Impact of previous cardiac function status assessed by echocardiography on the outcome of COVID-19.

Abstract

More than 91,000 fatalities due to Coronavirus Disease 2019 (COVID-19) have occurred in Spain. Several factors are associated with increased mortality in this disease, including cardiovascular risk factors (CVRF). However, information on the cardiac function of patients prior to the onset of COVID-19 is scarce and the potential impact it may have is uncertain. The aim of the EchoVID study was to describe the potential association between CVRF and cardiac function status prior to SARS-CoV-2 infection and in-hospital mortality. We studied clinical characteristics and cardiac function of patients admitted during the first wave of COVID-19. All patients had a transthoracic echocardiogram performed in the previous 12 months prior to diagnosis; conventional systolic and diastolic function parameters were analyzed. Logistic regression analysis was performed to identify predictors of in-hospital mortality. We included 296 individuals. Median age was higher in the group of patients who died (81.0 vs 76.1 years; p = 0.007). No significant differences were found in CVRF. Survivors were more frequently receiving anticoagulation therapy (52.9% vs 70.8%; p = 0.003). LVEF, although preserved on average in both groups, was significantly lower in the group of deceased patients (56.9% vs 61.1%; p = 0.017). Average E/e' ratio was higher in the deceased group (11.1 vs 10.1; p = 0.049). Five variables were found to be independently associated with in-hospital mortality due to COVID-19: Age, male gender, LVEF, E/e' ratio and anticoagulation therapy. A model including these variables had an area under the ROC curve of 0.756 (CI 0.669-0.843). The echocardiographic variables included in the model significantly improved the discriminative power, compared to a model including only demographic data. Left ventricular ejection fraction and E/e' ratio prior to SARS-CoV-2 infection are two easily-obtained echocardiographic parameters that provide additional prognostic information over clinical factors when assessing patients admitted for SARS-CoV-2 infection.