Summary of findings
Description of the condition
Chronic obstructive pulmonary disease (COPD) is an important cause of morbidity and mortality worldwide. There are several therapeutic options to help people with COPD manage their symptoms, but to date, only smoking cessation and the provision of long-term oxygen therapy to hypoxic patients have been shown to prolong life (Crockett 2001). Usually, treatment is with bronchodilators and anti-inflammatory drugs (corticosteroids), although the latter is still controversial in people with COPD. When medication doses are optimal and people still have dyspnoea or an impaired exercise tolerance, pulmonary rehabilitation can be added to medical therapy (Lacasse 2006). In people with more severe COPD, bronchoscopic lung volume reduction (Slebos 2012), lung volume reduction surgery (Cooper 1997), and, in extreme cases, lung transplantation (Orens 2006), can be considered. In people with COPD with chronic hypercapnic respiratory failure nocturnal non-invasive positive pressure ventilation (nocturnal-NIPPV) might be beneficial.
Description of the intervention
In non-invasive positive pressure ventilation (NIPPV), the person receives ventilatory support through a non-invasive interface, such as a nasal mask, full-facemask or helmet. NIPPV is currently applied as evidence-based therapy in people with COPD admitted to hospital with acute hypercapnic respiratory failure due to an exacerbation. It has been shown that NIPPV reduces hospital deaths and complications associated with treatment and length of hospital stay (Ram 2004). NIPPV during an acute exacerbation is often applied intermittently or continuously for a few days to reduce the (life-threatening) ventilatory failure, after which the person is weaned from ventilation and treatment is ended. Chronic nocturnal-NIPPV, however, entails the use of NIPPV at home during the night for a longer period. Currently there is much discussion about the need for NIPPV in COPD, mainly because conflicting results have been published (Rossi 2000). There is consensus, but with little supportive evidence, that people with COPD who have substantial daytime hypercapnia and superimposed nocturnal hypoventilation are most likely to benefit from NIPPV (Hill 2004).
How the intervention might work
Several theories exist as to why nocturnal-NIPPV might be beneficial. First, nocturnal-NIPPV might rest chronically fatigued muscles. Periods of rest may lead to recovery of the inspiratory muscle function, thereby leading to an increased muscle strength and endurance capacity of the respiratory muscles during the daytime (Ambrosino 1990). Second, nocturnal-NIPPV has been shown to improve sleep time and efficiency (Meecham 1995), as people with severe COPD can experience poor sleep quality due to sleep-disordered breathing with episodes of hypoventilation associated with desaturation. Third, nocturnal-NIPPV may ameliorate nocturnal hypoventilation and allow the respiratory centre to be reset. In this way nocturnal-NIPPV may reduce daytime hypercapnia (Elliott 1991). Fourth, it is postulated that nocturnal-NIPPV decreases hyperinflation leading to an improvement in respiratory mechanics, such as an increase in forced expiratory volume in one second (FEV
Why it is important to do this review
Despite all these theories, the effect of nocturnal-NIPPV in people with stable severe COPD remains unclear and needs further investigation. This is an update of a Cochrane review first published in 2002. Since then, more studies have been reported, making an update necessary. As in 2002, we performed a systematic review and meta-analysis based on individual patient data (IPD). With IPD, we collected original research data for each participant from the original researchers for data checking, validation and re-analysis. This gives an advantage over the conventional meta-analysis based on summary statistics from published papers as more or different analyses are possible. IPD meta-analyses have greater power, enabling investigation of additional hypothesis related to individual characteristics, for example within subgroups or treatment across trials, or both.
To assess the effects of nocturnal-NIPPV at home via nasal mask or face mask in people with stable COPD by using a meta-analysis based on IPD.
Criteria for considering studies for this review
Types of studies
Randomised controlled trials (RCTs) in people with stable COPD comparing nocturnal-NIPPV at home plus standard therapy with standard therapy alone.
Types of participants
People with COPD according to the guidelines of American Thoracic Society (ATS 1995).
Types of interventions
NIPPV, applied through a nasal or face mask, for at least five hours during the night, for at least three consecutive weeks. Participants also received their usual standard COPD therapy, which comprised supplemental oxygen, bronchodilators, theophylline and corticosteroids.
The intervention in the control group was standard therapy alone. The control group did not receive nocturnal-NIPPV.
