Outcomes associated with phosphorus binders in men with non-dialysis-dependent CKD

Abstract

Background: Phosphorus binders are used to treat hyperphosphatemia in maintenance dialysis patients, in whom the use of these medications has been associated with lower mortality in some observational studies. It is not clear whether similar benefits can be seen in patients with non-dialysis-dependent chronic kidney disease (CKD).

Study design: Historical cohort.

Setting & participants: 1,188 men with moderate and advanced non-dialysis-dependent CKD at a single medical center.

Predictor: Administration of any phosphorus binder.

Outcomes & measurements: We examined associations of any phosphorus-binder administration with all-cause mortality and the slopes of estimated glomerular filtration rate using time-varying Cox models and mixed-effects models. Associations also were examined in intention-to-treat analyses and in 133 patient-pairs matched according to propensity scores.

Results: 344 patients were treated with a phosphorus binder; 658 patients died (mortality rate, 141 deaths/1,000 patient-years; 95% CI, 131-153) during a median follow-up of 3.1 years. Treatment with phosphorus binders was associated with significantly lower mortality (adjusted HR, 0.61; 95% CI, 0.45-0.81; P < 0.001). Results were similar when exposure was modeled in intention-to-treat analyses and examining propensity-matched patients. Phosphorus-binder use was not associated with significant changes in kidney function loss.

Limitations: Results may not apply to all patients with non-dialysis-dependent CKD.

Conclusions: Administration of phosphorus binders is associated with lower mortality in men with moderate and advanced non-dialysis-dependent CKD. Clinical trials are needed to determine the risks and benefits of phosphorus-binder use in this patient population.

Figure 1

Distribution of patients according to their treatment or non-treatment with phosphorus binders and according to the types of phosphorus binders administered.

Figure 2

Survival curves in 133 patients administered any type of phosphorus binder versus 133 patients on no phosphorus binder, matched by their propensity scores reflecting the likelihood of binder administration.

Figure 3

Hazard ratio (95% confidence intervals) of overall all-cause mortality in various subgroups of patients treated with a phosphorus binder vs. not treated with any binder. All results are adjusted for age, race, Charlson comorbidity index, diabetes mellitus, cardiovascular disease, blood pressure, body mass index, smoking status, enrollment period, the use of calcitriol, estimated GFR, serum albumin, bicarbonate, calcium, phosphorus, alkaline phosphatase, blood hemoglobin, white blood cell count, percentage of lymphocytes and 24 hour urine protein.

Figure 4

Mean eGFR (95% confidence intervals) in patients who were and who were not treated with binders, assessed at monthly intervals during follow-up.

Figure 5

Estimated mean slopes of eGFR vs. time in patients who were never treated with phosphorus binders (solid line) and in patients treated with phosphorus binders (dashed line). The arrow indicates the change in slope associated with the initiation of phosphorus binders in treated patients. Slopes were estimated from unadjusted mixed effect models in the respective subgroups.

Figure 6

Differences in post-treatment vs. pre-treatment eGFR slopes (ml/min/1.73m2 per year) associated with binder therapy in various subgroups of patients treated with a phosphorus binder. 95% confidence intervals are indicated (horizontal lines). Results are adjusted for age, race, Charlson comorbidity index, diabetes mellitus, cardiovascular disease, blood pressure, body mass index, smoking status, enrollment period, the use of calcitriol, serum albumin, bicarbonate, calcium, phosphorus, alkaline phosphatase, blood hemoglobin, white blood cell count, percentage of lymphocytes and 24 hour urine protein and the presence or absence of a death or end stage renal disease event.