Role of sodium bicarbonate in resuscitation of out-of-hospital cardiac arrest

Transcription

1 Hong Kong Journal of Emergency Medicine Role of sodium bicarbonate in resuscitation of out-of-hospital cardiac arrest CW Chung, CT Lui, KL Tsui Objective: To report the use of sodium bicarbonate (SB) in out-of-hospital cardiac arrest (OHCA) and to explore the role of sodium bicarbonate in resuscitation of OHCA in terms of restoration of spontaneous circulation (ROSC) and survival, and existence of dose-dependent relationship, especially in those with prolonged arrest. Design: Retrospective cohort study. Setting: Emergency department of two regional hospitals in a cluster of Hong Kong. Methods: Adult patients aged at least 18 years old who presented to the study centres with non-traumatic OHCA in the period between March 2013 and December 2013 were included. Cases in which resuscitations were considered medically futile or not actually performed were excluded. Those with do-not-resuscitate (DNR) order or advance directives, those who were death before arrival with postmortem changes, and those who developed ROSC before or within 15 minutes of arrival were excluded from the study. Patients with known poisoning from tricyclic antidepressant or other sodium channel blockers were excluded from analysis if any. The primary outcome of this study was ROSC. Other outcome variables included survival to hospital admission (STA) and survival to hospital discharge (STD). Results: A total of 489 patients were included during the study period for analysis. We found that patients who received sodium bicarbonate injection (SB group) during CPR had a higher percentage of ROSC than those who did not (no-sb group) (60.8% vs 22.5%; p<0.001). The survival-to-admission rate was higher in the SB group (56.8%) compared with the contrary (21.4%). The difference of survival-to-discharge between the two groups did not reach statistical significance (4.1% in SB group and 2.9% in the no-sb group; p=0.484). The rate of ROSC was found to be dose dependent, being higher in higher dose of SB administration. The ROSC rate increased from no-sb (22.4%) to 57.1% if given 50 ml SB, and further to 64.1% if given 100 ml SB. Conclusions: Our study shows that the use of sodium bicarbonate in the CPR of OHCA is beneficial in ROSC. The effect is dose dependent, with better results in higher dose (>100 ml) of sodium bicarbonate; however, we fail to demonstrate its benefit for prolonged CPR cases (>30 minutes). (Hong Kong j.emerg. med. 2015;22: ) ( >18 ) Correspondence to: Lui Chun Tat, FHKCEM, FHKAM(Emergency Medicine) Tuen Mun Hospital, Department of Accident and Emergency, Tsing Chung Koon Road, Tuen Mun, N.T., Hong Kong ectlui@yahoo.com.hk Chung Chi Wai, MBBS(HK), MCEM(Emergency Medicine) Tsui Kwok Leung, FRCSEd, FHKCEM, FHKAM(Emergency Medicine)

2 282 Hong Kong j. emerg. med. Vol. 22(5) Sep % (60.8%) (56.8%) 50 ml 22.4% 57.1% 100 ml (>30 ) Keywords: Buffers, cardiopulmonary resuscitation, heart arrest, treatment outcome Introduction The use of sodium bicarbonate (SB) during cardiopulmonary resuscitation (CPR) was originally based on the belief of its action in correcting metabolic acidosis in cardiac arrest and hence improved the outcomes in cardiac arrest. 1 But the 1986 update of the AHA Guidelines had raised major reservations regarding the necessity, efficacy and side-effects of SB, 2 and the 1992, and versions had deemphasized its use. Some studies even showed hypernatraemia, alkalosis, and excess carbon dioxide production had been associated with sodium bicarbonate injection during CPR. 6 As a result, the 2010 American Heart Association guidelines for advanced cardiac life support did not recommend the routine use of sodium bicarbonate during CPR, except for pre-existing metabolic acidosis, hyperkalaemia, and tricyclic antidepressant intoxication. 5 All those studies and guidelines caused a marked decline in the use of SB, one of the major changes in the conduction of ACLS over the past 15 years. 7 The fact that practically all recent animal experiments demonstrated its beneficial effect on both short- and long-term outcome 8,9 had been overlooked. Despite the fact that the use of sodium bicarbonate during cardiopulmonary resuscitation had been controversial for years, studies delineating the role of SB in resuscitation of cardiopulmonary arrest were still scanty, particularly evidence from the Asian-pacific area. The role of SB in promoting restoration of spontaneous circulation (ROSC) and in improving long-term outcome has not been evaluated in a welldesigned controlled trial. 