Systematic review on the clinical effectiveness of continuous positive airway pressure

Alison McMillan; Daniel J Bratton; Rita Faria; Magda Laskawiec-Szkonter; Susan Griffin; Robert J Davies; Andrew J Nunn; John R Stradling; Renata L Riha; Mary J Morrell

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

McMillan A, Bratton DJ, Faria R, et al. A multicentre randomised controlled trial and economic evaluation of continuous positive airway pressure for the treatment of obstructive sleep apnoea syndrome in older people: PREDICT. Southampton (UK): NIHR Journals Library; 2015 Jun. (Health Technology Assessment, No. 19.40.)

Cover of A multicentre randomised controlled trial and economic evaluation of continuous positive airway pressure for the treatment of obstructive sleep apnoea syndrome in older people: PREDICT

A multicentre randomised controlled trial and economic evaluation of continuous positive airway pressure for the treatment of obstructive sleep apnoea syndrome in older people: PREDICT.

Show details

Contents

< Prev Next >

Appendix 5Systematic review on the clinical effectiveness of continuous positive airway pressure

Methods

A systematic review on the clinical effectiveness of CPAP was conducted to identify RCTs comparing CPAP with sham CPAP, BSC/usual care and dental devices in a patient population with an average age of 60 years or over with OSAS and capacity to give informed consent. Therefore, studies in patients with dementia were excluded. The literature searches updated the searches conducted for the McDaid et al.²⁰ report on CPAP and dental devices but focused on the period from January 2006 to March 2012. The studies identified in this review were checked for their relevance for our systematic review.

Results

The updated searches found 4872 titles, which were added to the library with the results of the cost-effectiveness searches and de-duplicated. In total, the searches identified 3560 unique titles. Of these, 67 studies appeared to be potentially relevant for the systematic review on clinical effectiveness. Figure 32 presents the flow diagram of identification and selection of studies. The searches retrieved six potentially relevant reviews on the use of CPAP.²⁰^,²⁰⁹^–²¹² The studies included in these reviews were examined. Only one study met our inclusion criteria;⁸⁸ however, since it was a conference abstract referring to another title included in the review,⁸⁸ it was subsequently excluded. From the other 60 titles identified, three studies met our inclusion criteria.⁸⁸^–⁹⁰ The systematic review in the previous HTA report had identified 48 relevant studies out of 6325 potentially relevant titles.²⁰ Given that the average age across the 48 studies ranged from 44 years to 58 years, none met our inclusion criteria and therefore none was included in our review. In sum, the systematic review on clinical effectiveness of CPAP identified three relevant RCTs comparing CPAP with sham CPAP, BSC/usual care and dental devices in a patient population with an average age of 60 years or over with OSAS.

FIGURE 32

Flow diagram showing number of studies identified and included in the review of clinical effectiveness of CPAP.

Table 69 summarises the characteristics of the studies included in the systematic review of clinical effectiveness. Full data extraction tables can be found in the section Data extraction tables.

TABLE 69

Characteristics of included studies

The three studies included in the systematic review compared CPAP therapy with sham CPAP,⁸⁸ or no CPAP⁸⁹^,⁹⁰ for OSAS in the secondary care setting in patients with cardiovascular conditions. None of the studies was conducted in the UK. The studies varied in duration: 1 month in Ruttanaumpawan et al.,⁸⁹ 3 months in Egea et al.⁸⁸ and 24 months in Parra et al.⁹⁰ The primary outcome was different for each study: left ventricular ejection fraction,⁸⁸ baroreflex sensitivity⁸⁹ and a number of neurological, quality of life, sleep-related and mortality outcomes.⁹⁰ Common secondary outcomes examined by two or more studies were ESS score,⁸⁸^–⁹⁰ BP⁸⁸^,⁸⁹ and quality of life with SF-36.⁸⁸^,⁹⁰

