Bye-bye Bicarb? Not So Fast

The role of sodium bicarbonate (“bicarb”) in cardiac arrest has long been a controversial topic with limited evidence. The latest guidance from the American Heart Association does not support routine use of bicarb in cardiac arrest, but could there still be a time and a place for this medication? Like the answer to many questions—it probably depends.

A recent study published in Resuscitation used data from more than 1,300 EMS agencies in the ESO Data Collaborative to dig deeper into the effects of bicarb use in cardiac arrest. But first, let’s talk about how we got here.

Why the Controversy?

Early advanced life support guidelines recommend the use of sodium bicarb to correct metabolic acidosis in patients with cardiac arrest. This medication was used frequently until the mid-1980s. In 1986, The American Heart Association (AHA) first raised concerns about the effectiveness of bicarb when treating patients with cardiac arrest. Guidelines released in 1992 and 2000 discouraged use of bicarb, while the 2010 and 2020 guidelines did not recommend its use, except for under the special circumstances of hyperkalemia or tricyclic antidepressant overdose or cases of severe cardiotoxicity.

Some previous work has suggested that early and more frequent use of bicarb leads to favorable short and long-term outcomes. Meanwhile, other reports failed to establish beneficial effects of bicarb in cardiac arrest.

A 2021 systematic review of bicarb use during resuscitation identified 14 studies (4 randomized controlled trials and 10 observational studies)—five of which were from the United States. The most recent RCT was conducted in 2015 by Ahn et. al., in Korea and included 50 patients who arrived at the ED and did not achieve return of spontaneous circulation (ROSC) after 10 minutes of CPR and who had severe metabolic acidosis. This study found no difference in survival for patients who received bicarb compared to those who did not. The most recent observational study was conducted by Kawano et. al., using data from 2005–2016 in Canada. This study found no difference in survival to hospital discharge for patients who received bicarb; however, the analysis was not separated by initial ECG rhythm.

New Research Linked With Hospital Data

The study used the ESO Data Collaborative to analyze electronic health records from more than 23,000 patients and 1,300 agencies between 2019–2020. Researchers considered cases of patients with cardiac arrest and resuscitations that lasted 5–40 minutes. They stratified the analysis based on presenting electrocardiogram (ECG) rhythms: ventricular fibrillation or ventricular tachycardia (VF/VT), pulseless electrical activity (PEA), or asystole.  

Sara Niederberger is a medical student at the University of Pittsburgh and led the research on the piece in collaboration with ESO’s Remle Crowe and University of Pittsburgh researchers David Salcido and James Menegazzi.

“There are dozens of papers, all with very different results, and a lot of different methodologies—some of which provide stronger evidence than others,” Niederberger said. “It was valuable to have data that looked at the length of resuscitation attempts at different time points. We were able to control for things that a lot of these other studies did not account for.”

Since the data did not come from a randomized controlled trial, the researchers needed a way to control for other variables that could explain differences in ROSC and survival.

In this study, Niederberger used a novel analysis approach known as propensity score matching to create fair comparisons by matching patients from the group that received bicarb to the group that did not receive bicarb based on characteristics such as age, gender, race, witnessed status, bystander CPR, pre-arrival instructions, any defibrillation attempt, use of CPR feedback devices, impedance threshold device (ITD) use, any attempted ventilation, length of resuscitation, and epinephrine given. In contrast to previous research, the researchers also stratified the analysis based on the type of presenting rhythm.

“What we found was that after we split up the groups based on what their presenting rhythm was during the cardiac arrest, we got different results," said Niederberger. "So, interestingly, patients who were presenting with non-shockable rhythms—asystole and PEA—showed benefit with the bicarb use and had significantly greater odds of ROSC. We didn't see any difference with bicarb use for patients who presented in shockable rhythms.”

Implications and Limitations

“While these findings are exciting, it is important to recognize that the study was not a perfectly controlled experiment, so this study can’t establish causality or say definitively that bicarb causes the better outcomes,” said Crowe, ESO’s director of clinical and operational research. “Nevertheless, such a large study coming from so many EMS agencies and with the advanced analysis methods used does tell us there is more to be learned about bicarb use in resuscitation.”

Overall, this research supports the notion that giving bicarb indiscriminately in cardiac arrest is not likely helpful; however, there may be cases when bicarb use is appropriate and is associated with better outcomes—specifically for patients presenting with non-shockable rhythms and longer resuscitation times. The underlying mechanisms for this finding remain to be explored and may be more complicated than correcting metabolic acidosis as the leading hypothesis states. Nevertheless, identifying ways to improve survival for patients presenting in non-shockable rhythms, which historically have been associated with dismal survival rates, will only increase in importance as the epidemiology of cardiac arrest continues to evolve.

Key Points

• Out of more than 23,000 adult patients with cardiac arrest treated by ALS units, bicarb was administered in 28% of cases. The median time to bicarb administration was almost 18 minutes.  
• In propensity-matched analysis, patients who received bicarb were more likely to survive hospital discharge in the asystole (3.3% vs 2.4%) and PEA (8.1% vs. 5.4%) groups. 

Request the full manuscript here.