Handling time elements for in-hospital cardiac arrest

 

This commentary refers to ‘Cardiac arrest in COVID-19: characteristics and outcomes of in- and out-of-hospital cardiac arrest. A report from the Swedish Registry for Cardiopulmonary Resuscitation’, by P. Sultanian et al., doi:10.1093/eurheartj/ehaa1067 and the discussion piece ‘In-hospital resuscitation of COVID-19 patients is impeded by serious delays, but the problem is obscured by poor time data’, by J.A. Stewart, doi:10.1093/eurheartj/ehab160.

This is a response to Stewart,¹ commenting on the handling of time data intervals in our publication, Cardiac arrest in COVID-19: characteristics and outcomes of in- and out-of-hospital cardiac arrest. A report from the Swedish Registry for Cardiopulmonary Resuscitation.²

We would like to address and comment on the points being raised by Stewart.

During the COVID-19 pandemic, personal protective equipment (PPE) has become a necessity in the daily work of healthcare personnel (HCP).³ As a consequence, subsequent delays in time response to a cardiac arrest are possible. However, one could also suggest that the pandemic has brought about an unprecedented readiness in the entire healthcare system, especially in the in-hospital setting, where HCP have monitored and cared for patients in an unforeseen manner. Thus, while one could argue that PPE may have caused some delay, one could also argue that patients have been more closely monitored than before. Both these are speculations but reasonably plausible.

The Swedish Registry for Cardiopulmonary Resuscitation has monitored management and outcomes for cardiac arrests in Sweden since 1990. The registry complies with the Utstein style of reporting for variables and outcomes.⁴ Time intervals are registered upon witnessed collapse, followed by the attachment of automated external defibrillators (AED), with recording of the built-in clock for these devices. This could possibly lead to a difference in reporting time elements (from witnessed collapse until attached AED). However, the elapsed time from witnessed arrest until attached AED should not be longer than 1 min. The Swedish Resuscitation Council updated guidelines for management and personal protection for COVID-19 in the emergence of cardiac arrest. The updated recommendations for HCP were to use PPE, including face visor, face mask, protective gloves, and apron. Donning of aforementioned PPE should not account for delaying intervention more than necessary, and definitely not for several minutes.

Moreover, our main analyses are all adjusted for age, sex, and initial rhythm. Adjustment for time delays did not result in any material difference in the results. Claiming that this is due to an inherent flaw in the quality of data is speculative. Also, as in any investigation—be it a randomized trial or observational study—we do not always know which mediators that convey the effect of an initial exposure or risk factor and that does not invalidate the findings; if the lethality of COVID-19 is due to treatment delays, or differences in the underlying aetiology, etc., is of course an important piece of the puzzle, but it does not reduce the importance of the finding that COVID-19 is a hazardous infection that warrants close monitoring and readiness.

Analyzing only monitored arrests, as suggested by Stewart, would lend us to bias caused by cases with pulseless electrical activity, in addition to the fact that half of the study population would be excluded.

Nevertheless, we agree with Stewart that we should always strive for better and more accurate ways of handling time elements for in-hospital cardiac arrest.

Data availability

Data is available upon reasonable request and approval from the Swedish Ethical Review Authority.

Conflict of interest: none declared.