats with return of spontaneous circulation (ROSC), compared with those in rats without ROSC (16.5±3.07 mmHg vs. 13.2±2.62 mmHg, P=0.004). In the asphyxial group, the values of PETCO2after 2 minutes of PC in rats with ROSC were signi fi cantly higher than those in rats without ROSC (20.8±3.24 mmHg vs. 13.9±1.50 mmHg, P=0.000). Receiver operator characteristic (ROC) curves of PETCO2showed signi fi cant sensitivity and speci fi city for predicting ROSC in VF versus asphyxial cardiac arrest.

CONCLUSIONS:The initial values of PETCO2immediately after CPR may be helpful in differentiating the causes of cardiac arrest. Changes of PETCO2during CPR can predict outcomes of CPR.

Partial pressure of end-tidal carbon dioxide; Cardiac arrest; Cardiopulmonary resuscitation; Return of spontaneous circulation; Rats

INTRODUCTION

Partial pressure of end-tidal carbon dioxide (PETCO2) is a noninvasive monitor of physiological variable which has been widely used in clinical patients, especially in critically ill patients. Under constant ventilation, PETCO2reflects the level of tissue circulation and the state of body metabolism. The 2010 American Heart Association Guidelines for Cardiopulmonary Resuscitation (CPR) and Emergency Cardiovascular Care (ECC) recommend quantitatively analyzing capnogram to monitor efficacy of CPR.[1]During CPR in patients, a sudden increase in the values of PETCO2signals the return of spontaneous circulation (ROSC).[2,3]PETCO2values also predict the outcomes of CPR in cardiac arrest patients.[4–6]

Two clinical studies, which were conducted to assess the pattern of PETCO2changes during CPR in patients with primary cardiac arrest and those with asphyxial cardiac arrest, demonstrated the initial values of PETCO2could be useful in differentiating the causes of cardiac arrest in a prehospital setting. In asphyxial cardiac arrest, the initial values of PETCO2couldn't predict the outcomes of CPR, as they could do in ventricular fibrillation/ pulseless ventricular tachycardia (VF/VT) arrest.[6,7]In the first study, investigators found PETCO2values after 5 minutes of CPR could predict the outcomes of CPR in VF/ VT and asphyxial cardiac arrest patients, as PETCO2values after 1 minute of CPR could do in the second study.

An experimental study was undertaken to compare changes in PETCO2during CPR in VF versus asphyxial cardiac arrest in piglets.[8]The results showed that the values of PETCO2during the first five breaths of CPR were much higher after asphyxial cardiac arrest than those of VF. The high initial values of PETCO2were correlated with ROSC. After 1 minute of CPR, the values of PETCO2could predict the outcomes of CPR in patients with asphyxial cardiac arrest. However, in cardiac arrest caused by VF, the relationship between PETCO2and ROSC was not further investigated. Most of the other experimental studies evaluated the pattern of PETCO2changes during CPR either in VF arrest or in asphyxial arrest.[9–13]Studies on the patterns of PETCO2changes during CPR in cardiac arrest rats are rare at present.

In this study, we aimed to compare changes in PETCO2during CPR in VF versus asphyxial cardiac arrest in rats and evaluated the relation of PETCO2to ROSC and the mean aortic pressure (MAP).

METHODS

Animal preparation

Experimental studies were approved by the Institutional Animal Ethics Committee of Sun Yat-Sen University. All experiments were performed on male healthy Sprague-Dawley (SD) rats from the Experimental Animal Center of Sun Yat-Sen University, weighing 300–400 g. The rats were randomly divided into an asphyxial group (n=32) and a VF group (n=30). After an overnight fast except for free access to water, the rats were anesthetized by intraperitoneal injection of 45 mg/kg pentobarbital. Additional doses of 10 mg/kg were administered at intervals of approximately 1 hour if necessary. The trachea was orally intubated with a 14 gauge cannula (Abbocath-T, USA).

