0
Original Research: SLEEP DISORDERS |

Flexible Pressure Delivery Modification of Continuous Positive Airway Pressure for Obstructive Sleep Apnea Does Not Improve Compliance With Therapy: Systematic Review and Meta-analysis FREE TO VIEW

Jessie P. Bakker, PhD; Nathaniel S. Marshall, PhD
Author and Funding Information

From the WellSleep Sleep Investigation Centre (Dr Bakker), Department of Medicine, University of Otago, Wellington, New Zealand; and National Health and Medical Research Council (NHMRC) Centre for Integrated Research and Understanding of Sleep (CIRUS) (Dr Marshall), Woolcock Institute of Medical Research, University of Sydney, Sydney, NSW, Australia.

Correspondence to: Jessie P. Bakker, PhD, Department of Medicine, University of Otago, PO Box 7343, Wellington 6242, New Zealand; e-mail: jessie.bakker@otago.ac.nz


Funding/support: Dr Bakker’s salary that was used to support this project was provided internally from the University of Otago, New Zealand. Dr Marshall’s salary was provided by the Australian government through the NHMRC-funded CIRUS (NHMRC #571421).

For editorial comment see page 1266

Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (http://www.chestpubs.org/site/misc/reprints.xhtml).


© 2011 American College of Chest Physicians


Chest. 2011;139(6):1322-1330. doi:10.1378/chest.10-2379
Text Size: A A A
Published online

Background:  Continuous positive airway pressure (CPAP) is the first-line therapy for obstructive sleep apnea (OSA), but patient compliance is a major barrier to long-term effectiveness. Flexible pressure delivery of PAP reduces pressure during early exhalation with the aim of improving comfort and, therefore, compliance, leading to subsequent symptoms improvement.

Methods:  We undertook a systematic literature search of PubMed (January 1, 2000, to July 11, 2010) to identify all randomized trials comparing flexible and standard CPAP in adult patients with OSA with at least 1-week follow-up. Either we or the original trial investigators extracted means, SEs, and sample sizes for all relevant outcome measures. We then performed meta-analyses quantifying improvements in objective compliance and symptoms as measured by the Epworth Sleepiness Scale (ESS), the Maintenance of Wakefulness Test (MWT), and the Psychomotor Vigilance Task (PVT).

Results:  We found 10 relevant trials (599 patients). Meta-analysis of the seven trials where we could extract compliance data (514 patients) indicated that flexible pressure did not improve compliance compared with CPAP in either the parallel (0.16 h; 95% CI, −0.09-0.42; P = .21) or the crossover trials (0.20 h; 95% CI, −0.26-0.66; P = .39). Flexible pressure caused no improvement over CPAP in any secondary outcome (ESS, MWT, PVT, and residual OSA, all P > .05).

Conclusions:  Flexible pressure modification neither significantly improves compliance with CPAP in patients with OSA nor significantly improves patient outcomes beyond the effects of CPAP. Unfortunately, we were unable to locate compliance data in the correct format for three out of the 10 suitable trials.

Figures in this Article

Obstructive sleep apnea (OSA) is a condition affecting around 24% and 9% of adult men and women, respectively,1 in which the upper airway is intermittently partially or completely obstructed during sleep.2 The consequences of OSA include daytime sleepiness, neurocognitive decline, and, in the long term, increased cardiovascular morbidity and mortality.35

Continuous positive airway pressure (CPAP)6 is regarded as the first-line treatment of moderate to severe OSA7 because it eliminates obstructive apneic/hypopneic events,8 resulting in improved daytime symptoms9 and possibly reduced adverse cardiovascular outcomes.4,10 As with all medical treatments, the effectiveness of CPAP can be limited by suboptimal compliance and high rejection rates.11

Various technologic modifications of CPAP have been developed with the rationale of improving compliance; these include the addition of pressure ramping at sleep onset and humidification of applied air, which are now standard in most CPAP devices. Automatically adjusting positive airway pressure (AutoPAP) and flexible pressure devices also have been marketed as improving compliance. Flexible pressure contrasts with the constant pressure delivery of CPAP by reducing pressure for the first part of exhalation then restoring it to a therapeutic level for the latter part of expiration and subsequent inhalation.12 It has been reported that there is no difference in inspiratory flow limitation between flexible pressure and CPAP, but flexible pressure may lead to a decrease in expiratory time compared with CPAP.13

Although ResMed Ltd (Sydney, New South Wales, Australia), Hoffrichter GmbH (Schwerin, Germany), and Breas Medical AB (Molnlycke, Sweden) all have released a flexible pressure delivery mode (Expiratory Pressure Relief [EPR], Flexline, and eAdapt, respectively), the first market entrant with this technology was Philips Respironics (Murrysville, Pennsylvania). As such, the majority of published research has used its devices, branded as C-Flex and A-Flex (flexible pressure incorporated into an AutoPAP device). These technologies reduce expiratory pressure in proportion to expiratory flow.12

One previous Cochrane meta-analysis investigated the compliance-improving effects of a range of CPAP modifications, which included a subanalysis of flexible pressure.14 In five randomized controlled trials of flexible pressure (267 patients), a nonsignificant improvement in compliance was observed (parallel studies, 0.20 h; P = .20; crossover studies, 0.09 h; P = .92). Since that literature search in September 2008, a number of additional trials with almost 300 more patients have been published, including one of our own.1517 The previous meta-analysis also focused much more broadly on any PAP modifications for improving compliance, including AutoPAP, bilevel pressure ventilation, and humidification. As such, it did not delve in detail into the potential secondary effects of flexible pressure on patient-focused outcomes. An updated systematic review and meta-analysis focused on the effects of flexible pressure is now required so that clinicians are able to make up-to-date, evidence-based treatment decisions.

