LAUSR

BACKGROUND In the course of SARS-CoV-2 infection, early prognostic evaluation is important since clinical symptoms may worsen rapidly and may be fatal. Inflammation plays an important role in the pathogenesis of COVID-19 and can cause myocardial damage which is common in severe COVID-19 patients. Therefore, novel inflammatory indexes and myocardial damage may be predictive of prognosis in patients with COVID-19. The aim of the study was to evaluate the role of cardiac troponin I (cTnI), modified Glasgow prognostic score (mGPS), systemic immune inflammation index (SII), prognostic nutritional index (PNI), and CRP to albumin ratio (CAR) in the outcome estimation of COVID-19 and to develop a risk model predicting the survival probability of COVID-19 survivors during early post-discharge. METHODS This was a single-center, observational, retrospective cohort study. Laboratory confirmed COVID-19 patients (n = 265) were included and grouped according to in-hospital mortality. ROC curve analysis was performed and Youden's J index was used to obtain optimal cutoff values for inflammatory indexes in discriminating survivors and non-survivors. Cox regression analysis was performed to assess the possible predictors of in-hospital mortality. A nomogram was constructed based on the Cox regression model, to calculate 7- and 14-day survival. RESULTS The area under the ROC curve (AUC) of the variables ranged between 0.79 and 0.92 with the three highest AUC values for albumin, PNI, and cTnI (0.919, 0.918, and 0.911, respectively). Optimal threshold value for cTnI was 9.7 pg/mL. Univariate analysis showed that gender, albumin, CRP, CAR, PNI, SII, cTnI, and mGPS were significantly related to in-hospital mortality. The Cox regression analysis indicated that mGPS (p = 0.001), CRP (p = 0.026), and cTnI (p = 0.001) were significant prognostic factors. CONCLUSIONS cTnI should not be considered merely as an indicator of myocardial damage. It also reflects the inflammatory phase and, along with other inflammatory markers, it should be included in risk models as a prognostic factor for COVID-19.