Abstract
BACKGROUND: COVID-19, the most destructive pandemic of this century, caused the highest mortality rate among ICU patients. The evaluation of these patients is insufficient in lower-middle-income countries with limited resources during pandemics. As a result, our primary goal was to examine the characteristics of patients at baseline as well as their survival outcomes, and propose mortality predictors for identifying and managing the most vulnerable patients more effectively and quickly. METHODS: A prospective analysis of COVID-19 ICU-admitted patients was conducted in our healthcare centre in Iran, from 1 April until 20 May 2020. Ninety-three patients were included in the study, and all were unvaccinated. A multi-variate logistic regression was conducted to evaluate mortality-associated factors. RESULTS: There were 53 non-survivors among our ICU-admitted patients. The mean duration from symptoms' onset to hospitalization was 6.92 ± 4.27 days, and from hospitalization to ICU admission was 2.52 ± 3.61 days. The average hospital stay for patients was 13.23 ± 10.43 days, with 8.84 ± 7.53 days in the ICU. Non-survivors were significantly older, had significantly lower haemoglobin levels and higher creatine phosphokinase levels compared to survivors. They had marginally lower SpO2 levels at admission, higher vasopressor administrations, and were intubated more significantly during their ICU stay. The use of immunosuppressive drugs was also significantly higher in non-survivors. Logistic regression revealed that a one-point increase in APACHE II score at ICU admission increased mortality by 6%, and the presence of underlying diseases increased mortality by 4.27 times. CONCLUSION: The authors presented clinical mortality prediction factors for critically ill patients infected with COVID-19. Additional studies are necessary to identify more generalized mortality indicators for these patients in lower-middle-income countries.