Background: Since patients with severe COVID-19 who undergo invasive ventilation can take time to recover, it provides an opportunity to investigate the long-term prognosis and evaluate which factors influence outcomes.

Methods: Patients with severe COVID-19 who received mechanical ventilation were reviewed. On admission, comorbidities and laboratory findings were collected from clinical records. For patients transferred to another hospital, clinical information was obtained from the hospitals to which patients were transferred. Results and conclusions: Prognostic information 90 days after diagnosis was reviewed for 158 patients, 126 were male. Of the 158 patients, 72 were discharged, 26 were under inpatient care, and 60 died at 90 days. Age (65.5 vs 57.4yo, p<0.001) and Charlson Risk Index (1.78 vs 1.04, p<0.01) of those who died were significantly higher than the survival group. Furthermore, the number of patients with chronic respiratory disease (26.7% vs 9.2%, p<0.01) and chronic kidney disease (26.7% vs 6.1%, p<0.001) in those who died was significantly higher than in the survival group. Laboratory findings at admission showed significantly higher levels of blood lactate dehydrogenase (LDH), urea nitrogen, creatinine, and C-reactive protein (CRP), and significantly lower hemoglobin levels and platelet counts in those who died compared to the survival group. Conclusions: Fewer than half the patients treated with invasive ventilation were discharged 90 days after diagnosis. Older patients with respiratory and renal comorbidities, higher CRP and higher LDH, or renal impairment on admission appeared to be at higher risks for a poor prognosis.