Mortality and associated risk factors in patients hospitalized due to COVID-19 in a Peruvian reference hospital

OBJECTIVES: To determine the risk factors for in-hospital mortality in patients with COVID-19 from a Peruvian national hospital. METHODS: Retrospective cohort study of medical records of patients with COVID-19 hospitalized at Hospital Nacional Hipólito Unanue (HNHU) during the months of April to August 2020. The dependent variable was in-hospital mortality. Independent variables included sociodemographic and clinical characteristics, physical examination findings, oxygen saturation (SaO(2)) at admission, treatment received during hospitalization and laboratory results at admission. A Cox regression model was used to evaluate the crude and adjusted hazard ratios for associated factors. RESULTS: We included 1418 patients. Median age was 58 years (IQR 47–68 years) and 944 (66.6%) were male. The median length of hospitalization was 7 (4–13) days, and the mortality rate was 46%. The most frequent comorbidities were type 2 diabetes mellitus, hypertension, and obesity. In the adjusted analysis, mortality was associated with age (HR 1.02; 95%CI 1.02–1.03), history of surgery (HR 1.89; 95%CI 1.31–2.74), lower oxygen saturation at admission (HR 4.08; CI95% 2.72–8.05 for SaO(2)<70% compared to SaO(2)>94%), the presence of poor general condition (HR 1.81; 95% CI 1.29–2.53), altered state of consciousness (HR 1.58; 95%CI 1.18–2.11) and leukocyte levels (HR 1.01; 95%CI 1.00–1. 02). Treatment with ivermectin (HR 1.44; 95%CI 1.18–1.76) and azithromycin (HR 1.25; 95%CI 1.03–1.52) were associated with higher mortality. Treatment with corticosteroids at low to moderate doses was associated with lower mortality (HR 0.56 95%CI 0. 37–0. 86) in comparison to no steroid use. CONCLUSION: A high mortality was found in our cohort. Low oxygen saturation at admission, age, and the presence of hematological and biochemical alterations were associated with higher mortality. The use of hydroxychloroquine, ivermectin or azithromycin was not useful and was probably associated with unfavorable outcomes. The use of corticosteroids at moderate doses was associated with lower mortality.