Introduction Domiciliary NIV in COPD remains controversial with previous randomised controlled trials showing little clinical benefit. However, these trials have been criticised for using low pressure NIV and consequently nocturnal hypoventilation occurs. In contrast, there are genuine concerns that patients may not be able to tolerate the high pressure domiciliary NIV required to manage nocturnal hypoventilation effectively.

Method Patients admitted for acute hypercapnic respiratory failure due to an exacerbation of COPD with persistent hypercapnia (PaCO₂>7 kPa) 2–4 weeks following resolution of acute episode were offered participation into the trial. Patients were randomised to either home oxygen therapy (HOT) or home mechanical ventilation (HMV) and followed up at 6 weeks and 3 months. Patient assessment included anthropometrics, arterial blood gases and health related quality of life measures. Sleep disruption was assessed using actigraphy (Actiwatch spectrum, Philips-Respironics, Murrysville, Pennsylvania, USA) for 7 days following the assessment.

Results 36 patients have been recruited and randomised to date. Abstract S119 table 1 show the baseline data for 20 patients that have completed follow-up to 3 months. Discharge ventilator settings were IPAP 26±3 cmH₂O, EPAP 5±1 cmH₂O and back up rate 15±1 bpm in the HMV group. Total sleep time during the first 2 weeks of HMV was significantly shorter than in those patients receiving HOT (?88 min; 95% CI 5 to 172 min, p=0.04). However, by 6 weeks there was no difference between the groups (Δ66 min; 95% CI −62 to 194 min, p=0.3). This was sustained at 3 months (Δ51 min; 95% CI −100 to 202 min, p=0.5). There were no between group differences in wake after sleep onset, sleep efficiency or sleep latency at initial assessment or subsequent follow-up. Ventilator compliance at 6 weeks was 3 h 41 min ±1 h 41 min and at 3 months was 4 h 30 min ±1 h 44 min.

Conclusion A reduction in total sleep time between patients receiving HMV compared with those receiving HOT was demonstrated during the initial period of acclimatisation to HMV. However, despite increasing ventilator usage during the follow-up period the difference in sleep duration between treatment groups reduced suggesting improved tolerability to NIV.