Types of outcome measures
- Arterial blood gas tensions (partial pressure of carbon dioxide in the blood (PaCO
2), partial pressure of oxygen in the blood (PaO 2)).
- Six-minute walking distance (6MWD).
- Health status (health-related quality of life (HRQoL) measurements).
- Lung function (FEV
1and forced vital capacity (FVC)).
- Respiratory muscle function (muscle strength, including maximal inspiratory pressure (PImax)).
- Sleep efficiency (time asleep as percentage of total time in bed.
Search methods for identification of studies
We identified trials from the Cochrane Airways Group Specialised Register (CAGR), which is maintained by the Trials Search Co-ordinator for the Group. The Register contains trial reports identified through systematic searches of bibliographic databases including the Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE, EMBASE, CINAHL, AMED and PsycINFO, and handsearching of respiratory journals and meeting abstracts (see Appendix 1 for further details).
For the original version of this review we carried out a search on all COPD records in the Register using the terms: nasal ventilat* OR positive pressure OR NIPPV. For this update, a search was done in all COPD records in the Register using the following search string:
(nasal OR mechanical OR noninvasive OR non-invasive or "non invasive" or positive OR intermittent OR bi-level OR "bi level" OR airway* OR controlled OR pressure OR support AND (ventilat*)) OR (NIPPV).
We conducted this most recent search in August 2012.
Searching other resources
We searched the bibliographies of each RCT for additional papers that may have contained RCTs. We contacted authors of identified RCTs for other published and unpublished studies.
Data collection and analysis
Selection of studies
For the 2002 version of this review, two review authors (PJW, RSG) independently assessed all identified abstracts and for the 2013 update this was done by PJW and FMS. When we selected an abstract, full papers were retrieved and read in detail by the same two review authors and disagreements were resolved by discussion with a third review author.
Data extraction and management
After identification of studies, we contacted trial authors to ask for the IPD including anthropometric data and follow-up data of the identified outcome variables. We requested missing data from the included primary studies from the authors. We checked supplied data against study publications after which we copied raw data from all included studies to one main database.
Assessment of risk of bias in included studies
Two review authors (FMS and PJW) assessed risk of bias of each study independently (for details see 'Risk of bias' table in Characteristics of included studies). We used criteria for assessment of risk of bias as provided in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We considered potential for bias using the following domains:
- sequence generation;
- allocation concealment;
- blinding of participants, personnel and outcome measures;
- incomplete outcome data;
- selective outcome reporting; and
- other sources of bias.
We judged each domain as 'high', 'low' or 'unclear' risk of bias. If insufficient details were reported, the judgement was 'unclear' of risk of bias.
Measures of treatment effect
The principal investigators of all the trials included in the meta-analysis kindly provided the individual data for each of their study subjects. We therefore conducted an individual data meta-analysis. We expressed study outcomes in the same natural units across the trials. For each individual and for each outcome, we calculated an absolute difference in score that defined treatment effect. An overall treatment effect (mean difference and associated 95% confidence interval (CI)) was obtained from the difference in scores under each study condition (NIPPV minus controls). Different proportions of participants contributed to the different outcomes.
Unit of analysis issues
In the case of cross-over trials, we considered only the first study period (prior to the cross-over).
Dealing with missing data
We contacted trialist and obtained IPD for all trials.
Assessment of heterogeneity
To consider the homogeneity among trials, a random factor was defined in the statistical models. Statistical significance (P value < 0.05) in the test of homogeneity suggested that the observed difference in the treatment effects was in part attributable to the study effect.
Assessment of reporting biases
We planned to inspect a funnel plot visually if we were able to meta-analyse 10 or more trials for an outcome.
We analysed IPD using a linear mixed model to compare the treatment effects. Treatment and time of follow-up (3 and 12 months) were analysed with interaction terms as fixed factors. We performed all the analyses using SAS version 9.3 (SAS Institute, Caru, NC).
Subgroup analysis and investigation of heterogeneity
We considered subgroup analyses if sufficient numbers of studies and a large enough sample size were to be included in the analysis and if we found significant heterogeneity among the outcomes of the trials. We identified a priori potential sources of heterogeneity among the primary and secondary outcomes. We postulated the following sources of heterogeneity:
- the more hypercapnic patients might benefit more from NIPPV;
- the benefits of NIPPV might be greater among those who used it for longer periods;
- people who received higher levels of inspiratory airway pressure (IPAP) might have a greater benefit of NIPPV.
