Association of Continuous Positive Airway Pressure Treatment With Sexual Quality of Life in Patients With Sleep Apnea: Follow-up Study of a Randomized Clinical Trial

Key Points

Question  What is the association of long-term continuous positive airway pressure with sexual quality of life in patients with sleep apnea?

Findings  In a cohort study of 182 patients with sleep apnea, a significant improvement in sexual quality of life measures was observed among patients who used continuous positive airway pressure vs those who did not at long-term follow-up.

Meaning  In patients with sleep apnea, successful continuous positive airway pressure use may be associated with better improvements in sexual quality of life.

Importance  Obstructive sleep apnea reduces sexual quality of life (QOL) as a result of reduced libido and intimacy, erectile dysfunction, and several other mechanisms. Treatment for obstructive sleep apnea may improve sexual QOL.

Objective  To test the association of long-term continuous positive airway pressure (CPAP) treatment with sexual QOL for patients with obstructive sleep apnea.

Design, Setting, and Participants  Prospective cohort study at a single, tertiary medical center of patients with newly diagnosed obstructive sleep apnea who were prescribed CPAP treatment from September 1, 2007, through June 30, 2010 (follow-up completed June 30, 2011). The statistical analysis was performed from February 1 through December 31, 2017.

Exposures  Use of CPAP treatment objectively measured by the number of hours per night. Users of CPAP were defined as patients who used CPAP treatment for more than 4 hours per night, and nonusers were defined as patients who used CPAP treatment for fewer than 0.5 hours per night.

Main Outcomes and Measures  Data were collected from eligible patients before CPAP treatment was prescribed and 12 months later by using the validated Symptoms of Nocturnal Obstruction and Related Events–25 (SNORE-25) QOL instrument. The 2 sex-specific items used to create the sexual QOL domain were taken from the SNORE-25. The sexual QOL domain was scored in a range from 0 to 5 (higher score is worse). The difference in sexual QOL between CPAP users and nonusers was analyzed using a paired, 2-tailed t test and multivariable linear regression adjusted for potential confounders.

Results  Of the 182 participants in the cohort, 115 (63.2%) were men (mean [SD] age, 47.2 [12.3] years) with severe OSA (mean [SD] apnea-hypopnea index, 32.5 [23.8] events per hour). At the 12-month follow-up, 72 CPAP users (mean [SD] use, 6.4 [1.2] hours per night) had greater improvement than 110 nonusers (0 [0] hours per night) in sexual QOL scores (0.7 [1.2] vs 0.1 [1.1]; difference, 0.54; 95% CI, 0.18-0.90; effect size, 0.47). A moderate treatment association was observed after adjustment for age, sex, race/ethnicity, marital status, income level, educational level, body mass index, apnea-hypopnea index, and the Functional Comorbidity Index (adjusted difference, 0.49; 95% CI, 0.09-0.89; effect size, 0.43). Subgroup analysis revealed a large treatment association for women (adjusted difference, 1.34; 95% CI, 0.50-2.18; effect size, 0.87) but not for men (adjusted difference, 0.16; 95% CI, −0.26 to 0.58; effect size, 0.19).

Conclusions and Relevance  Successful CPAP use may be associated with improved sexual QOL. Subgroup analysis revealed a large improvement in women but no improvement in men. Further study is warranted to test other measures of sexual QOL and other treatments.

Trial Registration  ClinicalTrials.gov Identifier: NCT00503802

Obstructive sleep apnea (OSA) is a disorder of symptomatic, repetitive upper airway obstruction during sleep that is associated with reduced function and health-related quality of life (QOL). In particular, there is growing evidence that OSA negatively affects sexual function and sexual QOL. In men, OSA is associated with erectile dysfunction, ejaculation disorder, and decreased sexual frequency.^1-6 Similarly, OSA in women is associated with genitopelvic pain, insufficient vaginal lubrication, and decreased sexual interest.^7-11 For both sexes, OSA is associated with impaired orgasmic function, decreased sexual satisfaction, and reduced partner intimacy.^12,13

