In-hospital mortality in SARS-CoV-2 stratified by hemoglobin levels: A retrospective study.

This study is to estimate in-hospital mortality in severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) patients stratified by hemoglobin (Hb) level. Patients were stratified according to hemoglobin level into two groups, that is, Hb <100 g/L and Hb >100 g/L. A total of 6931 patients were included. Of these, 6377 (92%) patients had hemoglobin levels >100 g/L. The mean age was 44 ± 17 years, and 66% of the patients were males. The median length of overall hospital stay was 13 days [2; 31]. The remaining 554 (8%) patients had a hemoglobin level <100 g/L. Overall mortality was 176 patients (2.54%) but was significantly higher in the group with hemoglobin levels <100 g/L (124, 22.4%) than in the group with hemoglobin levels >100 g/L (52, 0.82%). Risk factors associated with increased mortality were determined by multivariate analysis. The Kaplan-Meier survival analysis showed hemoglobin as a predictor of mortality. Cox proportional hazards regression coefficients for hemoglobin for the HB ≤ 100 category of hemoglobin were significant, B = 2.79, SE = 0.17, and HR = 16.34,  < 0.001. Multivariate logistic regression showed Hb < 100 g/L had a higher cumulative all-cause in-hospital mortality (22.4% vs. 0.8%; adjusted odds ratio [aOR], 0.33; 95% [CI]: [0.20-0.55];  < 0.001). In this study, hemoglobin levels <100 g/L were found to be an independent predictor of in-hospital mortality.