Switching from conventional to nocturnal hemodialysis can improve sleep apnea.OBJECTIVE
To determine whether nocturnal peritoneal dialysis (NPD) is associated with improvement in sleep apnea compared with continuous ambulatory peritoneal dialysis (CAPD).DESIGN AND INTERVENTION
Two studies were conducted. In the first, all consenting patients from the prevalent NPD population of a Hong Kong dialysis center underwent overnight polysomnography between September 2001 and January 2004. Overnight polysomnography was then performed in stable prevalent CAPD patients from the same center, who were matched to the NPD patients for demographic characteristics, comorbid conditions, peritoneal transport properties, BMI and dialysis dose adequacy. The choice of peritoneal dialysis modality was made by the patient. Inclusion criteria were clinical euvolemia, baseline serum sodium level 135-145 mmol/l and absence of peritonitis during the previous 3 months. The second study was carried out from January 2004 to April 2006. Patients who were receiving cycler-assisted NPD for 6-8 weeks while awaiting CAPD training underwent polysomnography, which was repeated after they had begun CAPD. Clinical euvolemia and baseline serum sodium level 135-145 mmol/l were prerequisites for inclusion. The goals of both NPD and CAPD were a weekly urea Kt/V of 1.8-2.1 and euvolemia.OUTCOME MEASURE
The primary endpoint was prevalence of sleep apnea, based on assessment of polysomnograms, and expressed as an apnea-hypopnea index (AHI; mean number of apnea and hypopnea episodes per hour of sleep).RESULTS
In the first study, there were no significant differences in baseline demographic and clinical attributes between the NPD and CAPD groups (n=23 for each group). Significantly more CAPD than NPD patients had sleep apnea (91% vs 52%; P=0.007) when an AHI ≥15 was used as the definition. Mean AHI was 31.6 in the NPD group and 50.9 in the CAPD group (P=0.025). Multivariate regression analysis revealed that peritoneal dialysis modality was independently associated with AHI (r2=0.351; P=0.017). The second study recruited 24 patients (mean age 50.8 years; 46% female). The prevalence of sleep apnea was significantly lower among NPD patients than among CAPD patients at AHI cut-off points of 15 (4.2% vs 33.3%; P=0.016), 10 (12.5% vs 41.7%; P=0.016) and 5 (16.7% vs 50%; P=0.008). Mean AHI was 3.4 during NPD and 14.0 during CAPD (P<0.001). Bioelectrical impedance analysis revealed that patients (n=15) had significantly lower total body water content during NPD than during CAPD (32.8 l vs 35.1 l; P=0.004). Total body water and hydration fraction declined significantly more during sleep while patients were on NPD than while they were on CAPD (declines of 2.81 l vs 1.34 l; P=0.015 and 5.49% vs 1.07%; P=0.005).CONCLUSION
In Chinese patients, NPD might provide better volume control during sleep than CAPD, and could thus be associated with lower prevalence of sleep apnea.