Assessment of the Modified CHA2DS2VASc Risk Score in Predicting Mortality in Patients Hospitalized With COVID-19

Since the modified CHA2DS2VASC (M-CHA2DS2VASc) risk score includes the prognostic risk factors for COVID-19; we assumed that it might predict in-hospital mortality and identify high-risk patients at an earlier stage compared with troponin increase and neutrophil-lymphocyte ratio (NLR). We aimed to investigate whether M-CHA2DS2VASC RS is an independent predictor of mortality in patients hospitalized with COVID-19 and to compare its discriminative ability with troponin increase and NLR in terms of predicting mortality. A total of 694 patients were retrospectively analyzed and divided into 3 groups according to M-CHA2DS2VASC RS which was simply created by changing gender criteria of the CHA2DS2VASC RS from female to male (Group 1, score 0-1 (n = 289); group 2, score 2-3 (n = 231) and group 3, score ≥4 (n = 174)). Adverse clinical events were defined as in-hospital mortality, admission to intensive care unit, need for high-flow oxygen and/or intubation. As the M-CHA2DS2VASC RS increased, adverse clinical outcomes were also significantly increased (Group 1, 3.8%; group 2, 12.6%; group 3, 20.8%; p <0.001 for in-hospital mortality). The multivariate logistic regression analysis showed that M-CHA2DS2VASC RS, troponin increase and neutrophil-lymphocyte ratio were independent predictors of in-hospital mortality (p = 0.005, odds ratio 1.29 per scale for M-CHA2DS2VASC RS). In receiver operating characteristic analysis, comparative discriminative ability of M-CHA2DS2VASC RS was superior to CHA2DS2VASC RS score. Area under the curve (AUC) values for in-hospital mortality was 0.70 and 0.64, respectively. (AUC(M-CHA2DS2-VASc) vs. AUC(CHA2DS2-VASc) z test = 3.56, p 0.0004) In conclusion, admission M-CHA2DS2VASc RS may be a useful tool to predict in-hospital mortality in patients with COVID-19.