Table 1

Non-invasive ventilation randomised controlled trials for COAD

No of subjectsType of ventilation and locationMean (SD or range) duration of ventilationMajor outcomes*Conclusions
*Statistically significant results only: p≤0.05; comparison: intervention v standard therapy. APO = acute pulmonary oedema; BiPAP = bilevel positive airway pressure; COAD = chronic obstructive airways disease; HR = heart rate; Pao2 = arterial oxygen tension; PSV = pressure support ventilation; RR = respiratory rate.
Kraemer et al2931 totalNasal BiPAP v standard therapy in intensive care unit14.4 (2.2) hours/day31% v 73% intubation rate overallNon-invasive ventilation decreased the intubation rate, particularly in COAD
23 COAD3.8 (1.4) days9% v 67% for COAD subgroup
2 APOWeaning not considered before 6 hoursImproved HR, RR at 1 hour and Pao2 at 1–6 hoursImproved some clinical parameters
6 other
Brochard et al3085Face PSV v standard therapy in intensive care unit≥6 hours/day26% v 74% intubatedNon-invasive ventilation decreased intubation rate, hospital mortality, and length of stay and improved some clinical parameters
4 (4) days16% v 48% complication rate
9% v 29% hospital mortality
23 (17) v 35 (33) days length of stay
Improved Pao2, pH, and RR at 1 hour
Çelikel et al3130Face BiPAP v standard therapy in intensive care unit26.7 (16.1) hours7% v 30% failure rate. 33% failure rate for BiPAP used on those who had already failed standard therapyNon-invasive ventilation sped improvement and decreased both intubation rate and length of stay. Improved some clinical parameters
11.7 (3.5) v 14.6 (4.7) days length of stay
Improved pH and RR at 1 and 6 hours
Better Pao2 at 6 hours
Bott et al2260Nasal volume cycled v standard therapy in intensive care unit7.63 (1–23) hours/day4% v 30% 30 day mortality (efficacy)Non-invasive ventilation decreased mortality and improved some clinical parameters
6.0 (2–9) daysNot significant on an intention-to-treat basis
Improved pH, Paco2, and breathlessness at 1 hour