Admission Predictors of In-Hospital Mortality and the Combined Outcome of Death or Invasive Mechanical Ventilation in Patients with COVID-19 During the Pre-Vaccination Era: A Retrospective Cohort Study

Abstract Background Reliable identification of early predictors of adverse outcomes was essential during the pre-vaccination phase of the COVID-19 pandemic. Few studies have comprehensively integrated clinical presentation, laboratory parameters including arterial blood gas analysis, and chest computed tomography (CT) findings within a single well-characterized cohort, particularly in underrepresented regions of Brazil. Methods This retrospective cohort study included 482 consecutive adults (median age 61 years [IQR 49–73]; 64.3% men) with RT-PCR–confirmed SARS-CoV-2 infection hospitalized at a tertiary cardiac center in Central-West Brazil between March 2020 and January 2021. Demographic, clinical, laboratory (including arterial blood gas analysis), and chest CT data obtained within 48 hours of admission were analyzed. Univariable logistic regression was performed for 76 variables. Multivariable models were constructed using an a priori variable selection strategy based on clinical relevance, representation of distinct pathophysiological domains, and adherence to events-per-variable principles. Complete-case analyses were performed without imputation. Results In-hospital mortality was 9.3% (45/482). Invasive mechanical ventilation was required in 74 patients (15.4%), with a mortality rate of 58.1% among those ventilated. In univariable analysis, 42 variables were associated with mortality (p < 0.05). In multivariable analysis (n = 438), five independent predictors of death were identified: age (adjusted OR 1.66 per 10 years; 95% CI 1.19–2.32; p = 0.003), arterial pH (adjusted OR 0.47 per 0.1-unit increase; 95% CI 0.25–0.89; p = 0.021), neutrophil-to-lymphocyte ratio (adjusted OR 1.30; 95% CI 1.18–1.44; p < 0.001), number of comorbidities (adjusted OR 1.59; 95% CI 1.25–2.02; p < 0.001), and serum creatinine (adjusted OR 1.37; 95% CI 1.05–1.77; p = 0.019). The model demonstrated good calibration (Hosmer–Lemeshow p > 0.05) and moderate-to-high explanatory power (Nagelkerke R² = 0.43). For the composite outcome of death or invasive mechanical ventilation (74 events; 15.4%), four predictors remained independently associated; serum creatinine showed a non-significant trend (p = 0.069). On chest CT (n = 424), analyzed descriptively and in univariable models only, pulmonary involvement > 50% was associated with increased odds of death (OR 2.87; 95% CI 1.42–5.79; p = 0.003). Conclusions Five admission variables representing distinct pathophysiological domains—age, arterial pH, neutrophil-to-lymphocyte ratio, comorbidity burden, and serum creatinine—were independently associated with in-hospital mortality in this pre-vaccination cohort. Arterial pH provided independent prognostic information beyond inflammatory and renal markers. These findings support early risk stratification using routinely available clinical data.