State-of-the-Art Management of Hyperphosphatemia in Patients With CKD: An NKF-KDOQI Controversies Perspective

Abstract

Phosphate binders are among the most common medications prescribed to patients with kidney failure receiving dialysis and are often used in advanced chronic kidney disease (CKD). In patients with CKD glomerular filtration rate category 3a (G3a) or worse, including those with kidney failure who are receiving dialysis, clinical practice guidelines suggest "lowering elevated phosphate levels towards the normal range" with possible strategies including dietary phosphate restriction or use of binders. Additionally, guidelines suggest restricting the use of oral elemental calcium often contained in phosphate binders. Nutrition guidelines in CKD suggest<800-1,000mg of calcium daily, whereas CKD bone and mineral disorder guidelines do not provide clear targets, but<1,500mg in maintenance dialysis patients has been previously recommended. Many different classes of phosphate binders are now available and clinical trials have not definitively demonstrated the superiority of any class of phosphate binders over another with regard to clinical outcomes. Use of phosphate binders contributes substantially to patients' pill burden and out-of-pocket costs, and many have side effects. This has led to uncertainty regarding the use and best choice of phosphate binders for patients with CKD or kidney failure. In this controversies perspective, we discuss the evidence base around binder use in CKD and kidney failure with a focus on comparisons of available binders.

Keywords: Phosphate binder; calcium; calcium acetate; chronic kidney disease (CKD); dialysis; ferric citrate; hemodialysis; hypercalcemia; kidney failure; lanthanum carbonate; medical nutrition therapy (MNT); mortality; non–calcium-based binder; phosphorus; sevelamer; sucroferric oxyhydroxide.

Figure 1.. Schematic Model Demonstrating Potentially Violated Assumptions of Current Surrogate Outcomes in Trials of Phosphate Binders.

Panel A depicts the idealized surrogate outcome in which the intervention influences the patient-centered or clinical outcome exclusively via the surrogates. In this scenario the intervention is correlated with the outcome but is also a cause of the patient-centered or clinical outcome. The intervention does not affect the outcome through any other pathways. Panel B depicts several more likely scenarios in the conceptualization of phosphate binder trials. First multiple confounding pathways (red arrows) likely exist between the surrogate outcomes and the patient-centered or clinical outcomes which may induce correlation without causation (i.e. the green arrow may not exist). Additionally, alternative pathways between the intervention and the patient-centered clinical outcome, both favorable (blue arrows) and unfavorable (purple arrows), may also exist such that the surrogate outcome cannot capture the full effect of the intervention on the patient-centered or clinical outcome. Other violations, not depicted here are also possible, including a variety of pathways between the surrogate outcomes and the patient-centered or clinical outcomes that could be a mix of favorable and unfavorable effects.

Figure 1.. Schematic Model Demonstrating Potentially Violated Assumptions of Current Surrogate Outcomes in Trials of Phosphate Binders.

Panel A depicts the idealized surrogate outcome in which the intervention influences the patient-centered or clinical outcome exclusively via the surrogates. In this scenario the intervention is correlated with the outcome but is also a cause of the patient-centered or clinical outcome. The intervention does not affect the outcome through any other pathways. Panel B depicts several more likely scenarios in the conceptualization of phosphate binder trials. First multiple confounding pathways (red arrows) likely exist between the surrogate outcomes and the patient-centered or clinical outcomes which may induce correlation without causation (i.e. the green arrow may not exist). Additionally, alternative pathways between the intervention and the patient-centered clinical outcome, both favorable (blue arrows) and unfavorable (purple arrows), may also exist such that the surrogate outcome cannot capture the full effect of the intervention on the patient-centered or clinical outcome. Other violations, not depicted here are also possible, including a variety of pathways between the surrogate outcomes and the patient-centered or clinical outcomes that could be a mix of favorable and unfavorable effects.