In the early years of hemodialysis, the number of patients treated was small, dialysis techniques were in their infancy, and patient survival was often short. As dialysis techniques and patient survival improved, the realization that untreated tap water was not adequate for the preparation of dialysate led to the release in 1982 of the American National Standard for Hemodialysis Systems (1) by the A d a t i o n for the Advancement of Medical Instrumentation (AAMI). This standard focused primarily on chemical contaminants, because of the accumulated data demonstrating their toxicity. The standard also limited the total viable microbial count in the water used to prepare the dialysate to 200 colony-forming units (&)/mi and in the final dialysate to 2000 cfu/ ml.While the AAMI standard has remained essentially unchanged for the past 8 years, hemodialysis therapy has not, and the question arises as to whether or not this standard is adequate for today's hemodialysis therapies. The advent of high-flux and bicarbonate hemodialysis, and increased understanding of the effects of bacterial products, has made the adequacy of the AAMI standard for microbiological contamination an issue of particular concern. In the remainder of this commentary, we examine the case for a more stringent microbiological standard for dialysate and the water used to prepare it.The microbiological purity of water or dialysate is commonly assessed in two ways. The first is by the determination of the number of viable organisms per unit volume and the second is by measuring the concenmtion of endotoxin. The number of viable organisms is determined by culturing the water to be tested in nutrient medium under standardized conditions, and then counting the number of colonies growing on the surface of the medium. The colony count obtained depends on the nature of the organisms present, the type of medium used, and the length and conditions of the incubation. The AAMI standard recommends culturing water samples for 48 hours at 37'C using tryptic soy agar, standard methods agar, or blood agar as the culture medium. These three media have been developed for detecting organisms which grow in nutrient-rich media, such as blood or other body fluids. In contrast, organisms which contaminate purified water thrive in nutrientpoor conditions and may not grow well in these media. Indeed, recent studies (2,3) have confirmed that standard clinical laboratory techniques are inadequate for quantitation of bacteria in water and show that use of nutrient-poor media, such as Reasoner's 2A medium, and lower culture temperatures (2428°C) allow more sensitive determination of contaminating organisms in water.Bacteria produce a number of biologically active products. These include exotoxins, toxic substances synthesized and released from intact bacteria, and endotoxins, toxic substances generally produced from the cell wall on bacterial lysis. Endotoxins are lipopolysaccharides and are usually quantitated using an assay based on the Limulus amebocyte lysate (LAL) reaction. ...