10 The outcome of cardiac arrest and CPR, not responding to early defibrillation and requiring ACLS, is still very poor. 11 Bar-Joseph et al reported that the resuscitation outcomes in emergency medical systems improved in correlation with the increased use of sodium bicarbonate during CPR. 12 Further studies and randomised controlled trials (RCTs) by Vukmir et al failed to demonstrate the benefit of buffer therapy in out-of-hospital cardiac arrest (OHCA), but it showed a trend towards an improvement in the outcomes of prolonged resuscitative efforts associated with the use of sodium bicarbonate. 13 In this study, we aimed to report the use of SB in OHCA in local emergency department and to explore the role of SB in resuscitation of OHCA in terms of ROSC and survival, and existence of dose-dependent relationship, especially in those with prolonged arrest. Methods Study design and setting This was a retrospective observational study conducted at the emergency department of two regional hospitals in a cluster of Hong Kong. The cluster served a population of more than a million people. The daily visits of the emergency departments were approximately 600 and 350 respectively. All cardiopulmonary arrest patients would be directed to either hospital according to the proximity. Prehospital

3 Chung et al./sodium bicarbonate in resuscitation 283 management of cardiopulmonary arrest was carried out by the prehospital personnel of the Fire Service Department who were trained for Basic Life Support (BLS) and specific skills in management of the cardiac arrested. All prehospital resuscitation team had members of EMA-2. Automated external defibrillators (AED) were attached to all patients in cardiac arrest. In the two emergency departments of study centres, resuscitations were carried out based on the ACLS 2010 guideline, with adjustments according to the clinical decision of the attending physician. The study was approved by the hospital ethics committee. The study protocol was reviewed and approved for exemption of informed consent. Patient inclusion We included all adult patients aged at least 18 years old who presented to the study centres with nontraumatic OHCA in the period between March 2013 and December Patients in which resuscitation would be considered medically futile by the in-charge physician and those who had not been resuscitated were excluded. Those with do-not-resuscitate (DNR) order or advance directives, those who were death before arrival with postmortem changes, and those who developed ROSC before or within 15 minutes of arrival were excluded from the study. Patients with known poisoning from tricyclic antidepressant or other sodium channel blockers were excluded from analysis if any. Figure 1 illustrated the patient enrollment of our study. Data collection For all patients with cardiac arrest, the attending physician would acquire the necessary data of prehospital management and resuscitation and submit to the local cardiac arrest registry. Data were collected prospectively in Utstein style. Data collected includes patients' demographics, witnessed collapse, bystander CPR, time factors, prehospital rhythms and management. EMS response time was defined as the time of ambulance call to the time of arrival of ambulance crews to patient's side. Time to advanced life support (ALS) was defined as the time of ambulance call to the arrival time to emergency department. Patient would be classified as presumed cardiac aetiology of cardiac arrest if this was a witnessed arrest, shockable rhythms, preceding chest pain or history of unstable cardiac diseases. Patient would be classified as presumed non-cardiac aetiology if cardiac arrest was related to respiratory diseases, intracranial lesion, poisoning or other known non-cardiac aetiology. The in-hospital resuscitation details including the use of sodium bicarbonate, the dose and timing of administration, the acid-base and electrolyte status of the patients were retrieved and recorded. Whether the attending physician got experience or qualification of fellowship of emergency medicine was also retrieved to analyse the prescription practice. Patients' outcomes were traced by electronic hospital records. Outcome measures The primary outcome of this study was ROSC, which was defined as sustainable spontaneous circulation