Background: Recovery kinetics following cardiopulmonary exercise testing (CPET) have yet not been described in people with OSA and comorbid insomnia (COMISA). Our objectives were to i) describe heart rate recovery (HRR), oxygen recovery slopes (VO2/t) and time to half recovery of VO2max (T½) following CPET in COMISA; and ii) investigate their response to either one of the following: 16 weeks of exercise training (Ex) or 8 week-relaxation control intervention followed by 8 weeks of exercise and cognitive-behavioural therapy for insomnia (CBTi-Ex).Ex consisted of 3 weekly sessions of aerobic and resistance training. The CBTi-Ex program included the same exercise plan in addition to four 60 minutes modules (each offered every other week) emphasizing proper sleep behavioural principles.

Methods: Participants with COMISA underwent 3 maximal exercise tests on a cycle ergometer pre- (n=17), mid- (Ex n=6, CBTi-Ex n=8), post-intervention (Ex n=7, CBTi-Ex n=8). HRR and VO2/t (at minutes 1, 2 and 5), and T½ were calculated post- peak exercise intensity. Descriptive statistics [Mean (SD)] and repeated measures ANOVA were used.

Results: At baseline, HRR at minutes 1, 2 and 5 were 20(7), 35(8), 49(13) bpm, respectively, VO2/t slopes were 0.731(0.521), 0.590(0.288) and 0.222(0.096) L/min2 respectively and T½ was 97(64) sec. Significant time effects with large effect sizes were found pre- to post- intervention for VO2/t 2 (p = 0.045; partial eta squared (ηp2) = 0.275) and VO/t 5 (p = 0.017; ηp2 = 0.367). Large effect size of time by group interaction was found pre- to mid- intervention for HRR 5 (ηp2 = 0.232).

Conclusion: Despite people with COMISA showing poor exercise recovery kinetics initially, encouraging improvements in oxygen kinetics obtained through Ex or CBTi-Ex confirm the relevance of those interventions on COMISA, and should motivate similar studies including larger sample size and stricter recovery protocols to be done.