Am J Kidney Dis. 2009 Oct;54(4):619-37. Epub 2009 Aug 18
To determine whether newer agents deliver improved biochemical and patient-level outcomes, with particular reference to musculoskeletal and cardiovascular cardiovascular morbidity, hospitalization, and mortality
Study design and study population:
Meta analysis of randomized controlled trials (RCTs) and quasi-RCTs (trials that use a method of allocating participants to different forms of care that are not truly random, such as allocation by date of birth, alternate medical records, day of the week, or other forms of alternation) in adult patients (age >18 years) with CKD stages 3 to 5 and 5D (dialysis). The quality of RCTs was assessed by using a checklist that included allocation concealment; blinding of participants, investigators, outcome assessors, and data analysts; use of intention-to-treat analyses; and completeness of follow-up.
Intervention or observation:
Phosphate binders in people with CKD, alone or in combination with other nonrandomized co interventions (eg, vitamin D compounds).
Primary end point, secondary end point:
Outcomes when reported: all cause mortality, cardiovascular mortality, nonfatal cardiovascular events, vascular calcification by any imaging modality, end-of-treatment PTH concentration (intact PTH and PTH 1-84 as reported in the studies), serum calcium , serum phosphorus, serum calcium-phosphorus product, alkaline phosphatase, serum bicarbonate, total cholesterol, bone mineral density, bone mineral content, bone histomorphometry, occurrence of hypercalcemia, and treatment-related toxicity.
Forty trials with 6406 subjects were included, but only 3301 subjects were in trials of more than 6 months duration. There was no significant reduction in all cause mortality or hospitalization when comparing different binders. Hypercalcemia was reduced with Sevelamer and Lanthanum. There was a significant increase in GI side effects with Sevelamer. Serum PO4 and iPTH levels were higher at end of study with Sevelamer, but calcium levels were lower with both Sevelamer and Lanthanum.
Effect of Sevelamer compared with calcium salts on all-cause mortality in people with chronic kidney disease.
Methodological assessment:Threats to validity of Meta analysis include publication bias, study heterogeneity, trends over time that alter treatment effects, and poor quality of included studies.
Problems of Internal Validity in RCTs include attrition bias or loss to follow-up, especially if differential loss and loss is greater than 20 %. The DCOR study which accounts for 46 % of the weight in the meta-analysis had significant problems with loss to follow up and would be considered low quality. The authors acknowledge this in their discussion, and state that excluding this study resulted in a similar, but smaller risk of all cause mortality.
Impact on practice:
Given the current funding models in place this analysis will not result in changes in prescribing. This analysis is a good example of how therapies become standard of care (Calcium binders) without rigorous evaluation, and how newer agents face significant challenges as the levels of evidence required have changed significantly. The recurring mantra of “more studies are needed” remains undiminished.
- Reviewed by Dr. Steven Soroka