Egea et al.⁸⁸ and Ruttanaumpawan et al.⁸⁹ included patients with chronic heart failure referred to the sleep clinic whose AHI was greater than 10 events/hour⁸⁸ or equal or greater than 20 events/hour,⁸⁹ while Parra et al.⁹⁰ included patients admitted with first ever ischaemic stroke with Apnoea–Hypopnoea Index equal or greater than 20 events/hour.⁹⁰ Egea et al.⁸⁸ included patients with Cheyne–Stokes apnoea (17% of the study population) but presented results for the subgroup of patients with confirmed OSAS. Participants were mostly male and overweight or obese, as indicated by an average BMI of at least 28 kg/m². The AHI and ESS measures at baseline suggest that patients across the three studies suffered from moderate to severe OSAS.

Egea et al.⁸⁸ observed a statistically significant improvement in left ventricular ejection fraction and in ESS score in the CPAP group but no statistically significant differences were recorded for BP and quality of life. Results were similar for the subgroup of patients with confirmed OSAS. Ruttanaumpawan et al.⁸⁹ also observed a statistically significant improvement in baroreflex sensitivity, Apnoea–Hypopnoea Index, heart rate and systolic BP. In Parra et al.,⁹⁰ the CPAP group experienced a statistically significantly higher improvement in the neurological outcomes at 1 month, which was not sustained throughout follow-up. There were no statistically significant differences in SF-36 scores at any of the data collection points (1, 3, 12 and 24 months).

Across the three studies, CPAP appears to improve sleep function, cardiovascular outcomes and quality of life in patients over 60 years of age. However, the limitations of the studies prevent definitive conclusions. First, all three studies included only patients with cardiac conditions, who are unlikely to represent all patients over 60 years of age with OSAS. Second, two studies had short follow-up and small samples sizes. Given that Parra et al.⁹⁰ found that statistically significant differences at 1 month were not sustained at longer follow-ups, there are doubts on whether or not the results observed at 1 month⁸⁹ and at 3 months⁸⁸ can be extrapolated over longer time horizons. Third, none of the studies collected measures of HRQoL, such as EQ-5D or SF-6D, that can be used for cost-effectiveness analysis.

