ABSTRACT
BACKGROUND: The COVID-19 pandemic caused hospital reconversion throughout Mexico and it was scarce information about its development in second-level intensive care units (ICU). OBJECTIVE: To determine the clinical characteristics related to COVID-19 mortality in a second-level ICU. MATERIAL AND METHODS: Observational, cohort, retrospective, and analytical study. Demographic variables, medical history, as well as clinical, ventilatory and laboratory characteristics, and complications of patients admitted to ICU from March to November 2020 due to acute respiratory failure were recorded. Patients were divided into two groups: improvement or death. Lost data were imputed by normal multivariated regression. Descriptive statistics and inferencial analysis were made to determine the risk of significant variables against the death outcome with Cox regression. RESULTS: 60% of patients were male. In-hospital mortality was 55%. An older age (44.4 ± 12.1 vs. 50.7 ± 12.1, p = 0.01), higher APACHE II score (8 (10-13) vs. 15 (11-21), p < 0.001), larger onset-symptom time to ICU (10.1 ± 4.0 vs. 12.0 ± 5.3 days, p = 0.049) and a lower oxygen saturation (78.2 ± 16.%5 vs. 71.1 ± 17.9%, p = 0.017) were significantly asociated characteristics to mortality. Average of stay at ICU was 8 days. CONCLUSIONS: A higher age, more days from beginning of symptoms to hospital admission, and lower oxygenation at admission were pre-admission determining factors for risk of death, while cardiovascular, renal complications and hyperglycemia were the in-hospital determinants.