Systematic review and meta-analysis of sevelamer versus calcium-based binders for the treatment of hyperphosphatemia in chronic kidney disease

Introduction and Aims: A number of recently published studies have evaluated the efficacy and safety of the non-calcium-based binder (CBB) sevelamer versus CBBs. The aim of this study was to conduct an updated meta-analysis of sevelamer versus CBBs for the treatment of hyperphosphatemia in adult patients with CKD stages 3 to 5D.

Methods: MEDLINE and the Cochrane Central Register of Controlled Trials were searched to identify RCTs or quasi-RCTs comparing sevelamer with CBBs. Patient-level outcomes assessed included all-cause mortality, cardiovascular (CV) events and mortality, hospitalization, adverse effects and quality of life (QOL). Intermediate outcomes included vascular calcification (VC) and changes to bone. Biochemical outcomes included values of serum phosphorus and calcium, hypercalcemia, the calcium-phosphorus (Ca × P) product, parathyroid hormone (PTH), alkaline phosphatase (ALP), bicarbonate, 1,25-dihydroxyvitamin D, fetuin A and lipids. We followed Cochrane guidelines for the conduct and reporting of this systematic review.

Results: A total of 24 studies (4,330 participants) were included. The majority (87.5%) enrolled dialysis patients while 12.5% enrolled stage 3 to 5 CKD-ND patients. For patients treated with sevelamer vs. CBBs there were significant reductions in all-cause mortality (risk ratio [RR] 0.54, 95% confidence interval [CI] 0.32 to 0.93), serum cholesterol (mean difference [MD] -20.22 mg/dL, 95% CI -25.95 to -14.50), LDL-C (MD -21.64, 95% CI -27.88 to -15.41), hypercalcemia (RR 0.30, 95% CI 0.19 to 0.48), calcium (MD -0.43 mg/dL, 95% CI -0.64 to -0.22) and bicarbonate (MD -1.51 mEq/l, 95% CI -2.34 to -0.69). No significant between group differences were observed for CV mortality, serum phosphorus, ALP, PTH, the Ca × P product, HDL-C, 1,25-dihydroxyvitamin D, or the incidence of nausea/vomiting, constipation, diarrhea, or abdominal bloating. The incidence of combined gastrointestinal adverse events was higher with sevelamer vs. CBBs (RR 1.42, 95% CI 0.97 to 2.08). There were insufficient data for a formal meta-analysis of CV events, hospitalization, VC, bone outcomes, or QOL. A subgroup analysis of patients with CKD-5D revealed similar results to the primary analysis, except for all-cause mortality, which showed a non statistically significant 46% overall reduction (RR 0.54; 95% CI 0.29 to 1.01).Additional subgroup analyses were performed to investigate potential sources of heterogeneity between studies included in the analysis of all-cause mortality. In the analysis that included only studies where all-cause mortality was a pre-defined outcome, the risk of all-cause mortality for sevelamer vs. CBBs was no longer statistically significant (RR 0.53, 95% CI 0.26 to 1.09). The results for all-cause mortality were also sensitive to study duration, number of study participants, dialysis duration, and the presence of VC.

Conclusions: For patients with CKD stages 3-5D, sevelamer vs. CBBs was associated with a 46% reduction in all-cause mortality that reached statistical significance. Sevelamer vs. CBBs was also associated with significant reductions in serum calcium, the risk of hypercalcemia, total and LDL-cholesterol and bicarbonate. A subgroup analysis that included only patients with CKD-5D showed a 46% overall reduction in mortality that did not reach statistical significance. Insufficient data were available to establish the comparative efficacy of sevelamer vs. CBBs for CV mortality, CV events, hospitalization, VC, bone outcomes, 1,25-dihydroxyvitamin D, or QOL.