Background: Guidelines for the management of cardiac arrest play a crucial role in guiding clinical decisions and care. We examined the strength and quality of evidence underlying these recommendations in order to elucidate strengths and gaps in knowledge.

Methods: Using the 2020 American Heart Association (AHA) Guidelines for Adult CPR, we subdivided all recommendations into advanced life support (ALS), basic life support (BLS), and recovery after cardiac arrest, as well as a more granular categorization by topic (i.e. the intervention or evaluation recommended). The Class of Recommendation (COR) and Level of Evidence (LOE) for each were reviewed. Additionally, we reviewed the 2023 guidelines to ensure the inclusion of the most recent updates.

Results: We noted 254 recommendations, of which 181 were ALS, 69 were BLS, and 4 were recovery after resuscitation. In total, only 2 (1%) had the most robust evidence (LOE A), while 23% were at LOE B-NR (Non-Randomized), 15% at LOE B-R (Randomized), 50% at LOE C-LD (Limited Data), and 12% relied on expert opinion LOE C-EO (Expert Opinion). Despite the strength of ALS recommendations (Class 1, 2a, or 2b), none had LOE A. In BLS, no recommendations were supported by LOE A. For BLS, 7% of recommendations had LOE C (C-LD or C-EO). The evidence for specific BLS topics, such as airway management, was notably low. Among ALS topics, neurological prognostication had relatively stronger evidence.

Conclusions: Only 26 out of the 81 COR 1 recommendations (32%) were supported by LOE A or B, indicating a strong discrepancy between the strength of recommendation and the underlying evidence in cardiac arrest guidelines. The findings underscore a pressing need for more rigorous research, particularly randomized trials.