Description of studies
Results of the search
The search conducted up to August 2012 identified 687 records of which we retrieved 17 full-text papers for further examination. We reduced the publications to 14 potentially eligible papers. We excluded five studies for the following reasons: not randomised (Clini 1998; Kamei 1999); duration of NIPPV too short (less than four hours per night and at daytime) (Diaz 1999; Renston 1994); training of NIPPV too short (less than three weeks) (Lin 1996). One study is awaiting classification as the authors failed to respond to clarify important issues (Xiang 2007). This update identified three new trials (Clini 2002; McEvoy 2009; Sin 2007), and together with the four trials from the original review (Casanova 2000; Gay 1996; Meecham Jones 1995; Strumpf 1991), we included seven studies in the meta-analysis.
Seven studies met the inclusion criteria for the review (Casanova 2000; Clini 2002; Gay 1996; McEvoy 2009; Meecham Jones 1995; Sin 2007; Strumpf 1991). The Characteristics of included studies table shows full details of the included studies. Five studies provided data on NIPPV after three months and are classified as 'short term' (Casanova 2000; Clini 2002; Gay 1996; Meecham Jones 1995; Sin 2007), while two studies provide data after 12 months and are defined as 'long term' (Clini 2002; McEvoy 2009). We obtained IPD for each of these studies from the trial authors. We provided summary details below.
The seven included studies were based in different countries: Spain (Casanova 2000), Italy (Clini 2002), USA (Gay 1996; Strumpf 1991), Australia (McEvoy 2009), UK (Meecham Jones 1995), and Canada (Sin 2007). Five studies compared nocturnal-NIPPV with standard treatment (Casanova 2000; Clini 2002; McEvoy 2009; Meecham Jones 1995; Strumpf 1991), and two studies compared nocturnal-NIPPV with sham treatment in the form of continuous positive airway pressure (CPAP) at 2 and 4 cm H
Funding of trial
Six studies were funded by their National Respiratory Society/Foundation (Casanova 2000; Clini 2002; Gay 1996; McEvoy 2009; Meecham Jones 1995; Sin 2007), from which two were also partly funded by an industrial company (Clini 2002; McEvoy 2009). One study was funded by an industrial company alone (Strumpf 1991).
The Characteristics of excluded studies table provides full details of the excluded studies.
Risk of bias in included studies
|Figure 1. Risk of bias summary: review authors' judgements about each methodological quality item for each included study.|
All studies were described as randomised and described the method of randomisation adequately. Allocation concealment was also deemed to be adequate in all studies, four of them describing a centralised randomisation office or independent person. In the other three cases, sequentially numbered, opaque, sealed envelopes were used.
Given the nature of the intervention it can be difficult to blind participants; however, two studies used a sham-device (Gay 1996; Sin 2007). In one study, personnel were also blinded for the treatment allocation making this the only study that we could classify as a low risk of bias in this area (Sin 2007). Three studies had blinding for all physiological measurements (Casanova 2000; Clini 2002; Strumpf 1991), two had no blinding (McEvoy 2009; Meecham Jones 1995), and one was unclear (Gay 1996). However, we judged that outcome measurement was not likely to be influenced by lack of blinding for personnel and, therefore, overall we judged all studies as low risk of bias.
Incomplete outcome data
In one of the cross-over studies, only seven out of the 19 randomised participants completed both arms (Strumpf 1991). Another study reported that four out of seven participants randomised to NIPPV completed the trial, as opposed to all six in the sham group (Gay 1996). Both studies were classified as high risk of attrition bias. Two other studies also had dropouts and did not perform intention-to-treat (ITT) analyses but as numbers due to intolerance were small these studies were classified as low risk (Meecham Jones 1995; Sin 2007). The three remaining studies all reported ITT analyses as well as per-protocol-analyses (or stated that "Inclusion of the patients who did not complete the trial (intent to treat) did not affect any of the outcomes") and were therefore considered as low risk of bias (Casanova 2000; Clini 2002; McEvoy 2009). In addition, in the long-term studies, reasons for missing data were not substantially different between both groups, for example; not being able to come for re-testing due to a worsening of the disease (Clini 2002; McEvoy 2009).
We could not find the original protocols to check if the prespecified outcomes were all reported in the articles, so in this area the risk of bias was unclear. However, all outcomes listed in the methods section of the studies were reported in the results section.
