Abstract
OBJECTIVES: To investigate the association between longitudinal trajectories of metabolic risk clusters and the risk of progression to end-stage kidney disease (ESKD) and major adverse kidney events (MAKEs) in patients with chronic kidney disease (CKD). DESIGN: Prospective registry-based cohort study. SETTING: Secondary and tertiary care settings in Taiwan, using data from a multidisciplinary pre-ESKD care programme. PARTICIPANTS: A total of 1494 adult patients with CKD stages 3b-5 enrolled in a structured pre-ESKD care programme. PRIMARY AND SECONDARY OUTCOME MEASURES: Time to initiation of dialysis (primary outcome) and time to MAKE, defined as a composite of dialysis initiation or all-cause mortality (secondary outcome). Group-based multitrajectory modelling was used to categorise longitudinal trajectories of metabolic risk clusters, including systolic blood pressure, fasting blood glucose and low-density lipoprotein (LDL) cholesterol. RESULTS: Four trajectory groups were identified: Group I had controlled blood pressure and glucose but elevated LDL (dialysis incidence: 19.5 per 1000 person-years); Group II had borderline-high blood pressure and elevated glucose (33.6 per 1000 person-years); Group III had controlled glucose and low LDL but borderline-high blood pressure (38.8 per 1000 person-years) and Group IV had controlled glucose but elevated blood pressure and LDL (46.7 per 1000 person-years). Compared with the other groups, Group I exhibited significantly longer dialysis-free and MAKE-free survival (log-rank test, p<0.001). With covariates adjusted for, Groups II (HR 1.02, 95% CI 0.52 to 2.00) and III (HR 1.46, 95% CI 0.85 to 2.52) showed non-significant elevations in dialysis risk, whereas Group IV (HR 1.96, 95% CI 1.01 to 3.78) exhibited a significantly higher risk compared with Group I. Similar patterns were observed for MAKE outcome, confirming the association between trajectory groups and adverse kidney events. CONCLUSION: Longitudinal trajectories of metabolic risk cluster are associated with differential risks of CKD progression to ESKD and death. Our findings provide valuable insights into the monitoring of metabolic risk profiles over time in patients with CKD.