A gauge 4F polyethylene (PE) catheter (C-PMS-401J, Cook Critical Care, USA) was advanced through the right external jugular vein into the right atrium. A precurved guidewire supplied with the catheter was then advanced through this catheter into the right ventricle until an endocardial electrogram was con fi rmed. The guidewire was used for inducing VF. The surgical procedure of the right external jugular vein was not conducted for the rats in the asphyxia group. Through the left femoral artery, a 23 gauge PE-50 catheter (Abbocath-T, USA) was advanced into the thoracic aorta. Another 23 gauge PE-50 catheter was also advanced through the left femoral vein into the inferior vena cava for infusion. Aortic pressure was measured with a pressure transducer (BD, Germany). Prior to insertion, the catheters were filled with physiological salt solution containing 5 IU/mL of heparin. A rectal thermometer continuously monitored rectal temperature and rectal temperature was maintained at 36.5±0.5 oC with the use of an infrared thermolamp. Electrocardiogram lead II was also continuously monitored. All data were recorded in a six channel recorder (Windaq acquisition system, USA). The rats were mechanically ventilated with a rodent ventilator of our own design. Minute ventilation with a FiO2of 0.21 was maintained at a tidal volume of 0.65 mL/100 g body weight and a frequency of 100 breaths/min. Peak PETCO2measured with a side stream infrared CO2analyzer (CAPSTAR-100, CWE Inc, USA) was maintained between 30 and 40 mmHg. Baseline PETCO2(prior to cardiac arrest), PETCO2after 2 and 6 minutes of VF, and initial PETCO2immediately after PC were recorded.

Cardiac arrest model induced by VF

VF was induced with a 2–5 mA alternating current (60 Hz) delivered to the right ventricular endocardium through the guidewire. The current flow was continued for 3 minutes to prevent spontaneous reversion of VF to a supraventricular rhythm. Mechanical ventilation was discontinued after onset of VF. Six minutes after onset of VF, precordial compression (PC) was initiated with an electrically driven mechanical chest compressor developed in our laboratory. Coincident with the start of PC, mechanical ventilation was resumed. FiO2was increased to 1.0. Compression rate was maintained at a rate of 200/min and synchronized with a compressionventilation ratio of 2:1 with an equal compressionrelaxation duration. The depth of compression was adjusted to maintain a 1/3 decrease in the anteriorposterior chest diameter.[14]Defibrillation was delivered with a 2J direct current at 6 minutes after compression for a maximum of three shocks. If ROSC was not observed, PC was resumed and maintained for 30 seconds before delivering a second set of up to three shocks. ROSC was de fi ned as return of a supraventricular rhythm with MAP of about 60 mmHg lasting for at least 5 minutes.Based on our prior study in which mean ROSC time (an interval from PC to ROSC) in the fi brillatory group was 447 seconds, so PETCO2after 1, 2, 3 and 6 minutes of PC was recorded before defibrillation. The rats regaining ROSC within 6 minutes were excluded in this study.

Cardiac arrest model induced by asphyxia

The prearrest and CPR protocols were similar to the fibrillatory cardiac arrest. However, instead of inducing VF, cardiac arrest was induced by clamping the endotracheal tube. Cardiac arrest was determined by loss of aortic pulsation, de fi ned as MAP equal or less than 20 mmHg. PC and mechanical ventilation were initiated at 6 minutes after onset of cardiac arrest. Based on our prior study in which mean ROSC time in the asphyxial group was 144 seconds, so PETCO2after 1 and 2 minutes of PC was recorded. The rats regaining ROSC within 2 minutes were excluded in this study.

