Long-term prevalence of cardiac dysfunction according to acute COVID-19 infection severity

Introduction Myocardial injury has been observed during the acute phase of COVID-19 infection. Whether this persists at long-term follow-up and if there is an association with acute COVID-19 severity is unclear. Aim To investigate the long-term effects of acute COVID-19 infection severity on cardiac structure and function. Method This study included patients with prior confirmed COVID-19 infection who underwent a standardized clinical follow-up approximately 4 years after infection, including transthoracic echocardiography, lung ultrasound (LUS), and symptom questionnaire. Acute COVID-19 severity was categorized as: mild (managed at home), moderate (hospitalised), or severe (intensive care unit (ICU) admission). Associations were examined using linear and logistic regression models adjusted for age, sex, body mass index, and comorbidities, with tests for linear trend across increasing severity class. Results A total of 695 patients were included (52.5% female, median age 61 years [51;70]), distributed in the following severity groups: mild (n=468), moderate (n=179), and severe (n=48). Prevalent comorbidities included hypertension (37.3%), dyslipidaemia (36.7%), asthma (17.1%), atrial fibrillation (11.1%), diabetes (7.9%), and ischemic heart disease (7.8%). (Table 1). The prevalence of ≥3 B-lines on LUS and NYHA ≥II was higher with increasing COVID-19 severity (home, hospital, ICU) (6.4%, 14.4%, 25.6%, p <0.05, and 28.2%, 74.1%, 69.2% p <0.001) (Table 1). Increasing COVID-19 severity was associated with a higher prevalence of left ventricular systolic dysfunction; GLS < 16% (9.7%, 18.0%, 33.3%, p <0.05), LVEF (40-49%) (17.8, 27.7, 30.3%, p <0.05) and LVEF (<40%) (5.9%, 11.8%, and 12.1%, p<0.05) (Figure 1). Prevalence of LV diastolic dysfunction (E/e’ ratio >14) was similar across groups (p=0.35) but became significantly different after adjustment (p <0.05). Right ventricular (RV) function was worse in higher severity groups according to RV fractional area change (42.5% (10.4), 38.5% (11.0), 35.2% (9.1), p <0.001), with association persisting after adjustment. Left atrial (LA) parameters were also worse in higher severity groups (peak atrial longitudinal strain (36.2 [30;44], 32.9 [26;41], 27.5 [24;36], p < 0.001), and LA conduit strain (17.9 [13;24], 14.7 [10;20], 11.0 [9;16], p <0.001), however these trends were attenuated after adjustment. Conclusion Increasing severity of acute COVID-19 infection was associated with long-term echocardiographic signs of worse RV, LA, and LV systolic function, and higher number of B-lines on LUS.Figure 1Table 1