Continuous positive airway pressure (CPAP), the first-line treatment for OSA, may improve sexual QOL. In men with OSA, several studies reported that CPAP is associated with improvements in partner intimacy and in sexual function, such as erectile dysfunction, orgasmic function, and sexual satisfaction.^6,14-25 However, limited data exist for women. A recent case series of women with OSA indicated that CPAP adherence was associated with a modest improvement in sexual dysfunction but with no changes in isolated sexual difficulty, sexual distress, or overall satisfaction after 1 year of treatment.²⁶ Because that study had no control group, the implications of these findings are limited. For both men and women with OSA, there is a need for high-quality studies assessing the long-term effect of CPAP therapy on sexual QOL.

The purpose of this study was to evaluate the association of long-term CPAP therapy with sexual QOL in a cohort of patients with OSA. We wanted to test the hypothesis that long-term CPAP therapy is associated with improved sexual QOL in both men and women, as assessed by the sex-specific items of the validated Symptoms of Nocturnal Obstruction and Related Events–25 (SNORE-25) QOL instrument.²⁷

A prospective, observational cohort study was conducted to test the hypotheses. This study was approved by the University of Washington Institutional Review Board, Seattle, and all patients signed written informed consent prior to enrollment in the study.

Eligible patients were adults with a new polysomnographic diagnosis of OSA identified from September 1, 2007, to June 30, 2010, from a parent trial.²⁸ The diagnosis of OSA was defined as an apnea-hypopnea index of 5 or more events per hour on full-night, attended diagnostic polysomnography using the American Academy of Sleep Medicine accredited scoring criteria.²⁹ A total of 242 patients were enrolled in this parent trial, and patient follow-up was completed on June 30, 2011.

The present study analyzed data from CPAP users and nonusers at 12 months after the initiation of CPAP treatment. Users of CPAP were defined as patients with more than 4 hours per night of CPAP use (objectively measured with an embedded CPAP pressure-on recorder) during the 4 weeks immediately before the 12-month follow-up visit. Nonusers of CPAP were defined as patients with fewer than 0.5 hours per night of CPAP use (or self-reported nonuse of CPAP if the data were not available) in the 4 weeks immediately before the 12-month follow-up visit. The type of CPAP machine prescribed to patients was chosen at the discretion of the physicians participating in the parent trial. However, the vast majority of patients received the REMstar Auto M Series or the REMstar Pro M Series (Phillips Respironics, Inc). Additional inclusion criteria included being aged 18 to 80 years, English language fluency, receipt of a medical recommendation to start CPAP therapy, the ability to give informed consent, and the ability and willingness to complete the study protocol. Patients were excluded if they did not have a telephone, had a previously diagnosed sleep disorder, or planned to move locations during the study period.

Study participants were seen in a research clinic at the University of Washington within 3 weeks of their diagnosis of OSA but prior to the CPAP titration and initiation, and then again 12 months later. Baseline data were obtained at the initial research clinic visit using paper and computer-based questionnaires. Body mass index (calculated as the weight in kilograms divided by the height in meters squared) was calculated from measurements obtained by research staff at the initial visit. The remainder of the data were elicited by patient self-report. Outcome data were obtained at the 12-month visit, and CPAP use was downloaded from the embedded data card on the CPAP device. Follow-up data were available for 235 of 242 patients (97.1%) enrolled in the parent trial.

The SNORE-25 instrument is a 25-item, OSA-specific QOL instrument that has been shown to be valid, reliable, and responsive to QOL change in OSA.^27,30 Each item is scored (on a scale from 0 [no problem] to 5 [problem as bad as it can be]) based on how “bad” that OSA-related symptom has been over the previous 2 weeks. The total instrument score is the mean of the scored items. We created a sexual QOL domain from the following 2 sex-specific items: “Because of medical problem, unable to have sexual relations” (item 14); and “Lack of desire for sexual relations” (item 21). We calculated the mean of the 2 items to produce a sexual QOL score ranging from 0 to 5 (higher is worse), which is the outcome of interest for this study.

