P11 Long-term home non-invasive ventilation

A meta-analysis of 21 RCTs of domiciliary non-invasive ventilation (NIV) concluded that domiciliary NIV does not reduce mortality in patients with stable COPD or in patients post admission for an exacerbation of COPD (Dretzke 2016) [evidence level I]. The authors noted significant differences in patient populations and trial designs.   Note that the 2017 trial by Murphy et al mentioned below is not included in this meta-analysis.

Given the heterogeneity of the results the four largest trials should be considered in detail:

An Australian trial of hypoxaemic COPD patients with hypercapnia (n=144) randomised to long term oxygen therapy alone or with NIV found a small mortality benefit in the NIV group, at the expense of worse quality of life (McEvoy 2009) [evidence level II]. Mean pressures used were inspiratory positive airways pressure (IPAP) 13 cmH2O, EPAP 5 cmH2O. No significant reduction in PaCO2 was observed.

In 2014 two further RCTs of long term NIV in patients with COPD were published. A major difference from the McEvoy trial is the significantly higher inspiratory positive pressures used. Kohnlein et al (Kohnlein 2014) randomised 195 patients with clinically stable, severe COPD and hypercapnia to NIV or usual care. Both groups were admitted to hospital every 3 months for 1 year for ‘treatment optimisation’. NIV was titrated to target a reduction in PaCO2 by 20% or achieve PaCO2 of less than 48mmHg. There was a significant reduction in 1-year mortality. 12% mortality in the intervention group and 33% in the control group was reported; hazard ratio 0·24 (95% CI 0·11 to 0·49; p=0·0004) [evidence level II]. Mean pressures used were IPAP 22cmH2O, EPAP 5cmH2O. Mean reduction in PaCO2 was 7mmHg. In direct contrast to this finding, Struik et al 2014 found no mortality or exacerbation rate difference in a 12 month randomised controlled trial (n =201) of NIV and usual care in patients with severe COPD who had been admitted to hospital with an exacerbation of COPD and hypercapnia (Struik 2014). Mean pressures used were IPAP 19cmH2O, EPAP 5cmH2O. Mean reduction in PaCO2 was 3.8mmHg.

In 2017, Murphy et al reported on a trial of long term NIV in patients who remained hypercapnic and hypoxic two to four weeks after resolution of respiratory acidosis due to an exacerbation of COPD (Murphy 2017). Murphy excluded patients with a BMI >35 kg/m2 and patients with obstructive sleep apnoea. 116 patients were randomised to continuous home oxygen or continuous home oxygen plus home NIV.  It should be noted that over 2,000 patients were screened for inclusion. Similar to Kohnlein, high NIV pressures were used with a median IPAP pressure of 24cmH2O and a median EPAP pressure of 4cmH2O. Median NIV adherence at 12 months was high at 7.6 hours. The 12-month risk of readmission or death was 63.4% in the home oxygen plus home NIV group versus 80.4% in the home oxygen alone group, absolute risk reduction of 17.0% (95% CI 0.1-34%). There was no mortality difference at 12 months.

Comparison between the trials above is difficult as they used different treatment algorithms and NIV pressure settings. Furthermore, inclusion criteria and patient characteristics also differed significantly. For example, McEvoy et al performed a baseline diagnostic sleep study and excluded patients with OSA. A baseline PSG was not performed in the other trials. It is also unclear if the results from the Kohnlein trial are generalisable to an Australasian patient population given that all participants were electively admitted three monthly.

Despite the significant heterogeneity of results, patient characteristics and methodology, there is mounting evidence that long term NIV should be considered in patients with severe stable COPD, hypercapnia and recent hospitalisation.  Referral for specialist opinion at an institution with expertise in this area should be sought.