The investigators hypothesize that continuation of non-invasive ventilation (NIV) at home after an episode of acute hypercapnic respiratory failure (AHRF) treated by NIV in COPD patients would reduce the likelihood of death and recurrent AHRF requiring NIV or intubation. The investigators designed this study in a way that recruited COPD patients would be started on home NIV or sham treatment after an episode of AHRF requiring acute NIV. The patients are acclimatised to NIV application after a few days of acute use. The investigators chose occurrence of life-threatening event (recurrent AHRF and death) as the primary endpoint.

Non-invasive ventilation (NIV) has been shown in randomised controlled trials to improve arterial blood gases, reduce intubation and mortality rates in patients suffering from exacerbations of chronic obstructive pulmonary disease (COPD) complicated by acute hypercapnic respiratory failure (AHRF) [1-7]. Despite success of NIV in AHRF of COPD, survivors of this group of patients might suffer from further episodes of AHRF after discharge. It has been found in a recent study that COPD patients who survived AHRF after treatment with acute NIV had a high risk of readmission and life-threatening events in the ensuing year [8]. At one year after discharge, 80% had been readmitted for respiratory diagnoses, 63% had another life-threatening event and 49% had died, mainly due to respiratory failure. Survivors spent a median of 12% time hospitalized in the subsequent year. A significant proportion of survivors required repeated NIV for recurrent AHRF. Another study showed that in COPD patients who declined intubation for AHRF and were treated with acute NIV, these outcomes were even worse, with a 1-year survival of only 30% [9]. It becomes imperative to find ways to reduce the occurrence of life-threatening events in the survivors.

There has been uncontrolled data to suggest that home NIV might reduce both hospital admissions and clinic visits in severe COPD with hypercapnic respiratory failure. In 11 severe stable COPD patients with chronic hypercapnia who did not respond to conventional treatment, Jones et al put them on home NIV [10]. Hospital admissions and clinic visits were halved in the subsequent year, together with a sustained improvement in arterial blood gases. Cost saving was demonstrated with home NIV in severe COPD in another study [11]. However, results from randomized controlled studies (RCTs) are conflicting. Three early studies suggested that home NIV was not superior to standard treatment in stable severe COPD [12, 13, 14]. On the other hand, Meecham-Jones et al found that NIV with long term oxygen therapy (LTOT) significantly improved daytime blood gases, nocturnal gas exchange and sleep quality in severe COPD [15]. A long-term RCT on home NIV in severe COPD showed that home NIV significantly reduced dyspnoea ratings, improved psychomotor coordination and decreased hospital admissions at 3 month, though reduction in hospital admissions was no longer evident by 12 months [16]. However, these randomized studies have been criticized for including chronic stable COPD patients who were not hypercapnic enough to benefit from home NIV, using inadequate inflation pressures, inadequate patient acclimatization time and not selecting the optimal outcome variables [17]. In the most recent RCT [18], home NIV with LTOT was shown to significantly improve gas exchange, dyspnoea score and quality of life; there was also a trend to reduced hospital and ICU admissions. However, the study was only powered to detect improvement in daytime PaCO2 in the NIV group.

We hypothesize that continuation of NIV at home after an episode of AHRF treated by NIV in COPD patients would reduce the likelihood of death and recurrent AHRF requiring NIV or intubation. We design this study in a way that recruited COPD patients would be started on home NIV or sham treatment after an episode of AHRF requiring acute NIV. The patients are acclimatised to NIV application after a few days of acute use. We choose occurrence of life-threatening event (recurrent AHRF and death) as the primary endpoint.

Inclusion Criteria:

• COPD patient with AHRF who survives after treatment with NIV and successfully weaned off for at least 48 hours.
• Patients who have been intubated and mechanically ventilated can be included if they have also been treated with NIV in the same hospital admission
• Significant obstructive sleep apnoea ruled out by overnight polysomnography done after successful weaning of NIV (Apnoea-hypopnoea index, AHI, < 10/hr
• Patients willing to give their written informed consent to participate in the study
• Patients understand that he/she would be randomised into receiving home NIV or sham ventilation
• Patients who are able to use the home pressure support ventilators after a period of acclimatisation and training before discharge from the hospital

Exclusion Criteria:

• Patients with non-COPD causes of AHRF (e.g. asthma, bronchiectasis or lobar pneumonia, fibrothorax, acute pulmonary oedema, etc)
• Patients who have contraindications to NIV and those who refused or failed NIV during an initial 15-minute acclimatization period
• Active smoker
• An increase of ≥ 15% in FEV1 after inhaled salbutamol (200μg)
• Obstructive sleep apnoea (OSA) with apnoea-hypopnoea index (AHI) of ≥ 10/hr
• Other significant co-morbid conditions that in the investigators' view, would confer an adverse prognosis during the study period, e.g., congestive heart failure, uncontrolled diabetes mellitus, tuberculosis, neoplasms, peripheral vascular disease threatening organ functions
• Adverse psycho-social circumstances not conducive to home NIV treatment (Appendix 2)
• On long-term systemic steroid (prednisolone ≥ 7.5 mg per day for ≥ 3 months)