Obstructive Sleep Apnea and Reverse Dipping

Obstructive sleep apnea (OSA), one of the most common sleep-related breathing disorders, affects up to one-fifth of the adult population in the United States [1], and its prevalence is estimated to be on the rise, probably because of escalating obesity rates. OSA exerts unfavourable effects on the cardiovascular system through a number of hemodynamic and chemo-mediated mechanisms, mainly triggered by the occurrence of wide swings in blood pressure (BP), heart rate, and oxygen saturation as a result of alternating obstructive apnea and hyperventilation episodes during sleep, and convergent towards the common phenotype of hypertension [2]. As a matter of fact, repeated episodes of apnea and hypoxia lead to sleep interruptions, thus indefinitely multiplying the arousal responses driven by the autonomous nervous system. As a consequence, subjects with repeated events of apnea and hypoxia during sleep are prone to develop intermittent bouts of acute BP changes during nocturnal arousals, with increases in systolic BP up to 20 mmHg within few minutes [3]. Moreover, the acute and recurrent overstimulation of autonomous nervous system also leads to a long-lasting sympathetic activation [4]. OSA may also induce a sustained increase in BP through other mechanisms regulating BP, such as endothelin-1 release [5], low-grade systemic chronic inflammation [6], increased arterial stiffness [7], and endothelial dysfunction [8]. As a confirmatory finding, a number of observations suggest that the reversal of OSA, obtained through the use of continuous positive airway pressure [9] or pharyngoplasty [10], significantly reduces BP, and may be considered as a valid tool for improving BP control in hypertensive patients with OSA. Observational studies show a continuous, graded relationship between OSA and office BP [11], with progressive increases in BP values paralleling the increased numbers of apnea-hypoxia events per night. However, office or home BP measurements may underestimate BP-related cardiovascular risk because of inherent problems in measuring BP during sleep, and fully automated 24-h ambulatory BP monitors or the recently available automated home BP monitoring devices with nocturnal readings enabled [12] may better evaluate the BP-related cardiovascular risk in OSA and its association with target-organ damage. In fact, given that acute and repeated changes in BP occur during the night, nocturnal hypertension and absent or reverse BP dipping are frequent in OSA [13] and can be considered key features of hypertension in such patients. In longitudinal studies, OSA has been associated to the future occurrence of nocturnal hypertension and reduced or inverted BP dipping patterns in children [14], adults [15], and the elderly [16]. Day–night BP reduction, known as the dipping phenomenon [17], represents another sleep-related event with considerable clinical and prognostic implications. A reduced or inverted day–night BP pattern is a well known independent predictor of adverse cardiovascular outcomes in hypertension [18] and in the general population [19]. Interestingly, the predictive power of the night–day BP ratio appears to be largely dependent on the worse prognosis of reverse dippers [20]. Sleep apnea is one of the mechanisms that may explain in part the higher nocturnal BP and associated worse outcome, along with nocturnal autonomic dysfunction, disturbed baroreflex sensitivity, abnormal sodium handling, and nocturnal volume overload. Therefore, the guidelines for the management of hypertension published in 2013 by the European Society of Hypertension and the European Society of Cardiology suggest that OSA should be suspected in subjects with unexplained nocturnal hypertension or absent or reverse dipping patterns on 24-h ambulatory BP monitoring [21]. The close association between OSA and acute BP changes was also confirmed in other studies. Steinhorst et al.[22] showed that the short-term variability of nocturnal BP, assessed through appropriate methods such as the time-rate index and the first derivative of blood pressure over time, was higher in subjects with moderate-to-severe forms of OSA than in controls, whereas daytime BP variability did not differ, thus confirming a close link of the number and severity of the apnea-hypoxia episodes with prognostically adverse changes of the BP profile during the night. In another study, oxygen-triggered BP monitoring with a variable threshold was able to detect severe OSA-related BP surges more effectively than the usual nocturnal BP measurement at fixed time intervals [23]. In the present issue of the Journal, Lee et al.