Noninvasive ventilation in patients with do-not-intubate orders: medium-term efficacy depends critically on patient selection

Objective: Randomized clinical trials demonstrating benefits of noninvasive ventilation (NIV) systematically exclude patients with do-not-intubate (DNI) orders, but in daily clinical practice these patients are frequently treated with NIV. A recent North American study found a 43% hospital survival rate in patients with DNI orders. Our hypothesis was that, due to the very different social and cultural setting, written DNI orders in a southern European country would be restricted to a population with a poor outcome, independently of whether they receive NIV, and we analyzed hospital survival in patients receiving NIV and the impact of DNI orders on survival. Design and setting: Retrospective cohort study in a general ICU in a university-affiliated hospital. Patients and methods: All 233 patients treated with NIV during 2002-2004. We recorded clinical characteristics on admission, mortality risk by APACHE II and ICU and hospital outcome, and 6-month outcome. Results: Hospital survival was 66%. Survival was better in the 199 patients without DNI orders than in the 36 with DNI orders both during hospitalization (74% vs. 26%, OR 7.9) and after 6 months (64% vs. 15%, OR 10.2). In both groups the presence of COPD was associated with better prognosis during hospitalization, but not in the medium-term. Conclusions: Our study suggests that NIV offers low expectations for medium-term survival in DNI patients.