The value of non-contrast chest CT in the prediction of myocardial injury in patients with the COVID-19 Omicron variant

Abstract

The Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) Omicron variant associated myocardial injury seriously affected the patient's health. Chest computed tomography (CT) is an essential imaging diagnostic tool for evaluating lung diseases in these patients, but its value in the diagnosis of myocardial injury remains unknown. The purpose of this study was to evaluate the lung lesions in patients with Omicron infection with or without myocardial injury, and to evaluate the predictive value of non-contrast chest CT in such patients with myocardial injury. We enrolled 122 consecutive hospitalized patients with laboratory-confirmed coronavirus disease 2019 (COVID-19) for non-contrast chest CT examination. These patients were divided into two groups according to whether myocardial injury occurred. Myocardial injury was defined as a Troponin I level above the 99th-percentile upper reference limit (0.04 ng/mL). The imaging manifestations of the patients’ lungs were evaluated. Myocardial CT value, left atrium (LA) size, long diameter of left ventricular (LV), and cardiothoracic ratio (CTR) were recorded. Multivariate logistic analysis was performed to identify the predictive factors associated with myocardial injury. Of 122 patients, 61 patients (50%) had myocardial injury. Compared with patients without myocardial injury, there was worse NYHA class, more critical patients, higher incidence of bronchial meteorology, larger area and percentage of lung lesions, diameters of LA, and lower myocardial CT value in the myocardial injury group (P < 0.05). Troponin I concentration in patients with myocardial injury group showed negative correlation with myocardial CT value (r = − 0.319, P = 0.012). The multivariable logistic regression analysis showed that disease severity status (OR 2.279; 95% CI 1.247–4.165, P = 0.007), myocardial CT value (OR 0.849; 95% CI 0.752–0.958, P = 0.008), neutrophil count (OR 1.330; 95% CI 1.114–1.587, P = 0.002) were independent predictors of myocardial injury. The discrimination of the model was good (C-statistic = 0.845, 95% CI 0.775–0.914) and well calibrated with a Hosmer–Lemeshow test for goodness of fit (P = 0.476). Patients infected with Omicron with myocardial injury had more severe lung disease than those without myocardial injury. Non-contrast chest CT can be a useful method of detection of myocardial injury in Omicron infection patients.