Authors' objectives:

The objectives were to perform a systematic review of the efficacy and harm of sevelamer, and to conduct a primary economic evaluation and budget impact analysis of its use when compared with calcium-based phosphate binders, in patients with end-stage renal disease (ESRD), who are on dialysis.

Authors' results and conclusions: We identified 10 primary publications of RCTs with a total of 3,025 participants eligible for efficacy analysis. Eight RCTs reported serum phosphate and calcium levels, and four reported all-cause mortality. In analyses pooling, the eight studies reporting on serum phosphate and calcium levels (2,445 participants), the overall control of phosphate was better with calcium-based phosphate binders by 0.09 mmol/L [95% confidence interval (CI) 0.02 to 0.16]. On-treatment calcium-phosphate product was not significantly different in patients receiving calcium-based phosphate binders [weighted mean difference (WMD) 0.09 mmol2/L2, -0.07 to 0.25]. The overall WMD in serum calcium was lower with sevelamer therapy by 0.10 mmol/L (-0.12 to -0.07). In the four RCTs that reported all-cause mortality (2,302 participants), the duration of follow-up varied from two to 45 months. The overall risk difference for all cause mortality in these four trials was not significantly different (-1%, 95% CI -4 to 2). The finding of improved survival among patients >65 years old, and those treated for >2 years in the DCOR study secondary analyses requires confirmation in a subsequent randomized trial. We also identified 28 publications of prospective trials (16 RCTs and 12 single-arm trials) with a total of 3,983 participants eligible for harm analysis. In three RCTs that reported serious adverse events (SAEs) (2,185 participants), there was a non-significant lower risk of SAEs in patients receiving calcium-based phosphate binders (13% lower, 95% CI -2 to 29). We found no studies examining the efficacy of sevelamer specifically in ESRD patients with concomitant hyperphosphatemia and hypercalcemia=the type of patient for whom sevelamer is funded in many provinces. There are no data indicating that such a strategy will reduce mortality or morbidity in these patients.

Authors' recommendations: Compared with calcium-based phosphate binders, there was no evidence that sevelamer reduced all-cause mortality, cardiovascular mortality, hospitalization, or the frequency of symptomatic bone disease, and no evidence that sevelamer improved quality of life. Based on available evidence, the use of sevelamer, in comparison with calcium, in patients with ESRD who are on dialysis, is associated with similar to slightly higher phosphate levels, similar calcium phosphate product levels, and slightly lower serum calcium levels. The only study that investigated whether sevelamer reduces mortality and morbidity compared with calcium had methodological limitations, including lack of blinding, and substantial loss to follow-up. Overall mortality was not significantly different between sevelamer and calcium-based phosphate binders in this study, and the methodological concerns raise questions about the validity of the purported benefit in the pre-specified group of patients aged = or >65 years. The results of these secondary analyses require confirmation in future randomized studies. Based on the available evidence, the use of sevelamer results in similar to slightly higher phosphate levels; similar calcium phosphate product levels; and slightly lower calcium levels, compared with calcium-based phosphate binders. The risk of SAEs was non-significantly lower in patients receiving calcium-based phosphate binders (13% lower, 95% CI -2 to 29), compared with those receiving sevelamer. The routine use of sevelamer in patients with ESRD is not supported by available data. Even assuming that sevelamer is associated with a clinical benefit, as we did in our primary economic evaluation, the cost per QALY gained for treating all patients with ESRD, with sevelamer compared to calcium-based phosphate binders, may be perceived as excessive depending on what a decision maker is willing to pay for a QALY. Although lower, the cost per QALY gained for treating easily identifiable patient groups (i.e., age = or >65 years) exceeds $100,000 per QALY, and is based on considerable clinical uncertainty. If the effectiveness of sevelamer were confirmed in this group in a subsequent study, then its use in patients >65 years old should be reconsidered. Many provinces fund sevelamer for patients with ESRD, with specific abnormalities in mineral metabolism (most commonly the co-existence of hyperphosphatemia and hypercalcemia). There are no data to support that such a strategy will reduce mortality or morbidity in these patients, and the cost effectiveness of this approach is unknown. Because calcium-based phosphate binders reduce serum phosphate levels to the same extent as sevelamer, it is unlikely that sevelamer will be more cost effective in this patient group.

Authors' methods: Systematic review

Project Status: Completed

URL for project: http://www.cadth.ca/media/pdf/HTA_349_sevelamer_tr_e.pdf

Year Published: 2006

English language abstract: An English language summary is available

Publication Type: Not Assigned

Country: Canada

Organisation Name: Canadian Agency for Drugs and Technologies in Health

Contact Address: 600-865 Carling Avenue, Ottawa, ON K1S 5S8 Canada. Tel: +1 613 226 2553; Fax: +1 613 226 5392;

Contact Name: requests@cadth.ca

Contact Email: requests@cadth.ca

Copyright: Canadian Agency for Drugs and Technologies in Health (CADTH)