with palpable pulse for at least 10 minutes. Other outcome variables included survival to hospital admission (STA) and survival to hospital discharge (STD). Statistical tests The statistical software package IBM SPSS version 20 for Windows was employed for analysis. Ages was shown as mean and standard deviations within normal distribution and were compared with independent sample t-test for the two groups. Skewed continuous variables such as time factors were described with median and interquartile range (IQR). Values were compared with Mann-Whitney U test. Categorical variables were shown in proportions and percentages that were compared with Chi-square test. P value of less than 0.05 was taken to imply statistical significance. Dose-dependent relationship was shown in plots for various doses of SB on the outcome. Predictors of survivals were entered together with the use of SB to a binominal logistic regression model for confounding control. Potential predictors of ROSC in OHCA which was biologically plausible and reported in the literature were included in the model. Defibrillations were included in the model while presumed aetiology and initial rhythms were omitted due to significant overlap. SB was categorised with different doses of 50 ml and 100 ml or above. Patients

4 284 Hong Kong j. emerg. med. Vol. 22(5) Sep 2015 who were not given SB were assigned as the reference group. Adjusted odd ratios were calculated along their confidence intervals. P values were calculated for individual predictors. To illustrate the effect of SB in the outcomes for patients with short versus long down time, relative risk of ROSC were calculated for two groups with arrest to ED arrival time <30 minutes and 30 minutes. Relative risk reduction, absolute risk reduction and the number need to treat of using SB to gain one ROSC were calculated along with the 95% confidence intervals. Results A total of 585 patients were recorded OHCA in the cardiac arrest registry from March 2013 to December Ninety-six patients were excluded with the study flow as shown in Figure 1. A total of 489 patients were included during the study period for analysis, with a male predominance (59.1%) and a mean age of 71.8 years. Patients were divided into two groups based on whether they were treated (SB group, n=74) or not treated (no-sb group, n=415) with sodium bicarbonate. Comparing the two groups (Table 1), there were differences in the demographic characteristics shown by statistical test. Patients with younger age, male gender, presumed cardiac aetiology, initial rhythm of VT/VF, those received defibrillation, and those being treated by fellow physician were more likely to be given SB. For the prescription practice of SB by emergency physicians for OHCA, we found that patients with younger age were more likely to be given SB during CPR (65.9 vs 72.8 years old; p=0.025) (Table 1). For those patients with presumed cardiac cause of death, a higher proportion of them received SB during CPR Figure 1. Patient enrollment of the study.

5 Chung et al./sodium bicarbonate in resuscitation 285 (52.7% compared with non-cardiac 30.4%; p=0.001). SB was also more likely to be given for those OHCA patients with initial rhythm being VF / pulseless VT and for those who had received defibrillation (Table 1). We also found that the in-charge physician with fellowship in EM specialty had higher tendency to give SB to OHCA patients during CPR (47.3% in fellows vs 34.7% in non-fellows; p=0.038). For those 74 patients who were given SB during CPR, 65 of them (87.8%) had blood gas and electrolytes taken (Table 2). The median ph value was 6.92 and the median serum potassium level was 4.5. The mean dose of 8.4% SB was 81.9 ml while there were 39 patients (52.7%) received more than 100 ml SB (Table 2). In the univariate analysis, patients who received sodium bicarbonate injection during CPR had a higher percentage of ROSC than those who did not (60.8% vs 22.5%; p<0.001). The survival-to-admission rate was higher in the SB group (56.8%) compared with the contrary (21.4%). The difference of survival-todischarge between the two groups did not reach statistical significance (4.1% in SB group and 2.9% in the no-sb group; p=0.484). The rate of ROSC was found to be dose dependent, being higher in higher dose of SB administration. The ROSC rate increased from no-sb (22.4%) to 57.1% Table 1. Baseline characteristics and outcome of patients All patients (N=489) SB given (N=74) SB not given (N=415) P value Age (mean, SD) 71.8, , , Male gender 289 (59.1%) 52 (70.3%) 237 (57.1%) Presumed cause of arrest Cardiac 165 (33.7%) 39 (52.7%) 126 (30.4%) Non cardiac 95 (19.4%) 12 (16.2%) 83 (20.0%) Unknown 229 (46.8%) 23 (31.1%) 206 (49.6%) Witnessed arrest 228 (46.6%) 46 (62.2%) 182 (43.9%) Bystander CPR 109 (22.3%) 10 (13.5%) 99 (23.9%) First rhythm in prehospital <0.001 VF/pulseless VT 53 (10.8%) 17 (23.0%) 36 (8.7%) PEA/asystole 436 (89.2%) 57 (77.0%) 379 (91.3%) First rhythm in A&E <0.001 VF/pulseless VT 16 (3.3%) 9 (12.2%) 7 (1.7%) PEA/asystole 473 (96.7%) 65 (87.8%) 408 (98.3%) Defibrillation 103 (21.1%) 30 (40.5%) 73 (17.6%) <0.001 Prehospital 75 (15.3%) 23 (31.1%) 52 (12.5%) <0.001 A&E 54 (11.0%) 19 (25.7%) 35 (8.4%) <0.001 Time factors (median, IQR) EMS response time, min 8, 5 9, 5 8, Time to ALS, min 18, 6 18, 6 18, In-charge physician Fellow 179 (36.6%) 35 (47.3%) 144 (34.7%) ROSC 138 (28.2%) 45 (60.8%) 93 (22.5%) <0.001 Survival to admission 131 (26.8%) 42 (56.8%) 89 (21.4%) <0.001 ICU admission 22 (16.8%) 7 (16.7%) 15 (16.8%) Survival to discharge 15 (3.1%) 3 (4.1%) 12 (2.9%) SD=standard deviation; CPR=cardiopulmonary resuscitation; VF=ventricular fibrillation; VT=ventricular tachycardia; PEA=pulseless electrical activity; A&E=accident and emergency department; IQR=interquartile range; EMS=emergency medical service; ALS=advanced life support; ROSC=restoration of spontaneous circulation; ICU=intensive care unit

6 286 Hong Kong j. emerg. med. Vol. 22(5) Sep 2015 if given 50 ml SB, and further to 64.1% if given 100 ml SB (Figure 2). In the logistic regression predicting ROSC (Table 3), witnessed arrest (OR 3.30 p<0.001) and EMS response time (OR 0.94; p=0.026) were independent predictors of ROSC. In addition, with reference group of no-sb given, the odds ratio of ROSC in group of patients received 50 ml 8.4% SB was 2.87 (95% CI ; p=0.022). The odds ratio was 4.9 (95% CI ; p<0.001) when 100 ml 8.4% SB was given which was compatible with the dose-dependent relationship (Figure 2). Subgroup analysis on SB injection during prolonged CPR (arrest to ED arrival time 30 minutes) was shown in Table 4. In both groups ROSC rate was higher when SB was given. The relative ratio of ROSC was 2.41 ( ) in the prolonged arrest group while the value was 2.85 ( ) in the contrary. The number need to treat (NNT) was comparable in both groups (3.2 and 2.5 respectively). Discussion After over 40 years of modern CPR, the outcome of OHCA was still very poor. The overall rate of ROSC of OHCA was less than 30%, with the survival to discharge rate of only about 3%. Local studies also showed the outcome of OHCA had been poor Cardiopulmonary arrest acutely decreases cardiac output, coronary blood flow and respiration. Perfusion failure results in tissue hypoxia, depletion of high energy phosphate stores and anaerobic metabolism. All these conditions contribute to a mixed respiratory and metabolic acidosis during cardiopulmonary arrest. Bicarbonate functions as a physiological buffer. 20 Administration of sodium bicarbonate was regarded as an essential component in CPR for many years after modern CPR was introduced in clinical practice in the early 1960s. 1 Patients in cardiac arrest were often found to be acidotic. Multiple old animal studies demonstrated that myocardial function was compromised by acidosis 21,22 and that Adrenaline was less effective in acidosis. 23,24 Those facts made physiological sense that correction of acidosis should be beneficial in successful resuscitation. However administration of sodium bicarbonate during CPR simply reflected the presumption that the acidosis originating from ischaemic injury of cardiac arrest was of itself detrimental for successful resuscitation. 25 Levine et al had demonstrated that in some situations where there was acidosis, the acidosis was actually protective. 26 Table 2. Sodium bicarbonate use pattern Variables Results Taking blood gas and electrolytes 65/74 (87.8%) Serum potassium Median, IQR 4.5, 2.0 K> (25.7%) Base deficit (Median, IQR) 17.0, 4.9 Serum bicarbonate (Median, IQR) 16.5, 5.9 ph (Median, IQR) 6.92, 0.21 Time from A&E arrival to first dose of SB, 21.0, 11.3 minutes (Median, IQR) Dose of SB 8.4%, in ml (Mean, SD) 81.9, 30.5 Dose >100 ml 8.4% SB 39/74 (52.7%) IQR=interquartile range; A&E=accident and emergency department; SD=standard deviation Figure 2. Dose dependent relationship of sodium bicarbonate on rate of survival-to-admission.

