CMR Outcomes in Patients Recovered From COVID-19
- In a cohort of unselected patients recovered from COVID-19, abnormal CMR findings were present in 78%, indicative of myocardial edema and/or inflammation in most and pericardial involvement in some.
- While these findings could explain common symptoms in COVID-19 convalescents, including chest pain and dyspnea, further study is needed to determine how CMR should be employed in clinical practice.
What are the cardiovascular magnetic resonance (CMR) findings in unselected patients with recent coronavirus disease 2019 (COVID-19)?
This prospective observational cohort study enrolled patients recovered from COVID-19, identified from the University of Frankfurt COVID-19 Registry between April and June 2020. Patients were considered eligible if a minimum of 2 weeks had elapsed following initial diagnosis, respiratory symptoms had resolved, and upper respiratory swab was negative for SARS-CoV-2 at the end of the isolation period. Patients referred for clinical CMR were excluded. Comparisons were made with age- and sex-matched healthy controls and with risk factor-matched patients, with comorbidities including hypertension, diabetes, and coronary artery disease. All participants underwent blood sampling for high-sensitivity troponin T (hsTnT) and N-terminal pro-B-type natriuretic peptide (NT-proBNP). The CMR protocol included native T1 mapping (to assess for diffuse fibrosis and/or edema), native T2 mapping (to assess for edema), and late gadolinium enhancement (LGE) imaging (indicative of inflammation and/or scar).
The cohort included 100 patients (53% male, median age 49 years, 33% requiring hospitalization, 2% requiring mechanical ventilation, 17% undergoing noninvasive ventilation with positive airway pressure). During hospitalization, 15% had significantly elevated hsTnT values (≥13.9 pg/ml). Median time between positive upper respiratory swab and CMR was 71 days. Persistent symptoms at the time of CMR included atypical chest pain (17%), palpitations (20%), and shortness of breath and general exhaustion (36%). hsTnT was detectable (3 pg/ml or higher) in 71% and significantly elevated in 5%.
Abnormal CMR findings were present in 78% of patients: increased myocardial native T1 (73%), increased myocardial native T2 (60%), myocardial LGE (32%, of whom 12/32 had an ischemic LGE pattern), and pericardial enhancement (22%). Native T1 and native T2 were correlated with hsTnT (r = 0.35, p < 0.001 and r = 0.22, p = 0.03, respectively). Time from COVID-19 diagnosis was not correlated with hsTnT, native T1, or native T2. Endomyocardial biopsy was performed in three patients with severe abnormalities (including high T1 and T2, LGE, and left ventricular ejection fraction <50%), revealing active lymphocytic inflammation with no detectable viral genome. In receiver operating characteristic analyses, native T1 and T2 were best for distinguishing recovered COVID-19 patients from controls (native T1 area under the curve 0.71, p < 0.001, cutoff value 1140 ms; native T2 area under the curve 0.73, p < 0.001, cutoff value 40 ms).
In this cohort of patients recently recovered from COVID-19, a majority had abnormal CMR findings, illustrating a spectrum of disease including myocardial edema and/or inflammation, as well as pericardial involvement.
In these patients, many of whom had relatively mild COVID-19 disease, native T1 and T2 mapping were more frequently abnormal than LGE imaging. Importantly, T1 and T2 mapping are not included in standard cardiomyopathy CMR protocols at most institutions. T2 mapping is routinely employed in myocarditis protocols to assess for myocardial edema, while T1 mapping is still emerging in clinical settings and is not well standardized. It is tempting to speculate that myocardial inflammation and edema could be contributing to symptoms such as chest pain and dyspnea in COVID-19 convalescents. However, further study will be needed to determine if and how abnormal CMR findings should impact clinical management and follow-up in this patient population.
Clinical Topics: COVID-19 Hub, Heart Failure and Cardiomyopathies, Noninvasive Imaging, Pericardial Disease, Prevention, Atherosclerotic Disease (CAD/PAD), Acute Heart Failure, Heart Failure and Cardiac Biomarkers, Magnetic Resonance Imaging, Hypertension
Keywords: Chest Pain, Contrast Media, Coronary Artery Disease, Coronavirus, COVID-19, Diabetes Mellitus, Diagnostic Imaging, Edema, Fibrosis, Gadolinium, Heart Failure, Hypertension, Inflammation, Magnetic Resonance Imaging, Myocarditis, Natriuretic Peptide, Brain, Pericardial Effusion, Respiration, Artificial, Risk Factors, Secondary Prevention, Stroke Volume, Troponin T, Ventilators, Mechanical
< Back to Listings