Potential confounding variables selected for adjustment include age, sex, race/ethnicity (white vs nonwhite), body mass index, apnea-hypopnea index, the Functional Comorbidity Index, marital status (single vs married/coupled), annual income level (<$75 000 vs ≥$75 000), and educational level (college degree vs no college degree). The Functional Comorbidity Index is an 18-item index of comorbidities that are associated with reduced physical function and QOL³¹ and has been validated for comorbidity adjustment in QOL studies for OSA.³² A higher score indicates a greater burden of comorbid diseases affecting functional status and QOL. The social and socioeconomic variables (marital status, income level, and educational level) were added to the original model based on study feedback.

Descriptive data are reported as mean (SD) or percentage frequency. The differences in sexual QOL between baseline and 12 months were tested for the CPAP users and nonusers, each with the paired, 2-tailed t test. Change in sexual QOL scores were calculated for each study patient by subtracting the baseline score from the score at 12 months, with scores that have a positive result representing improved sexual QOL. Mean (SD) sexual QOL change scores, 95% CIs, and effect sizes (ESs) (defined as the mean change score divided by the baseline SD) were calculated for each CPAP user group.³³ Differences in the sexual QOL change scores were tested between CPAP user groups using the unpaired, 2-tailed t test and multivariable linear regression adjusted for potential confounders. Subgroup analyses were performed separately for the men and women. All analyses were performed with Stata/SE, version 12 (Stata Inc). The statistical analysis was performed from February 1 through December 31, 2017.

The cohort consisted of 182 adult patients with newly diagnosed OSA who were prescribed CPAP treatment. Of the 182 participants in the cohort, 115 (63.2%) were men (mean [SD] age, 47.2 [12.3] years), and 137 (75.3%) were of white race/ethnicity. The cohort was obese (mean [SD] body mass index, 31.9 [6.7]), had severe OSA (mean [SD] apnea-hypopnea index, 32.5 [23.8] events per hour), and had QOL comorbidities (mean [SD] Functional Comorbidity Index, 2.2 [1.7]) (Table 1).

This cohort included 72 CPAP users (mean [SD] use, 6.4 [1.2] hours per night) and 110 nonusers (mean [SD] use, 0 [0] hours per night). An additional 60 patients from the parent trial had CPAP use between 0.5 and 4 hours per night and were not included in this analysis. Between CPAP users and nonusers, there were differences in the potential confounding variables of age, sex, race/ethnicity, marital status, income level, educational level, and apnea-hypopnea index (Table 1).

Both CPAP users and nonusers had a clinically important baseline deficit in sexual QOL (Table 2). At 12 months, CPAP users experienced a mean improvement in sexual QOL, with a moderate to large ES and a 95% CI consistent with an important clinical association (mean [SD] change, 0.7 [1.2]; 95% CI, 0.4-1.0; ES, 0.60). In contrast, nonusers had no significant change in sexual QOL at 12 months (mean [SD] change, 0.1 [1.1]; 95% CI, –0.1 to 0.4; ES, 0.11).

In unadjusted analysis, CPAP users had greater improvement than nonusers on sexual QOL with a moderate ES (unadjusted difference, 0.54; 95% CI, 0.18-0.90; ES, 0.47) (Table 3). The association persisted after adjustment for age, sex, race/ethnicity, marital status, income level, educational level, body mass index, apnea-hypopnea index, and the Functional Comorbidity Index (adjusted difference, 0.49; 95% CI, 0.09-0.89; ES, 0.43). Adjusted and unadjusted subgroup analysis in men revealed no significant difference in sexual QOL improvement between CPAP users and nonusers (unadjusted difference, 0.14; 95% CI, −0.23 to 0.51; ES, 0.17; adjusted difference, 0.16; 95% CI, −0.26 to 0.58; ES, 0.19). However, adjusted and unadjusted subgroup analysis in women concluded that CPAP users had a greater improvement than nonusers in sexual QOL at 12 months, with a large ES indicative of a clinically important difference (unadjusted difference, 1.24; 95% CI, 0.51-1.96; ES, 0.80; adjusted difference, 1.34; 95% CI, 0.50-2.18; ES, 0.87) (Table 3).