7 Chung et al./sodium bicarbonate in resuscitation 287 More studies also demonstrated potential negative effects of sodium bicarbonate administration in CPR and postcardiac arrest conditions. 27 These included venous hypercapnoea with an increase of mixed venous CO2, 28 leading to a decrease of tissue intracellular ph, 29 circulatory congestion, hypernatraemia and hyperosmolality with brain damage. 30 It may also cause alkalosis, excess production of carbon dioxide, central venous acidosis with inactivation of adrenaline administration, and compromised coronary flow. 6,21 Thus the American Heart Association recommended that sodium bicarbonate should not be used routinely in CPR. 3 Several international critical care organisations even recommended that sodium bicarbonate should be given Class 3 status (inappropriate, without scientific evidence of efficacy) for CPR. 31 Disparate opinions concerning the appropriate indications for use of sodium bicarbonate in patients with acidosis in cardiac arrest become an important topic in modern CPR. The American Heart Association certainly did not assign sodium bicarbonate as Class 3. It is a Class 1 for preexisting hyperkalaemia, Class 2a (likely to be beneficial) for preexisting metabolic acidosis or tricyclic or barbiturates overdose, and Class 2b (acknowledging the lack of well-established evidence) for prolonged cardiac arrest or upon ROSC. 3 The 1992 National Conference on Cardiopulmonary Resuscitation and Emergency Cardiac Care classified sodium bicarbonate as a Class 2b intervention that could be used after prolonged intervals of cardiac arrest. 32 Table 3. Logistic regression of ROSC predicted by SB use and other variables Variables Odds ratio 95% confidence interval P value Age Defibrillation Witnessed arrest <0.001 Bystander CPR EMS response time NaHCO3 dose Not given (reference group) 1 50 ml ml <0.001 Hosmer-and-Lemeshow goodness-of-fit test p= CPR=cardiopulmonary resuscitation; EMS=emergency medical service Table 4. Effects of sodium bicarbonate (SB) on restoration of spontaneous circulation (ROSC) in arrest patients with various times from arrest to accident and emergency department (AED) arrival Arrest to AED arrival time <30 minutes Arrest to AED arrival time 30 minutes SB given No SB p value SB given No SB P value ROSC < Yes 26 (61.9%) 62 (21.8%) 15 (53.6%) 26 (22.2%) No RR of ROSC 2.85 ( ) 2.41 ( ) RRR 0.51 ( ) 0.59 ( ) ARR 0.40 ( ) 0.31 ( ) NNT 2.49 ( ) 3.19 ( ) RR=relative ratio; RRR=relative risk reduction; ARR=absolute risk reduction; NNT=number needed to treat.

8 288 Hong Kong j. emerg. med. Vol. 22(5) Sep 2015 The use of sodium bicarbonate seems to be fallen out of favour in CPR over the past 15 years. However it has never been studied in a prospective controlled clinical trial. Subsequent investigations to determine the effects of sodium bicarbonate administration in CPR have largely been uncontrolled, descriptive experimental an adult human studies. The only prospective, controlled clinical trial on buffer therapy in CPR was conducted by Dybvik et al who have compared Tribonat1 (a mixture of SB, tromethanol, disodium phosphate and acetate) to normal saline during CPR of OHCA. 33 Buffer therapy did not improve immediate resuscitability or long-term outcome. There were some published clinical studies retrospectively evaluated the effect of SB on CPR outcome. Most of them demonstrated poor outcome associated with its use. 34,35 However, these studies did not prove that SB worsened CPR outcome. They merely demonstrated that more SB was administered in more prolonged resuscitation, which were obviously associated with worse outcome. 36 The likelihood of successful defibrillation and resuscitation decreases as the duration of cardiac arrest increases. Prolonged cardiac arrest is also associated with more development of acidosis. 9 Our study showed the use of sodium bicarbonate in the CPR of OHCA was beneficial in ROSC, but failed to demonstrate its benefit during prolonged CPR. Vukmir et al and Leong et al showed in prolonged cardiac arrest, sodium bicarbonate administration improved the survival rate and neurological outcome in dogs. 8,9 Bar-Joseph et al reported that the resuscitation outcomes in emergency medical systems improved in correlation with the increased use of SB during CPR. 12 Further studies and RCTs by Vukmir et al failed to demonstrate the benefit of buffer therapy in OHCA, but it showed a trend towards an improvement in the outcomes of prolonged resuscitative efforts associated with the use of sodium bicarbonate. 