a b s t r a c t

Background: This study was performed to determine the effects of sodium bicarbonate injection during prolonged cardiopulmonary resuscitation (for N 15 minutes).

Methods: The retrospective cohort study consisted of adult patients who presented to the emergency department (ED) with the diagnosis of cardiac arrest in 2009. Data were retrieved from the institutional database.

Results: A total of 92 patients were enrolled in the study. Patients were divided into 2 groups based on whether they were treated (group1, n = 30) or not treated (group 2, n = 62) with sodium bicarbonate. There were no significant differences in demographic characteristics between groups. The median time interval between the administration of CPR and sodium bicarbonate injection was 36.0 minutes (IQR: 30.5-41.8 minutes). The median amount of bicarbonate injection was 100.2 mEq (IQR: 66.8-104.4). Patients who received a sodium bicarbonate injection during prolonged CPR had a higher percentage of return of spontaneous circulation, but not statistical significant (ROSC, 40.0% vs. 32.3%; P = .465). Sustained ROSC was achieved by 2 (6.7%) patients in the sodium bicarbonate treatment group, with no survival to discharge. No significant differences in vital signs after ROSC were detected between the 2 groups (heart rate, P = .124; systolic blood pressure, P = .094). Sodium bicarbonate injection during prolonged CPR was not associated with ROSC after adjust for variables by regression analysis (Table 3; P = .615; odds ratio, 1.270; 95% confidence interval: 0.501-3.219)

Conclusions: The administration of sodium bicarbonate during prolonged CPR did not significantly improve the rate of ROSC in out-of-hospital cardiac arrest.

(C) 2013

Introduction

Background

The use of buffer therapy in cardiac arrest is a source of debate. The use of sodium bicarbonate, which is the principal drug for buffer therapy, was originally based on the assumption that correcting metabolic acidosis could improve the outcomes in cardiac arrest. Bar- Joseph et al. reported that the resuscitation outcomes in emergency medical systems improved in correlation with the increased use of sodium bicarbonate during cardiopulmonary resuscitation (CPR) [1]. However, further studies and Randomized controlled trials failed to demonstrate the benefit of buffer therapy in out-of-hospital cardiac arrest (OHCA) [2-4]. Furthermore, hypernatremia, alkalosis, and excess carbon dioxide production have been associated with sodium bicarbonate injection during CPR [4-6]. Therefore, the 2010 American Heart Association guidelines for advanced cardiac life support did not recommend the routine use of sodium bicarbonate

? Conflict of interest statement: The authors have no conflicts of interest to declare.

* Corresponding author. Tel.: +886 3 3281 200×2505; fax: +886 3 3287 715.

E-mail address: [email protected] (S-Y. Chen).

during CPR, except for pre-existing metabolic acidosis, hyperkalemia, and tricyclic antidepressant intoxication [7].

In contrast to the routine use of sodium bicarbonate during CPR, the use of sodium bicarbonate during prolonged CPR could be beneficial. Animal studies have shown the benefits of buffer therapy during prolonged cardiac arrest [8-11]. In one RCT, Vukmir et al showed a trend towards an improvement in the outcomes of prolonged Resuscitative efforts (N 15 minutes) associated with the use of sodium bicarbonate [3]. In addition, the low prevalence of bystander CPR and the high proportion of patients with an Initial shockable rhythm may have contributed to prolonged arrest [12-15]. Although public education and rising awareness of the importance of bystander CPR is crucial [16,17], the role of buffer therapy during prolonged CPR should also be explored. Nevertheless, the number of human studies investigating the effects of sodium bicarbonate use during prolonged CPR is limited. There are currently no guidelines regarding the use of sodium bicarbonate in OHCA after prolonged CPR on admission to the emergency department (ED).

Goals of this investigation

The present study aimed to determine the effects of sodium bicarbonate during prolonged CPR (N 15 minutes) in OHCA [3]. We

0735-6757/$ - see front matter (C) 2013 http://dx.doi.org/10.1016/j.ajem.2012.11.001

Y-M. Weng et al. / American Journal of Emergency Medicine 31 (2013) 562-565 563

hypothesized that the use of sodium bicarbonate during prolonged CPR may improve the rate of return of spontaneous circulation (ROSC) and survival to discharge in patients with OHCA.

Methods

Study design and setting

This was a retrospective cohort study conducted at a university- affiliated teaching hospital with approximately 130,000 visits annually. Patients were treated by emergency physicians. An estimated ROSC, sustained ROSC and survival to discharge rate of conventional resuscitation standards were 30%, 15% and 6% in our city pre-hospital arrests, respectively. This study was approved by the Hospital Ethics Committee on Human Research. The study protocol was reviewed and qualified as exempt from the requirement to obtain informed consent.