Literature Search

Being a review of published data, ethical approval for this research was not required. A systematic literature search was undertaken using the PubMed database that included studies from January 1, 2000, until July 11, 2010. The search terms (in all search fields) were used as follows to identify randomized clinical trials comparing flexible pressure to standard, fixed-pressure CPAP for the treatment of OSA: (“C-Flex” or “flexible pressure” or “expiratory pressure relief” or “expiratory pressure reduction” or “pressure relief” or “pressure reduction” or “EPR” or “A-Flex”) and (“obstructive sleep apnea” or “obstructive sleep apnoea” or “sleep apnea” or “sleep apnoea”) and (“randomized” or “randomized” or “randomized controlled trial” or “randomised controlled trial” or “clinical trial”). Papers or abstracts referenced in these papers, those identified by the previous Cochrane review,14 and studies already known to the authors also were included and subjected to the selection criteria.

Study Selection Criteria

After first excluding qualitative review articles, both authors independently applied the following exclusion criteria in participants, interventions, comparators, outcomes, and study design (PICOS) order18 to all papers and abstracts identified: patients without OSA, children (aged < 18 years), treatment other than flexible pressure delivery, comparator other than CPAP or AutoPAP, no inclusion of at least one of the outcome measures of interest listed next, nonrandomized trial including any meta-analyses at this stage, and a < 1-week treatment arm duration (Fig 1). The outcome measures of interest were as follows: objectively measured compliance expressed as mean hours of use per night, objective and subjective daytime sleepiness (as measured by the Epworth Sleepiness Scale [ESS],19 Maintenance of Wakefulness Test [MWT],20 or Multiple Sleep Latency Test [MSLT]20), vigilance (as measured by the Psychomotor Vigilance Task [PVT]21), or sleep-related quality of life (as measured by the Functional Outcomes of Sleep Questionnaire [FOSQ]22). Based on reviewer comment, we also analyzed a measure of treatment efficacy or residual disease (apnea-hypopnea index [AHI] or respiratory disturbance index) recorded either by the PAP device or during a full, nontitration sleep study. We did not exclude papers published in a language other than English or peer-reviewed conference proceedings, where we were able to identify them. All descriptions of included studies were compared for common methodology to ensure that the study samples were not duplicated.

Figure Jump LinkFigure 1. Literature inclusion flowchart. The studies excluded on the basis of being non-C-Flex/A-Flex (Phillips Respironicss; Murrysville, Pennsylvania) mainly were concerned with alternative treatments, such as drug therapy, surgery, and mandibular advancement splints. AHI = apnea-hypopnea index; ESS = Epworth Sleepiness Scale; MWT = Maintenance of Wakefulness Test; OSA = obstructive sleep apnea; PAP = positive airway pressure; PVT = Psychomotor Vigilance Task; RCT = randomized control trial.Grahic Jump Location
Data Extraction

The following data were extracted from each included study: country and setting of study, trial structure, flexible pressure type (C-Flex, A-Flex, EPR, Flexline, or eAdapt), fixed-pressure type comparator (CPAP or AutoPAP), OSA inclusion criteria, titration protocol, duration of treatment arm, washout duration where applicable, method and level of blinding, method of randomization, criteria for diagnosing OSA, method of establishing treatment (full-night, split-night, or AutoPAP titration protocol), number of patients and dropouts or withdrawals, proportion of male patients, whether patients were naive to PAP at baseline, mean age, mean BMI, mean baseline AHI, and mean pressure requirement. The mean, SE, and sample size for the differences in all outcome measures (CPAP − flexible pressure) were extracted from each trial. If any of these data were absent or unclear, we attempted to contact the first, last, and corresponding authors of the relevant papers and abstracts and allowed at least 8 weeks response time. When we were unable to contact authors, we relied on data reported in the previous meta-analysis where possible.14

Data Synthesis and Statistical Analysis

Analyses were conducted by the first author using Review Manager (RevMan) version 5.0 software (Nordic Cochrane Centre; Copenhagen, Denmark). Fixed- and random-effects models can be used in meta-analysis. Fixed-effects models assume that all studies share the same true effect size and that any differences between studies are due to sampling error. Random-effects models assume that the true effect size varies between studies because of differences between samples and implementation of the intervention (clinical and methodologic diversity, respectively).

The authors of many published meta-analyses decide which model to use, using the significance of the Q test of heterogeneity. However, this method is inappropriate for our meta-analysis because it ignores the underlying assumptions of each model; the choice of model should depend on an understanding of whether it is likely that the studies in question all share a common, true effect size. Thus, in the majority of meta-analyses combining data available in the international literature, which are based on samples with different characteristics and different experimental protocols, a random-effects model is usually most appropriate. Most of the meta-analyses performed here, therefore, used a random-effects model, using DerSimonian and Laird methodology.23 The exception was when a very small number of studies were included in the meta-analysis: A random-effects model is still the most appropriate but yields little useful information in this scenario because the estimate of the effect size and its CI is poor.