Other potential sources of bias
We did not find any other sources of bias.
Effects of interventions
Table 1 shows the results of the meta-analysis based on IPD. There is a difference between the number of participants included in the studies and the number included in the meta-analysis. There were some dropouts in most studies, for the short-term follow-up this was often due to intolerance of the nose mask, intercurrent infections or participants no longer meeting the inclusion criteria after a stabilisation period. In both studies looking at long-term effects, this was often due to progression of the disease and reluctance of participants to return to hospital for follow-up measurements. Finally, not all parameters were measured in all participants and, therefore, there is a difference in number of participants per outcome. In total, 245 participants were included in the IPD meta-analysis.
Arterial blood gas tensions
Three-month follow-up: all five short-term studies contributed data towards this outcome (Casanova 2000; Gay 1996; Meecham Jones 1995; Sin 2007; Strumpf 1991), as well as the Italian long-term study (Clini 2002), which also provided data after three months. In total, 162 participants were analysed for blood gases. The 95% CI of PaO
Twelve-month follow-up: two studies (Clini 2002; McEvoy 2009), with 118 participants gathered data for this outcome. There was no significant difference in PaO
Six-minute walking distance
Three-month follow-up: three studies with 40 participants measured 6MWD (Gay 1996; Meecham Jones 1995; Sin 2007). Meta-analysis showed a moderate treatment effect on 6MWD (MD 27.7, 95% CI -11.0 to 66.3), but this difference was not statistically different. Exercise endurance was reported by one study and determined by measuring treadmill walking time and could not be included in the meta-analysis (Strumpf 1991).
Twelve-month follow-up: as only one study measured 6MWD after 12 months meta-analysis was not possible (Clini 2002).
Three-month follow-up: only one study measured HRQoL after three months using the St. George's Respiratory Questionnaire making meta-analysis impossible (Meecham Jones 1995).
Twelve-month follow-up: Both long-term studies measured HRQoL in 103 participants after 12 months using three different questionnaires (Short Form-36 item (SF-36) questionnaire by McEvoy 2009; Maugeri Respiratory Failure questionnaire-28 by Clini 2002; St. George's Respiratory Questionnaire by Clini 2002 and McEvoy 2009), making it possible to only combine results for the St. George's Respiratory Questionnaire. The overall treatment effect was very small and was not significant (MD 0.90, 95% CI -19.21 to 21.01) and was found to be heterogeneous (P value = 0.03).
Three- month follow-up: all five short-term studies with 83 participants provided data for FEV
Twelve-month follow-up: both long-term studies measured FEV
Respiratory muscle function
Three-month follow-up: three studies with 48 participants provided data for PImax and PEmax (Casanova 2000; Gay 1996; Strumpf 1991). The improvement in PImax was not statistically different (MD 4.87 cm H
Twelve-month follow-up: as only one study reported data on PImax (Clini 2002), and none on PEmax, no meta-analyses were undertaken for these outcomes.
Three-month follow-up: three studies with 24 participants provided data for sleep efficiency (Gay 1996; Meecham Jones 1995; Strumpf 1991) showing a small negative effect after three months (MD -9.11, 95% CI -38.09 to 19.86). This effect was heterogeneous. The study by Strumpf et al with only seven participants, reported a very broad 95% CI (MD 25.4, 95% CI -69.17 to 70.4). Subgroup analysis was not performed due to the low number of trials.
Twelve-month follow-up: sleep quality was measured differently by both long-term studies. One reported on sleep efficiency by measuring time asleep as percentage of total time in bed (McEvoy 2009), but only performed follow-up measurements in the NIPPV group. The other study also measured sleep quality but by means of a semi-qualitative multipoint scale of 1 to 4 (Clini 2002). No meta-analyses could be performed.
Three-month follow-up: dyspnoea was measured in two studies, but as they were measured with different scales (the Medical Research Council (MRC) scale and Borg scale by Casanova 2000 or Transitional Dyspnea Index (TDI) by Strumpf 1991), data could not be combined.
Twelve-month follow-up: one study (Clini 2002) measured dyspnoea after 12 months by means of a 6-point MRC score. No meta-analysis could be performed.