Patient selection

The study cohort consisted of adult patients aged more than 18 years who presented to the ED with the diagnosis of cardiac arrest between January 1 and December 31, 2009. Patients with International Classification of Diseases, 9th Edition (ICD-9) cardiac arrest Diagnosis codes 7981, 7982, and 7989 were included in the study. Patients for whom emergency medical services (EMS) were not activated, who had incomplete documentation or were missing blood gas tests or chemical studies, and those who developed ROSC in 15 minutes or received no resuscitation because of do-not-resuscitate (DNR) orders

were excluded from the study. Patients were assigned to group 1 or 2 according to whether they received sodium bicarbonate injection or not, respectively.

Methods and measurements

We enrolled OHCA patients for whom EMS was activated with standard CPR at the scene. Automatic electrical defibrillator (AED) was applied to all patients once emergency medical technicians (EMT) arrived and confirmed the cardiac arrest event within 5 minutes or immediately after one cycle of CPR. Patients were then transferred to our hospital by ambulance while receiving continuous CPR. Patients received advanced cardiac life support according to the 2005 American Heart Association guidelines that included chest compression and emergency endotracheal intubation with mechan- ical ventilator support, intravenous injection of adrenaline, and cardioversion at the ED [18]. A blood sample was collected and sent for laboratory tests, which included blood gas analysis, complete blood cell count, and serum biochemistry. ROSC was defined as any signs of circulation without chest compression. The term sustained ROSC was used when chest compressions were not required for 20 consecutive minutes and signs of circulation persisted [19]. Patients with sustained ROSC were then transferred to intensive care units for further care as indicated.

The first author performed the chart review and data abstraction using a standardized template with clear definitions and codes. The corresponding authors provided quality improvement feedback after data analysis during the study by holding periodic meetings. The variables collected from the EMS sheet included a witnessed collapse,

Two hundred and fourteen adults aged more than 18 years with major diagnosis of cardiac arrest*

Incomplete medical record, N = 33 DNR orders, N = 25

No blood tests available, N = 11

Ninety-two patients received prolonged CPR (>5 minutes)

One hundred and forty-five patient records were reviewed

ROSC within 15 minutes, N = 18 No EMS activated, N = 35

Group 2: 62 patients did not receive sodium bicarbonate injection during prolonged cardiac arrest (>15 minutes) ROSC, N = 20

Sustained ROSC (>20 minutes), N = 12 Survive to discharge, N = 2

Group 1: 30 patients received sodium bicarbonate injection during prolonged cardiac arrest (>15 minutes)

ROSC, N = 12

Sustained ROSC (>20 minutes), N = 2 Survive to discharge, N = 0

Abbreviations: ICD, international classification of disease code; DNR, Do-not-resuscitate; CPR, cardiopulmonary resuscitation; ROSC, return of spontaneous circulation.

Fig. Patients enrolled in the study. Numbers of patients included and excluded from the study. *Adults aged more than 18 years, with ICD-9 cardiac arrest diagnosis codes 7981, 7982, and 7989 were included. Patients for whom emergency medical services were not activated, who had incomplete documentation or lacked results of blood gas tests or chemical studies, and those who developed ROSC in 15 minutes did not receive CPR because of DNR orders were excluded from the study.

564 Y-M. Weng et al. / American Journal of Emergency Medicine 31 (2013) 562-565

bystander CPR, shockable rhythm by AED, and time from scene to hospital. Patient demographics (age, sex, cardiac rhythm at the ED,

time of sodium bicarbonate injection), and Laboratory test results during prolonged CPR
(blood gas analysis: pH, bicarbonate, serum potassium level) were	Group 1a		Group 2b	P
abstracted from the medical charts.	N = 30		N = 62
2.4. Outcome measures Male gender, N (%)	19 (63.3)		47 (75.8)	.213
Age in years; median (IQR)	65.0 (51.0-		70.5	.433
	80.5)		(51.5-
The primary outcome of this study was the rate of survival to			83.0)
discharge, which was defined as being discharged alive or able to be Witnessed collapse, N (%)	17 (56.7)		28 (45.2)	.301
transferred to a long-term care center. The secondary outcomes Bystander CPR, N (%)	5 (16.7)		16 (25.8)	1.000?
Shockable rhythm by AED, N (%)	1 (3.3)		2 (3.2)	.454?
included, percentage of ROSC, sustained ROSC, time to ROSC, and vital Time from scene to ED by EMS, minutes; median	21.5 (14.0-		20.0	.846
signs after ROSC (heart rate, systolic blood pressure). (IQR)	25.5)		(15.0-
			26.5)
2.5. Statistical tests Cardiac rhythm at ED, N (%) Asystole	23 (76.7)	57 (91.9)		.118
VF/pulseless VT	2 (6.7)		1 (1.6)

Table 1

Baseline characteristics of patients who received or did not receive buffer therapy

using the ?² test. The Mann-Whitney U test was used for continuous variables as indicated. In all analyses, P b .05 was taken to indicate statistical significance.