For each meta-analysis conducted, we produced a Forest plot using the pooled estimate and 95% CI as well as the estimates derived from each individual study. The Q statistic was calculated and considered significant at P ≤ .05, indicating heterogeneity. The I2 statistic was calculated to estimate the percentage of the observed variability due to heterogeneity rather than to chance. We decided a priori that when we had at least four trials and an I2 ≥ 50%24 that we needed to investigate reasons for heterogeneity. Sensitivity analyses were performed for each meta-analysis by removing one study at a time and observing the effect this had on the overall result. Where the number of trials permitted, additional sensitivity analyses were performed by excluding abstracts, trials that were not double blinded, and trials reporting high (> 15%) dropout rates. Finally, publication bias for each analysis was assessed visually using a funnel plot.

Identification and Description of Included Studies

The systematic literature search returned 132 articles. An additional three studies reported as abstracts were referenced in the recent Cochrane review.14 After applying the exclusion criteria, both authors agreed that 10 studies remained eligible (seven full articles and the three abstracts). The reasons for excluding the other 125 studies are shown in Figure 1.

The design characteristics of the included studies are summarized in Table 1, the patient characteristics from each study are summarized in Table 2, and the authors’ judgment as to the risk of bias in each study are summarized in Table 3. A full tabulation of our bias assessment can be obtained from the corresponding author. The included studies had a combined sample size of 599 patients. One abstract did not give adequate information to determine the trial structure.25 One article was written in German,26 and two sleep medicine researchers fluent in German determined that this study should be included and extracted the data independently from each other (see the “Acknowledgments” section). All studies compared C-Flex with CPAP. No included studies used A-Flex or AutoPAP, and none used flexible pressure devices produced by a manufacturer other than Philips Respironics. The OSA severity range was wide, although all studies had a mean baseline AHI in the severe range (≥ 30 events/h). Although only the CPAP pressures have been reported here, all parallel studies that titrated patients on the randomized device reported very similar mean CPAP and C-Flex pressures.

Table Graphic Jump Location
Table 1 —Design Characteristics of Included Studies

AHI = apnea-hypopnea index; AutoPAP = automatically adjusting positive airway pressure; CPAP = continuous positive airway pressure; ESS = Epworth Sleepiness Scale; n/a = not applicable; OSA = obstructive sleep apnea; REM = rapid eye movement.

Table Graphic Jump Location
Table 2 —Patient Characteristics in Included Studies

Data are presented as mean ± SD or No. (%), unless otherwise indicated. See Table 1 legend for expansion of abbreviations.

a 

Reported pressure in mbar units rather than cm H2O units.

Table Graphic Jump Location
Table 3 —Author Assessment of Risk of Bias for Each Study
a 

Single blinding was considered inadequate for this risk-of-bias assessment.

b 

Denotes author supplying additional information during correspondence.

Additional Data Obtained

We provided some additional data that were not completely reported in our own publications.17,27 One abstract author we contacted declined to provide further data.28 We thank Pépin et al,16 Wenzel et al,26 and Gfullner et al29 for providing additional data.

Meta-analyses
Compliance:

For our primary outcome, we were unable to locate data in the correct format for three trials that included 85 patients.15,25,30 Meta-analysis of compliance in four parallel studies16,17,26,31 indicated that C-Flex was not used significantly more than standard CPAP (0.16 h favoring C-Flex; 95% CI, −0.09-0.42; P = .21, random-effects model) (Fig 2A). Similarly, meta-analysis of three crossover studies26,28,29 indicated that C-Flex was not used significantly more than traditional CPAP (0.20 h favoring C-Flex; 95% CI, −0.26-0.66; P = .39, random-effects model) (Fig 2B). The Q tests for the parallel studies (Q = 2.11; degrees of freedom [df] = 3; P = .55) and crossover studies (Q = 0.04; df = 2; P = .98) did not indicate significant heterogeneity between studies. For both analyses, I2 = 0%, so subgroup analysis was not performed. Sensitivity analyses were performed by removing each study from the meta-analyses one at a time. When removed individually, no studies affected the significance of any of our results. We also selectively removed abstracts and full articles, unblinded/single-blinded and double-blinded studies, and studies reporting high (> 15%) and low (≤ 15%) dropout rates. These sensitivity analyses did not lead to a change in overall results.

Figure Jump LinkFigure 2. Forest plot of objective compliance data. A, In parallel trials. B, In crossover trials. The diamond signifies the summary effect combining data from the studies shown above it; the apex represents the overall measure of effect, whereas the width represents the 95% CI of this effect. Each individual study is represented by a square, the size of which indicates the proportional weight of that study in the combined analysis, and a horizontal line representing the 95% CI. In these plots, the 95% CIs of the individual studies and the two combined effect diamonds cross the vertical line at 0 (no effect), meaning that no individual study, or overall meta-analysis, found a significant difference in compliance between CPAP and C-Flex at this level of confidence. CPAP = continuous positive airway pressure.Grahic Jump Location
Treatment Efficacy:

Three studies reported a measure of treatment efficacy. Nilius et al30 (parallel design) and Leidag et al15 (crossover design) performed a full polysomnography on treatment and reported a residual AHI, and Bakker et al17 (parallel design) reported the residual AHI as interpreted by the PAP device averaged over the 88-day study period. Data from the two parallel studies were combined (n = 180), which indicated no significant difference in treatment efficacy between CPAP and C-Flex (−0.31 events/h nonsignificantly favoring CPAP; 95% CI, −1.33-0.72; P = .56, random-effects model) (see Forest plot provided in e-Figure 1).