Summary of main results
In this update, IPD from three new studies were added to the original review (Wijkstra 2002), and two of these studies were conducted over 12 months, which means we now have some long-term data on which to base our conclusions, including long-term information on quality of life outcomes. Nocturnal-NIPPV at home for at least three months in hypercapnic patients with stable COPD had no consistent clinically or statistically significant effect on gas exchange, exercise tolerance, lung function, respiratory muscle strength or sleep efficiency. Meta-analysis of the two new long-term studies did not show significant improvements in blood gases, HRQoL or lung function after 12 months of NIPPV.
Overall completeness and applicability of evidence
In this meta-analysis, we did not find statistically significant effects in any included outcomes. By adding data of PaCO
Although the improvement of 27.7 m in the 6MWD is not statistically significant, it could be clinically significant as it does reach the clinically minimal important difference of 26 m (Puhan 2011). This meta-analysis of 6WMD after three months, however, was only based on 40 participants and although not contributing to the meta-analysis, the long-term study of Clini 2002 looking at effects on 6MWD after 12 months of NIPPV only found a very small improvement of 3.2 m (95% CI -49.7 to 56.1) in 46 participants.
The upper limit of the CI of 66 m for the 6MWD in the meta-analysis suggests that it remains possible that NIPPV has beneficial effects on walking in at least some people, but it is not possible to identify these people a priori. Additional studies with larger sample sizes that address participant selection, ventilator settings, training and NIPPV compliance should clarify the role of this treatment.
Not all outcomes could be combined because of measurements with different scales. Dyspnoea was measured in three studies, but as this was measured with the 5-point MRC scale (Casanova 2000) and 6-point MRC scale (Clini 2002), Borg (Casanova 2000), or TDI (Strumpf 1991) and different lengths of follow-up, data could not be combined. The same applied to HRQoL; three studies examined this outcome but all used different questionnaires ranging from the more generic SF-36 questionnaire (McEvoy 2009) and the (lung) disease specific St. George's Respiratory Questionnaire (Clini 2002; McEvoy 2009; Meecham Jones 1995) to the Maugeri Respiratory Failure questionnaire-28 (MRF-28) (Clini 2002), designed for people with respiratory failure. For future studies, it would be of great benefit if the same questionnaires were to be used making comparison and pooling of data possible. The effect of NIPPV may not be captured by the HRQoL questionnaires currently analysed. Several studies have shown good reliability and validity of the Severe Respiratory Insufficiency (SRI) questionnaire and the MRF-28, specifically for people with COPD with hypercapnic respiratory failure (Duiverman 2008; Windisch 2003). It has been suggested to use both these questionnaires, as the SRI focuses more on psychological aspects and the MRF-28 focuses on restrictions of daily living. But moreover, a large trial also looking at the effect of NIPPV on survival, exacerbation frequency and admissions is needed.
Quality of the evidence
Seven (three new to this update) RCTs with individual data from 245 people were included in this systematic review. Meta-analysis was performed when trials used similar outcome measures. Short-term trials were analysed as one group, measuring effects of NIPPV after three months. Long-term trials comprised the second group, measuring effects of NIPPV after 12 months. All studies described how randomisation was performed and described adequate allocation concealment. Mainly the short-term studies had few participants (between 7 and 36 participants) and sometimes showed large CIs for some outcome measures such as PImax (Strumpf 1991). Two studies show a high dropout rate after randomisation because people were not able to tolerate NIPPV (Gay 1996; Strumpf 1991). Only results from the completers were reported, which could make outcomes susceptible to selection bias. In the other studies, dropout due to non-tolerance was considered small. The two larger studies with long-term follow-up had quite large dropout rates but mainly due to progression of the disease and unwillingness to repeat tests; these were similar amounts between groups. Both studies performed ITT analysis and per-protocol analysis. In this systematic update we only included complete data.
The meta-analysis in this review was performed based on IPD. Treatment effects can, therefore, be seen as more conservative (with wider CIs). They take into account not only interstudy variation, but also intrastudy variation.
Potential biases in the review process
Limitations regarding the setup of the meta-analyses
The design of this meta-analysis included only studies in which nocturnal-NIPPV was applied for at least five hours per night. This excluded two studies that reported beneficial effects from NIPPV administered for two hours during the day (Diaz 1999; Renston 1994). In keeping with the application of mechanical ventilatory support for people with thoracic restriction or neuromuscular conditions, we considered night-time ventilation to be the most appropriate clinical approach and reasoned that several hours would be required to achieve therapeutic goals. Furthermore, a minimum duration of three weeks was chosen, as from our own clinical experience we were aware that it might take up to two weeks just for mask fitting, adjustment and patient familiarisation with non-invasive ventilation. Therefore, one study in which NIPPV was assessed for only two weeks was excluded from the analysis (Lin 1996).
