Ultrafiltration, diffusion, osmotic shifts, blood-membrane interactions, and psychological factors have all been implicated in the pathogenesis of postdialysis fatigue (PDF). To identify responsible factors, we performed a prospective, randomized, crossover analysis of fatigue scores (scale, 0 to 4) in 12 maintenance hemodialysis subjects with PDF. Fatigue scores were evaluated on nondialysis days (baseline) and after the following procedures on midweek treatment days: standard dialysis using either a 135- to 140-mEq/L sodium bath (routine hemodialysis) or a 150- to 155-mEq/L sodium bath (hypernatric hemodialysis); isolated ultrafiltration; isolated diffusion; and sham procedures with (isolated membrane) or without (recirculation) exposure to a dialysis membrane. Maximal fatigue scores are expressed as mean and 95% confidence intervals (CIs). The highest scores were recorded by patients who had just undergone routine hemodialysis (mean score, 2. 6; 95% CI, 1.7 to 3.4), isolated ultrafiltration (mean score, 2.1; 95% CI, 1.3 to 2.9), and isolated diffusion (mean score, 2.4; 95% CI, 1.5 to 3.2). There were no significant differences in fatigue scores between baseline periods and isolated membrane and recirculation procedures (mean score, 1.3; 95% CI, 0.4 to 2.2). Fatigue scores after hypernatric hemodialysis occupied an intermediate position (mean score, 1.7; 95% CI, 0.8 to 2.6). These results suggest that rapid hydraulic and molecular flux have a greater role in the pathogenesis of PDF than psychological stress and blood-membrane interactions. Use of a high-sodium bath may ameliorate PDF. We conclude that appropriate adjustments in both ultrafiltration and sodium profiling remain the most important means for controlling PDF in patients on short-duration hemodialysis.