ABSTRACT
INTRODUCTION: The aim was to investigate the impact of different ventilator strategies (non-invasive ventilation (NIV); invasive MV with tracheal tube (TT) and with tracheostomy (TS) on outcomes (mortality and intensive care unit (ICU) length of stay) in patients with COVID-19. We also assessed the impact of timing of percutaneous tracheostomy and other risk factors on mortality. METHODS: The retrospective cohort included 868 patients with severe COVID-19. Demographics, MV parameters and duration, and ICU mortality were collected. RESULTS: MV was provided in 530 (61.1%) patients, divided into three groups: NIV (n = 139), TT (n = 313), and TS (n = 78). Prevalence of tracheostomy was 14.7%, and ICU mortality was 90.4%, 60.2%, and 30.2% in TT, TS, and NIV groups, respectively (p < 0.001). Tracheostomy increased the chances of survival and being discharged from ICU (OR 6.3, p < 0.001) despite prolonging ICU stay compared to the TT group (22.2 days vs. 10.7 days, p < 0.001) without differences in survival rates between early and late tracheostomy. Patients who only received invasive MV had higher odds of survival compared to those receiving NIV in ICU prior to invasive MV (OR 2.7, p = 0.001). The odds of death increased with age (OR 1.032, p < 0.001), obesity (1.58, p = 0.041), chronic renal disease (1.57, p = 0.019), sepsis (2.8, p < 0.001), acute kidney injury (1.7, p = 0.049), multiple organ dysfunction (3.2, p < 0.001), and ARDS (3.3, p < 0.001). CONCLUSIONS: Percutaneous tracheostomy compared to MV via TT significantly increased survival and the rate of discharge from ICU, without differences between early or late tracheostomy.