13 Mattar et al found that large amount of sodium bicarbonate administration in CPR resulted in severe hypernatraemia, hyperosmolality, increased lactic acidosis and 100% mortality rate. 37 But we found the rate of STA with the use of SB was dose dependent, being higher in higher dose of SB administration (about 100 to 200 ml 8.4% NaHCO3). However, the survival-to-discharge rate could not be assessed by this retrospective study with limited sample size. An observation about the prescription behaviour of SB in resuscitation is that generally physicians would be more likely to prescribe SB in those patients who were believed to have higher chance of survival e.g. younger patients, patients with cardiac cause of arrest, first rhythm in A&E being shockable, defibrillated, etc. The use of sodium bicarbonate in CPR and whether to discard it during CPR was still a controversial topic. Our study showed the use of sodium bicarbonate in the CPR of OHCA was beneficial in ROSC, which was also dose dependent. But we failed to demonstrate its benefit on survival-to-discharge. The fact that many different studies demonstrated contradictory results making whether to use sodium bicarbonate or to discard it during CPR a serious dilemma in the management of OHCA. In the modern days of evidence-based medicine, we may need a large prospective clinical trial involving multi-centre to provide a more convincing answer and approach to the role of sodium bicarbonate in the CPR of OHCA. Limitations We used the rate of ROSC as a surrogate marker for cardiac arrest. STD should be a better outcome variable. However it was not feasible in our study due to the low STD rate. The relatively small simple size also limited the general applicability of this study. Furthermore, our study design is a retrospective cohort study where there is still possibility of uncontrolled residual confounders. A randomised controlled trial with adequate sample size would be better to illustrate the causal relationship and the role of SB in resuscitation of OHCA. Conclusions The use of sodium bicarbonate in CPR of OHCA and whether to discard it during CPR is still a controversial

9 Chung et al./sodium bicarbonate in resuscitation 289 topic. There are still disparate opinions concerning the appropriate indications for use of sodium bicarbonate in the CPR of patients with cardiac arrest. Our study shows the use of sodium bicarbonate in the CPR of OHCA is beneficial in ROSC. The effect is dose dependent, with better results in higher dose (>100 ml) of SB. However, the effect is not demonstrated for cases with prolonged CPR (>30 minutes). As it is an important topic in modern CPR, a multi-centre large scale RCT should be done to give the answer. References 1. Standards for cardiopulmonary resuscitation (CPR) and Emergency cardiac care (ECC). JAMA 1974;227 (7): Standards and Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiac Care. JAMA 1986;255(21): Guidelines for cardiopulmonary resuscitation and emergency cardiac care. Emergency Cardiac Care Committee and Subcommittees, American Heart Association. Part III. Adult advanced cardiac life support. JAMA 1992;268(16): Guidelines 2000 for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Part 6: advanced cardiovascular life support: section 6: pharmacology II: agents to optimize cardiac output and blood pressure. The American Heart Association in collaboration with the International Liaison Committee on Resuscitation. Circulation 2000;102(8 Suppl):I Neumar RW, Otto CW, Link MS, Kronick SL, Shuster M, Callaway CW, et al. Part 8: adult advanced cardiovascular life support: 2010 American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Circulation 2010;122(18 Suppl.3): Kette F, Weil MH, Gazmuri RJ. Buffer solutions may compromise cardiac resuscitation by reducing coronary perfusion pressure. JAMA 1991;266(15): Levy RD, Rhoden WE, Shearer K, Varley E, Brooks NH. An audit of drug usage for in-hospital cardiopulmonary resuscitation. Eur Heart J 1992;13 (12): Vukmir RB, Bircher NG, Radovsky A, Safar P. Sodium bicarbonate may improve outcome in dogs with brief or prolonged cardiac arrest. Crit Care Med 1995;23(3): Leong EC, Bendall JC, Boyd AC, Einstein R. Sodium bicarbonate improves the chance of resuscitation after 10 minutes of cardiac arrest in dogs. Resuscitation 2001; 51(3): Weng