Sodium bicarbonate amount; mEq, median (IQR) 100.2 - -

(66.8-

The data were analyzed using SPSS 13.0 for Windows (SPSS,	PEA	0	4 (6.5)
Chicago, IL). Demographic characteristics and patient outcomes were compared between groups. Categorical variables were compared	Time from CPR by EMT to sodium bicarbonate injection, minutes; median (IQR)	36.0 (30.5- 41.8)	-	-

104.4)

Laboratory tests, median (IQR)

Results

A total of 214 patients with ICD-9 cardiac arrest diagnosis codes 7981, 7982, and 7989 were recruited during the study period. Of

Blood gas, pH 6.9 (6.8-

7.0)

Blood gas, bicarbonate, mm/L 14.5 (17.6-

20)

Potassium, mEq/L 6.3 (4.9-

8.9)

6.9 (6.7-

7.1)

17.7 (9.6-

21.9)

6.2 (4.8-

8.2)

.680

.522

.597

these, 122 patients were excluded from the study: 33 had incomplete records, 11 lacked results of blood gas tests or chemical studies, 25 had DNR orders, 35 had no EMS activated, and 18 developed ROSC within 15 minutes. The high exclusion rate is due to our strictly defined inclusion criteria, which minimized the influence of selection bias. Finally, 92 patients were included in the analysis, with a higher proportion of men (71.3%) and a median age of 52.0 years (interquartile range [IQR]: 34.0-70.0). Patients were divided into 2 groups based on whether they were treated (group1, n = 30) or not treated (group 2, n = 62) with sodium bicarbonate (Fig.). No significant differences in demographic characteristics were detected between groups (Table 1). The median time interval between the administration of CPR and sodium bicarbonate injection was 36.0 minutes (IQR: 30.5-41.8 minutes). The median amount of bicarbon- ate injection was 100.2 mEq (IQR: 66.8-104.4). Table 2 shows the outcomes of patients in the 2 groups. Patients who received sodium bicarbonate injection during prolonged CPR had a higher percentage of ROSC than those who did not (40.0% vs 32.3%), but this difference did not reach statistical significance (P = .465). In group 1, 2 (6.7%) patients achieved sustained ROSC and survival to discharge was not observed. In group 2, 12 (19.4%) patients had sustained ROSC and 2 (3.2%) survived to discharge. ROSC was delayed in group 1 compared with group 2 (median 48.0 vs. 38.5 minutes; P = .025). No significant differences in vital signs after ROSC were detected (median heart rate, 101.5 vs 116.5 beats per minute, P = .124; median systolic blood pressure, 81.5 vs 110.0 mmHg, P = .094). Sodium bicarbonate injection during prolonged CPR was not associated with ROSC after adjust for variables by regression analysis (Table 3; P = .615; odds ratio, 1.270; 95% confidence

interval: 0.501-3.219).

Discussion

The Effects of Sodium bicarbonate use during prolonged CPR

In the present study, the outcome of OHCA with prolonged CPR (N 15 minutes) was poor, with a rate of survival to discharge of 2.2% (2/92). Although the use of sodium bicarbonate during prolonged CPR increased the rate of ROSC (40.0 vs. 32.3 %), but

VF, ventricular fibrillation; VT, ventricular tachycardia; PEA, pulseless electrical activity. Variables were tested using the ?² test (male, Mode of transportation, witnessed collapse, cardiac rhythm at ED) and Mann-Whitney U test (age, duration of resuscitation intervention, laboratory tests).

^a Group 1: Patients received sodium bicarbonate infusion.

^b Group 2: Patients did not receive sodium bicarbonate infusion.