In the initial analyses, variables including age, sex, race/ethnicity, body mass index, apnea-hypopnea index, and the Functional Comorbidity Index were selected a priori as potential confounders for adjustment. In this initial model, CPAP users had a greater improvement than nonusers for sexual QOL, with a mild to moderate ES and a 95% CI that included a clinically important but not statistically significant difference (adjusted difference, 0.33; 95% CI, −0.06 to 0.71; ES, 0.29). Similar to the fully adjusted model, the subgroup analysis of men in this initial model revealed no difference in sexual QOL improvement at 12 months between CPAP users and nonusers (adjusted difference, 0.03; 95% CI, −0.38 to 0.43; ES, 0.04), whereas the subgroup analysis of women showed a greater improvement in sexual QOL at 12 months in CPAP users than nonusers (adjusted difference, 0.88; 95% CI, 0.08-1.68; ES, 0.57).

The fully adjusted analyses were performed after receiving the recommendation to include potential social and socioeconomic confounders. The addition of marital status alone to the initial model slightly weakened the association (adjusted difference, 0.31; 95% CI, −0.09 to 0.71); the addition of income level alone strengthened the association (adjusted difference, 0.55; 95% CI, 0.15-0.94); and the addition of educational level alone slightly weakened the association (adjusted difference, 0.35; 95% CI, −0.03 to 0.74). The subgroup analysis in men and women for each of these additional models (marital status, income level, and educational level) showed similar results to the initial and fully adjusted models.

This study supported the hypothesis that long-term CPAP therapy is associated with improved sexual QOL in a cohort of patients with OSA. Our results showed that CPAP users experienced a greater improvement of sexual QOL when compared with CPAP nonusers. This association was noted in the unadjusted and the fully adjusted models after including several important confounders. Subgroup analysis for CPAP users showed that men had no treatment outcome and that women had a pronounced treatment outcome in sexual QOL improvement, which differs from previous studies that observed a more pronounced treatment outcome in men.

The association of CPAP with sexual function has been extensively studied in men with OSA. Several studies have shown that CPAP is associated with improvements in erectile dysfunction.^14-20 The results of a randomized clinical trial¹⁹ showed large improvement in erectile dysfunction (using the validated International Index of Erectile Function questionnaire) in a group of men with OSA treated with CPAP compared with those treated with antidepressants alone. In contrast, another randomized clinical trial²⁰ comparing CPAP with an oral appliance device showed no significant improvement in erectile dysfunction or other measures of sexual function in either treatment group. Nonetheless, most studies support the notion that CPAP treatment improves erectile dysfunction in men with OSA.

Five previous studies evaluated the outcomes of CPAP treatment specifically for sexual function in men with OSA.^21-25 In a case series of 73 patients, Lai et al²¹ showed that adherence to CPAP treatment for 1 year was associated with improved bed partner intimacy (measured using the Functional Outcomes of Sleep Questionnaire) compared with pretreatment levels. Similarly, Petersen et al²² showed improvements in sexual satisfaction, measured using the Life Satisfaction Questionnaire (LISAT-11) and the Brief Sexual Function Inventory, in a series of men with OSA after 1 year of CPAP treatment. Other case series assessing the outcomes of CPAP treatment after shorter time frames (1-3 months) showed similar improvements in sexual function in men with OSA.^23-25

Only 1 study to date has evaluated the association of CPAP treatment with sexual function and QOL in women. In a case series of 44 women with OSA, Petersen et al²⁶ showed that CPAP treatment was associated with improvements in sexual dysfunction (measured by the Manifest Female Sexual Dysfunction questionnaire) but no changes in isolated sexual difficulty (measured by the Female Sexual Functional Index), sexual distress (using the Female Sexual Distress Scale), or overall satisfaction (measured using LISAT-11) after 1 year of treatment. While Petersen et al noted that CPAP may improve sexual QOL in women with OSA,²⁶ this improvement was less pronounced compared with their parallel study of men with OSA.²²