Data extraction tables

TABLE 70

Egea et al. data extraction table

Study details	Egea et al. (2008)⁸⁸
Intervention	CPAP therapy
Comparator(s)	Sham CPAP therapy
Study setting	Spain
Design	Randomised multicentre controlled trial
Duration	3 months
Inclusion criteria	Patients with chronic heart failure referred to the sleep laboratory Diagnosis of heart failure with at least one episode of cardiac failure Left ventricular ejection fraction < 45% using radionucleotide ventriculography Clinically stable for at least 1 month prior to inclusion Optimum treatment with diuretics and/or beta-blockers and/or digoxin and/or angiotensin-converting enzyme inhibitors according to tolerance No change in treatment for 1 month prior to inclusion AHI (events/hour) > 10, measured by polysomnography
Exclusion criteria	Patients who had a previous diagnosis of sleep apnoea or who had had CPAP therapy Uncontrolled arterial hypertension Valvular or congenital cardiopathy, unstable angina, acute MI or cardiac surgery in the 3 months prior to enrolment Severe somnolence in situations of activity Present or past medical history of clinically significant renal, liver or pulmonary disease Untreated hypothyroidism Clinically significant kyphoscoliosis Morbid obesity with a BMI > 41 kg/m² Concomitant use of morphine, hypnotics and sedatives, theophylline, acetazolamide (Diamoz^®, AMCo) or home oxygen therapy
Outcomes	Primary outcome: left ventricular ejection fraction
Outcomes	Secondary outcomes: BP (systolic and diastolic) Subjective daytime sleepiness as measured by ESS score Quality of life as measured by SF-36 Dyspnoea using the Borg scale, New York Heart Association scale and the 6-minute walking test
Pre-defined subgroups	Patients with Cheyne–Stokes apnoea defined as: Cheyne–Stokes apnoeas reported in over 50% of total sleep time
Pre-defined subgroups	Respiratory events were classified as: Obstructive apnoea: airflow cessation with thoraco-abdominal motion lasting 10 seconds Obstructive hypopnoea: discernible airflow reduction with thoracoabdominal motion lasting 10 seconds or more associated with arousal and/or a 3% cyclical SaO₂ dip Cheyne–Stoke apnoea: absence of flow without thoracoabdominal motion followed by a clear-cut crescendo–decrescendo pattern of breathing
Participants: number randomised	73 patients randomised: 35 for CPAP 38 for sham CPAP
Participants: number of withdrawals with reasons	7 patients randomised for CPAP withdrawn: 2 refused for personal reasons; 5 CPAP intolerant
Participants: number of withdrawals with reasons	6 patients randomised for sham CPAP withdrawn: 1 death 1 angina 4 sham CPAP intolerant
Baseline characteristics	Baseline characteristics (adapted from Table 1, p. 663)⁸⁸
			CPAP (n = 28)		Sham CPAP (n = 32)