Statistical analysis

SPSS 13.0 software was used for statistical analysis. Continuous variables were presented as means± standard deviation. Data between the two groups of each measurement time were compared by analysis of variance for repeated measures. Receiver operator characteristic (ROC) analysis for PETCO2was performed to determine its ability for predicting ROSC during CPR. The analysis identified threshold values for PETCO2in terms of predicting ROSC. The threshold values of the variable for predicting ROSC were defined as the values with the maximum area of sensitivity multiply speci fi city. Pearson's product-moment correlation coefficient was used for analysis of correlation of PETCO2and MAP during PC. A P value less than 0.05 was considered statistically signi fi cant.

RESULTS

The initial PETCO2immediately after PC was significantly lower in the fibrillatory group than in the asphyxial group (12.8±4.87 mmHg vs. 49.2±8.13 mmHg; P=0.000). There was no significant difference between the groups at baseline PETCO2and PETCO2after 1 and 2 minutes of PC (all P>0.05) (Figure 1).

In the VF group, PETCO2after 6 minutes of PC for the rats with ROSC was signi fi cantly higher than that for those without ROSC (16.5±3.07 mmHg vs. 13.2±2.62 mmHg; P=0.004). There was no significant difference among those with and without ROSC at baseline PETCO2, PETCO2after 2 and 6 minutes of VF, initial PETCO2immediately after PC, and PETCO2after 1, 2 and 3 minutes of PC (all P>0.05). Likewise, MAP after 6 minutes of PC for the rats with ROSC was signi fi cantly higher than that for those without ROSC (37.2±4.87 mmHg vs. 27.6±6.67 mmHg; P=0.000). There was no significant difference among those with and without ROSC at baseline MAP, MAP after 2 and 6 minutes of VF, initial MAP immediately after PC and MAP after 1, 2 and 3 minutes of PC (all P>0.05) (Figures 2 and 3).

Figure 1. Partial pressure of end-tidal carbon dioxide (PETCO2) during PC in VF and asphyxial cardiac arrests.

Figure 2. Partial pressure of end-tidal carbon dioxide (PETCO2) before and after VF cardiac arrest. PC: precordial compression; ROSC: return of spontaneous circulation.

Figure 3. MAP before and after VF cardiac arrest. PC: precordial compression; ROSC: return of spontaneous circulation.

Figure 4. Partial pressure of end-tidal carbon dioxide (PETCO2) before and after asphyxial cardiac arrest. PC: precordial compression; ROSC: return of spontaneous circulation.

Figure 5. MAP before and after asphyxial cardiac arrest. PC: precordial compression; ROSC: return of spontaneous circulation.

Figure 6. ROC curves for partial pressure of end-tidal carbon dioxide (PETCO2) in terms of predicting ROSC when measured at 6 minutes after precordial compression for VFCA rats (A) or at 2 minutes after precordial compression for ACA rats (B). The sensitivities and speci fi cities threshold values for predicting ROSC. ROSC: return of spontaneous circulation; VFCA: ventricular fi brillation cardiac arrest; ACA: asphyxial cardiac arrest.

In the asphyxial group, PETCO2after 2 minutes of PC for the rats with ROSC was significantly higher than that for those without ROSC (20.8±3.24 mmHg vs. 13.9±1.50 mmHg; P=0.000). There was no significant difference among those with and without ROSC at baseline PETCO2, initial PETCO2immediately after PC and PETCO2after 1 minute of PC (all P>0.05). Likewise, MAP after 2 minutes of PC for the animals with ROSC was significantly higher than that for those without ROSC (35.2±5.88 mmHg vs. 27.8±3.67 mmHg; P=0.000). There was no significant difference among those with and without ROSC at baseline MAP, initial MAP immediately after PC and MAP after 1 minute of PC (all P>0.05) (Figure 4 and 5).