YM, Wu SH, Li WC, Kuo CW, Chen CY, Chen JC. The effects of sodium bicarbonate during prolonged cardiopulmonary resuscitation. Am J Emerg Med 2013; 31(3): Becker LB. The epidemiology of sudden death. In: Paradis NA, Halperin HG, Nowak RM, eds. Cardiac Arrest: the Science and Practice of Resuscitation Medicine. Baltimore: Williams & Wilkins, 1996; Bar-Joseph G, Abramson NS, Kelsey SF, Mashiach T, Craig MT, Safar P. Improved resuscitation outcome in emergency medical systems with increased usage of sodium bicarbonate during cardiopulmonary resuscitation. Acta Anaesthesiol Scand 2005;49(1): Vukmir RB, Katz L. Sodium bicarbonate improves outcome in prolonged prehospital cardiac arrest. Am J Emerg Med 2006;24(2): Wong TW, Yeung KC. Out of hospital cardiac arrest: two and a half years experience of an accident and emergency department in Hong Kong. J Accid Emerg Med 1995;12(1): Leung LP, Lo CM, Tong HK. Prehospital resuscitation of out-of-hospital cardiac arrest in Queen Mary Hospital. Hong Kong J Emerg Med 2000;7(4): Fan KL, Leung LP. Prognosis of patients with ventricular fibrillation in out-of-hospital cardiac arrest in Hong Kong: prospective study. Hong Kong Med J 2002;8(5): Chung CH, Wong PCY. A six-year prospective study of out-of-hospital cardiac arrest managed by a voluntary ambulance organization. Hong Kong J Emerg Med 2005;12(3): Leung KL, Lui CT, Cheung KH, Tsui KL, Tang YH. Outcome and prognostic factors of patients in out-ofhospital cardiac arrests presenting with non-shockable rhythm in Hong Kong. Hong Kong J Emerg Med 2012; 19(1): Chan TH, Lui CT, Cheung KH, Tang YH, Tsui KL. Outcome predictors of patients in out-of-hospital cardiac arrests with pre-hospital defibrillation in Hong Kong. Hong Kong J Emerg Med 2013;20(3): Wyckoff MH, Perlman J, Niermeyer S. Medications during resuscitation what is the evidence? Seminars in Neonatalogy 2001;6(3): Adler S, Roy A, Relman AS. Intracellular acid-base regulation. I. The response of muscle cells to changes in CO2 tension or extracellular bicarbonate concentration. J Clin Invest 1965;44: Anderson MN, Mouritzen CV. Effect of acute respiratory and metabolic acidosis on cardiac output and peripheral resistance. Ann Surg 1966;163(2): Preziosi MP, Roig JC, Hargrove N, Burchfield DJ. Metabolic academia with hypoxia attenuates the hemodynamic responses to epinephrine during resuscitation in lambs. Crit Care Med 1993;21(12): Downing SE, Talner NS, Gardner TH. Cardiovascular responses to metabolic acidosis. Am J Physiol 1965;208: Gazmuri RJ. Buffer treatment for cardiac resuscitation:

10 290 Hong Kong j. emerg. med. Vol. 22(5) Sep 2015 putting the cart before the horse? Critical Care Medicine 1999;27(5): Levine RL. Ischemia: From acidosis to oxidation. FASEB J 1993;7(13): Arieff AI. Current concepts in acid-base balance: Use of bicarbonate in patients with metabolic acidosis. Curr Anaesth Crit Care 1996;7: Weil MH, Rackow EC, Trevino R, Grundler W, Falk JL, Griffel MI. Difference in acid-base state between venous and arterial blood during cardiopulmonary resuscitation. N Engl J Med 1986;315(3): Arieff AI, Leach W, Park R, Lazarowitz VC. Systemic effects of NaHCO3 in experimental lactic acidosis in dogs. Am J Physiol 1982;242(6):F Simmons MA, Adcock EW 3rd, Bard H, Battaglia FC. Hypernatremia and intracranial hemorrhage in neonates. N Engl J Med 1974;291(1): Dobb GJ. New guidelines for cardiopulmonary resuscitation and emergency cardiac care. Intensive Care World 1993;9: Guidelines for cardiopulmonary resuscitation and emergency cardiac care. Emergency Cardiac Care Committee and Subcommittees, American Heart Association. Part III. Adult advanced cardiac life support. JAMA 1992;268(16): Dybvik T, Strand T, Steen PA. Buffer therapy during out-of hospital cardiopulmonary resuscitation. Resuscitation 1995;29(2): vanwalraven C, Stiell IG, Wells GA, Herbert PC, Vandemheen K for the OTAC Study Group. Do advanced cardiac life support drugs increase resuscitation rates from in-hospital cardiac arrests? Ann Emerg Med 1998;32(5): Roberts D, Landolfo K, Light RB, Dobson K. Early predictors of mortality for hospitalized patients suffering cardiopulmonary arrest. Chest 1990;97(2): Aufderheide TP, Martin DR, Olson DW et al. Prehospital bicarbonate use in cardiac arrest: a 3-year experience. Am J Emerg Med 1992;10(1): Mattar JA, Weil MH, Shubin H, Stein L. Cardiac arrest in critically ill. II. Hyperosmolar states following cardiac arrest. Am J Med 1974;56(2):162-8.