	Age (years), mean (SE)		64 (0.9)		63 (1.6)
	Sex (% male)		96		91
	BMI (kg/m²), mean (SE)		31.7 (2.4)		30.5 (1.6)
	Daily snoring (%)		83		69
	Snoring three or more times/week (%)		88		76
	ESS score, mean (SE)		8.0 (0.7)		7.3 (0.8)
	Minimum SaO₂ (%), mean (SE)		76.9 (2.0)		77.4 (2.1)
	AHI (events/hour), mean (SE)		43 (4.4)		41 (5.6)
	Systolic BP (mmHg), mean (SE)		123 (3.7)		126 (2.9)
	Diastolic BP (mmHg), mean (SE)		76 (2.3)		75 (2.1)
	Pretibial oedema (%)		19		13
	Jugular ingurgitation (%)		8		0
	Left ventricular ejection fraction (%), mean (SE)		28.0 (0.5)		28.1 (1.5)
	Statistically significant results at p = 0.05 in bold
Results	Results at 3 month follow-up (adapted from Table 3, page 665)⁸⁸
		Baseline	3 months	Baseline	3 months
	Left ventricular ejection fraction (%), mean (SE)	28.0 (1.5)	30.5 (0.8)	28.1 (1.5)	28.1 (1.7)
	ESS score, mean (SE)	8.0 (0.7)	4.8 (0.6)	7.1 (0.8)	5.3 (0.7)
	Systolic BP (mmHg), mean (SE)	123.0 (3.7)	123.0 (4.1)	124.2 (2.8)	120.5 (2.6)
	Diastolic BP (mmHg), mean (SE)	76.3 (2.3)	75.3 (2.3)	74.8 (2.1)	75.2 (3.1)
	SF-36 physical, mean (SE)	41.8 (1.8)	45.1 (1.4)	42.1 (1.7)	41.3 (1.9)
	SF-36 mental, mean (SE)	47.8 (2.4)	49.9 (2.0)	47.0 (2.3)	49.8 (1.8)
	New York Heart Association (% class I–II)	75	82	65	74
	New York Heart Association (% class II)	64	71	58	74
	6-minute walking test (m), mean (SE)	424 (20)	420 (19)	394 (20)	405 (22)
	Statistically significant results at p = 0.05 in bold
Results for subgroups	Results at 3-months follow-up for the patients with obstructive apnoea or hypopnoea (adapted from Table 4, page 665)⁸⁸
		CPAP (n = 20)		Sham CPAP (n = 25)
		Baseline	3 months	Baseline	3 months
	Left ventricular ejection fraction (%), mean (SE)	28.8 (1.6)	31.0 (1.6)	27.2 (1.6)	26.7 (1.7)
	ESS score, mean (SE)	8.6 (0.8)	5.0 (0.8)	6.9 (5.2)	5.2 (0.8)
	Systolic BP (mmHg), mean (SE)	124.3 (4.2)	124.3 (4.9)	125 (2.7)	123.4 (2.8)
	Diastolic BP (mmHg), mean (SE)	75.6 (2.3)	76.0 (2.8)	75.8 (2.4)	77.0 (3.7)
	SF-36 physical, mean (SE)	41.4 (2.0)	44.9 (1.8)	42.0 (2.1)	40.7 (2.1)
	SF-36 mental, mean (SE)	46.4 (3.0)	48.8 (2.3)	45.8 (2.7)	48.7 (2.2)
	New York Heart Association (% class I–II)	60	70	50	67
	6-minute walking test (m), mean (SE)	403 (21)	406 (21)	381 (23)	393 (24)
	Statistically significant results at p = 0.05 in bold
Conclusions	CPAP therapy increases left ventricular ejection fraction in patients with associated sleep-related disordered breathing and severe chronic heart failure, however, this improvement was not translated into an improvement in quality of life or cardiac function
Limitations	Small sample size Difficult to generalise given that patients were predominantly male and with moderate to severe OSA Long-term outcomes, namely mortality, not evaluated
Conflicts of interest	None