ROC analysis was performed and areas under the ROC curves calculated for the PETCO2at 6 minutes after PC in VF cardiac arrest or at 2 minutes after PC in asphyxial cardiac arrest (Figure 6). Areas under the ROC curves were 0.824 for PETCO2in VF cardiac arrest and 0.982 for PETCO2in asphyxial cardiac arrest. The sensitivity and specificity for predicting ROSC at threshold value≥13.2 mmHg in VF cardiac arrest was 91.7% and 66.7% of the PETCO2, respectively. Also, the sensitivity and specificity for predicting ROSC at threshold value≥17.1 mmHg in asphyxial cardiac arrest was 93.8 and 100% of the PETCO2, respectively. During PC, PETCO2was positively correlated with MAP in spite of cardiac arrest induced by VF or asphyxia ( fi brillatory arrest: r=0.618, P=0.000; asphyxial arrest: r=0.832, P=0.000).

DISCUSSION

In this study, our results are consistent with earlier evidence in clinical trial that initial PETCO2is signi fi cantly higher in asphyxial cardiac arrest than that in fi brillatory cardiac arrest.[15]We further af fi rm the fact that initial PETCO2in cardiac arrest may be helpful in differentiating between the causes of cardiac arrest. During CPR in asphyxial cardiac arrest, PETCO2levels are initially high, then fast decrease to subnormal levels and promptly increase again to above-normal or near-normal levels after the onset of ROSC.[8,9]The pattern of PETCO2changes in asphyxial arrest is different from that in fi brillaroty arrest. After the onset of cardiac arrest induced by VF, PETCO2levels rapidly decrease to nearly zero and then begin to increase after providing effective CPR. Upon ROSC, PETCO2levels sharply increase again.[15]During asphyxial arrest, pulmonary blood flow from several minutes of continued cardiac output continues for some period of time. The CO2produced in the tissues during the period will continue to be delivered to the lungs where it diffuses into the alveoli, thereby increasing alveolar CO2. At last, PETCO2levels rapidly increase once mechanical ventilation is resumed.[8,9]

The clinical trials of Lah and Grmec in which initial PETCO2values in primary VF/VT cardiac arrest had a prognostic value for outcomes of ROSC contrasted with the present study.[6,7]Our data demonstrated that initial PETCO2values in fibrillatoty cardiac arrest could not predict the outcomes of ROSC; however, the prognostic value for ROSC was achieved after 6 minutes of CPR in fibrillatory arrest. This may be a result of high-quality CPR delivered to patients with ROSC in primary cardiac arrest, which caused higher initial PETCO2values. The extrapolation has been supported by the evidence that the measurement of PETCO2during CPR serves as a noninvasive monitor of the effective CPR.[16–18]In asphyxial cardiac arrest, initial values of PETCO2do not have a prognostic value for outcomes of ROSC since initial values of PETCO2in asphyxial cardiac arrest generally reflect alveolar CO2prior to CPR. The pattern of MAP changes in fi brillatory cardiac arrest was consistent with that of PETCO2changes, indicating that there was no difference in the effectiveness of PC before 6 minutes of CPR. So there was no difference in initial PETCO2values for rats with and without ROSC in the fi brillatory group. The signi fi cant difference in PETCO2values for rats with and without ROSC was achieved at 6 minutes after CPR in the fibrillatory group. In asphyxial cardiac arrest, the signi fi cant difference in PETCO2and MAP values for rats with and without ROSC was achieved at 2 minutes after CPR. Thereby, PETCO2values after 2 minutes of CPR had a prognostic value for ROSC outcomes in the asphyxial group.

ROC curves of PETCO2showed that threshold values for predicting ROSC in VF and asphyxial cardiac arrest were 13.2 and 17.1 mmHg, respectively. For rats in the group with fibrilllatory cardiac arrest, a PETCO2exceeding 13.2 mmHg after 6 minutes of CPR meant a great success of ROSC. However, if it was less than 13.2 mmHg, the outcome was likely poor. So, a PETCO2less than 13.2 mmHg after 6 minutes of CPR indicated that more aggressive strategies such as high-quality chest compressions would be carried out. The information may be instructive in light of the evidence that high-quality chest compressions improve outcomes of resuscitation.[19]For rats in the group with asphyxial cardiac arrest, if a PETCO2was more than 17.1 mmHg after 2 minutes of CPR, a good outcome would be predicted. Likewise, if it was less than 17.1 mmHg, the outcome would be also poor.

