Echocardiographic markers of mortality at 60 days in patients with COVID-19 in the intensive care unit

Abstract

Abstract Introduction Coronavirus disease (COVID)-19 predominantly produces its effects through lung damage, but an important component of multi-organ dysfunction is cardiac involvement. We have few reports that inform about the behavior of echocardiographic images of patients with the most severe forms of the disease. Purpose The present work aims to identify prognostic markers for 60-day mortality in patients hospitalized in intensive care based on echocardiographic findings. Methodology A single-center retrospective cohort was conducted. Hospitalized patients were included in one of the nine intensive care units for COVID-19 confirmed by RT-PCR from May to October 2020. Patients with previous conditions that determined a limitation of the therapeutic effort, those who died before 24 hours and pregnant women were excluded. Portable echocardiograms were performed by two expert cardiologists following the recommendations for isolation and personal protection. The time to death was evaluated as outcome. A Cox proportional hazards model was constructed, HR and 95% confidence intervals with their p values. The study was approved by the institutional ethics committee. Results Of 326 patients included, 153 patients had an echocardiogram. The mean age was 60.7 years, 47 (30.7%) were female and 67 (44.7%) had positive troponin. 91 patients (59.5%) not survive, the mean long of stay was 8.4 (SD: 4.2) days. 111 (72.5%) had shock, 128 (83.7%) severe ARDS (PaO2 / FiO2 <100 mmHg), 142 (92.8%) required invasive ventilatory support, and 86 (56.2%) acute kidney injury. 27 (17.6%) patients had acute pulmonary embolism, 16 (10.4%) acute myocardial infarction and 9 (5.9%) myocarditis. The mean right ventricular ejection fraction was 37%, TAPSE was decreased in 16 cases (10.4%). 41 cases (26.8%) had right diastolic dysfunction. 34/48 (71%) cases had pulmonary hypertension. The average LVEF was 59.3% and 74 (48.4%) had some left ventricular diastolic dysfunction. 12 (7.8%) had left ventricular segmental wall motion abnormality and 16 (10.4%) had pericardial effusion. Univariate analysis identified TAPSE, PSAP, acute cor pulmonale and right ventricular dilatation as variables related to the outcome of mortality. The multivariate Cox model (Table 2) documented that acute cor pulmonale with a HR of 12.8 (95% CI 3.51 - 46.63, p<0.001) and right ventricular dilation with a HR of 4, 87 (95% CI 1.36–17.46, P 0.016) were associated with mortality. Conclusions In patients hospitalized in the intensive care unit for COVID-19, acute cor pulmonale and right ventricular dilatation behaved as independent predictors of in-hospital death. Funding Acknowledgement Type of funding sources: None. Table 1. Baselines characteristicsTable 2. Multivariate analisys