: AHI, Apnoea–Hypopnoea Index.

TABLE 71

Rutanaumpawan et al. data extraction table

Study details	Rutanaumpawan et al. (2008)⁸⁹
Intervention	CPAP
Comparator(s)	Usual care for heart failure
Study setting	Canada
Design	RCT
Duration	1 month
Inclusion criteria	Patients with heart failure due to ischaemic or non-ischaemic dilated cardiomyopathy of more than 6 months’ duration Left ventricular ejection fraction of ≤ 45% by nuclear angiography or echocardiography Stable condition for at least 3 months prior to the study Moderate to severe OSA, defined as AHI ≥ 20 events/hour, with more than 50% of events obstructive
Exclusion criteria	Primary valvular heart disease Atrial fibrillation or a paced cardiac rhythm Ventricular premature beats > 15 beats for 100 heart beats Unstable angina, MI or cardiac surgery within the 3 months before to the study
Outcomes	Primary outcome: baroreflex sensitivity
Outcomes	Secondary outcomes: Sleep-related: AHI, mean and lowest oxyhaemoglobin saturation, arousal index, total sleep time, stages I and II sleep, slow-wave sleep, rapid eye movement sleep Cardiovascular outcomes: left ventricular ejection fraction; heart rate; systolic and diastolic BP
Pre-defined subgroups	None
Participants: Number randomised	34 enrolled
Participants: number of withdrawals with reasons	1 patient excluded because of technical problems in assessing baroreflex sensitivity 33 patients: 14 randomised to control and 19 to CPAP
Baseline characteristics	Baseline characteristics (adapted from table 1, page 1165)⁸⁹
	Selected variables		CPAP (n = 19)		Control (n = 14)
	Age (years), mean (SE)		59.0 (7.8)		60.5 (10.3)
	Sex (% male)		94.7		85.7
	Cause of heart failure
	Ischaemic (%)		63		64
	Non-ischaemic (%)		37		36
	New York Heart Association functional class, mean (SE)		2.4 (0.6)		2.3 (0.4)
	BMI (kg/m²), mean (SE)		30.3 (5.8)		32.3 (8.6)
	Left ventricular ejection fraction (%), mean (SE)		29.0 (11.4)		30.8 (8.9)
	SBP (mmHg), mean (SE)		122 (15)		131 (24)
	Diastolic BP (mmHg), mean (SE)		66 (12)		64 (14)
	Data expressed as means (SE). Statistically significant results at p = 0.05 in bold
Results	Results at 1-month follow-up (adapted from tables 2 and 3 pages 1165–6)⁸⁹
	Selected variables	CPAP (n = 19)		Control (n = 14)
		Baseline	1 month	Baseline	1 month
	Sleep-related
	AHI events per hour, mean (SE)	36.2 (18.1)	9.3 (8.7)	51.3 (15.6)	47.4 (19.1)
	Mean SaO₂ (%), mean (SE)	94.7 (1.6)	96.1 (1.6)	94.3 (2.1)	94.1 (2.0)
	Cardiovascular
	Left ventricular ejection fraction (%), mean (SE)	29.0 (11.4)	36.1 (10.6)	30.8 (8.9)	29.4 (8.0)
	Heart rate (bpm), mean (SE)	66 (8)	62 (8)	66 (11)	66 (9)
	SBP (mmHg), mean (SE)	122 (15)	113 (12)	131 (24)	136 (28)
	Diastolic BP (mmHg), mean (SE)	67 (12)	61 (9)	64 (14)	63 (12)
	Baroreflex sensitivity
	Slope (ms/mmHg), median (IQR)	5.4 (2.2; 8.3)	7.9 (4.4–9.4)	4.9 (3.1–8.7)	4.7 (2.9–7.4)
	Slope in +SBP/+R–R sequences, median (IQR)	4.7 (2.4; 8.5)	8.1 (4.5–12.6)	5.5 (3.3–10.0)	4.0 (3.2–6.9)
	Slope in SBP/–R–R sequences, median (IQR)	5.4 (2.2; 8.8)	5.6 (4.3–10.0)	5.5 (4.1–11.2)	5.0 (2.7–7.7)
	Between-groups statistically significant results at p = 0.05 in bold
Results for subgroups	None
Conclusions	Treatment of OSA in patients with heart failure improves baroreflex sensitivity and cardiovascular signs
Limitations	Small sample size Short follow-up No quality-of-life assessment
Conflicts of interest	None