Subgroup analysis of our cohort, in contrast, revealed a significant association of CPAP treatment with sexual QOL for women with OSA, but no association for men with OSA. After adjustment for several confounders, women who were CPAP users had a more than 1-point improvement on the sexual QOL domain than women who were CPAP nonusers. A 1-point improvement corresponds to a change in 1 category level of severity (ie, from a moderate problem [score of 3] to a mild or slight problem [score of 2], or from a mild or slight problem [score of 2] to a very mild problem [score of 1]). Unexpectedly, no difference was seen in the men in our cohort between the CPAP users and nonusers after the adjustment of confounders.

The design of this study might partly explain the discrepancy between the results of our study and other published studies. Our prospective clinical cohort study design allowed the measurement of change in sexual QOL after 1 year for CPAP users in direct comparison with the change in sexual QOL for a control group of CPAP nonusers. Many of the other studies that assessed CPAP treatment and sexual QOL (in both men and women) were case series with no control group. The study design of a case series cannot account for the natural course or the variability of OSA and its association with the treatment outcome. Self-reported outcomes, such as sexual function and sexual QOL, may be more vulnerable to systematic biases. Thus, a control group in our study may have helped mitigate these risks and potentially provided a more accurate representation of the association of CPAP treatment with sexual QOL. Our controlled study findings were consistent with 1 of 2 randomized clinical trials in men.²⁰

Another potential reason for the discrepancy of our study with other published studies was the degree of adjustment for confounding. Many of the previous studies on CPAP outcomes had limited the adjustment for potential confounders of sexual QOL, while other studies were limited to subgroup analysis (without further adjustment for confounding) or to no accounting for confounders. The association between OSA and sexual dysfunction was strong for both sexes, but there were many prevalent characteristics and comorbidities associated with OSA that may have confounded this association (eg, depression, diabetes, and other diseases). The studies that adjusted for confounders frequently adjusted for age, obesity, and/or OSA severity; adjustment for comorbidities was rare. Comorbidities may also significantly affect sexual QOL in OSA patients; this highlights the need for adequate comorbidity adjustment in sexual QOL studies. We attempted to adjust for potential confounding by using the robust Functional Comorbidity Index. This index captures several important potential confounders of QOL (and sexual QOL) into a composite index and has been validated as a robust predictor of QOL, function, and general health in patients with OSA.³²

A major strength of this study was the measurement of CPAP use, which resulted in accurate designations of CPAP users and nonusers, and accurate assessments of treatment outcomes. Additional strengths were its prospectively collected metrics, the use of a validated QOL instrument, and the use of a general patient population with OSA. While some of the previous studies selected patients with OSA and sexual dysfunction to be included in their studies, our study included a broad sample of patients with OSA who were not selected on the basis of baseline sexual dysfunction. These features helped mitigate selection bias that might exaggerate the treatment outcomes and strengthened the external validity of our findings by making the results more generalizable to a broad patient population with OSA.

There are important limitations to this study. The measurement of sexual QOL derived from the sex-related questions in the SNORE-25 was 1 key limitation. While the SNORE-25 is a validated instrument for measuring QOL related to OSA, we did not use a dedicated sexual QOL instrument. The study question of whether CPAP affects sexual QOL was posed after the parent trial was conducted and all prospective data were collected. In the parent trial, the SNORE-25 instrument was administered to all the patients, but data from dedicated sexual QOL instruments were not collected. A dedicated sexual QOL instrument may have been more sensitive for measuring sexual dysfunction and may partially explain these discrepant results compared with previous studies. Nonetheless, the use of a sexual domain derived from the SNORE-25 instrument still has merit. According to item response theory, there is evidence to suggest that responses to individual items from a validated QOL instrument can be combined into new valid domains.³⁴ Furthermore, the individual items in the SNORE-25 have substantial face validity because the questions were developed from multiple semistructured interviews with patients with OSA to identify the cogent issues of OSA and its treatment, further supporting that the sex-related questions from the SNORE-25 are relevant to patients and their sexual QOL.²⁷ Although this study did not use a dedicated sexual QOL instrument or independently validate a sexual QOL domain, the use of a sexual QOL domain from the SNORE-25 instrument, while imperfect, provides useful insight of the association of OSA and its treatment with sexual QOL.

