Abstract
BACKGROUND: Hyperuricemia influences lipid metabolism, yet relationships between urate-lowering therapy with allopurinol, serum urate, and lipid levels in patients with chronic kidney disease remain underexplored. METHODS: This was a post-hoc analysis of 1970 participants of the CAN AIM to PREVENT who had pre-dialysis chronic kidney disease and were not receiving lipid-lowering therapy or febuxostat. Joint generalized structural equation modeling was used to investigate associations between allopurinol use (yes or no), serum urate [as a continuous or categorical variable (target if <6 mg/dl or high if ≥6 mg/dl)], and lipid levels [total cholesterol, low-density lipoprotein-cholesterol (LDL-C), high-density lipoprotein cholesterol, and triglycerides) assessed every 6 months for up to 3 years, along with time-to-event outcomes (death or initiation of renal replacement therapy), adjusting for demographic and clinical factors. Mediation analysis was used to determine allopurinol's direct and indirect effects (via urate) on lipid levels. RESULTS: Allopurinol use independently predicted lower total cholesterol (-7.94%, 95% CI: -12.13% to -3.54%, p < 0.001) and LDL-C [-13.84% (-21.14 to -5.87), p = 0.001]. Serum urate independently predicted a small increase in LDL-C [0.02% per mg/dl (0.009 to 0.03), p < 0.001]. Patients on allopurinol with target urate had lower LDL-C compared to those not on allopurinol with target urate [-4.46% (-8.25 to -0.50), p = 0.027] and those on allopurinol with high urate [-10.15% (-13.16 to -7.04), p < 0.001]. Mediation analysis showed that serum urate indirectly mediated only 24% of the effect of allopurinol on LDL-C. CONCLUSION: Allopurinol use predicted lower total and especially LDL cholesterol independently of serum urate in this cohort of patients with pre-dialysis chronic kidney disease. Future studies could investigate underlying mechanisms, evaluate clinical implications, and confirm these findings in this and other populations.