Factors associated with mortality in mechanically ventilated patients with severe acute respiratory syndrome due to COVID-19: a multicenter retrospective cohort

Abstract

Introduction: Severe Acute Respiratory Syndrome (SARS) caused by COVID-19 is a multisystemic condition, which since 2020 has provided high mortality rates worldwide. Predictors of worse outcomes collected at patient admission to invasive mechanical ventilation can provide useful information to support clinical and public health decisions. However, these predictors of worse outcomes in mechanically ventilated patients still remain uncertain. Objective: To assess factors associated with mortality in mechanically ventilated SARS patients. Methods: A retrospective observational cohort study, in line with the STROBE Checklist, longitudinally followed 425 adult patients, mechanically ventilated due to the evolution of COVID-19, hospitalized in 4 Intensive Care Units in two Brazilian states. Using abstraction from standardized medical records, data were collected on predictors, including clinical data at admission to invasive mechanical ventilation, sequential assessment of organ failure (SOFA), characteristics of standardized ventilatory mechanics that were classified as high respiratory system compliance (Csr) > 45 mL/cmH2O, intermediate compliance if Csr 30> Csr <45 mL/cmH2O and low compliance if Csr <30 mL/cmH2O. Risk factors for death were analyzed using Cox regression to estimate risk ratios (hazard ratios, HR) and their respective 95% confidence intervals (95%CI), to establish predictors related to mortality in patients with mechanically ventilated SARS. Results: The body mass index (BMI) (HR 1.17 [95%CI 1.11-1.20, p<0.001], SOFA (HR 1.4 [95%CI 1.31-1.49, p<0.001], Driving pressure (HR 1.24[1.21-1.29, p<0.001], respiratory system compliance (HR 0.92 [0.90-0.93, p<0.001] and PaO2/FiO2 (HR 0.94 [95%CI 0.91-0.98), p<0.001] are the major independent factors associated with mortality in patients with SARS by mechanically ventilated COVID-19. Comparative analysis of survival curves shows that patients with low respiratory system compliance (Csr <30 mL/cmH2O) are more likely to die at 28d and 60d when compared to those with intermediate compliance (P<0.001 ) and high compliance (P<0.001), respectively. Conclusion: Obese patients (BMI >32 kg/m2), with mechanical characteristics of low respiratory system compliance (Csr <30 mL/cmH2O), with Driving pressure >14 cmH2O and with SOFA > 5,8 immediately after starting invasive ventilatory assistance, have worse outcomes in the segment, being risk factors independent baits for mortality in this population.