* Variables were tested using the Fisher’s exact test.

not statistical significant. Only 6.7% of the patients had sustained ROSC (N 20 minutes) and survival to discharge was 0% in buffer therapy group. Despite the small sample size with limited outcomes of sustained ROSC or survival to discharge, our study showed no additional benefit of ROSC with sodium bicarbonate injection during prolonged CPR. Further large-scale study might reveal a statistical Survival benefit. The median time from CPR to ROSC was significantly longer in patients treated with sodium bicarbonate during prolonged CPR. Previous animal studies showed that buffer therapy was beneficial in cardiac arrest, especially in prolonged arrest [8-11]. However, most of these studies measured ROSC, the level of acidosis, or coronary perfusion pressure as primary outcomes instead of sustained ROSC or long-term survival. Vukmir et al. performed a prospective double-blinded RCT evaluating the effect of sodium bicarbonate (1 mEq/kg) injection in addition to advanced cardiac life support by EMT in the pre- hospital setting [3]. The results of this study showed a trend

Table 2

Study outcomes

	Group 1a		Group 2b	P
	N = 30		N = 62
ROSC, N (%)	12 (40.0)		20 (32.3)	.465
Sustained ROSC (N 20 minutes), N (%)	2 (6.7)		12 (19.4)	.134?
Survival to discharge, N (%)	0 (0)		2 (3.2)	1.000?

Variables were tested using the ?² test.

^a Group 1: Patients received sodium bicarbonate infusion.

^b Group 2: Patients did not receive sodium bicarbonate infusion.

* Variables were tested using the Fisher’s exact test.

Y-M. Weng et al. / American Journal of Emergency Medicine 31 (2013) 562-565 565

Table 3 Sodium bicarbonate injection during prolonged CPR was not associated with ROSC after adjusting for variables by regression analysis

	Odds ratio (95% CI)	P
Sodium bicarbonate infusion during prolonged CPR	1.270 (0.501-3.219)	.615
Witness collapse	2.443 (0.899-6.640)	.080
Bystander CPR	0.834 (0.245-2.842)	.771

toward improvement in the outcome of prolonged (N 5 minutes) cardiac arrest (32.8% vs 15.4%; P = .007). The outcome was defined as the presence of a pulse on admission to the ED. Our study also showed an improvement in the rate of ROSC, but not statistical significant. Furthermore, the use of sodium bicarbonate in anticipation of a prolonged arrest and resuscitation effort is not warranted. In conclusion, the results or our study indicate that the use of sodium bicarbonate during prolonged CPR is not recom- mended. In cases requiring prolonged CPR, the focus should be the CPR delivery and identification of the cause of arrest rather than the administration of buffer therapy.

Possible explanations of the study results

There are several possible explanations for the results of the present study. First, the high mortality rate was probably independent of the transient ROSC during prolonged CPR. A vicious cycle of metabolic acidosis caused by the arrest and prolonged CPR may have further compromised the cardiac response and delayed the ROSC. Although buffer therapy improved the acidosis and increased the chance of achieving ROSC, the mortality rate remained high. In a double-blind RCT, Dybvik et al concluded that in patients resuscitated after OHCA, who had metabolic acidosis, buffer therapy did not improve the outcome [2]. Second, patients who did not achieve ROSC and required prolonged CPR were more likely to have an unidentified cause of arrest. Despite the fact that ROSC occurred after the administration of sodium bicarbonate during prolonged CPR, sus- tained ROSC or survival to discharge were not achieved without identifying and correcting the cause of the arrest. In cases in which the cause of the arrest was identified and ROSC was achieved during prolonged CPR, the use of sodium bicarbonate showed no additional benefit on sustained ROSC and survival to discharge. Third, several adverse effects have been associated with the use of sodium bicarbonate, which may increase the severity of the post-cardiac arrest condition. This may explain why sustained ROSC was achieved in a limited number of cases and none showed survival to discharge. The adverse effects associated with the use of sodium bicarbonate include alkalosis, hypernatremia, excess production of carbon dioxide, central venous acidosis with inactivation of epinephrine administra- tion, and compromised coronary flow [4-6]. The influence of survival outcomes of the adverse effects associated with sodium bicarbonate use is unknown to date. Weight-based mEq/kg dosing may mitigate adverse effects of excessive bicarbonate dosing. However, the evidence of optimized dosing is frail under the concern of the benefit of sodium bicarbonate use in such condition.

