Systematic Review and Meta-Analyses of the Effects of Phosphate-Lowering Agents in Nondialysis CKD

Abstract

Background: Benefits of phosphate-lowering interventions on clinical outcomes in patients with CKD are unclear; systematic reviews have predominantly involved patients on dialysis. This study aimed to summarize evidence from randomized controlled trials (RCTs) concerning benefits and risks of noncalcium-based phosphate-lowering treatment in nondialysis CKD.

Methods: We conducted a systematic review and meta-analyses of RCTs involving noncalcium-based phosphate-lowering therapy compared with placebo, calcium-based binders, or no study medication, in adults with CKD not on dialysis or post-transplant. RCTs had ≥3 months follow-up and outcomes included biomarkers of mineral metabolism, cardiovascular parameters, and adverse events. Outcomes were meta-analyzed using the Sidik-Jonkman method for random effects. Unstandardized mean differences were used as effect sizes for continuous outcomes with common measurement units and Hedge's g standardized mean differences (SMD) otherwise. Odds ratios were used for binary outcomes. Cochrane risk of bias and GRADE assessment determined the certainty of evidence.

Results: In total, 20 trials involving 2498 participants (median sample size 120, median follow-up 9 months) were eligible for inclusion. Overall, risk of bias was low. Compared with placebo, noncalcium-based phosphate binders reduced serum phosphate (12 trials, weighted mean difference -0.37; 95% CI, -0.58 to -0.15 mg/dl, low certainty evidence) and urinary phosphate excretion (eight trials, SMD -0.61; 95% CI, -0.90 to -0.31, low certainty evidence), but resulted in increased constipation (nine trials, log odds ratio [OR] 0.93; 95% CI, 0.02 to 1.83, low certainty evidence) and greater vascular calcification score (three trials, SMD, 0.47; 95% CI, 0.17 to 0.77, very low certainty evidence). Data for effects of phosphate-lowering therapy on cardiovascular events (log OR, 0.51; 95% CI, -0.51 to 1.17) and death were scant.

Conclusions: Noncalcium-based phosphate-lowering therapy reduced serum phosphate and urinary phosphate excretion, but there was an unclear effect on clinical outcomes and intermediate cardiovascular end points. Adequately powered RCTs are required to evaluate benefits and risks of phosphate-lowering therapy on patient-centered outcomes.

Keywords: cardiovascular disease; mineral metabolism; phosphate; phosphate binders.

Figure 1.

Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram showing selection of studies.

Figure 2.

Risk of bias assessment of the included studies according to the Cochrane Collaboration tool.

Figure 3.

Effect of noncalcium phosphate binders versus placebo on change in serum phosphate. Forest plot showing the effect of noncalcium phosphate binders compared with placebo on serum phosphate (mg/dl) at last measurement.

Figure 4.

Effect of noncalcium phosphate binders on change in urinary phosphate, PTH, c-terminal FGF23 (cFGF23), and intact FGF23 (iFGF23). (A) Forest plot showing the effect of noncalcium phosphate binders compared with placebo on urinary phosphate at last measurement. (Units for urinary phosphate: * mg/day; # mmol/L; and mmol/day; % 24hr phosphate:creatinine mg/g; $spot urine phosphate:creatinine ratio mg/mg). (B) Forest plot showing the effect of noncalcium phosphate binders compared with placebo on PTH (pg/ml) at last measurement. (C) Forest plot showing the effect of noncalcium phosphate binders compared with placebo on cFGF23 (RU/ml) at last measurement. (D) Forest plot showing the effect of noncalcium phosphate binders compared with placebo on iFGF23 (pg/ml) at last measurement.

Figure 5.

Effect of noncalcium phosphate binders on change in PWV, in vascular calcification, on mortality, and on cardiovascular events. (A) Forest plot showing the effect of noncalcium phosphate binders compared with placebo on PWV (m/s) at last measurement. (B) Forest plot showing the effect of noncalcium phosphate binders compared with placebo on vascular calcification score at last measurement. (C) Forest plot showing the effect of noncalcium phosphate binders compared with placebo on mortality. (D) Forest plot showing the effect of noncalcium phosphate binders compared with placebo on nausea.

Figure 6.

Effect of noncalcium phosphate binders on constipation. Forest plot showing the effect of noncalcium phosphate binders compared with placebo on constipation.