Predictors of noninvasive mechanical ventilation weaning failure in the emergency department

To analyze factors related to the failure of noninvasive mechanical ventilation (NIV) weaning in a hospital emergency department (ED).

Prospective, observational cohort study with enrolled a sample of consecutive patients who required NIV during ED care. The dependent variable was NIV weaning failure, defined by the need to restart NIV in the ED after a first attempt to withdraw the respirator.

Of a total of 675 candidates, we included 360 patients (53.4%). Exclusions were 100 patients (31.7%) who were on NIV at home; 58 (18.4%) in whom NIV initially failed; and 157 (49.9%) in whom weaning was attempted outside the ED. Seventy-two (17.3%) cases of weaning failure in the ED were observed. Factors independently associated with failure were the bicarbonate (HCO3) concentrations before attempted weaning (adjusted odds ratio [aOR], 1.06; 95% CI, 1.01-1.12; P = .014), time on NIV in hours (aOR, 1.10; 95% CI, 1.04-1.16; P .001), and a pH less than 7.35 before weaning (aOR, 2.48; 95% CI, 1.16-5.31; P = .019).

Weaning failure occurs in 17% of ED patients on NIV. Time on NIV, HCO3 concentration, and a pH less than 7.35 before weaning are independently associated with failure to wean from the respirator.