Experiences from introducing non-invasive ventilation in the intensive care unit:: a 2-year prospective consecutive cohort study

Objective: To describe 2 years of experience and staff learning curves after the introduction of non-invasive ventilation (NIV). Methods: A prospective, consecutive, strictly observational, 2-year cohort study of all patients treated with NIV in a county general hospital intensive care unit (ICU), with no interventions, was performed. Results: One hundred and fifty-seven patients with 15 different diagnoses were treated with NIV. An increasing number of patients were treated in the second year and, probably as a result of increased staff experience, the NIV treatment time and overall time spent in the ICU were less in the second year of the study period (30 h vs. 19 h and 55 h vs. 34 h, respectively; P < 0.05). Patients were also intubated earlier if NIV failed during the second year. Of the 157 patients, 119 had a full treatment option (i.e. including the possibility of invasive mechanical ventilation) and 26% died; their Acute Physiology and Chronic Health Evaluation II (APACHE II)- and Simplified Acute Physiology II (SAPS II)-predicted death rates were 31% and 32% respectively (not significant, NS). The overall mortality rate in all NIV patients was 38%, compared with predicted death rates of 36% and 33%, respectively (NS). 'Do-not-intubate' orders were issued for 38 of the 157 patients, and 11 of these (29%) left the hospital alive. Patients treated successfully with NIV had significantly lower APACHE II scores than those in whom it failed (18.8 vs. 22, P = 0.01). Conclusion: With increasing staff experience, more patients may be treated with NIV and the treatment period decreases significantly. Weaning from NIV has almost exclusively been taken over by nurses. Unselected cohorts of patients with chronic obstructive pulmonary disease can be treated successfully with NIV, and NIV offers a treatment option even for some patients with a 'do-not-intubate' order.