Limitations regarding the included studies
We included RCTs that determined the effects of NIPPV versus normal medical treatment. However, it is questionable whether NIPPV with IPAP pressures below 14 cm H
One published, long-term RCT that compared the effects of NIPPV, not to standard care, but in addition to rehabilitation (Duiverman 2011), showed a significant decrease in PaCO
Agreements and disagreements with other studies or reviews
In 2007, one systematic review on NIPPV in people with stable COPD was published (Kolodziej 2007). An important difference with this systematic review is the inclusion of non-RCTs. They found improvements in blood gasses, hyperinflation and work of breathing, but only in a combined analysis of non-randomised trials with evident heterogeneity. This heterogeneity was probably the result of including studies with different lengths of follow-up, usages per day and types of ventilation (during the day or at night).
Implications for practice
Nocturnal-non-invasive positive pressure ventilation at home for 3 and 12 months in hypercapnic patients with COPD had no clinically or statistically significant effect on gas exchange, exercise tolerance, quality of life, lung function, respiratory muscle strength or sleep efficiency.
Implications for research
Future research should focus on adequate patient selection, ventilator settings, training and length of ventilation, as well as exacerbation frequency, admissions to hospital and survival. During ventilation, people should be monitored carefully, to observe more precisely the changes that are occurring with non-invasive ventilation. Long-term non-invasive ventilation for people with COPD should only be started in the context of a clinical trial, preferably with agreed upon common outcome parameters.
We gratefully thank the authors of the seven included studies; Ciro Casanova, Enrico Clini, Peter Gay, Doug McEvoy, Jeffrey Meecham Jones, Don Sin and David Strumpf (Carol Carlisle and Nicholas Hill) who provided individual patient data for this meta-analysis. We also wish to acknowledge the assistance provided by Serge Simard, biostatistician and the Cochrane Airways Review Group (for the original review: Steve Milan and Toby Lasserson, for the update: Emma Welsh and Susan Ann Hansen) and the statistical help from Dr. Chris Cates.
Data and analyses
This review has no analyses.
Appendix 1. Sources and search methods for the Cochrane Airways Group Specialised Register (CAGR)
Electronic searches: core databases
Handsearches: core respiratory conference abstracts
MEDLINE search strategy used to identify trials for the CAGR
1. Lung Diseases, Obstructive/
2. exp Pulmonary Disease, Chronic Obstructive/
4. (chronic$ adj3 bronchiti$).mp.
5. (obstruct$ adj3 (pulmonary or lung$ or airway$ or airflow$ or bronch$ or respirat$)).mp.
Filter to identify RCTs
1. exp "clinical trial [publication type]"/
2. (randomised or randomised).ab,ti.
11. 9 not (9 and 10)
12. 8 not 11
(The MEDLINE strategy and RCT filter are adapted to identify trials in other electronic databases)
Last assessed as up-to-date: 1 August 2012.
Protocol first published: Issue 2, 2001
Review first published: Issue 2, 2002
Contributions of authors
Update 2012: Struik and Wijkstra searched and reviewed relevant papers, collected IPD of the newly included RCTs, authors of the review. Lacasse: statistical analysis of IPD, co-author of the review. Goldstein and Kerstjens: review development, co-author of the review.
Original review: Wijkstra: searched and reviewed relevant papers, collected IPD of included RCTs, author of the review. Lacasse: statistical analysis of IPD, co-author of the review. Guyatt: co-author of the review. Goldstein: search and reviewing of relevant papers, co-author of the review.
Declarations of interest
Differences between protocol and review
We selected three of the primary outcomes from the last update as our primary outcomes in consultation with the Cochrane Airways Group.
Medical Subject Headings (MeSH)
Diaphragm [physiopathology]; Hypercapnia [etiology; physiopathology; *therapy]; Partial Pressure; Positive-Pressure Respiration [*methods]; Pulmonary Disease, Chronic Obstructive [complications; physiopathology; *therapy]; Pulmonary Gas Exchange; Quality of Life; Sleep [physiology]; Time Factors; Walking
MeSH check words