Patient Outcomes:

The weighted mean improvement of ESS scores in three parallel studies16,17,27 was 0.76 (out of 24 points), which nonsignificantly favored CPAP over C-Flex (95% CI, −3.82-2.30; P = .63, random-effects model) (Fig 3A). The Q test indicated significant heterogeneity between studies (Q = 9.04; df = 2; P = .01). The I2 was 78%, indicating that 78% of the observed variance came from real differences between the studies. However, because only three studies were available, we were unable to investigate potential sources of this heterogeneity. In the two crossover studies,26,29 the mean improvement of ESS scores was 0.57 (out of 24 points), nonsignificantly favoring C-Flex (95% CI, −1.55-2.69; P = .60, random-effects model) (Fig 3B). The I2 (0%) was not positive, and because there were only two studies, we could not reliably detect any heterogeneity32 and did not perform a subgroup analysis.

Figure Jump LinkFigure 3. Forest plot of ESS data. A, In parallel trials. B, In crossover trials. The diamond signifies the summary effect combining data from the studies shown above it; the apex represents the overall measure of effect, whereas the width represents the 95% CI of this effect. Each individual study is represented by a square, the size of which indicates the proportional weight of that study in the combined analysis, and a horizontal line representing the 95% CI. In these plots, the 95% CIs of the individual studies and the two combined effect diamonds cross the vertical line at 0 (no effect), meaning that no individual study, or overall meta-analysis, found a significant difference in ESS between CPAP and C-Flex at this level of confidence. See Figure 1 and 2 legends for expansion of abbreviations.Grahic Jump Location

Only two studies reported MWT and PVT data,17,27 and both used a parallel-arm structure and were conducted in the same patient population, so a fixed-effects model was used for these meta-analyses. The weighted mean improvement in MWT sleep latency was 12.8 s, nonsignificantly favoring C-Flex (95% CI, −299.7-325.4; P = .94). The weighted mean improvement in the average PVT reaction time was 30.9 milliseconds, nonsignificantly favoring C-Flex (95% CI, −13.2-74.9; P = .17). The I2 was 0% in both of these analyses; therefore, no subgroup analyses were performed. The Q test is not appropriate when only two studies are included in an analysis because it is very unlikely to obtain a significant result,32 so it was not performed here. Forest plots for MWT and PVT data are provided in e-Figures 2, 3.

No studies used the MSLT. FOSQ data were available for only one trial,17 and these, therefore, were not analyzable.

Evidence of Publication Reporting Bias

We found no significant evidence of bias when funnel plots for each meta-analysis were inspected. However, each analysis involved only a small number of trials, so funnel plots need to be interpreted with caution. Additional unpublished trials may exist.

Flexible pressure (C-Flex) does not significantly increase compliance with PAP therapy compared with standard CPAP. The numerical trend we observed was small (9.4 min in parallel studies; 12 min in crossover studies), did not approach statistical significance (both P > .20), and is probably of low clinical relevance.33 Despite a very promising initial report from a nonrandomized study that C-Flex may substantially aid compliance,34 no randomized clinical trial singly, or in combination, has been able to demonstrate any significant compliance advantage in > 500 patients.

We also found no significant advantage of C-Flex over standard CPAP for improving subjective sleepiness. The improvement in the ESS was < 1 out of a possible 24 points in both types of trials. Objective sleepiness (difference in MWT sleep latency of < 1 min) or objective vigilance (difference in PVT reaction time of 30.85 ms) were not positively influenced by C-Flex, although regrettably, only two trials were available for inclusion in these analyses. Unfortunately, we did not have sufficient data to perform meta-analyses on two other outcomes that we had hoped to investigate: the MSLT and the FOSQ. The comparison of treatment efficacy between traditional CPAP and C-Flex is really a question of equivalence rather than of one treatment being superior. So, although the limited data we located do not indicate a significant difference, we did not set up these analyses as a test of equivalence.

C-Flex costs more than standard CPAP because of restricted manufacturer choice and has not been demonstrated to confer additional benefits in any patient subgroup. Thus, cannot be recommended at this time for routine clinical care over standard CPAP. Our conclusions are in agreement with those of the Cochrane analysis by Smith and Lasserson,14 who also found no rationale to support the additional cost of C-Flex given the randomized trials available to them. It is possible that patient subgroups where C-Flex is superior to traditional CPAP exist, but these are yet to be identified.

Because we were able to include more than double the number of patients and analyze a greater range of patient-centered outcomes, we have extended the previous analysis.14 Other strengths of this meta-analysis include our attempt to address publication bias by including conference proceedings, our analysis of funnel plots, and the performance of sensitivity analyses that indicated that no single trial exerted undue influence on the overall result of each meta-analysis. We also tried not to introduce duplication or language bias.

Our meta-analysis does have a number of limitations that should be noted. First, despite efforts to correspond with authors, we were only able to include compliance data from seven out of the 10 trials suitable for meta-analysis because the data in three studies were not reported in an appropriate format. None of the missing trials reported a positive effect of C-Flex, and they contained a combined total of only 85 patients out of our total possible pool of 599.15,25,30 The conclusion regarding treatment efficacy relied on only three trials that recorded data through the device or by standard polysomnography. Although we visually tested for publication bias, it is always possible that further unpublished trials exist. However, unpublished trials tend to be biased toward negative findings, and such trials would probably serve to further confirm our findings. Finally, we did not attempt to perform meta-analyses for other outcomes, such as treatment comfort and satisfaction, because it was assumed that any difference between C-Flex and standard CPAP in these variables would be reflected in a difference in compliance.

