Sleep-disordered breathing and renal failure: A search for fundamental mechanisms☆

The most exciting phrase to hear in science, the one that heralds new discoveries, is not “Eureka!” (I found it!) but rather “Hmm…that’s funny.” Isaac Asimov

Sleep disorders are quite common in patients with renal failure. Studies have demonstrated increased prevalence of poor quality sleep, excessive daytime sleepiness, insomnia, restless legs syndrome and sleep disordered breathing [1]. Sleep disorders influence quality of life and contribute to the morbidity of renal disease and perhaps to decreased survival [2]. In particular, sleep-disordered breathing (SDB) is much more prevalent in patients with end-stage renal disease (ESRD) than in the general population [3], [4]. It would have been entirely reasonable to implicate comorbid and highly prevalent conditions such as obesity, cardiac disease, and diabetes as being responsible for this association or to simply blame upper airway edema from fluid excess and uremia. However, it seems likely that an Asimov “that’s funny” moment prompted some observers to look for less obvious mechanisms; this search may, in turn, lead to fundamental insights into the pathogenesis of complex combinations of sleep-related breathing events in patients with, or even without, ESRD.

Several recent observations with respect to SDB in ESRD seem capable of provoking a “that’s funny” moment in the clinician/scientist caring for such patients. To begin, it is now doubtful that comorbid diseases alone cause this high prevalence of SBD. A seminal study comparing patients undergoing thrice-weekly conventional hemodialysis (CHD) with matched controls from the Sleep Heart Health Study identified a strong association of CHD with severe SDB and nocturnal hypoxemia that was independent of age, BMI, and chronic disease [5]. In addition, while frank uremia or serious fluid overload may be somewhat involved in pathogenesis, they cannot now be thought of as entirely responsible. We now know that the prevalence of SDB is similar in patients before initiation of dialysis compared to patients receiving peritoneal dialysis or hemodialysis [6], [7], [8], and despite case reports indicating resolution of sleep apnea following renal transplantation, many patients do not improve [9], [10], [11]. Furthermore, SDB in ESRD (in patients untreated or on dialysis) can incorporate features of both central and obstructive sleep apnea, and a given individual may exhibit only one or the other mechanism or varying combinations of both [12]. The coexistence of obstructive, central, and mixed apneas, occurring in various patterns depending on sleep stage, sleeping posture, time of night and therapeutic interventions (e.g., titration of positive airway pressure) has been termed complex sleep apnea syndrome (CompSAS), and aberrations in ventilatory control are commonly invoked to explain the central events [13], [14]. Many sleep laboratories overlook sleep-onset and post-arousal central apneas, and clinical polysomnography is not capable of accurately discerning the difference between hypopneas generated by airway obstruction and those produced by decreased respiratory drive. These factors may have delayed the realization that central and complex SDB frequently occur in ESRD. Once recognized, the wide spectrum of SDB phenotypes in ESRD clearly implicates control of breathing as an important pathogenetic factor in these individuals.

Disturbances in control of breathing figured prominently in early theories of sleep apnea pathogenesis. The focus of attention subsequently shifted to upper airway mechanics after the recognition of positive airway pressure as an effective treatment for obstructive sleep apnea. There is now a developing consensus that both upper airway phenomena and perturbations of ventilatory control contribute to the development of apnea [15], [16]. Central apnea, including the central component of mixed apnea, CompSAS, and presumably sleep apnea associated with ESRD, occurs when arterial pCO2 falls below a putative apnea threshold. Theory holds that enhanced peripheral and central chemoreceptor sensitivity can contribute to an increase in respiratory control system “loop gain” and the development of unstable breathing during sleep [17], [18]. Indeed, there is considerable evidence, much of which is derived from studies of patients with congestive heart failure, that increased chemoreceptor responsiveness to CO2 is associated with central sleep apnea [19]. Furthermore, many studies of obstructive sleep apnea pathogenesis, particularly in severe disease, have demonstrated increased chemoreceptor responsiveness and unstable control of breathing that undoubtedly contribute to dysregulation of upper airway caliber [17], [20], [21]. It remains to be demonstrated whether increased chemoreceptor responsiveness is an acquired trait or represents a constitutional predisposition that is genetically programmed. Factors that might be associated with an acquired increase in chemoresponsiveness include: altitude, obesity, arterial blood gas abnormalities, medication, menstrual phase, cardiac output, pulmonary vascular congestion (interstitial edema leading to increased mechanoreceptor stimulation), and the status of the sympathetic nervous system, including anxiety/panic states [22], [23], [24].