: AHI, Apnoea–Hypopnoea Index; SBP, systolic blood pressure.

TABLE 72

Parra et al. data extraction table

Study details	Parra et al. (2011)⁹⁰
Intervention	Nasal CPAP in addition to usual stroke care
Comparator(s)	Usual stroke care, in accordance with the recommendations of the Spanish Cerebrovascular Study Group of the Spanish Society of Neurology
Study setting	Spain
Design	Randomised controlled multicentre study
Duration	24 months
Inclusion criteria	Patients who were admitted to hospital with first-ever ischaemic stroke Age < 75 years AHI ≥ 20 events/hour At least one of the following conditions: habitual snoring observed apnoeas history of hypertension or ischaemic heart disease
Exclusion criteria	Patients with consciousness impairment Patients previously diagnosed and treated for OSA
Outcomes	Assessments at baseline, 1, 3, 12 and 24 months after stroke Barthel index for functional abilities Canadian scale for neurological impairment Modified Rankin scale for outcomes from stroke SF-36 for quality of life Questionnaire for sleep–wake habits and symptoms ESS score for daytime sleepiness
Pre-defined subgroups	None
Participants: Number randomised	140 patients were randomised but only 126 were followed up: 71 patients randomised to CPAP 69 randomised to no treatment
Participants: Number of withdrawals with reasons	14 patients randomised to CPAP dropped-out because of machine discomfort
Baseline characteristics	Baseline characteristics (selected data from Table 1, page 1131)⁹⁰
		CPAP (n = 57)	Control (n = 69)
	Age (years), mean (SE)	63.7 (9.1)	65.5 (9.1)
	Sex (% male)	71.9	69.6
	BMI (kg/m²), mean (SE)	30.2 (4.6)	28.8 (4.0)
	Snoring, often or always (%)	94.7	85.5
	Observed apnoea at night, often or always (%)	70.2	46.4
	ESS score, mean (SE)	8.3 (3.3)	7.3 (4.1)
	AHI (events/hour), mean (SE)	38.4 (12.6)	38.4 (14.6)
	Physical component SF-36, mean (SE)	42.3 (11.1)	43.1 (7.8)
	Mental component SF-36, mean (SE)	47.1 (13.3)	48.2 (12.9)
	Statistically significant results at p = 0.05 in bold.
Results	Mean CPAP use was 5.3 hours (SE = 1.9 hours) per night during an average of 6.8 (SE = 0.6) nights a week
	Results (adapted from table 3 page 1133)⁹⁰
		CPAP (n = 57), mean (SE)	Control (n = 69), mean (SE)
	Barthel index
	Baseline	75.9 (27.9)	73.6 (27.0)
	3 months	95.0 (13.4)	92.8 (17.8)
	12 months	95.3 (10.0)	91.4 (17.8)
	24 months	94.3 (10.9)	93.1 (15.8)
	Canadian scale
	Baseline	8.3 (1.6)	8.0 (1.9)
	3 months	9.3 (1.0)	9.3 (1.3)
	12 months	9.4 (1.2)	9.4 (1.3)
	24 months	9.3 (1.3)	9.5 (1.0)
	Rankin scale
	Baseline	2.3 (1.3)	2.8 (1.3)
	3 months	1.6 (0.9)	2.0 (1.1)
	12 months	1.6 (0.9)	2.1 (1.2)
	24 months	1.8 (1.1)	2.2 (1.1)
	SF-36 physical
	Baseline	42.6 (10.2)	42.3 (11.8)
	3 months	44.9 (9.2)	44.8 (11.8)
	12 months	46.7 (8.8)	46.5 (11.7)
	24 months	45.8 (10.0)	46.0 (9.8)
	SF-36 mental
	Baseline	43.3 (13.2)	43.7 (14.1)
	3 months	46.9 (10.9)	46.3 (14.4)
	12 months	49.1 (14.0)	44.6 (12.8)
	24 months	47.6 (13.8)	47.8 (12.1)
	Statistically significant results at p = 0.05 in bold. The overall cardiovascular event-free survival rate after 24 months was 87.7% (50 out of 57 subjects) in the CPAP group and 88.4% (61 out of 69) in the control group (p = 0.911)
Results for subgroups	None
Conclusions	Early use of CPAP in patients with a first-ever ischaemic stroke and moderate to severe OSA is associated with a significant improvement in neurological function, but this improvement is not sustained at follow-up
Limitations	Limited generalisability owing to exclusion criteria (patients with altered consciousness or age > 75 years of age were excluded) Non-placebo controlled Per protocol analysis Neurological measures may not be sensitive enough to detect changes in patients with minor stroke
Conflicts of interest	None

: AHI, Apnoea–Hypopnoea Index.

Copyright © Queen’s Printer and Controller of HMSO 2015. This work was produced by McMillan et al. under the terms of a commissioning contract issued by the Secretary of State for Health. This issue may be freely reproduced for the purposes of private research and study and extracts (or indeed, the full report) may be included in professional journals provided that suitable acknowledgement is made and the reproduction is not associated with any form of advertising. Applications for commercial reproduction should be addressed to: NIHR Journals Library, National Institute for Health Research, Evaluation, Trials and Studies Coordinating Centre, Alpha House, University of Southampton Science Park, Southampton SO16 7NS, UK.

Included under terms of UK Non-commercial Government License.

Bookshelf ID: NBK299275

Contents

< Prev Next >

PubReader
Print View
Cite this Page
McMillan A, Bratton DJ, Faria R, et al. A multicentre randomised controlled trial and economic evaluation of continuous positive airway pressure for the treatment of obstructive sleep apnoea syndrome in older people: PREDICT. Southampton (UK): NIHR Journals Library; 2015 Jun. (Health Technology Assessment, No. 19.40.) Appendix 5, Systematic review on the clinical effectiveness of continuous positive airway pressure.
PDF version of this title (3.9M)

Systematic review on the clinical effectiveness of continuous positive airway pr...
Systematic review on the clinical effectiveness of continuous positive airway pressure - A multicentre randomised controlled trial and economic evaluation of continuous positive airway pressure for the treatment of obstructive sleep apnoea syndrome in older people: PREDICT

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

Bookshelf