Obstructive sleep apnea (OSA) is a clinical condition characterized by recurrent episodes of partial or complete upper airway obstructions during sleep, promoting intermittent hypoxia and sleep fragmentation (1). The prevalence of OSA among patients with end-stage renal disease (ESRD) varies according to the diagnostic criteria used and whether it was based on questionnaires (reaching 50–80%) (2, 3) or polysomnography (PSG) (16–46%) (4, 5) and is associated with mortality among patients on hemodialysis (HD) (6, 7).

Several factors contribute to the pathophysiology of OSA in patients with ESRD, such as uremia and associated comorbidities in ESRD (8). Volume overload, which is commonly observed in these patients, plays a pivotal role in this context. It leads to edema formation in upper airways and, consequently, to pharyngeal narrowing (9, 10). Hypervolemia also augments overnight spontaneous fluid shift, which occurs as the fluid accumulated in the legs during the day moves rostrally during bedtime, while on recumbent position. This is associated with overnight increase in neck circumference (NC) and the severity of OSA, not only in patients with OSA from the general population (11) but also particularly in patients with fluid retaining states, such as congestive heart failure (CHF) (12) and ESRD (4).

Avoiding fluid accumulation in the legs during the day by wearing compression stockings (CS) is an option to attenuate the overnight fluid shift as the compression exerted increases the pressure applied to the interstitium, thus reducing fluid filtration from capillaries in the legs (13). It has been demonstrated that wearing CS during the day attenuates OSA (14–16). It is still unclear why avoiding leg swelling leads to lower obstructive respiratory events during the night. A plausible mechanism is that the hydraulic pressure exerted upward leads to fluid retention in the trunk, thus increasing daytime diuresis and, consequently, lowering total body water (TBW). The effects associated with the wearing of CS in relation to fluid redistribution and severity of OSA in oliguric and anuric patients are unknown.

On the other hand, the standard treatment for OSA, the continuous positive airway pressure (CPAP), avoids airway collapsibility, by mechanically exerting pressure on airways, thus preventing some critical anatomic points to collapse, such as the retropalatal region of the oropharynx (17). However, another consequence of such pressure is avoiding fluid accumulation in upper airways, once relatively small amounts of fluid (100–200 ml) enlarge airway soft tissues and may cause airway constriction (18). Even though CPAP may reduce such local edema, it is not clear whether such local and “downward” pressure interferes in overall body fluid shift.

Therefore, this randomized crossover clinical trial was conducted to evaluate the short-term impact of wearing CS and CPAP on the severity of OSA and fluid redistribution between legs and trunk in a sample of oliguric or anuric patients on HD. We also evaluated the impact of these interventions on overnight changes in NC. We hypothesized that CS would interfere in overnight rostral fluid shift from the legs, with consequent improvement of OSA, while CPAP would act locally on upper airways and improve OSA regardless of fluid shift.

To our knowledge, this is the first study comparing overnight fluid shift and NC changes between CS and CPAP in a population on HD. We have also compared the effect of both therapies, CS and CPAP, on the severity of OSA. Our results revealed that (1) daytime use of below-the-knee CS for 1 week attenuated OSA severity in oliguric/anuric patients on HD although this effect was lower in comparison to CPAP; (2) the effects of CS were most likely attributed to displacement of fluid from the legs to the intracellular compartment of the trunk during the day, thus avoiding overnight rostral shift of extracellular fluid from the legs to the neck; (3) short-term CPAP therapy, on the other hand, did not avoid overnight fluid shift, yet it promoted a prominent reduction NC. Taken together, our findings suggest that CS may be an alternative and feasible treatment for attenuating OSA in HD patients. Furthermore, they suggest that CS and CPAP are effective on the treatment of OSA through distinct influences on overnight fluid distribution.

Obstructive sleep apnea is associated with a myriad of cardiovascular complications in the general population, such as coronary artery disease, stroke, CHF, and arterial hypertension (25–28). In patients with ESRD, who already exhibit an excessive cardiovascular risk (29), the impact of OSA can be even more devastating. Fluid overload is a possible link between high prevalence of cardiovascular disease and OSA in the setting of ESRD (30–32). Recently, hypervolemia has also been associated with more severe OSA in patients on HD (33). Indeed, there is an established background in literature in favor of fluid shift as a pathogenic factor for OSA in edematous states. In this regard, inflation of antishock trousers shifts fluid rostrally, mimicking spontaneous fluid shift and thus increasing pharyngeal resistance and narrowing the upper airway cross-sectional area in healthy individuals (34–36). In addition, magnetic resonance imaging study has shown that the severity of OSA correlates with jugular vein volume and the amount of mucosal water content surrounding the upper airway in patients on HD (37), suggesting an important role of hypervolemia in the pathogenesis of OSA in this population. In this regard, proof of concept was demonstrated by ultrafiltration without interference in uremic status in patients on HD, which attenuated OSA in this population (38).

