Effects of long-term non-invasive ventilation in stable chronic obstructive pulmonary disease: a systematic review of 16 randomized controlled trials

Background: The benefits of non-invasive ventilation (NIV) in chronic obstructive pulmonary disease (COPD) have been debated for many years due to the conflicting results observed in these patients. The aim of this metaanalysis was to assess the long-term effects of NIV in stable patients with COPD. Methods: The Cochrane Central Register of Controlled Trials, MEDLINE, EMBASE, and CINAHL databases were searched. Randomized, controlled trials and crossover studies were included. Mortality, hospitalization, gas exchange, exercise tolerance, heath-related quality of life (HRQoL), lung function and sleep efficiency were used as outcome measurements. These outcomes were pooled to yield mean differences (MDs), standardized mean differences (SMDs) and risk differences (RDs). Results: Fourteen randomized controlled trials and two crossover studies were included. Compared with the control group, NIV significantly affected partial pressure of oxygen in the blood (MDshort term 4.72, [2.66, 6.78]; MDlong term 2.34, [1.21, 3.47]) and partial pressure of carbon dioxide in the blood in long time (MDlong term -4.55, [-7.49, -1.62]). Compared to control group, the risk difference of mortality in long term group of NIV was -0.06 [-0.12, -0.01]. Three of the six long-term studies exerted significant effects during hospitalization, whereas one of the short-term studies demonstrated no significant effects. Two of the four short-term studies and all three long-term studies demonstrated no significant effects on HRQoL. Conclusion: NIV exerted no clinically or statistically significant effects on gas exchange, exercise tolerance, HRQoL, death, lung function or sleep efficiency in patients with stable COPD when utilized for less than 3 months. It might significantly improve blood gases, exercise tolerance and lung function after six months. However, long term NIV efficacy for stable COPD remains uncertain in terms of mortality and readmission. Long-term follow-up data of such patients is needed to reconfirm the benefits of NPPV.