Noninvasive ventilation after early extubation in patients recovering from hypoxemic acute respiratory failure: a single-centre feasibility study.

Vaschetto, Rosanna; Turucz, Emilia; Dellapiazza, Fabrizio; Guido, Stefania; Colombo, Davide; Cammarota, Gianmaria; Della Corte, Francesco; Antonelli, Massimo; Navalesi, Paolo

PURPOSE The use of noninvasive ventilation (NIV) to facilitate discontinuation of mechanical ventilation in patients with acute hypoxemic respiratory failure (hypoxemic ARF) has never been explored. This pilot study aims to assess the feasibility of early extubation followed by immediate NIV, compared conventional weaning, in patients with resolving hypoxemic ARF. METHODS Twenty consecutive hypoxemic patients were randomly assigned to receive either conventional weaning or NIV. The changes in arterial blood gases and respiratory rate were compared between the two groups at 1, 12, 24 and 48 h. Differences in the rate of extubation failure, ICU and hospital mortality, number of invasive-ventilation-free-days at day 28, septic complications, number of tracheotomies, days and rates of continuous intravenous sedation, and ICU length of stay were also determined. RESULTS No patient interrupted the study protocol. Arterial blood gases were similar during invasive mechanical ventilation, 1 h after NIV application following extubation, and after 12, 24 and 48 h. Respiratory rate was higher after 1 h in the NIV group, but no different after 12, 24 and 48 h. The number of invasive-ventilation-free-days at day 28 was 20 ± 8 (min = 0, max = 25) days in the treatment group and 10 ± 9 (min = 0, max = 25) days in the control group (p = 0.014). The rate of extubation failure, ICU and hospital mortality, tracheotomies, septic complications, days and rates of continuous sedation, and ICU length of stay were not significantly different between the two groups. CONCLUSIONS In a highly experienced centre NIV may be used to facilitate discontinuation of mechanical ventilation in selected patients with resolving hypoxemic ARF.