Evaluation of Modified ATRIA Risk Score in Predicting Mortality in Hospitalized Patients With COVID-19

Background: As the Modified Anticoagulation and Risk Factors in Atrial Fibrillation Risk Score (M-ATRIA-RS) encompasses prognostic risk factors of novel coronavirus-2019 (COVID-19), it may be used to predict in-hospital mortality. We aimed to investigate whether M-ATRIA-RS was an independent predictor of mortality in patients hospitalized for COVID-19 and compare its discrimination capability with CHADS, CHA2DS2-VASc, and modified CHA2DS2-VASc (mCHA2DS2-VASc)-RS. Methods: A total of 1,001 patients were retrospectively analyzed and classified into three groups based on M-ATRIA-RS, designed by changing sex criteria of ATRIA-RS from female to male: Group 1 for points 0-1 (n = 448), Group 2 for points 2 -4 (n = 268), and Group 3 for points >= 5 (n = 285). Clinical outcomes were defined as in-hospital mortality, need for high-flow oxygen and/or intubation, and admission to intensive care unit. Results: As the M-ATRIA-RS increased, adverse clinical outcomes significantly increased (Group 1, 6.5%; Group 2, 15.3%; Group 3, 34.4%; p <0.001 mortality for in-hospital). Multivariate logistic regression analysis showed that M-ATRIA-RS, malignancy, troponin increase, and lactate dehydrogenase were independent predictors of in-hospital mortality (p<0.001, per scale possibility rate for ATRIA-RS 1.2). In receiver operating characteristic (ROC) analysis, the discriminative ability of M-ATRIA-RS was superior to mCHA2DS2-VASc-RS and ATRIA-RS, but similar to that Charlson Comorbidity Index (CCI) score (AUC(M-ATRIA) vs AUC(ATRIA) Z-test=3.14 p = 0.002, AUC(M-ATRIA) vs. AUC(mCHA2DS2-VASc) Z-test=2.14, p = 0.03; AUC(M-ATRIA) vs. AUC(CCI) Z-test=1.46 p = 0.14). Conclusions: M-ATRIA-RS is useful to predict in-hospital mortality among patients hospitalized with COVID-19. In addition, it is superior to the mCHA2DS2-VASc-RS in predicting mortality in patients with COVID-19 and is more easily calculable than the CCI score.