Longitudinal cardiac imaging for assessment of myocardial injury in non-hospitalised community-dwelling individuals after COVID-19 infection: the Rotterdam Study

Heart. 2025 Jan 16:heartjnl-2024-324316. doi: 10.1136/heartjnl-2024-324316. Online ahead of print.

Abstract

Background: The aim of this study was to assess the presence of myocardial injury after COVID-19 infection and to evaluate the relation between persistent cardiac symptoms after COVID-19 and myocardial function in participants with known cardiovascular health status before infection.

Methods: In the prospective population-based Rotterdam Study cohort, echocardiography and cardiovascular magnetic resonance (CMR) were performed among participants who recovered from COVID-19 at home within 2 years prior to inclusion in the current study. Persistent cardiac symptoms comprised only self-reported symptoms of chest pain, dyspnoea or palpitations lasting >4 weeks after COVID-19 infection. We used linear regression and linear mixed models to estimate and test age-adjusted and sex-adjusted mean differences (95% CIs) of (1) post-COVID-19 CMR-derived and echocardiographic-derived parameters among participants with and without persistent post-COVID-19 symptoms and (2) pre-COVID-19 and post-COVID-19 echocardiographic assessments.

Results: 92 participants were included, with a mean age of 59±8 years of whom 52% were male. Normal post-COVID-19 CMR-derived left ventricular (LV) function and right ventricular ejection fraction were observed in 92% and 98% of participants, respectively. We observed normal native T1 relaxation times in 100%, normal extracellular volume in 98% and normal T2 relaxation times in 98% of the participants. Comparison of pre-COVID-19 and post-COVID-19 echocardiography revealed a significant but small decline in left ventricular ejection fraction (adjusted mean change -1.37% (95% CI -2.57%, -0.17%)) and global longitudinal strain (1.32% (95% CI 0.50%, 2.15%)). Comparing participants with and without persistent symptoms, there were no significant differences in adjusted CMR-derived ventricular volumes, LV function or presence of myocardial injury.

Conclusions: Almost all recovered non-hospitalised COVID-19 participants had normal CMR-derived ventricular volumes and function, without relevant myocardial injury.

Keywords: COVID-19; Echocardiography; Magnetic Resonance Imaging.