Predictors of Mortality in Patients with Severe COVID-19 Pneumonia—A Retrospective Study

Introduction: The novel coronavirus pandemic has caused significant mortality throughout the world. This study was done as there is scarce data on mortality predictors in severe COVID-19 pneumonia patients admitted to ICU in the Indian population. Material and methods: A retrospective study was conducted on COVID-19 pneumonia patients admitted to tertiary care center during June–October 2020. The records of patients admitted to ICU were collected and data included demography, symptoms, comorbidites and vital parameters. Laboratory parameters included complete hemogram, random blood sugar, S.Ferritin, S.LDH, renal function test, liver function test. Treatment-associated information such as the use of remdesivir, timing of initiating rem-desivir after the symptom onset, the use of steroids, use of anticoagulants, use of HFNC, NIV, ventilator were collected. 30 days mortality data post-discharge was collected via telephonic interview. Results: 4,012 confirmed cases of COVID-19 were admitted to hospital, of which 560 (13.95%) with severe pneumonia were included in the study. Mean age was 57.75 ± 13.96 years. The mortality rates were 54.64% among severe COVID-19 cases and 5% among mild to moderate COVID-19 cases. The Cox multinominal regression analysis identified SpO2/FiO2 < 400, age > 50 years, duration of symptom > 4 days, serum ferritin > 450 µg/L, respiratory rate > 23/min, the presence of comorbidities and non-usage of remdesivir were independently associated with increased mortality. Mortality rate at 30 days was 56.60%. Conclusion: Severe COVID-19 pneumonia is associated with very high mortality, especially in a resource-constrained setting. The use of remdesivir may have to be considered early in the course of disease to prevent excess mortality related to COVID-19.