The authors reply

We agree with Barros et al (1) that our findings are considerably different than theirs. In the first 1,094 consecutive hospitalized coronavirus disease 2019 (COVID-19) patients at our institution, the overall mortality was 13.6% (1), as we reported in our recently published article (2) in Critical Care Medicine. The attempted resuscitation was significantly higher, more than seven times in our hospitalized COVID-19 patients compared with the authors’ (5.8% vs 0.8%). Although the pre-arrest therapies and features of cardiac arrest events are similar, we are not sure if the baseline comorbidities of our patients are comparable to that of the authors’ institution. Our hospitalized COVID-19 patients have a higher burden of comorbidities compared with general U.S. population (3). The likelihood of success from cardiopulmonary resuscitation depends on the cause of arrest as well as on the health status of the patient (4). A number of prearrest and intra-arrest factors associated with poor survival after in-hospital cardiac arrest (IHCA) were present in our patients.

We will like to point out that the attempt by Barros et al (1) to compare our patient population to theirs is problematic for a number of reasons. First, there is significant difference in the total number of patients that underwent resuscitation (63 vs 7), which may account for some difference in survival. Second, authors mention that 66% of survivors had no neurologic deficits, however, even in their limited data, the overall survival with no neurologic deficits would be about 28% (2/7). Third, a significant number of patients at the authors’ institution had a do-not-resuscitate order entered 12 hours before their death. This will easily account for the difference in number of resuscitated patients between our institution and authors’ institution and possibly the rate of survival.

We agree with the authors that in select patients hospitalized with COVID-19 who experience cardiac arrest, attempts at resuscitation are appropriate and potentially lifesaving. As at the time we submitted our article, there was no data in the United States about the survival to discharge in COVID-19 patients suffering IHCA. Indeed, the title and conclusions of our article were meant to provoke more conversation and highlight the need for more data about IHCA in COVID-19 patients and not to label cardiopulmonary resuscitation futile. All critically ill patients should have early goals of care conversations to clarify treatment preferences regardless of code status.