Duration of positive airway pressure adherence: how much PAP is enough?

No medical therapy is effective in the presence of suboptimal adherence. In that regard, obstructive sleep apnea (OSA) management is no exception to the rule. Since the first utilization of continuous positive airway pressure (CPAP) to treat OSA by Dr. Sullivan in 1980,¹ regular use of CPAP continues to be a challenge. Despite the tremendous advances in positive airway pressure (PAP) technology (smaller and more quiet units, more comfortable interfaces, enhanced heating/humidity), suboptimal adherence to CPAP remains the main barrier to improved outcomes, with poor adherence ranging from 30% to 60%.² To further complicate matters, the definition of adequate adherence has been a topic of heated discussion. The most widely accepted criteria were established in 1993 by Kribbs et al,³ who defined “regular use” by at least 4 hours of CPAP administered on 70% of the days monitored.

After PAP therapy was first introduced to treat OSA, its proven efficacy resulted in authorization of thousands of units by the Centers for Medicare and Medicaid Services (CMS). In 2009, CPAP accounted for a total expenditure of $213 million and coverage for 2.6 million CMS-allowed services.⁴ While the population of patients on CPAP increased, there was a great deal of difficulty adapting to the novel therapy; many CPAP units were sitting unused in closets and on nightstands across the country. This resulted in the urgent need to establish adherence criteria to determine valid coverage and, consequently, mitigate the increased health care utilization and cost. Kribbs et al’s definition was adopted by the CMS, and since then has been the prevailing norm utilized for coverage policies⁵ and evaluation of CPAP efficacy on clinical outcomes.

However, the adoption of ≥ 4 hours of CPAP use on ≥ 70% of all nights did not emerge without discredit, mainly derived from the initial scarce evidence supporting that seemingly arbitrary cutoff. Over the last 2 decades, several studies tried to elucidate an acceptable definition of CPAP adherence based on clinical responses to therapy. In 2007, a study that included 149 participants determined that longer nocturnal CPAP use provides better clinical outcomes, including decreased excessive daytime sleepiness and improvements in functional status.⁶ Similar results supporting a CPAP dose–response have been demonstrated in controlled trials.⁷ Additionally, different levels of usage may be necessary to impact different outcome. Improvements on self-reported sleepiness measured by the Epworth Sleepiness Scale may require at least 4 hours of CPAP use per night, while normalization of objective sleepiness and functional status may require longer use—6 and 7.5 hours per night, respectively.⁶ Zimmerman et al⁸ evaluated the effects of therapeutic CPAP on memory performance in patients with OSA. After 3 months of CPAP use, memory-impaired participants with an average use of 6 hours per night were 8 times more likely to normalize memory function compared with those who used CPAP ≤ 2 hours.

The effects of CPAP adherence on cardiovascular outcomes have also been examined. A prospective observational study found lower cardiovascular mortality rates in women with OSA treated with 6 hours of CPAP compared with untreated and nonadherent patients (≤ 4 hours per day).⁹ In a randomized controlled trial, hypertensive patients with OSA required CPAP use for more than 5.6 hours per night to achieve significant reductions in blood pressure.

As a result of the published evidence, a dose–response relationship between CPAP use and different outcomes has been generally accepted. However, it is important to note that a significant number of patients who are optimally treated (≥ 7 hours per night) do not exhibit significant improvements or resolution of symptoms.^6,10 On the other hand, a minority of patients have reported benefit with less than 2 hours of CPAP use.⁶

In this issue of the Journal of Clinical Sleep Medicine Pascoe et al¹¹ examine the dose–response relationship between CPAP adherence and self-reported excessive daytime sleepiness. Using the CMS adherence definition and the Epworth Sleepiness Scale, 119 participants with moderate to severe OSA were evaluated at baseline and after 3 months of CPAP therapy. The authors determined that the percentage of nights with PAP use ≥ 4 hours predicted Epworth Sleepiness Scale change at 3 months. The percentage of nights with PAP use ≥ 4 hours and average PAP use provided the best discrimination for predicting Epworth Sleepiness Scale normalization. The results of the present study corroborate the previously published data that at least 4 hours of CPAP use per night are necessary to normalize symptoms of self-reported sleepiness.^6,10 The study by Pascoe et al provides additional support for the use of the current and widely utilized CMS definition of CPAP adherence. The authors also explored the percentage of sleep hours with PAP use per night; however, no significant associations were observed. This variable is an important consideration and worth exploring in future research. Adherence is simply defined as hours of CPAP use per night, which does not integrate the total sleep time duration. Four hours of CPAP use on 5 hours of sleep time may have a different clinical effect than on 10 hours of sleep duration.

It has been almost 30 years since the CPAP adherence definition of ≥ 4 hours on ≥ 70% of the nights first appeared. Although unanswered questions remain regarding the most appropriate definition of optimal adherence, the study by Pascoe et al has yielded further support to validate the current CPAP adherence criteria. Future research is needed to unravel the complex dose–response relationships between CPAP and different dimensions of clinical outcomes.

DISCLOSURE STATEMENT

All authors have seen and approved the manuscript. Work for this commentary was performed at The Keck Medical Center of University of Southern California. The authors report no conflicts of interest.