To prevent pyrogenic reactions during hemodialysis, it is recommended that bacteria and endotoxin in dialysate not exceed 100-200 colony forming units (CFU)/ml and 0.25-2 endotoxin units (EU)/ml, respectively. While these limits are adequate to prevent acute pyrogenic reactions, data are accumulating to suggest they may not prevent stimulation of chronic inflammation in hemodialysis patients. Fragments of endotoxin and other bacterial products capable of stimulating immune cells cross low-flux and high-flux membranes in vitro. In clinical studies, use of ultrapure dialysate (bacteria < 0.1 CFU/ml and endotoxin < 0.03 EU/ml) is associated with lower concentrations of inflammatory markers and acute phase reactants than are observed with dialysate meeting current quality recommendations. Moreover, observational studies suggest a link between clinical outcomes and dialysate purity. Treatment of patients with ultrapure dialysate is reported to improve nutritional status, increase responsiveness to erythropoietin, slow the decline in residual renal function, lessen cardiovascular morbidity, and decrease the incidence of beta(2)-microglobulin amyloidosis. To date, however, none of these studies has shown a cause-and-effect relationship between dialysate purity and outcome. Further, there are no data defining the concentration dependence of outcomes on dialysate purity and the relative importance of dialysate purity as a trigger of inflammation remains unclear. While the technology exists to routinely provide ultrapure dialysate, controlled clinical trials are still needed to answer the question of whether or not introducing ultrapure dialysate into routine clinical practice represents an efficient use of limited resources in terms of decreasing inflammation and improving outcomes in hemodialysis patients.