Abstract

Background:
Obstructive Sleep Apnea Syndrome (OSA) is a common sleep disorder characterized by partial or complete blockage of the pharyngeal airway during sleep, posing as a significant public health concern. OSA is often comorbid with chronic insomnia (COMISA). Insomnia disorder involves a difficulty initiating sleep, maintaining sleep and/or early awakenings, despite time and opportunity for sleep, causing impairments in daytime functioning and significant distress. Individuals with COMISA suffer from worse sleep and daytime functioning, compared to individuals presenting with either disorder in isolation. Due to their insomnia complaints, these individuals tend to be less compliant with continuous positive airway pressure (CPAP) therapy, the gold standard treatment for OSA. Cognitive behavioural therapy for insomnia (CBTi) has shown promising results in this population but can be difficult for individuals to access, and alone, cannot improve OSA severity. This unique group requires a therapy targeted to both disorders, which would potentially improve apnea-hypopnea index (AHI) and/or insomnia symptomology for a better sleep. Currently, no effective therapy tailored to these individuals has been found. Exercise is a potential non-pharmacological therapy that would simultaneously improve symptoms of both disorders. Studies examining the effects of exercise as a therapy for COMISA are lacking, and research examining exercise as a therapy for either of the two disorders alone yield mixed results likely due to variation in methodology and treatment approach.

Objectives:
(1) To determine if an 8-week exercise intervention would improve insomnia severity in patients with comorbid OSA and insomnia when compared to an 8-week active control condition of relaxation therapy. (2) To examine the effects of the exercise and relaxation interventions on cardiorespiratory fitness both within-groups and between-groups. (3) To examine if improvements in cardiorespiratory fitness would be associated with improved changes in objective and subjective sleep quality.

Methods:
Sixteen participants (10 female, 54.9  13.4 years of age) were randomised to 3 weekly sessions of exercise training or self-guided relaxation. The exercise sessions consisted of 60 minutes of moderate-intensity aerobic and resistance exercise, with 1 weekly session supervised and the remaining two unsupervised at home or in the community. The protocol included a screening polysomnography (PSG) night, a cardiopulmonary exercise test, and an overnight PSG, as well as questionnaires (including the Insomnia Severity Index (ISI)) before and after the 8 weeks. All PSG recordings were sampled at 512 Hz (Somnomedics, Germany) and sleep stages were scored offline according to AASM guidelines. ISI and sleep efficiency (SE) (extracted from PSG) were assessed as primary outcomes. VO2peak (ml/kg/min), VO2 at the ventilatory threshold (VT), heart rate (HR) at 50% isotime (ISO) pre- and post- intervention were assessed as secondary outcomes. Bivariate correlations were also performed to examine any potential relationships between change in cardiorespiratory fitness and change in subjective and objective sleep parameters.

Results:
Results revealed a significant effect of time (pre, post 8-weeks) on ISI score (F(1,14)= 12.315; p=.003), but no significant effect of condition (exercise, relaxation) or time*condition interaction. Both exercise and relaxation had large effects on ISI, with exercise showing a larger effect size (Cohen’s d= 3.88) than relaxation (Cohen’s d= 0.184). No significant effects were found for SE. A significant time*condition interaction was found for VO2peak (ml/kg/min) (F(1,15)= 10.724; p=.006), with the exercise condition showing improvements. Spearman correlations indicated a non-significant association for change in ISI and change in VO2peak (ml/kg/min) (r=0.59; p>.05).

Conclusions:
Both exercise and relaxation reduced insomnia severity in people with COMISA, with exercise having a larger effect. Objective measures of sleep efficiency did not improve significantly in either group. Cardiorespiratory fitness improved with the partly home-based, moderate-intensity aerobic and resistance training intervention used in this study. Larger trials are warranted to confirm these findings.

Significance:
This is the first randomized controlled trial (RCT) to examine the effects of an exercise intervention on sleep in this specific COMISA population.