Under constant ventilation, PETCO2reflects cardiac output and pulmonary blood flow during CPR.[20–22]A previous study[23]showed that MAP was a determinant of perfusion during CPR. Positive correlations between PETCO2, MAP, coronary perfusion pressure and cardiac output have been shown in several studies.[11,21,24]In the two cardiac arrest models, PETCO2was also positively correlated with MAP during PC. It should be noted that baseline and post-ROSC values of PETCO2and MAP were not included when the relationship between PETCO2and MAP was analyzed in our experimental study.

Limitation of the present study includes the fact that coronary perfusion pressure is not analyzed. Coronary perfusion pressure is an important determinant of cardiac perfusion during CPR and PETCO2is signi fi cantly correlated with coronary perfusion pressure. Consequently, coronary perfusion pressure should be included in further studies. Another potential limitation is that healthy animals were used in our study. In fact, VF in cardiac arrest patients is mostly caused by ischemic heart disease, and asphyxial cardiac arrest is usually accompanied with intoxication and trauma. Therefore, the findings of this study should be interpreted with caution when applied into clinical practice.

In conclusion, initial PETCO2was signi fi cantly higher in asphyxial cardiac arrest than that in fi brillatory cardiac arrest in our study. In fibrillatory cardiac arrest, PETCO2values after 6 minutes of CPR could predict ROSC outcomes. In asphyxial cardiac arrest, PETCO2values after 2 minutes of CPR could predict ROSC outcomes. Initial PETCO2might be helpful in differentiating the causes of cardiac arrest.

Funding:This study was supported in part by grants from the National Natural Science Foundation of China (30700303), and the National Clinical Key Subject Construction Project.

Ethical approval:The study was approved by the Animal Care and Use Committee of Sun Yat-sen University, Guangzhou, China.

Con fl icts of interest:The authors declare that there is no con fl ict of interest.

Contributors:Lin QM proposed the study, analyzed the data and wrote the fi rst draft. All authors contributed to the design and interpretation of the study and to further drafts.

REFERENCES

1 Field JM, Hazinski MF, Sayre MR, Chameides L, Schexnayder SM, Hemphill R, et al. Part 1: executive summary: 2010 American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Circulation 2010; 122: S640–56.

2 Pokorná M, Necas E, Kratochvíl J, Skripsky R, Andrlík M, Franek O. A sudden increase in partial pressure end-tidal carbon dioxide (PETCO2) at the moment of return of spontaneous circulation. J Emerg Med 2010; 38: 614–621.

3 Sehra R, Underwood K, Checchia P. End tidal CO2is a quantitative measure of cardiac arrest. Pacing. Clin Electrophysiol 2003; 26 (1 Pt 2): 515–517.

4 Kolar M, Krizmaric M, Klemen P, Grmec S. Partial pressure of end-tidal carbon dioxide successful predicts cardiopulmonary resuscitation in the fi eld: a prospective observational study. Crit Care 2008; 12: R115.

5 Weil MH. Partial pressure of end-tidal carbon dioxide predicts successful cardiopulmonary resuscitation in the fi eld. Crit Care 2008; 12: 90.

6 Lah K, Kri?mari? M, Grmec S. The dynamic pattern of end-tidal carbon dioxide during cardiopulmonary resuscitation: difference between asphyxial cardiac arrest and ventricular fibrillation/ pulseless ventricular tachycardia cardiac arrest. Crit Care 2011; 15: R13.

7 Grmec S, Lah K, Tusek-Bunc K. Difference in end-tidal CO2between asphyxia cardiac arrest and ventricular fibrillation/ pulseless ventricular tachycardia cardiac arrest in the prehospital setting. Crit Care 2003; 7: R139–144.

