BACKGROUND: Uremic toxicity is a major concern in the dialysis population. There is keen interest in techniques that increase the removal of larger uremic molecules. We examined the short-term impact of a new, more porous, super-flux Helixone membrane (FX-E) versus the conventional high-flux Helixone membrane (FX-60) on beta(2)-microglobulin (beta2M) reduction and nutritional and inflammatory parameters. STUDY DESIGN: Randomized, double blind, crossover, pilot trial. SETTING & PARTICIPANTS: A single freestanding dialysis center. 30 stable hemodialysis patients. INTERVENTION: Patients were treated with FX-60 and FX-E membranes for a treatment period of 6 weeks each, with a 2-week washout period in between. OUTCOME & MEASUREMENTS: Primary outcome was change in beta2M concentrations from baseline to end of treatment. Serum samples were obtained predialysis and postdialysis at 0, 2, and 6 weeks, and dialysate albumin samples were collected continuously throughout dialysis sessions. RESULTS: Mean postdialysis beta2M concentrations at the end of 6 weeks of treatment were 6.73 mg/L for FX-E versus 8.22 mg/L for FX-60, which was significantly lower overall by 0.69 mg/L (95% confidence interval [CI], -1.09 to -0.29; P = 0.001). beta2M reduction ratios were greater overall with FX-E by 4.83% (95% CI, 2.78 to 6.89; P < 0.001), with mean values of 57% for FX-60 versus 66% for FX-E at the end of treatment. Median dialysate albumin loss with FX-E was 1.23 g (range, 0.22 to 4.83 g) compared with 0.17 g (range, 0.0017 to 2.69 g) with FX-60, which was greater by 1.52 g (95% CI, 1.11 to 1.93; P < 0.001). Serum albumin concentrations were slightly lower with FX-E by 0.1 g/dL (0.55 g/L; 95% CI, -1.04 to -0.07; P = 0.03), but prealbumin concentrations were not significantly different at 8.53 mg/L (95% CI, -23.76 to 6.71; P = 0.3). There were no differences in inflammatory cytokine concentrations or small-solute removal. LIMITATIONS: Short-term pilot study. CONCLUSION: In this stable dialysis population, removal of beta2M was more efficient with the Helixone super-flux FX-E membrane, with only a small decrease in albumin concentrations despite increased albumin loss. Large trials with longer treatment periods are required to evaluate the impact of the FX-E membrane on clinical outcomes.