All studies analyzed here had a mean baseline AHI within the severe range (from 35.4 to 77.6 events/h), so we cannot comment on the use of C-Flex in patients with mild or probably even moderate OSA (AHI, 5-15 and 15-30, respectively). It is also possible that there remains a role for C-Flex in patients with very high pressure requirements. However, one trial used a reasonably high mean pressure (13.3 ± 2.7 cm H2O) and found no significant compliance benefit.27 There may still exist other important patient subpopulations that would benefit from C-Flex therapy (eg, those with nasal congestion; those with CPAP-induced aerophagia; or those struggling with CPAP, particularly as a rescue therapy); however, this needs to be demonstrated in rigorously designed randomized trials. At present, the use of C-Flex in such groups is a matter of clinical judgment alone. We also do not have data on the updated Philips Respironics C-Flex Plus device, the EPR device manufactured by ResMed Ltd, the Flexline machine manufactured by Hoffrichter GmbH, or the eAdapt machine manufactured by Breas Medical AB. These too should be subject to robust clinical trials, although based on the manufacturers’ descriptions of the technology, we have no reason to suspect that these devices would be either inferior or superior to Philips Respironics C-Flex technology.

In conclusion, this systematic review and meta-analysis has found no significant evidence that C-Flex provides any benefit over standard CPAP in terms of compliance, subjective or objective sleepiness, or psychomotor vigilance in patients with moderate to severe OSA. Treatment efficacy did not appear to be different. Because flexible pressure costs more than standard therapy but provides no additional benefit, there is no evidence-based rationale for its routine use as a first-line therapy.

Author contributions: Drs Bakker and Marshall had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Dr Bakker: contributed to the research inception, design, interpretation, and writing of the manuscript and performed all statistical analyses.

Dr Marshall: contributed to the research inception, design, interpretation, and writing of the manuscript.

Financial/nonfinancial disclosures: The authors have reported to CHEST the following conflicts of interest: Dr Bakker has received funding from Philips Respironics for the clinical trial included in this meta-analysis on which she was first author. Dr Marshall has received in-kind support only for the clinical trial included in this meta-analysis on which he was first author. For that trial, Respironics International Inc (as it was then known), through its New Zealand suppliers Care Medical, provided six C-Flex devices.

Other contributions: We thank the authors who provided additional data for inclusion in these analyses as well as Megan Crawford, BSc (Hons), and Margo van den Berg, BA, for translation and data extraction of the article written in German and Brendon Yee, PhD, for useful comments regarding the manuscript.

Role of sponsors: The funding sources had no role in the design, analysis, or decision to publish the study.

Additional information: The e-Figures can be found in the Online Supplement at http://chestjournal.chestpubs.org/content/139/6/1322/suppl/DC1.

AHI

apnea-hypopnea index

AutoPAP

automatically adjusting positive airway pressure

CPAP

continuous positive airway pressure

df

degrees of freedom

EPR

Expiratory Pressure Relief

ESS

Epworth Sleepiness Scale

FOSQ

Functional Outcomes of Sleep Questionnaire

MSLT

Multiple Sleep Latency Test

MWT

Maintenance of Wakefulness Test

OSA

obstructive sleep apnea

PVT

Psychomotor Vigilance Task

Young T, Palta M, Dempsey J, Skatrud J, Weber S, Badr S. The occurrence of sleep-disordered breathing among middle-aged adults. N Engl J Med. 1993;32817:1230-1235. [CrossRef] [PubMed]
 
American Academy of Sleep Medicine Task ForceAmerican Academy of Sleep Medicine Task Force Sleep-related breathing disorders in adults: recommendations for syndrome definition and measurement techniques in clinical research. The Report of an American Academy of Sleep Medicine Task Force. Sleep. 1999;225:667-689. [PubMed]
 
Marshall NS, Wong KK, Liu PY, Cullen SR, Knuiman MW, Grunstein RR. Sleep apnea as an independent risk factor for all-cause mortality: the Busselton Health Study. Sleep. 2008;318:1079-1085. [PubMed]
 
Young T, Finn L, Peppard PE, et al. Sleep disordered breathing and mortality: eighteen-year follow-up of the Wisconsin sleep cohort. Sleep. 2008;318:1071-1078. [PubMed]
 
Marin JM, Carrizo SJ, Vicente E, Agusti AG. Long-term cardiovascular outcomes in men with obstructive sleep apnoea-hypopnoea with or without treatment with continuous positive airway pressure: an observational study. Lancet. 2005;3659464:1046-1053. [PubMed]
 
Sullivan CE, Issa FG, Berthon-Jones M, Eves L. Reversal of obstructive sleep apnoea by continuous positive airway pressure applied through the nares. Lancet. 1981;3178225:862-865. [CrossRef]
 
Kushida CA, Littner MR, Hirshkowitz M, et al; American Academy of Sleep Medicine American Academy of Sleep Medicine Practice parameters for the use of continuous and bilevel positive airway pressure devices to treat adult patients with sleep-related breathing disorders. Sleep. 2006;293:375-380. [PubMed]
 
Engleman HM, Kingshott RN, Wraith PK, Mackay TW, Deary IJ, Douglas NJ. Randomized placebo-controlled crossover trial of continuous positive airway pressure for mild sleep Apnea/Hypopnea syndrome. Am J Respir Crit Care Med. 1999;1592:461-467. [PubMed]
 
Montserrat JM, Ferrer M, Hernandez L, et al. Effectiveness of CPAP treatment in daytime function in sleep apnea syndrome: a randomized controlled study with an optimized placebo. Am J Respir Crit Care Med. 2001;1644:608-613. [PubMed]
 