8 Berg RA, Henry C, Otto CW, Sanders AB, Kern KB, Hilwig RW, et al. Initial end-tidal CO2is markedly elevated during cardiopulmonary resuscitation after asphyxial cardiac arrest. Pediatr Emerg Care 1996; 12: 245–248.

9 Luo XR, Zhang HL, Chen GJ, Ding WS, Huang L. Active compression-decompression cardiopulmonary resuscitation (CPR) versus standard CPR for cardiac arrest patients: a metaanalysis. World J Emerg Med 2013; 4: 266–272.

10 Trevino RP, Bisera J, Weil MH, Rackow EC, Grundler WG. Endtidal CO2as a guide to successful cardiopulmonary resuscitation: a preliminary report. Crit Care Med 1985; 13: 910–911.

11 Von Planta M, von Planta I, Weil MH, Bruno S, Bisera J, Rackow EC. End tidal carbon dioxide as an haemodynamic determinant of cardiopulmonary resuscitation in the rat. Cardiovasc Res 1989; 23: 364–368.

12 Kern KB, Sanders AB, Voorhees WD, Babbs CF, Tacker WA, Ewy GA. Changes in expired end-tidal carbon dioxide during cardiopulmonary resuscitation in dogs: a prognostic guide for resuscitation efforts. J Am Coll Cardiol 1989; 13: 1184–1189.

13 Morimoto Y, Kemmotsu O, Murakami F, Yamamura T, Mayumi T. End-tidal CO2changes under constant cardiac output during cardiopulmonary resuscitation. Crit Care Med 1993; 21: 1572–1576.

14 Fang X, Tang W, Sun S, Huang L, Huang Z, Weil MH. Mechanism by which activation of delta-opioid receptor reduces the severity of postresuscitation myocardial dysfunction. Crit Care Med 2006; 34: 2607–2612.

15 Sanders AB, Kern KB, Berg RA. Searching for a predictive rule for terminating cardiopulmonary resuscitation. Acad Emerg Med 2001; 8: 654–657.

16 Fries M, Weil MH, Chang YT, Castillo C, Tang W. Microcirculation during cardiac arrest and resuscitation. Crit Care Med 2006; 34: S454–457.

17 Higgins D, Hayes M, Denman W, Wilkinson DJ. Effectiveness of using end tidal carbon dioxide concentration to monitor cardiopulmonary resuscitation. BMJ 1990; 300: 581.

18 Morimoto Y, Kemmotsu O, Morimoto Y, Gando S. End-tidal carbon dioxide and resuscitation. Curr Opin Anaesthesiol 1999; 12: 173–177.

19 Wu JY, Li CS, Liu ZX, Wu CJ, Zhang GC. A comparison of 2 types of chest compressions in a porcine model of cardiac arrest. Am J Emerg Med 2009; 27: 823–829.

20 Morley PT. Improved cardiac arrest outcomes: as time goes by? Crit Care 2007; 11: 130.

21 Shibutani K, Muraoka M, Shirasaki S, Kubal K, Sanchala VT, Gupte P. Do changes in end-tidal PCO2quantitatively reflect changes in cardiac output? Anesth Analg 1994; 79: 829–833.

22 White RD, Goodman BW, Svoboda MA. Neurologic recovery following prolonged out-of-hospital cardiac arrest with resuscitation guided by continuous capnography. Mayo Clin Proc 2011; 86: 544–548.

23 Gudipati CV, Weil MH, Bisera J, Deshmukh HG, Urban G, Rackow EC, et al. Mean aortic pressure: a monitor of successful CPR. Chest 1986; 89: 446S.

24 Sanders AB, Atlas M, Ewy GA, Kern KB, Bragg S. Expired CO2as an index of coronary perfusion pressure. Am J Emerg Med 1985; 3: 147–149.

Received November 22, 2013

Accepted after revision April 12, 2014

World J Emerg Med 2014;5(2):116–121

10.5847/ wjem.j.issn.1920–8642.2014.02.007