Cardiac Involvement After Recovering From COVID-19

To the Editor I read with interest the article by Puntmann et al¹ regarding the cardiac magnetic resonance (CMR) imaging findings in a group of 100 patients with asymptomatic to a severe course of coronavirus disease 2019 (COVID-19) after a median of 2 to 3 months from diagnosis. I was surprised by the high frequency of cardiac involvement still present in that group, including signs of fibrosis in 78% of patients and ongoing inflammation in 60% of them. The study drew large media attention around the world and brings serious clinical concerns regarding the potential need for in-depth cardiologic screening in all patients after recovering from COVID-19.

The recent study by my coauthors and me² on persistent subclinical changes in CMR imaging after a median of 7 months from myocarditis in children revealed ongoing active inflammation in 28% of patients and healed myocarditis with persistent scars in 44% of patients. Therefore, a 2-fold lower frequency of persistent myocardial changes, detected later from the onset of the disease, was observed in our group of adolescents with confirmed myocarditis. We did not use T1 and T2 mapping techniques, which are able to detect a subtler myocardial injury or ongoing myocardial inflammation,³ but the difference was still remarkably large, especially considering asymptomatic or mildly affected patients composed a significant amount of the post–COVID-19 group.²

Therefore, I would like to raise some comments regarding the study by Puntmann et al,¹ which might explain the high frequency of cardiac involvement detected. First, the application of the modified Lake Louise criteria,⁴ requiring the presence of both T1 and T2 relaxation time elevation to diagnose an ongoing inflammation, was not reported in the article, and although T1 and T2 usually rise in parallel, on that occasion, the frequency of ongoing myocarditis, at least theoretically, could range from 38% to 60%. Second, Puntmann et al¹ used reference criteria for cutoff values of elevated T1 and T2 from studies performed on a 3-T Philips scanner, while their study was done on a 3-T Siemens scanner. In my opinion, these values are not interchangeable. The reference ranges reported for a 3-T Philips scanner are much lower than reference values for a 3-T Siemens scanner, which have been summarized in our report² and are higher by approximately 100 milliseconds for T1 and 10 milliseconds for T2.⁵ Therefore, a cutoff point could have been set too low, leading to many false-positive findings. However, only 3 patients with the most severe course had a myocardial biopsy to verify that hypothesis.

Corresponding Author: Łukasz A. Małek, MD, PhD, Department of Epidemiology, Cardiovascular Disease Prevention and Health Promotion, National Institute of Cardiology, Niemodlińska str 33, 04-635 Warsaw, Poland (lmalek@ikard.pl).

Published Online: October 28, 2020. doi:10.1001/jamacardio.2020.5276

Conflict of Interest Disclosures: None reported.