Limitations

The present study should be interpreted in the context of the following limitations. First, our study was retrospective, and the data were collected from a computer database and by chart review. Although we made every effort to remain objective, possible errors may have been introduced. Second, there might be selection bias due to the retrospective design and the large proportion of excluded patients who presented without activation of EMS, complete documentation or blood gas tests, or resuscitation because of DNR orders, and those who developed ROSC in 15 minutes. In accordance

with such strict and clear defined criteria, all enrolled patients had non-traumatic OHCA with activation of EMS and prolonged CPR. In addition, none of the patients had incomplete documentation or blood test results. There were no significant differences in the backgrounds of the groups. Third, there might have been unmeasured confounding factors in our study. However, variables that potentially correlated with our study results according to previous reports were considered in the data analysis. Fourth, this study was conducted in a university- affiliated teaching hospital with a small sample size, which may limit the general applicability of our findings. A comparative and large- scale study would be of interest. Fifth, the adult body weight was not routine checked in our ED. Limited patients who admitted to intensive care unit had their body weight taken. Therefore, we could not do analysis of weight versus non-weight based bicarbonate effect.

Conclusions

Our study failed to demonstrate the benefit of using sodium bicarbonate during prolonged CPR.The administration of sodium bicar- bonate did not significant improve the rate of ROSC in such condition.

References

1. 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;49:6-15.
2. Dybvik T, Strand T, Steen PA. Buffer therapy during out-of-hospital cardiopulmo- nary resuscitation. Resuscitation 1995;29:89-95.
3. Vukmir RB, Katz L. Sodium bicarbonate improves outcome in prolonged prehospital cardiac arrest. Am J Emerg Med 2006;24:156-61.
4. Aufderheide TP, Martin DR, Olson DW, Aprahamian C, Woo JW, Hendley GE, et al. Prehospital bicarbonate use in cardiac arrest: a 3-year experience. Am J Emerg Med 1992;10:4-7.
5. Kette F, Weil MH, Gazmuri RJ. Buffer solutions may compromise Cardiac resuscitation by reducing coronary perfusion pressure. JAMA 1991;266:2121-6.
6. Graf H, Leach W, Arieff AI. Evidence for a detrimental effect of bicarbonate therapy in hypoxic lactic acidosis. Science 1985;227:754-6.
7. 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:S729-67.
8. Liu X, Nozari A, Rubertsson S, Wiklund L. Buffer administration during CPR promotes cerebral reperfusion after return of spontaneous circulation and mitigates post-resuscitation cerebral acidosis. Resuscitation 2002;55:45-55.
9. Bar-Joseph G, Weinberger T, Castel T, Bar-Joseph N, Laor A, Bursztein S, et al. Comparison of sodium bicarbonate, Carbicarb, and THAM during cardiopulmo- nary resuscitation in dogs. Crit Care Med 1998;26:1397-408.
10. 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;23:515-22.
11. Leong EC, Bendall JC, Boyd AC, Einstein R. Sodium bicarbonate improves the chance of resuscitation after 10 minutes of cardiac arrest in dogs. Resuscitation 2001:51309-15.
12. Herlitz J, Engdahl J, Svensson L, et al. Factors associated with an increased chance of survival among patients suffering from an out-of-hospital cardiac arrest in a national perspective in Sweden. Am Heart J 2005;149:61-6.
13. Rea TD, Cook AJ, Stiell IG, et al. Predicting survival after out-of-hospital cardiac arrest: role of the Utstein data elements. Ann Emerg Med 2010;55:249-57.
14. Ko PC, Ma MH, Yen ZS, et al. Impact of community-wide deployment of biphasic waveform automated external defibrillators on out-of-hospital cardiac arrest in Taipei. Resuscitation 2004;63:167-74.
15. Hu SC, Tsai J, Lu YL, et al. EMS characteristics in an Asian metropolis. Am J Emerg Med 1996;14:82-5.
16. Iwami T, Kawamura T, Hiraide A, Berg RA, Hayashi Y, Nishiuchi T, et al. Effectiveness of bystander-initiated cardiac-only resuscitation for patients with out-of-hospital cardiac arrest. Circulation 2007;116:2900-7.
17. SOS-KANTO Study Group. Cardiopulmonary resuscitation by bystanders with chest compression only (SOS-KANTO): an observational study. Lancet 2007;369:920-6.
18. ECC Committee, Subcommittees and Task Forces of the American Heart Association. 2005 American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Circulation 2005;112:IV 1-IV 203.
19. Jacobs I, Nadkarni V, Bahr J, et al. Cardiac arrest and cardiopulmonary resuscitation outcome reports: update and simplification of the Utstein templates for resuscitation registries. A statement for healthcare professionals from a task force of the International Liaison Committee on Resuscitation (American Heart Association, European Resuscitation Council, Australian Resuscitation Council, New Zealand Resuscitation Council, Heart and Stroke Foundation of Canada, Inter American Heart Foundation, Resuscitation Council of Southern Africa). Resusci- tation 2004:233-49.