An additional limitation was the scope of the QOL assessment. Although this study used a validated measure of QOL, it did not specifically include an evaluation of sexual partner perceptions concerning CPAP use with regard to sexual QOL. Sexual partner data may be useful because some studies have suggested that men with OSA who receive CPAP treatment may improve the sexual QOL of the women partners, whereas other studies have suggested that the CPAP treatment itself may adversely affect intimacy.^35-37 While these metrics are important for having a thorough understanding of how CPAP affects sexual QOL, sexual partner data as well as data from dedicated sexual QOL instruments were not prospectively collected as part of the parent trial.

Potential future directions include a larger, multicenter trial comparing CPAP with treatments that are less cumbersome each night (eg, surgery) to assess outcomes for sexual QOL. Such studies could include comprehensive assessments of sexual QOL in patients with OSA (and their sexual partners) with validated sexual QOL instruments as well as physiologic measures tracked longitudinally to allow for comparisons of benefits and sequelae of different treatment modalities. A few studies have compared CPAP with surgery and oral appliances with mixed results.^20,38,39 A larger, more detailed assessment of sexual QOL resulting from treatment by these different modalities would ultimately help to inform patients of the potential QOL benefits, including benefits in sexual QOL, resulting from OSA treatment.

Long-term sexual QOL may be improved in women with OSA who are using CPAP treatment for at least 4 hours a night compared with those women not using CPAP therapy. This study does not support improvement in long-term sexual QOL in men with OSA who are using CPAP therapy.

Accepted for Publication: March 23, 2018.

Corresponding Author: Sebastian M. Jara, MD, Department of Otolaryngology–Head and Neck Surgery, University of Washington, 1959 NE Pacific St, Box 356515, Seattle, WA 98195-6515 (sebastian.jara@gmail.com).

Published Online: May 24, 2018. doi:10.1001/jamaoto.2018.0485

Author Contributions: Dr Weaver had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: All authors.

Acquisition, analysis, or interpretation of data: Jara, Weaver.

Drafting of the manuscript: Jara, Weaver.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Jara, Weaver.

Obtained funding: Weaver.

Administrative, technical, or material support: Weaver.

Study supervision: Hopp, Weaver.

Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported.

Funding/Support: This work was supported by grants R01 HL084139 (Dr Weaver), T32 DC000018 (Dr Jara), and F32 HL136004 (Dr Jara) from the National Institutes of Health and by resources from the Veterans Affairs Puget Sound Health Care System.

Role of the Funder/Sponsor: The funding organizations had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Meeting Presentation: This work was presented at the American Academy of Otolaryngology–Head & Neck Surgery Foundation Annual Meeting, September 12, 2017; Chicago, Illinois.

Additional Contributions: The sleep medicine physicians at the following recruitment sites provided input on the parent trial and access to their patients for the parent trial: University of Washington Sleep Center at Harborview Medical Center (Vishesh Kapur, MD, MPH; Nathaniel Watson, MD; and Theodore Bushnell, MD) and the Sleep Center at the Virginia Mason Medical Center (William DePaso, MD, and Matthias Lee, MD). We acknowledge the parent trial study coordinators, who were led by Kathryn T. James, PA, MPH (University of Washington). We acknowledge the faculty and resident members of the University of Washington Otolaryngology Outcomes Research Group, who provided feedback on this study. None of the individuals or groups mentioned was financially compensated for their contribution.