ABSTRACT
Introduction: COVID-19 can be related with a poor clinical outcome. ECG abnormalities in COVID-19 have been widely described, but literature on the predictive value of a 12-lead ECG at hospital admission and normalization of these abnormalities after infection is limited. Purpose: To describe the predictive value of ECG abnormalities on admission and after recovery of COVID-19. Methods: After informed consent patients older than 18 years admitted with COVID-19 between March and July 2020 were included in a prospective registry. Diagnosis was confirmed by PCR-assay or based on suggestive clinical and radiological presentation. Demographic and clinical data were collected by review of the electronic medical record. All ECGs from admission until last follow-up were assessed lead by kead for repolarization abnormalities. The index ECG was defined as first ECG available after admission, a post-COVID ECG was obtained after hospital discharge in the absence of acute pathology. Minor abnormalities included iso-electric T-waves and ST-depression ≤2 mm. Major abnormalities were ST-depression >2 mm, ST-elevation, biphasic T-waves and T-wave inversion. Myocardial regions were defined as anterior (V1-V4), lateral (I, aVL, V5, V6) and inferior (II, III, aVF). Patients with a ventricular pacemaker were excluded. Results: A total of 283 patients were included, median age 65 years and 64.7% were male. The 30-day mortality rate was 20.5%. In 96.8% of patients an ECG was available within 48 hours after admission. Repolarization abnormalities were observed in 48.8% of patients. In 27.2% this was limited to minor abnormalities. Abnormal repolarization was related to age, cardiovascular medical history, renal function, high-sensitive troponin-T and NT-proBNP levels. There were no significant differences in clinical presentation, ICU admission, need for ventilation or ECMO. On Kaplan-Meier analysis (figure) the presence (p < 0.001) and the extent of repolarization abnormalities (p < 0.001) were associated with 30-day mortality. Forward Cox regression modelling identiefied age (per year, HR 1.07, 95% CI 1.05-1.09), history of heart failure (HR 2.12, 95% CI 1.08-4.52), neurological disorders (HR 2.47, 95% CI 1.36-4.51), active oncological disease (HR 2.13, 95% CI 1.01-4.50) and the extent of repolarization abnormalities (per region, HR 1.37, 95% CI 1.05-1.79) as independent predictors. A post-COVID ECG was available in 172 patients (60.8%), the median time between index and post-COVID ECG was 63.3 days. There was 1 new first-degree AV-block and 1 new RBBB. Repolarization abnormalities were present in 32 patients (11.3%);however, only 3 patients (1.7%) had new abnormalities, 2 of whom died during further follow-up. Conclusions: The extent of repolarization abnormalities on an ECG at admission for COVID-19 is an independent predictor of 30-day mortality. New ECG abnormalities after COVID-19 infection are uncommon but may be associated with adverse outcome.