Martínez-García MA, Soler-Cataluña JJ, Ejarque-Martínez L, et al. Continuous positive airway pressure treatment reduces mortality in patients with ischemic stroke and obstructive sleep apnea: a 5-year follow-up study. Am J Respir Crit Care Med. 2009;1801:36-41. [CrossRef] [PubMed]
 
Weaver TE, Grunstein RR. Adherence to continuous positive airway pressure therapy: the challenge to effective treatment. Proc Am Thorac Soc. 2008;52:173-178. [CrossRef] [PubMed]
 
Rühle KH, Domanski U, Happel A, Nilius G. Analysis of expiratory pressure reduction (C-Flex method) during CPAP therapy. Pneumologie. 2007;612:86-89. [CrossRef] [PubMed]
 
Canisius S, Kesper K, Jerrentrup L, et al. C-Flex technology: effects on breathing parameters and inspiratory flow limitation. Respiration. 2009;782:168-176. [CrossRef] [PubMed]
 
Smith I, Lasserson TJ. Pressure modification for improving usage of continuous positive airway pressure machines in adults with obstructive sleep apnoea. Cochrane Database Syst Rev. 2009;4:CD003531
 
Leidag M, Hader C, Keller T, Meyer Y, Rasche K. Mask leakage in continuous positive airway pressure and C-Flex. J Physiol Pharmacol. 2008;59suppl 6:401-406. [PubMed]
 
Pépin JL, Muir JF, Gentina T, et al. Pressure reduction during exhalation in sleep apnea patients treated by continuous positive airway pressure. Chest. 2009;1362:490-497. [CrossRef] [PubMed]
 
Bakker J, Campbell A, Neill A. Randomized controlled trial comparing flexible and continuous positive airway pressure delivery: effects on compliance, objective and subjective sleepiness and vigilance. Sleep. 2010;334:523-529. [PubMed]
 
Liberati A, Altman DG, Tetzlaff J, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. Ann Intern Med. 2009;1514:W65-W94. [PubMed]
 
Johns MW. A new method for measuring daytime sleepiness: the Epworth sleepiness scale. Sleep. 1991;146:540-545. [PubMed]
 
Littner MR, Kushida C, Wise M, et al; Standards of Practice Committee of the American Academy of Sleep Medicine Standards of Practice Committee of the American Academy of Sleep Medicine Practice parameters for clinical use of the multiple sleep latency test and the maintenance of wakefulness test. Sleep. 2005;281:113-121. [PubMed]
 
Dinges DF, Powell JW. Microcomputer analyses of performance on a portable, simple visual RT task during sustained operations. Behav Res Methods Instrum Comput. 1985;176:652-655. [CrossRef]
 
Weaver TE, Laizner AM, Evans LK, et al. An instrument to measure functional status outcomes for disorders of excessive sleepiness. Sleep. 1997;2010:835-843. [PubMed]
 
DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials. 1986;73:177-188. [CrossRef] [PubMed]
 
Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. BMJ. 2003;3277414:557-560. [CrossRef] [PubMed]
 
Loube DL, Ball N. Comparison of compliance with proportional positive airway pressure (C-Flex) to continuous positive airway pressure treatment of obstructive sleep apnea. Sleep. 2004;27suppl:A228
 
Wenzel M, Kerl J, Dellweg D, Barchfeld T, Wenzel G, Köhler D. Expiratory pressure reduction (C-Flex Method) versus fix CPAP in the therapy for obstructive sleep apnoea. Pneumologie. 2007;6111:692-695. [CrossRef] [PubMed]
 
Marshall NS, Neill AM, Campbell AJ. Randomised trial of compliance with flexible (C-Flex) and standard continuous positive airway pressure for severe obstructive sleep apnea. Sleep Breath. 2008;124:393-396. [CrossRef] [PubMed]
 
Modrak J, Greenblatt D. A prospective randomized crossover trial to obtain objective comparison between initial continuous positive airway pressure (CPAP) use with and without expiratory pressure relief in patients with obstructive sleep apnea (OSA). Sleep. 2007;29suppl:A206
 
Gfullner F, Montalvan S, Weber G, et al. Randomized crossover study of treatment compliance and satisfaction in non sleepy patients with obstructive sleep apnea: CPAP with pressure relief during exhalation vs. conventional CPAP. Sleep. 2007;30suppl:A185
 
Nilius G, Happel A, Domanski U, Ruhle KH. Pressure-relief continuous positive airway pressure vs constant continuous positive airway pressure: a comparison of efficacy and compliance. Chest. 2006;1304:1018-1024. [CrossRef] [PubMed]
 
Dolan DC, Okonkwo R, Gfullner F, Hansbrough JR, Strobel RJ, Rosenthal L. Longitudinal comparison study of pressure relief (C-Flex) vs. CPAP in OSA patients. Sleep Breath. 2008;131:73-77. [CrossRef] [PubMed]
 
Borenstein M, Hedges L, Higgins J, et al. Introduction to Meta-Analysis. 2009; Chichester, England John Wiley & Sons Ltd
 
Weaver TE, Maislin G, Dinges DF, et al. Relationship between hours of CPAP use and achieving normal levels of sleepiness and daily functioning. Sleep. 2007;306:711-719. [PubMed]
 
Aloia MS, Stanchina M, Arnedt JT, Malhotra A, Millman RP. Treatment adherence and outcomes in flexible vs standard continuous positive airway pressure therapy. Chest. 2005;1276:2085-2093. [CrossRef] [PubMed]
 