So far, three studies have evaluated the impact of CS in individuals with preserved renal function and reported reductions in AHI from 24 to 36% (14–16). Redolfi et al. evaluated the impact of wearing CS for 1 week in patients with venous insufficiency (15), showing that despite the same body weight in both control group and after CS period, there was a 150-ml reduction in leg fluid volume after wearing CS and, more importantly, a 600-ml reduction in TBW. Therefore, less fluid was available to move rostrally, which reduced NC increase. Neither trunk volumes nor diuresis was evaluated. Nevertheless, in light of their and our findings, we can postulate that CS may avoid translocation of fluid from the trunk to the legs during the day, which increases diurnal diuresis, thus reducing TBW. In our study, patients were oliguric or anuric, and TBW was not different at baseline or after wearing CS. Still, there was a marked difference regarding fluid arrangement, once CS prevented fluid from shifting from the trunk to the legs during the day and also led to an increased intracellular TW.

To explain why intracellular TW increases after wearing CS, it is important to understand how fluid shifts according to body position. In healthy individuals, moving from recumbent to standing position, extracellular volume increases while intracellular water decreases (39). While standing up, gravitational forces exert pressure on the liquid column of the body and, consequently, hydrostatic pressure increases in the lower limbs (40). The capillary pressure in the legs when standing (90–120 cmH₂O) exceeds the pressure needed for moving fluid to interstitial compartment (15–20 cmH₂O) (41, 42). In addition, intravascular volume decreases 300–400 ml while standing (43, 44). Therefore, fluid shifts from intravascular and intracellular spaces, increasing extracellular volume in the lower limbs.

In this study, wearing CS prevented fluid movement from the trunk to the lower limbs and also from the intracellular to the extracellular space of the trunk. Therefore, the upward pressure exerted by CS partially counteracted gravity forces, avoiding leg swelling and keeping more fluid in the intracellular space, which is less likely to move freely to other regions of the body. Consequently, there was a reduction of the total amount of fluid reaching the neck, which partially prevented the edema buildup in upper airway during the night. This is the most likely mechanism associated with the reduction of NC and AHI while wearing CS.

Besides fluid distribution with CS, another novelty of this study was the analysis of body fluid kinetics during CPAP titration exam. We demonstrated that positive pressure exerted on upper airways had no effect on fluid shift from the legs. Still, NC reduced significantly during the night, suggesting an important reduction in upper airways edema. Therefore, regardless of excess body water volume or nocturnal fluid shift, CPAP therapy effectively abolishes obstructive events by exerting local pressure that moves fluid away from the neck, thus avoiding the upstream volume to reach upper airways, and as shown in Figures 2 and 3, CPAP performed significantly better than CS for both reduction of AHI and overnight change in NC. Finally, the mechanisms involved in the reduction of both AHI and NC by CS and CPAP are completely different: the former interferes in fluid redistribution during the night, while the latter prevents fluid reaching the neck.

This study has strengths and limitations. The main strengths were the randomized and crossover study design, as well as the use of multifrequency BIS, which allowed us to precisely estimate total, extracellular, and intracellular fluid volumes in each body segment. We also provided a novelty by demonstrating that CS can alleviate OSA in patients on HD. In addition, new insights regarding the physiology involved in the overnight fluid shift contributed to elucidate the mechanisms of action of CS and CPAP therapy. One of the limitations of this study was the small sample size, which makes the study prone for both type I and II errors. Another limitation was the non-precise assessment of residual diuresis volume, although most patients were anuric. Even though the referred urinary output was estimated (and not measured) to be lower than 200 ml per episode, this information is likely to be accurate since there is a trend toward a lower bladder reservoir in patients with long dialysis vintage (45). The NC measurement is also prone to error, although there is no gold standard method to access it. Finally, this study is a short-term clinical trial. Our main focuses were the mechanisms of fluid kinetics during the night in three different situations. To ascertain the clinical outcomes of both CPAP and CS therapies in this population, a longer treatment period for each of these treatments would be necessary. Therefore, further studies are required to evaluate adherence and long-term effect of CS and CPAP on OSA in HD patients.

In conclusion, CS and short-term CPAP can alleviate OSA in patients on HD by two different predominant mechanisms. While CS exert pressure on the lower limbs, maintaining fluid in the intracellular compartment of the trunk, CPAP therapy had no action on fluid shift, yet prevented it from reaching the neck. Even though the effect of CS is inferior to the standard CPAP treatment, CS may be an alternative and feasible treatment to attenuate OSA in HD patients. Moreover, the association of these two treatments can be useful to treat OSA and requires further investigation on ESRD patients.

This study was carried out in accordance with the recommendations of CAPPesq (Comissão de Ética para Análises de Projetos de Pesquisa) with written informed consent from all subjects. All subjects gave written informed consent in accordance with the Declaration of Helsinki. The protocol was approved by CAPPesq.

Conception and design; analysis: BS and RE. Data acquisition: BS and RS. Interpretation of data; drafting the work or revising it critically for important intellectual content: BS, LD, FC, and RE. Final approval of the version to be published; BS, RS, LD, FC, and RE.

Financial arrangements: FAPESP (Fundação de Amparo à Pesquisa do Estado de São Paulo) provided financial support for polysomnography exams, compression stockings, and CPAP device. Non-financial arrangements: nothing to declare.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.