Nocturnal APD vs. CAPD in Non-diabetic ESKD

Randomization and Masking Eligible participants were randomly assigned in a 1:1 ratio to one of two treatment groups using a centralized block randomization system, stratified by study center. Group A consisted of 12 weeks of nocturnal APD followed by 12 weeks of CAPD; Group B consisted of the reverse order. The randomization code was generated by an independent statistical center.

The trial was open-label with respect to treatment assignment, as the nature of the interventions precluded blinding of participants and treating clinicians. However, outcome adjudicators and data analysts were masked to treatment allocation throughout the analysis period. All outcome data were reviewed by an independent clinical events committee whose members were unaware of treatment assignment.

Trial Procedures The trial comprised a screening period, two 12-week treatment periods, and a follow-up period. After screening and baseline assessments, participants initiated their assigned treatment sequence. Scheduled outpatient visits occurred every 4 weeks (±5 days) for safety monitoring and routine laboratory testing. Comprehensive outcome assessments were performed at week 12 (end of the first period) and week 24 (end of the second period). Additional unscheduled visits were arranged as needed in the event of adverse events or clinical changes.

Nocturnal APD was delivered using automated cyclers (Dongze Medical, Fuzhou, China). CAPD was performed with three to five manual exchanges per day. All participants used standard peritoneal dialysis solutions (Huaren Pharmaceutical, Qingdao, China), including dextrose-based solutions (1.5%, 2.5%, and 4.25% glucose concentrations). The choice of dialysate concentration and dwell volume was determined by the treating physician according to clinical needs, with the goal of achieving adequate dialysis (weekly total Kt/V ≥ 1.7) and maintaining fluid balance. Icodextrin-based solutions were not used during the trial.

Participants recorded daily dialysate inflow and outflow volumes in patient logs, from which net ultrafiltration was calculated. Adherence to the prescribed treatment regimen was monitored through patient logs and cycler download data at each scheduled visit.

Primary Outcome The primary outcome was the steady-state weekly total Kt/V. For the first treatment period, it was defined as the arithmetic mean of weekly total Kt/V values measured at weeks 8 and 12; for the second treatment period, as the arithmetic mean of values measured at weeks 20 and 24. If one measurement was missing within a period, the single available value was used as the steady-state value for that period.

Weekly total Kt/V was calculated as (24-hour dialysate urea clearance + 24-hour urine urea clearance) / (serum urea nitrogen concentration × urea distribution volume) × number of dialysis days per week. Urea distribution volume was estimated using the Watson formula.

Secondary Outcomes Secondary outcomes were assessed at baseline, week 12, and week 24, with the exception of sleep parameters (assessed at weeks 12 and 24) and ultrafiltration (recorded daily, with analysis based on the mean of the final 7 days of each treatment period).

Sleep architecture was measured using a non-contact vital signs monitoring device, providing objective parameters including total sleep time, awakening duration, number of body movements, and apnea episodes.

Health-related quality of life was assessed using the Kidney Disease Quality of Life-36 (KDQOL-36) instrument, which yields five subscale scores (symptom/problems, effects of kidney disease on daily life, burden of kidney disease, physical component summary, and mental component summary), each transformed to a 0-to-100 scale, with higher scores indicating better quality of life.

Social functioning was assessed using the World Health Organization Social Disability Screening Schedule (SDSS), a 10-item instrument validated in Chinese populations, with a total score ranging from 0 to 20, where higher scores indicate greater social disability.

Twenty-four-hour ambulatory blood pressure was recorded using validated monitors. Daytime (08:00 to 22:00) and nighttime (22:00 to 08:00) periods were defined according to European Society of Hypertension guidelines. For each participant, mean systolic blood pressure (SBP), SBP variability (standard deviation), and the nocturnal SBP reduction rate (percentage decline from daytime to nighttime) were calculated. The nocturnal SBP dipping pattern was categorized as dippers (≥10% reduction), non-dippers (0% to <10%), or reverse dippers (<0%).

Mean daily ultrafiltration volume was calculated as the average of daily net ultrafiltration (total effluent volume minus total infused volume) during the final 7 days of each 12-week treatment period.

Detailed descriptions of the KDQOL-36 subscales, SDSS domains, and blood pressure parameters are provided in the Supplementary Appendix.

Safety Assessment Safety was assessed by monitoring adverse events and serious adverse events throughout the study period. Laboratory measurements were performed at baseline and at scheduled visits. The safety population included all participants who received at least one dose of the study treatment.

Statistical Analysis Baseline characteristics were summarized for the full analysis set by randomized sequence group (Group A: nocturnal APD first then CAPD; Group B: CAPD first then nocturnal APD). Continuous variables were reported as mean (SD) or median (IQR), and categorical variables as number (%). No formal statistical comparisons between sequence groups were performed.

Primary analysis was conducted in the primary efficacy set, which included participants who completed both treatment periods and had evaluable steady-state weekly total Kt/V measurements for both periods (defined as the arithmetic mean of values measured at weeks 8 and 12 for the first period and at weeks 20 and 24 for the second period; if one measurement was missing, the single available value was used). The primary hypothesis was tested using a paired t-test to compare the within-participant difference in steady-state weekly total Kt/V between nocturnal APD and CAPD. Non-inferiority was established if the lower bound of the two-sided 95% confidence interval for the mean difference (nocturnal APD minus CAPD) was greater than the prespecified non-inferiority margin of -0.1, corresponding to a one-sided alpha of 0.025.

Sample size was calculated based on a two-period crossover non-inferiority design. Assuming a within-participant standard deviation of the difference of 0.51, a non-inferiority margin of -0.1, and a true treatment difference of 0, 110 participants were required to achieve 90% power at a one-sided significance level of 0.025. Accounting for an anticipated 10% dropout rate, we planned to enroll 124 participants, with 62 assigned to each sequence.

Sensitivity analyses were prespecified to assess the robustness of the primary findings. These included: (1) a per-protocol analysis that excluded participants with major protocol violations; (2) an analysis restricted to participants with complete weekly total Kt/V measurements at all time points (weeks 8, 12, 20, and 24); and (3) a linear mixed-effects model with fixed effects for baseline weekly total Kt/V, center, sequence, period, prior treatment, and treatment, and a random intercept for participant, to assess potential carryover effects.

Secondary outcomes were analyzed using paired t-tests for continuous variables and chi-square or Fisher's exact tests for categorical variables, based on measurements obtained at week 12 and week 24, with two-sided significance tests at alpha = 0.05. No adjustment for multiple comparisons was made; these analyses were considered exploratory.

All analyses were performed using R software, version 4.1.1 (R Foundation for Statistical Computing).