Figures

Figure Jump LinkFigure 1. Literature inclusion flowchart. The studies excluded on the basis of being non-C-Flex/A-Flex (Phillips Respironicss; Murrysville, Pennsylvania) mainly were concerned with alternative treatments, such as drug therapy, surgery, and mandibular advancement splints. AHI = apnea-hypopnea index; ESS = Epworth Sleepiness Scale; MWT = Maintenance of Wakefulness Test; OSA = obstructive sleep apnea; PAP = positive airway pressure; PVT = Psychomotor Vigilance Task; RCT = randomized control trial.Grahic Jump Location
Figure Jump LinkFigure 2. Forest plot of objective compliance data. A, In parallel trials. B, In crossover trials. The diamond signifies the summary effect combining data from the studies shown above it; the apex represents the overall measure of effect, whereas the width represents the 95% CI of this effect. Each individual study is represented by a square, the size of which indicates the proportional weight of that study in the combined analysis, and a horizontal line representing the 95% CI. In these plots, the 95% CIs of the individual studies and the two combined effect diamonds cross the vertical line at 0 (no effect), meaning that no individual study, or overall meta-analysis, found a significant difference in compliance between CPAP and C-Flex at this level of confidence. CPAP = continuous positive airway pressure.Grahic Jump Location
Figure Jump LinkFigure 3. Forest plot of ESS data. A, In parallel trials. B, In crossover trials. The diamond signifies the summary effect combining data from the studies shown above it; the apex represents the overall measure of effect, whereas the width represents the 95% CI of this effect. Each individual study is represented by a square, the size of which indicates the proportional weight of that study in the combined analysis, and a horizontal line representing the 95% CI. In these plots, the 95% CIs of the individual studies and the two combined effect diamonds cross the vertical line at 0 (no effect), meaning that no individual study, or overall meta-analysis, found a significant difference in ESS between CPAP and C-Flex at this level of confidence. See Figure 1 and 2 legends for expansion of abbreviations.Grahic Jump Location

Tables

Table Graphic Jump Location
Table 1 —Design Characteristics of Included Studies

AHI = apnea-hypopnea index; AutoPAP = automatically adjusting positive airway pressure; CPAP = continuous positive airway pressure; ESS = Epworth Sleepiness Scale; n/a = not applicable; OSA = obstructive sleep apnea; REM = rapid eye movement.

Table Graphic Jump Location
Table 2 —Patient Characteristics in Included Studies

Data are presented as mean ± SD or No. (%), unless otherwise indicated. See Table 1 legend for expansion of abbreviations.

a 

Reported pressure in mbar units rather than cm H2O units.

Table Graphic Jump Location
Table 3 —Author Assessment of Risk of Bias for Each Study
a 

Single blinding was considered inadequate for this risk-of-bias assessment.

b 

Denotes author supplying additional information during correspondence.

References

Young T, Palta M, Dempsey J, Skatrud J, Weber S, Badr S. The occurrence of sleep-disordered breathing among middle-aged adults. N Engl J Med. 1993;32817:1230-1235. [CrossRef] [PubMed]
 
American Academy of Sleep Medicine Task ForceAmerican Academy of Sleep Medicine Task Force Sleep-related breathing disorders in adults: recommendations for syndrome definition and measurement techniques in clinical research. The Report of an American Academy of Sleep Medicine Task Force. Sleep. 1999;225:667-689. [PubMed]
 
Marshall NS, Wong KK, Liu PY, Cullen SR, Knuiman MW, Grunstein RR. Sleep apnea as an independent risk factor for all-cause mortality: the Busselton Health Study. Sleep. 2008;318:1079-1085. [PubMed]
 
Young T, Finn L, Peppard PE, et al. Sleep disordered breathing and mortality: eighteen-year follow-up of the Wisconsin sleep cohort. Sleep. 2008;318:1071-1078. [PubMed]
 
Marin JM, Carrizo SJ, Vicente E, Agusti AG. Long-term cardiovascular outcomes in men with obstructive sleep apnoea-hypopnoea with or without treatment with continuous positive airway pressure: an observational study. Lancet. 2005;3659464:1046-1053. [PubMed]
 
Sullivan CE, Issa FG, Berthon-Jones M, Eves L. Reversal of obstructive sleep apnoea by continuous positive airway pressure applied through the nares. Lancet. 1981;3178225:862-865. [CrossRef]
 
Kushida CA, Littner MR, Hirshkowitz M, et al; American Academy of Sleep Medicine American Academy of Sleep Medicine Practice parameters for the use of continuous and bilevel positive airway pressure devices to treat adult patients with sleep-related breathing disorders. Sleep. 2006;293:375-380. [PubMed]
 
Engleman HM, Kingshott RN, Wraith PK, Mackay TW, Deary IJ, Douglas NJ. Randomized placebo-controlled crossover trial of continuous positive airway pressure for mild sleep Apnea/Hypopnea syndrome. Am J Respir Crit Care Med. 1999;1592:461-467. [PubMed]
 
Montserrat JM, Ferrer M, Hernandez L, et al. Effectiveness of CPAP treatment in daytime function in sleep apnea syndrome: a randomized controlled study with an optimized placebo. Am J Respir Crit Care Med. 2001;1644:608-613. [PubMed]
 
Martínez-García MA, Soler-Cataluña JJ, Ejarque-Martínez L, et al. Continuous positive airway pressure treatment reduces mortality in patients with ischemic stroke and obstructive sleep apnea: a 5-year follow-up study. Am J Respir Crit Care Med. 2009;1801:36-41. [CrossRef] [PubMed]
 
