Association of single and serial measures of serum phosphorus with adverse outcomes in patients on peritoneal dialysis: results from the international PDOPPS.

Abstract

While high serum phosphorus levels have been related to adverse outcomes in hemodialysis patients, further investigation is warranted in persons receiving peritoneal dialysis (PD). Longitudinal data (2014-17) from the Peritoneal Dialysis Outcomes and Practice Patterns Study (PDOPPS), a prospective cohort study, were used to examine associations of serum phosphorus with all-cause mortality and major adverse cardiovascular events via Cox regression adjusted for confounders. Serum phosphorus levels were parameterized by four methods: (i) baseline serum phosphorus; (ii) mean 6-month serum phosphorus; (iii) number of months with serum phosphorus >4.5mg/dL; and (iv) mean area-under-the-curve of 6-month serum phosphorus control. The study included 5847 PD patients from seven countries; 9% of patients had baseline serum phosphorus <3.5mg/dL, 24% had serum phosphorus ≥3.5 to ≤4.5mg/dL, 30% had serum phosphorus >4.5 to <5.5mg/dL, 20% had serum phosphorus ≥5.5 to <6.5mg/dL, and 17% had serum phosphorus ≥6.5mg/dL. Compared with patients with baseline serum phosphorus ≥3.5 to ≤4.5mg/dL, the adjusted all-cause mortality hazard ratio (HR) was 1.19 (0.92,1.53) for patients with baseline serum phosphorus ≥5.5 to <6.5mg/dL and HR was 1.53 (1.14,2.05) for serum phosphorus ≥6.5mg/dL. Associations between serum phosphorus measurements over 6 months and clinical outcomes were even stronger than for a single measurement. Serum phosphorus >5.5mg/dL was highly prevalent (37%) in PD patients, and higher serum phosphorus levels were a strong predictor of morbidity and death, particularly when considering serial phosphorus measurements. This highlights the need for improved treatment strategies in this population. Serial serum phosphorus measurements should be considered when assessing patients' risks of adverse outcomes.