Heart rate recovery in obstructive sleep apnea: obesity or not?
To the editor:
I read with great interest the invited commentary by Micha T. Maeder concerning the clinical utility of exercise heart rate recovery in obstructive sleep apnea (OSA) . I have read much of what Dr. Maeder has published on this topic and commend him for his work.
Attenuated heart rate recovery following exercise has been established as a powerful independent predictor of cardiovascular and all-cause mortality risk, both in apparently health and patient populations [2–4]. It has been thought to be a result of decreased parasympathetic activity [2, 5]. With OSA, sympathetic activity has been shown to be heightened both during sleep and normal waking hours [6, 7] but may also impact parasympathetic activity as well . Autonomic dysfunction is a key mediator in the increase in cardiovascular disease risk that is seen in those who suffer from OSA. Several recent studies have demonstrated that in individuals with OSA, post exercise heart rate is attenuated compared to similar subjects without OSA [9–12]. This would seem to indicate that the autonomic dysfunction seen in OSA manifests, not only at rest, but during exercise as well. A frequent complication in understanding OSA and autonomic function is that obesity, which is highly prevalent in OSA, also leads to alterations in autonomic function .
I agree, however, with Dr. Maeder that, at present, the use of heart rate recovery (HRR) as a clinical tool is not practical due to the heterogeneity of the study protocols used to investigate this issue . Another question that needs further examination relates to whether age might be yet another variable that influences the hypothesized relationship of interest between HRR and OSA. As discussed in previous studies [12, 14], HRR in OSA does appear to be impacted by age. Might the utility of exercise testing in OSA screening become more evident upon completion of larger scale studies, in which HRR and age, among other variables, are included in the prediction model?
Dr. Maeder, in his editorial, cites three of the previous studies that have examined HRR in OSA [10–12]. He states that the average body mass index (BMI) of subjects in these studies was >30 kg/m2 and concluded that, as a result, it is difficult to interpret whether the attenuated HRR seen in these studies was due to OSA or obesity . Our study , however, did in fact have a normal weight control group without OSA (mean BMI = 22.0 kg/m2 ± 1.3), in addition to obese groups with (BMI = 32.0 kg/m2 ± 3.7) and without OSA (BMI = 31.4 kg/m2 ± 3.7). Results from our study clearly showed only the obese group with OSA had an attenuated HRR. Further, we found no difference in HRR values between the normal weight and obese non-OSA group. These findings support that, at least in younger adult males, attenuated HRR does not appear to be mediated by obesity.
Thus, our study, has in fact provided evidence that, in young adult men, HRR is more likely to be suppressed by OSA and not obesity. Obviously, studies with large subject groups that rigorously differentiate groups by OSA and obesity factors are needed. Inclusion of subject groups that are of normal weight with OSA would aid in this investigation. Given the prevalence of obesity and other comorbid conditions with OSA, however, presents a significant challenge in subject recruitment.
Conflict of interest