Obstructive sleep apnea (OSA) is a prevalent sleep disorder that, in addition to numerous other adverse health effects, significantly increases the risk of cardiovascular disease [1]. Positive airway pressure (PAP) remains the standard therapy for OSA. However, the impact of PAP on cardiovascular events among those with OSA remains equivocal [2]. In fact, recent research suggests that the effects of PAP in some individuals with OSA may counteract the expected cardiovascular benefits of PAP [3, 4].

Exercise, often in conjunction with dietary modification, is a common alternative or complementary treatment option for OSA [5, 6]. Moreover, exercise significantly reduces cardiovascular risk [7] and factors that promote CVD such as inflammation [8]. Consistent evidence from small experimental trials suggests that exercise training, even without weight loss, reduces OSA severity [9, 10]. Often, though, the magnitude of reduction in OSA severity is modest, ranging between 25 and 40% reduction from baseline [10]. Because exercise rarely results in complete resolution of OSA, many of the chronic insults that link OSA to increased cardiovascular risk remain, including intermittent hypoxia and recurrent arousals. Therefore, it is possible that residual OSA may attenuate exercise-induced improvements in cardiovascular health.

In this issue of Sleep and Biological Rhythms, Dobrosielski and colleagues examined whether the effects of exercise training on inflammation differed based on baseline OSA severity [11]. In secondary analyses from a single-arm experimental trial focused on vascular function [12], 22 participants were categorized into two groups: no or mild OSA (apnea–hypopnea index [AHI] < 15, n = 10) or moderate/severe OSA (AHI ≥ 15, n = 12). Participants completed a 6-week supervised aerobic and resistance exercise training intervention at a dose that aligned with public health recommendations. Before and after the intervention, multiple blood-based markers of inflammation (i.e., C-reactive protein [CRP], interleukin-4 [IL-4], interleukin-8 [IL-8], tumor-necrosis factor-alpha [TNF-α]) were assessed along with the adipokines adiponectin and leptin. In addition, cardiorespiratory fitness and body composition were objectively assessed at both timepoints.

The authors found no changes in inflammatory markers as a result of the exercise intervention in either the no/mild OSA or moderate/severe OSA groups. As the authors noted, the lack of observed change may have been due to the exercise intervention being of insufficient duration and/or volume. The authors did observe a significant reduction in leptin following exercise training, but this reduction was only found in the moderate/severe OSA group. Importantly, this finding was contrary to their hypothesis that improvements would be blunted in the moderate/severe OSA group. The authors postulated that severe OSA may have a stronger influence on leptin and that leptin levels may respond to exercise training in a dose–response manner.

Standing on its own, little can be concluded from the study by Dobrosielski and colleagues based on its limitations (many of which are identified by the authors themselves). The small sample size prevented examination of exercise-induced changes in inflammation across more discrete categories of OSA severity (i.e., none vs. mild vs. moderate vs. severe) as well as the interaction that sex differences may have. Moreover, the use of AHI to categorize participants could be questioned since AHI has significantly less predictive ability for CVD than hypoxemia-based measures [13]. The study was also limited by its lack of OSA severity assessment at post-intervention, preventing a mechanistic examination of whether changes in inflammatory markers were related to the magnitude of exercise-induced reduction in AHI.

Nevertheless, we commend Dobrosielski and colleagues for taking the step forward and addressing an important topic that we hope spurs additional investigation by them and others. Part of the appeal of exercise for a role in the management of OSA resides in its potential ability to reduce OSA severity and reduce cardiovascular risk, presumably independent from its effects on OSA. Dobrosielski and colleagues state that, in theory, the cardiovascular adaptations conferred by exercise may be blunted in those with underlying OSA. To date, their work has not supported this hypothesis when investigating outcomes of inflammation [11], vascular function [12], and executive function [14]. This area of research remains in its infancy though, and future research should attempt to more comprehensively address (in larger and diverse samples) whether individuals with OSA should expect similar cardiovascular adaptations from exercise training as those without OSA.