Combined Versus Multicomponent Training in People With Chronic Kidney Disease Stages 3a and 3b

Design: single-center, parallel-group (1:1:1) randomized controlled trial with blinded outcome assessors and a blinded statistical analyst. Randomization will be stratified by sex, in randomly sized blocks, with allocation concealed in sequentially numbered, sealed, opaque envelopes. Assessments will be performed at baseline and after the 12-week intervention across three visits (resting blood pressure, anamnesis, KDQOL-SF v1.3 and 30-second chair stand; fasting blood and urine collection, anthropometrics and DXA; handgrip strength, 8-Foot Up and Go, Mini-BESTest and maximal treadmill test with gas exchange analysis under physician supervision and 12-lead ECG monitoring). Falls will be recorded prospectively with monthly calendars in all groups. Statistical analysis: generalized estimating equations (GEE) with group, time, and group × time interaction, adjusted for sex, CKD stage (3a/3b), and type 2 diabetes; working correlation structure selected by QIC. Intention-to-treat is the primary analysis and per-protocol (≥75% adherence) the supportive analysis. Type I error across the five primary outcomes will be controlled by a pre-specified fixed-sequence (gatekeeping) procedure in the following order: VO2peak, resting systolic blood pressure, resting diastolic blood pressure, KDQOL-SF v1.3 physical component summary, and eGFR; each outcome is tested at alpha = 0.05 only if the previous one is significant. Pairwise comparisons will use GEE contrasts with Tukey adjustment. Effects will be reported as between-group differences in change with 95% confidence intervals (Mancl-DeRouen small-sample correction) and as standardized mean differences in change. Missing data will be addressed in a sensitivity analysis by multiple imputation by chained equations (20 datasets). Secondary outcomes are exploratory.