In this issue of Sleep Medicine, Beecroft and colleagues [24] report a study of ventilatory control in azotemic patients converted from standard hemodialysis to nocturnal hemodialysis (NHD). This group of investigators has previously demonstrated that conversion to NHD dramatically improved sleep apnea [25]. They have also shown that patients with ESRD and sleep apnea possess increased sensitivity of chemoreceptor reflexes [26]. In the present paper [24], twenty-four patients underwent nocturnal polysomnography on a non-dialysis night before and after conversion to NHD. Ventilatory response to CO2 during hypoxic and hyperoxic conditions using a modified Read re-breathing technique [27] was assessed prior to each polysomnogram. Seven patients were identified as having moderate to severe OSA at baseline, and four of these patients demonstrated reduction in AHI (by >50% or to <15/h) on NHD. Nocturnal oxygenation improved in these responders, and these patients had slightly (but significantly) lower arterial pH compared with non-responders and patients without mild to moderate OSA. Ventilatory response to CO2 was higher in apneic than in non-apneic patients, and AHI correlated with this response slope in the patients with OSA. Ventilatory response on NHD fell in the apneics (mainly in those with a fall in AHI on NHD) and increased in those without OSA. These changes were significant only under hyperoxic conditions, suggesting predominant involvement of the central chemoreflex, [28] and were associated with a significant correlation between change in AHI and change in ventilatory response. The authors conclude that improvement in AHI upon conversion of some azotemic patients from conventional hemodialysis to NHD may be due to a reduction in central chemoreceptor response, resulting in more stability of ventilatory control. To make matters more “complex,” this same group of researchers previously demonstrated that patients with ESRD have a decreased pharyngeal cross-sectional area compared to normal controls [29] and that the decreased pharyngeal size remits after conversion from conventional hemodialysis to NHD [30].

The mechanism by which NHD could change ventilatory control in ESRD patients remains obscure, but it is becoming clear that this modality has significant advantages over CHD with respect to a variety of outcome metrics. Several small studies have suggested that NHD produces substantial improvement in blood pressure, left ventricular hemodynamics, biochemical profile, cognitive function, erythropoietin responsiveness, and quality of life compared to CHD [31], [32], [33], [34]. Walsh and colleagues conducted a systematic review of this literature in 2005 and concluded that further data on mortality and cardiovascular morbidity were necessary to justify the investment of resources necessary to support widespread implementation of NHD [35]. Subsequently, a randomized controlled trial demonstrated that NHD resulted in improved left ventricular mass, reduced the need for blood pressure medications, improved some measures of mineral metabolism, and improved selected measures of quality of life [36]. Presumably, a clue to how NHD affects respiratory control is buried somewhere in this constellation of improved outcomes, and future investigations will hopefully provide clarity.

While the data reported by Beecroft and colleagues [24] are derived from relatively small studies and consequently should be interpreted with caution, this avenue of investigation could shed new light upon the causative factors for SDB associated with chronic renal failure that may also be relevant for other populations of sleep apnea patients. For example, the apparent reduction in SDB severity associated with NHD provides investigators with an “experiment of nature” with the potential for considerable scientific utility. By “connecting the dots” between NHD-induced improvements in specific physiological functions and changes in ventilatory control, important insights into the pathogenesis of all types of SDB might be derived. Furthermore, altered control of breathing from NHD may affect individual patients with various backgrounds of predisposition and comorbidity in different ways [37], and this heterogeneity could lead to other important discoveries.

Finally, the importance of SDB as a contributing factor to the morbidity and mortality of patients with renal insufficiency should become a major area of interest for both clinicians and investigators in sleep medicine. It is highly probable that many of these patients have not been evaluated for sleep disorders in the past and will benefit from enhanced collaboration between sleep specialists and nephrologists. Such an alliance could identify appropriate candidates for new modalities of renal replacement therapy as well as enhanced management techniques for complex sleep apnea syndromes [38].