A Brief History of Fluid and Sleep

Obstructive sleep apnea (OSA) is not a new disease. However, it is only in the relatively recent past that a fuller understanding has been gained of the myriad phenotypic factors that interact to produce the final clinical syndrome. This has allowed the development of novel therapeutic approaches for carefully selected cohorts of patients with OSA, such as the use of electrical stimulation with hypoglossal nerve (1) and transcutaneous stimulation (2). Despite these advances, the vast majority of patients with OSA are likely to continue being treated with nocturnal continuous positive airway pressure therapy and to experience its associated issues with patient acceptance and compliance (3). Hence, there is a need to better characterize contributory factors in patients with OSA, potentially leading to the development of much-needed novel therapeutic strategies.

A significant part of our knowledge about sleep apnea is derived from its association with certain predisposing anatomical or pathophysiological features. Visceral adiposity, upper airway anatomy, age, and sex are recognized as important contributors, but the pathophysiology of sleep apnea is multifactorial and complex. Changes in posture can have immediate effects on respiratory mechanics (4) and neural respiratory drive (5) and slower effects on intra- and extravascular volumes as a result of plasma movement (6). However, the role of fluid, and in particular nocturnal fluid shifts, in OSA remains a relatively underexplored area. That fluid dynamics have a relationship with sleep-disordered breathing is well established, at least in the context of central sleep apnea/Cheyne-Stokes respiration (CSA/CSR). This condition is highly prevalent in subjects with severe or decompensated congestive heart failure and is likely contributed to by pulmonary vascular congestion triggering a degree of hyperventilation, with the consequent reduction of the partial pressure in carbon dioxide (Pa_CO2) below the apneic threshold (7). Aggressive medical therapy of heart failure, leading to improvements in pulmonary hemodynamics, can lead to a significant reduction of CSA/CSR (8).

In this issue of the Journal, the article by Lyons and colleagues (pp. 1287–1294) is the latest in a growing body of work evaluating the role of fluid dynamics in subjects with sleep-disordered breathing (9). Prior data have suggested that rostral fluid shifts may contribute to the occurrence and severity of OSA (10) and that the use of strategies to attenuate this fluid shift, such as graded compression stockings (11, 12), may measurably reduce the degree of sleep-disordered breathing seen. This most recent study elegantly demonstrates that reductions in fluid volume, as opposed to any changes in metabolic parameters, lead to marked improvements in sleep-disordered breathing and sleep quality in subjects with end-stage renal disease and sleep apnea.

The striking feature of these findings is that ultrafiltration has an effect on obstructive and central aspects of sleep apnea, and although the pathogenesis is different, this defines a common pathway of the pathophysiology of sleep apnea in patients with end-stage renal disease. The question arises whether a binary classification of sleep apnea, central or obstructive, gives justice to this complexity or whether a categorization considering a spectrum of more obstructive or more central respiratory events is more appropriate.

It has long been recognized that there is considerable overlap between the pathophysiology of central and obstructive sleep apnea (13). However, methodological limitations can contribute to misclassification of obstructive and central respiratory events, particularly as inspiratory effort is not routinely assessed with more sensitive invasive measures, such as esophageal manometry or diaphragm electromyography, during polysomnography (14). The direct measurement of neural respiratory drive in this patient cohort could quantify the variability and instability of an arousal threshold (15) and help to better phenotype the underlying pathophysiology.

In the article by Lyons and colleagues, the observed reduction in the apnea–hypopnea index of 36% (28–44%) after a single fluid removal of 2.17 ± 0.45 L is significant, although moderate. Fluid removal led to significant improvement in sleep quality with increased sleep efficiency, as well as increases in total, slow wave, and rapid eye movement sleep time (9). However, it remains uncertain whether manipulating fluid dynamics can achieve measureable benefits in general sleep apnea populations; the use of compression stockings to diminish nocturnal rostral fluid shift in a cohort of patients with moderate to severe OSA led to a statistically significant reduction in apnea–hypopnea index but had less effect on patient-reported outcomes (11). Whether more aggressive fluid removal strategies could be of benefit in such cohorts remains unexplored.

Nonetheless, the results of this well-designed physiological study are encouraging and need to be confirmed in larger controlled trials, focusing on the differentiation of mechanistic models of the etiology of sleep apnea in this patient group. Intelligent solutions of nocturnal fluid removal in patients with end-stage renal disease might further offer the chance to support these patients, potentially without recourse to established therapies such as continuous positive airway pressure.