Weaver TE, Grunstein RR. Adherence to continuous positive airway pressure therapy: the challenge to effective treatment. Proc Am Thorac Soc. 2008;52:173-178. [CrossRef] [PubMed]
 
Rühle KH, Domanski U, Happel A, Nilius G. Analysis of expiratory pressure reduction (C-Flex method) during CPAP therapy. Pneumologie. 2007;612:86-89. [CrossRef] [PubMed]
 
Canisius S, Kesper K, Jerrentrup L, et al. C-Flex technology: effects on breathing parameters and inspiratory flow limitation. Respiration. 2009;782:168-176. [CrossRef] [PubMed]
 
Smith I, Lasserson TJ. Pressure modification for improving usage of continuous positive airway pressure machines in adults with obstructive sleep apnoea. Cochrane Database Syst Rev. 2009;4:CD003531
 
Leidag M, Hader C, Keller T, Meyer Y, Rasche K. Mask leakage in continuous positive airway pressure and C-Flex. J Physiol Pharmacol. 2008;59suppl 6:401-406. [PubMed]
 
Pépin JL, Muir JF, Gentina T, et al. Pressure reduction during exhalation in sleep apnea patients treated by continuous positive airway pressure. Chest. 2009;1362:490-497. [CrossRef] [PubMed]
 
Bakker J, Campbell A, Neill A. Randomized controlled trial comparing flexible and continuous positive airway pressure delivery: effects on compliance, objective and subjective sleepiness and vigilance. Sleep. 2010;334:523-529. [PubMed]
 
Liberati A, Altman DG, Tetzlaff J, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. Ann Intern Med. 2009;1514:W65-W94. [PubMed]
 
Johns MW. A new method for measuring daytime sleepiness: the Epworth sleepiness scale. Sleep. 1991;146:540-545. [PubMed]
 
Littner MR, Kushida C, Wise M, et al; Standards of Practice Committee of the American Academy of Sleep Medicine Standards of Practice Committee of the American Academy of Sleep Medicine Practice parameters for clinical use of the multiple sleep latency test and the maintenance of wakefulness test. Sleep. 2005;281:113-121. [PubMed]
 
Dinges DF, Powell JW. Microcomputer analyses of performance on a portable, simple visual RT task during sustained operations. Behav Res Methods Instrum Comput. 1985;176:652-655. [CrossRef]
 
Weaver TE, Laizner AM, Evans LK, et al. An instrument to measure functional status outcomes for disorders of excessive sleepiness. Sleep. 1997;2010:835-843. [PubMed]
 
DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials. 1986;73:177-188. [CrossRef] [PubMed]
 
Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. BMJ. 2003;3277414:557-560. [CrossRef] [PubMed]
 
Loube DL, Ball N. Comparison of compliance with proportional positive airway pressure (C-Flex) to continuous positive airway pressure treatment of obstructive sleep apnea. Sleep. 2004;27suppl:A228
 
Wenzel M, Kerl J, Dellweg D, Barchfeld T, Wenzel G, Köhler D. Expiratory pressure reduction (C-Flex Method) versus fix CPAP in the therapy for obstructive sleep apnoea. Pneumologie. 2007;6111:692-695. [CrossRef] [PubMed]
 
Marshall NS, Neill AM, Campbell AJ. Randomised trial of compliance with flexible (C-Flex) and standard continuous positive airway pressure for severe obstructive sleep apnea. Sleep Breath. 2008;124:393-396. [CrossRef] [PubMed]
 
Modrak J, Greenblatt D. A prospective randomized crossover trial to obtain objective comparison between initial continuous positive airway pressure (CPAP) use with and without expiratory pressure relief in patients with obstructive sleep apnea (OSA). Sleep. 2007;29suppl:A206
 
Gfullner F, Montalvan S, Weber G, et al. Randomized crossover study of treatment compliance and satisfaction in non sleepy patients with obstructive sleep apnea: CPAP with pressure relief during exhalation vs. conventional CPAP. Sleep. 2007;30suppl:A185
 
Nilius G, Happel A, Domanski U, Ruhle KH. Pressure-relief continuous positive airway pressure vs constant continuous positive airway pressure: a comparison of efficacy and compliance. Chest. 2006;1304:1018-1024. [CrossRef] [PubMed]
 
Dolan DC, Okonkwo R, Gfullner F, Hansbrough JR, Strobel RJ, Rosenthal L. Longitudinal comparison study of pressure relief (C-Flex) vs. CPAP in OSA patients. Sleep Breath. 2008;131:73-77. [CrossRef] [PubMed]
 
Borenstein M, Hedges L, Higgins J, et al. Introduction to Meta-Analysis. 2009; Chichester, England John Wiley & Sons Ltd
 
Weaver TE, Maislin G, Dinges DF, et al. Relationship between hours of CPAP use and achieving normal levels of sleepiness and daily functioning. Sleep. 2007;306:711-719. [PubMed]
 
Aloia MS, Stanchina M, Arnedt JT, Malhotra A, Millman RP. Treatment adherence and outcomes in flexible vs standard continuous positive airway pressure therapy. Chest. 2005;1276:2085-2093. [CrossRef] [PubMed]
 
NOTE:
Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging & repositioning the boxes below.

Find Similar Articles
CHEST Journal Articles
PubMed Articles
Guidelines
  • CHEST Journal
    Print ISSN: 0012-3692
    Online ISSN: 1931-3543