Phosphate Kinetics During High - Flux Hemodialysis 1 ’ 2

Phosphate clearance by polysulfone (PS) and cuprophane (CU) membranes and the relationship of peridialytic changes in serum phosphate with those of serum-ionized calcium, parathyroid hormone (PTH), and insulin were studied in six stable patients undergoing chronic hemodialysis (HD). Dietary phosphate intake was 25.7 mmol/day, and total dose of elemental calcium was 3.2 g/day. Patients were dialyzed for 2 to 4 h, once with each membrane. Serum phosphate levels fell precipitously during the first hour of HD with both dialyzers, from I .42 and I .49 mM to nadirs of 0.53 and 0.69 mM for PS and CU, respectively. Phosphate levels began to increase either late in HD or at the end of HD and, by 4 h postHD, did not differ from predialysis values. Total mass transfer of phosphate was greater during the first hour (3.4 and 3.2 mmol for PS and CU, respectively) than during the remainder of HD (2.4 and 2.6 mmol/h). There were no significant differences in intradialytic serum phosphate changes, postdialytic phosphate rebound, or total phosphate removal between the two dialyzers. Ionized calcium increased by 0.11 mM, and PTH was suppressed to 40 to 50% of baseline values during dialysis with either membrane. Although phosphate removal continued for the duration of dialysis, serum phosphate did not continue to decrease, either reaching an apparent steady state or beginning to rebound, even during dialysis. This suggests active phosphate mobilization from a pool other than the extracellular fluid and demonstrates the inadequacy of a one-compartment model to explain these data. Further, these data do not support the regulation of intradialytic phosphate mobiI Rcelved July 23, 1992. Accepted June 7. 1993. 2 ThIs work was presentd, In part. at the 26th Annual Meeting of the American society of Nephrology. 3 Correspondence to Dr. J.G. Umans, Section of Nephrology, The University of ChIcago. 5841 South Maryland Avenue, MC-5100, chicago, IL 60637. 1046-6673/0405-1214503.00/0 Journal of the American Society of Nephrolo9y CopyrIght C 1993 by the American society of Nephrology S ustained hyperphosphatemia, common in patients with ESRD, leads to secondary hyperparathyroidism and, ultimately, to renal osteodystrophy. Control of serum phosphate depends on dietary phosphate restriction and the use of phosphate-binding agents. These approaches are inadequate in many patients. Phosphate-restricted diets may lead to protein-caborle malnutrition . Aluminum-based phosphate binders often result in bone, central nervous system, and hematobogic toxicity; calcium-containing binders may lead to hypercalcemia. Interest in phosphate removal by hemodiabysis (HD) has increased with the advent of more efficient dialysis techniques (1). Serum phosphate indeed declines significantly during HD (21 1 ). There is, in addition, a marked postdiabytic rebound of serum phosphate levels (25,7-9, 1 2). The clinical significance of dialytic phosphate removal is unclear, because it does not appear to influence prediabysis bevels (1 1 , 1 2), although it may affect calcium homeostasis (5). Bicarbonatecontaining diabysate may facilitate phosphate removal, though conflicting results have been reported (6,7, 12, 13). Because newer synthetic dialysis membranes cxhibit high phosphate clearances in vitro, we sought to compare phosphate clearance by polysulfone (PS) and cuprophane (CU) membranes, in vtvo. In these studies, conducted in patients with comparable diets and dialysis prescriptions, we also attempted to assess the relationship of peridialytic changes in serum phosphate with those of serum-ionized calcium, parathyroid hormone (PTH), and insulin.