[Impact of noninvasive ventilation failure upon patient prognosis. Subanalysis of a multicenter study].

OBJECTIVE Noninvasive ventilation (NIV) constitutes first-line treatment for the exacerbation of obstructive pulmonary disease and cardiogenic lung edema. Several studies suggest that NIV failure could increase the risk of mortality, mainly due to the delay in tracheal intubation. We aimed to evaluate the negative impact of NIV failure in routine practice among Spanish ICUs. PATIENTS A subanalysis was made of the multicenter validation of the Sabadell Score study, extracting patients with acute respiratory failure requiring either invasive or noninvasive mechanical ventilation, with the exclusion of patients presenting "do not resuscitate and/or do not intubate" orders. VARIABLES We recorded demographic parameters, ICU-specific treatments and the development of acute renal failure or infections during ICU stay. Patients were followed-up on until hospital discharge or death. The statistic analysis included Cox multiple logistic regression. RESULTS We analyzed 4132 patients, of whom 1602 (39%) received only invasive mechanical ventilation (IMV), while 529 (13%) received NIV. The latter succeeded in 50% of the patients, but the other 50% required intubation. NIV failure was more common in neurological and postsurgical patients. Mortality was lower than predicted in NIV patients (22% vs. 33%) and similar to predicted in IMV patients (27% vs. 29%). Mortality was lower than predicted in patients in whom NIV proved successful (12% vs. 28%), and in those in whom NIV failed (32% vs. 38%). CONCLUSION NIV failure and the need of intubation as routinely used do not seem to imply a poorer patient prognosis.