Background and objective: There is concern regarding long-term excess calcium intake in end-stage renal disease populations. Because calcium carbonate is an over-the-counter (OTC) medication, few studies have been able to track its use. The Veterans Health Administration (VA) tracks national pharmacy data for both OTC and prescription drugs. We thus compared survival in incident dialysis patients on sevelamer and calcium carbonate phosphate binders.
Methods: This was a retrospective cohort study of veterans initiating haemodialysis using existing VA databases. Patients were divided into calcium only (n = 769) and sevelamer only (n = 608) groups, then followed for up to 2 years until FY03 end. Survival was modelled using Cox regression adjusting for age, gender, race, marital status, service-connected disability, region, diabetes, hypertension and Charlson index. Stability of findings was examined using propensity score analysis.
Results: Sevelamer only vs. calcium only subjects were younger (respective mean ages 59.6 and 63.0, P < 0.001) with fewer comorbidities (Charlson index 3.8 and 4.5, P < 0.001). By study end, 24% of sevelamer and 30% of calcium subjects had died. Comparing sevelamer to calcium, the unadjusted hazard ratio for death was 0.62 (95% CI 0.50-0.76); the adjusted hazard ratio was 0.67 (CI 0.54-0.84). Propensity score analysis revealed similar results, with a hazard ratio of 0.65 (CI 0.54-0.80).
Conclusions: In a national incident dialysis cohort, sevelamer treatment was associated with improved survival compared with calcium carbonate. Further research should investigate whether the worse survival with calcium is a long-term consequence of increased calcium accumulation.