Our objective was to determine whether patients with chronic renal failure requiring maintenance hemodialysis retain intrinsic renal function longer when using reprocessed polysulfone (PS) membrane hemodialyzers or single-use cellulose acetate (CA) membrane hemodialyzers. Fifty consecutive patients with residual renal function (urea clearance > 2.0 mL/min) using PS dialyzers were compared with a retrospective, disease- and time-matched population of patients using CA dialyzers. Endogenous urea clearance was measured every 3 months in all patients with remaining residual function. Other data collected included age, sex, cause of chronic renal failure, use of angiotensin-converting enzyme inhibitors or calcium channel blockers, and hemodynamic stability during hemodialysis. All patients were observed for at least 6 months while using a single type of dialyzer. Study end points included loss of residual renal function (urea clearance < 1.0 mL/min), death, transplant, transfer to peritoneal dialysis, or change of dialyzer. The PS and CA groups of patients were well matched for sex, age, initial renal clearance, predialysis blood pressure, and hemodynamic stability during hemodialysis. The PS patients had a higher delivered Kt/V (1.34 +/- 0.30 [mean +/- SD]) than the CA patients (1.06 +/- 0.20). The PS group had a higher average urea clearance than the CA group after 4 to 9 months of dialysis (2.8 +/- 2.6 mL/min v 1.7 +/- 1.6 mL/min, respectively), after 10 to 15 months of chronic dialysis (2.0 +/- 2.4 mL/min v 1.1 +/- 1.5 mL/min, respectively), and after 16 to 21 months of dialysis (1.3 +/- 1.9 mL/min v 0.5 +/- 1.1 mL/min, respectively; all P < 0.03, t-test). After 22 to 24 months of dialysis, the difference between the two groups was not significant. When comparing patients with identical causes of chronic renal failure, there were no differences between the PS and CA groups for those with diabetes mellitus, tubulointerstitial disease, or polycystic disease. Patients with parenchymal renal disease (glomerulonephritis or nephrosclerosis) had markedly better retention of intrinsic renal function with PS than with CA dialyzers (all P < 0.01). Kaplan-Meier analysis for retention of intrinsic renal function showed that PS patients with parenchymal renal disease had a mean of 23 months before loss of intrinsic renal function, whereas for CA patients the mean was 11 months before loss of intrinsic renal function (P = 0.0005). Cellulose acetate patients lost renal function at an average rate of 0.27 +/- 0.22 mL/min/mo, whereas for PS patients the rate was 0.14 +/- 0.56 mL/min/mo (P = 0.06, rank sum). CA patients with parenchymal renal disease lost renal function at a rate of 0.29 +/- 0.22 mL/min/mo, whereas for PS patients the rate was 0.0 +/- 0.8 mL/min/mo (P = 0.004, rank sum). Age, sex, and the use of either angiotensin-converting enzyme inhibitors or calcium channel blockers did not have an effect on the loss of intrinsic renal function. Patients with nondiabetic parenchymal renal disease receiving chronic hemodialysis with hydrogen peroxide/peroxyacetic acid-reprocessed PS dialyzers and a higher Kt/V lose residual renal function at a slower rate than disease-matched patients using single-use CA dialyzers. Our findings provide further evidence that the choice of dialyzer membrane may have an effect on intrinsic renal function. (C) 1997 by the National Kidney Foundation, Inc.
