Incidence and outcomes of extubation failure in mechanically ventilated patients with cirrhosis: a post-hoc analysis of a prospective multicenter study.
Yassir Aarab, Joris Pensier, Fanny Garnier, Clement Monet, Ines Lakbar, Gerald Chanques, Audrey de Jong, Mathieu Capdevila, Samir Jaber, FreeRéa study group
Abstract
Open AccessBACKGROUND: Patients with comorbidities who are liberated from invasive mechanical ventilation could be at risk of extubation failure in the intensive care unit (ICU). Incidence and associated outcomes of reintubation in patients with cirrhosis have been poorly studied. We aimed to evaluate the incidence, causes and mortality of reintubation in patients with cirrhosis. METHODS: We conducted a post hoc analysis of the French prospective multicenter observational trial (FREE-REA) evaluating the incidence and risk factors of extubation failure in 26 ICUs. The primary outcome was the incidence of extubation failure defined as the need for reintubation within 7 days after extubation. Secondary outcomes were the incidence of reintubation at 48 h, the causes and risk factors of extubation failure, ICU length of stay and in-hospital mortality. We compared patients with cirrhosis and patients without cirrhosis. RESULTS: Of the 1,443 analyzed patients, 165 (11%) had cirrhosis. The incidence of reintubation within 7 days was 21% (34/165) in patients with cirrhosis and 13% (167/1278) in patients without cirrhosis (p < 0.01). Reintubation at 48 h was not significantly different between patients with cirrhosis and patients without cirrhosis (9% versus 10%, p = 0.55). Admission for shock was identified as the only independent risk factor for extubation failure in multivariate analysis [OR 3.24, 95% CI (1.24-8.44), p = 0.02]. In patients with extubation failure, ICU length of stay was significantly longer in patients with cirrhosis compared to those without (28 ± 25 versus 18 ± 12 days, p < 0.01); In-hospital mortality was higher in extubation failure patients with cirrhosis in comparison to patients without cirrhosis without reaching significance (16/34 (47%) versus 51/167 (31%), p = 0.06). CONCLUSION: Extubation failure was significantly higher in patients with cirrhosis compared to patients without cirrhosis. A trend for higher in-hospital mortality was observed in reintubatedpatients with cirrhosis. Neurologic failure was the main cause for reintubation at 48 h in patients with cirrhosis. CLINICAL TRIALS: The study was registered on clinicaltrials.gov (identifier no. NCT02450669). Registered 01/12/2013.