The progression of coronary artery calcification in predialysis patients on calcium carbonate or sevelamer

Russo D,   Miranda I,  Ruocco C,  Battaglia Y,  Buonanno E,  Manzi S,   Russo  L,  Scafarto A  and  Andreucci
Kidney Intl. 72:1255-1261, 2007

Study hypothesis

In dialysis patients the prevalence of coronary artery calcification (CAC) is greater than that seen in individuals with normal renal function, even those with established coronary artery disease; the severity of CAC at entry to hemodialysis is a strong predictor of long-term survival. The progression of CAC with time on dialysis is believed to be strongly linked to hyperphosphatemia, and further accelerated by the use of calcium-based dietary phosphate binders (CBPB’s). Thus, limiting the use of CBPB’s by utilising  low phosphate diets, and/or  non-calcium-based phosphate binders may limit the development of CAC in advanced renal failure, before dialysis is initiated.

Study design and study population

100 subjects were selected from an out-patient population of  chronic renal failure patients ( Stage 3-5). Inclusion criteria: on a “constant low dietary phosphate intake”, with stable biochemical markers of CKD-MBD. Exclusion criteria included patients with diabetes, on statins, with any history of coronary artery disease.

Intervention or observation

90 subjects,( average age 55 yr.), consented to enter the study, and were randomised to receive either calcium carbonate (2gm/d), sevelamer (1.6 gm/d)  - both interventions in fixed doses, - or no phosphate binder (Controls). All continued with a low phosphate diet (apparently < 1000 mg/d). 6 subjects were lost to follow-up, and the analysis was performed on the completers (CaC03, n=28; sevelamer, n=27; diet alone, n=29). The study was intended as a 2 yr. trial, but the observation period was reported as 24 ± 4.2 (SD) mon.

CAC was assessed by spiral CT, at the baseline and end of study. Coronary artery scores were summed to obtain a Total Calcification Score (TCS).

Primary endpoint, secondary endpoint

Primary endpoint was either an absolute change in TCS between baseline and 2 yr., or the annualised change in TCS.

Biochemical and GFR measures were made at intervals and reported.

Little information was reported on statistical methods. No methodology was provided about monitoring dietary phosphate intakes over time.


Mean GFR was about 26 ml/min in the 2 treatment arms, and 33 ml/min in the controls; GFR remained essentially stable over time. Mean serum Pi was < 1.5 mmol/L and PTH was < 15 pmol/L at baseline. Mean baseline dietary phosphate intakes were respectively  694, 690, and 682 mg. for the CaC03, sevelamer and Control groups; small changes in intake were observed over the course of study (-5%, +14%, +15% respectively). Baseline urine phosphate excretion was 496, 490 and 367 mg./d for the CaC03, sevelamer and Control groups; urine phosphate declined significantly by 17% and 16% in the CaC03 and sevelamer groups, but increased significantly by 40% in the Control group.

The final TCS was significantly greater than the initial TCS in Controls (369 ± 115 (mean ± SEM.) vs 547 ± 175; P <0.001) and in the CaC03 group (340 ± 38 vs 473 ± 69; P <0.001); in contrast, the final TCS was not significantly different from the initial TCS among the sevelamer group (415 ± 153 vs 453 ± 127; NS). Annualized progression of TCS was 205 ± 82 in Control, 178 ± 40 in the CaC03, and36 ± 32 in the sevelamer groups. No significant between-group change was found in either the mean absolute or the annualized absolute values of TCS.

Methodological assessment

This was a randomised, controlled, (but open-label) trial, with few drop-outs, and apparently blinded assessment of the primary outcome – the KDIGO working group rated the quality of the evidence as “B”. However the exclusion of diabetics makes generalisability somewhat limited. The cohort size was small and the variation in TCS large; thus there were no significant differences in CAC between the 3 groups, although it can be concluded that progression in CAC was only minimal in the group treated with sevelamer. No evidence was submitted on adherence to diet and assigned treatment – which could have been sub-optimal.

Impact on practice

CKD patients with 25-35% residual renal function have evidence for CAC even with minimal elevations of serum Pi, and this CAC will increase significantly despite modest dietary Pi restriction (<1000 mg./d). CaC03 had little impact on urine phosphate excretion, and there was no suggestion that its use retarded the progression of CAC, or improved 20 hyperparathyroidism. This does not provide support that early intervention with CBPB’s in the CKD clinic will prevent later progression of either CKD-MBD or cardiovascular outcomes.

The trial does provide evidence to support the use of sevelamer in the pre-dialysis phase of ESRD, if the goal is simply to retard progression of CAC. However it was performed in a relatively young population, who had no pre-existing cardio- or cerebro-vascular disease, and were not diabetic.

- Reviewed by Reviewed by Dr. Anthony Hodsman

Return to bone and mineral resource home