[24] report on the relationship between abnormal circadian BP profile, OSA and white matter hyperintensities (WMHs) on brain MRI in a cohort of normotensive Korean subjects enrolled within the Korean Genome and Epidemiologic Study (KoGES), a large community-based study. WMH, defined as hyperintensities in periventricular or subcortical areas at T2-weighted or fluid-attenuated inversion recovery sequences of brain MRI, are considered a sensitive marker of BP-related cerebral organ damage. In a cohort of 72 elderly subjects, increased ambulatory systolic BP, but not clinic systolic BP, from baseline to 24-month follow-up was associated with increased WMH volume over that same period, as well as measures of executive function/processing speed [25]. Their pathological correlates include myelin pallor and loss, tissue rarefaction, and gliosis [26]. Most importantly, WMH portend an increased risk for incident stroke, dementia, and mortality independent of vascular risk factors [27], thus representing an attractive surrogate marker for both cerebrovascular complications and cognitive decline. In a previous report from the KoGES database, a close correlation had been found between OSA severity and the number of WMH [28]. The association persisted also after adjustment for the presence of hypertension, thus suggesting that the brain damage induced by OSA may be in part independent from its effects on BP. In the current article, Lee et al.[24] examined 703 subjects with an average age of 59 years. Subjects with dementia and prevalent cardiac or cerebrovascular disease, as well as those receiving BP-lowering medications, were excluded from the analysis. All participants underwent 24-h ambulatory BP monitoring to assess day–night BP behaviour, brain MRI for the evaluation of WMH, and domiciliary respiratory multichannel recording with a portable device for the estimation of OSA on an overnight basis. WMH were evaluated both qualitatively (defined as ≥5 mm in periventricular or subcortical areas) and quantitatively, whereas OSA, defined as ≥5 apnea/hypopnea episodes per hour, was classified as mild (5–14 episodes per hour) and moderate-to-severe (≥15 episodes). A large proportion of the participants had a reduced or inverted nocturnal BP fall, with 45% of the subjects being classified as nondippers (day–night systolic BP reduction 0–9%) and 13% as reverse dippers (nocturnal systolic BP > diurnal systolic BP). Reverse dippers showed a significantly higher prevalence of WMH (54%) than either dippers (30%) or nondippers (32%). The odds of having WMH in subjects with reverse dipping remained significant after adjusting for multiple confounders, such as age, sex, BMI, diabetes, smoking, C-reactive protein levels, cholesterol levels, and alcohol consumption. In this relatively elderly population, the prevalence of any OSA was 40%, and that of moderate-to-severe OSA was 9%. Among reverse dippers, the prevalence of moderate-to-severe OSA was significantly higher than in the other groups (13.5 vs. 9.0% in nondippers and 7.5% in dippers). Interestingly, the presence of moderate-to-severe forms of OSA was an effect modifier of the relationship between reverse dipping status and WMH (P value for interaction = 0.0118). The odds of having WMH in the presence of a reverse dipping pattern were nearly triple in the subjects with moderate-to-severe OSA and reverse dipping (4.71, 95% confidence interval 1.21–18.36) than in those with no or mild OSA and reverse dipping (1.74, 95% confidence interval 1.01–3.00). The above findings suggest that the presence of moderate-to-severe OSA significantly amplifies the effects of reverse BP dipping on WMH in apparently healthy individuals. A major limitation of the study is that it cannot demonstrate cause-and-effect relationships because of its cross-sectional design. As such, the study can at most provide hypotheses regarding the causal pathways linking the examined variables. In a further effort to clarify the direction of the associations between OSA, reverse dipping status, and WMH, the authors applied a mediational statistical model to test the hypothesis that reverse dipping (independent variable) influences the dependent variable (WMH) not directly, but by influencing a mediator variable (OSA). In such analysis, OSA was not found to be a significant mediator of the association between reverse dipping and WMH. It should be noted, however, that a more plausible biological pathway is that OSA determines an abnormal circadian BP profile, which in turn leads to organ damage, rather than the opposite, as hypothesized and tested by the authors [24]. Other limitations include the suboptimal reproducibility of a single 24-h BP monitoring session in assessing day–night BP changes [29], and the fact that unattended, portable monitoring records fewer physiologic variables than full in-laboratory polysomnography, which remains the gold-standard technique for evaluating OSA. Despite the above limitations, the intriguing findings of the study by Lee et al.[24] suggest that OSA and absent nocturnal BP reduction might interact synergistically in determining the risk of brain damage. Individuals with both moderate-to-severe OSA and an absent or inverted nocturnal BP reduction are at high risk for having WMH, and may be at particularly risk of future cognitive decline, stroke, and mortality [27]. As OSA may lead to vascular injury and cerebrovascular changes from hypoperfusion not only by increasing BP but also through a number of BP-dependent mechanisms [2], a multiplicative interaction of nocturnal BP and OSA may not be an unexpected finding. In these subjects, vigorous efforts to correct cardiovascular risk factors are recommended, and due consideration should be given to specific therapeutic measures aimed at controlling OSA and at reducing nocturnal BP. ACKNOWLEDGEMENTS Funding source: The position of G.P. as an Assistant Professor at the University of Perugia is funded by a grant from the Fondazione Cassa di Risparmio di Terni e Narni. Conflicts of interest There are no conflicts of interest.