The concept of biocompatibility as a clinically important parameter in dialysis therapy is challenged. The failure to establish clear causally related pathogenetic associations between clinically relevant dialysis problems and measurements of complement activation and leucopenia have hindered intellectual acceptance of the value of measuring so called indices of biocompatibility. In an effort to find a link between these parameters and clinical problems, the interleukin hypothesis was proposed 5 years ago. The background to the interleukin hypothesis is reexamined. In view of the the availability of recombinant forms of IL-1 and tumor necrosis factor and the subsequent development of more specific radio-immune assays, it is suggested that the original hypothesis be extended to various cytokines manufactured by monocytes. This is made necessary by the failure to detect raised levels of IL-1 β in plasma both in normals and ESRD patients following endotoxin challenge. However, tumor necrosis factor is raised under these conditions as is IL-6. Futhermore, the observation that these substances act synergistically in procuring the inflammatory response is emphasised. Thus, although using more specific methodology, IL-1 is not the plasma messenger originally thought, tumor necrosis factor and IL-6 still allow the possible relation to stimulation of the inflammatory response by hemodialysis, either by endotoxin contaminated dialysate, acetate dialysate, or membrane activation of C5a. The association between monocyte cytokine production and β
2-microglobulin generation and possibly the deposition of this protein in amyloid form in tissues is also discussed. It is recommended that measures to reduce the pyrogen content of dialysate are introduced and that acetate dialysis be replaced by bicarbonate. The question of abandoning the use of membranes that cause C5a generation is also debated. In view of the rapid growth of knowledge in the last 5 years (Table 1), it is not surprising that a hypothesis launched in 1983-IL-1 Hypothesis has had to be modified and currently should be known as the monokine hypothesis. The hypothesis evoked the possibility that the repeated contact of blood with dialysis membranes and dialysate would stimulate the human blood monocyte adhering to the dialysis membrane. The stimulants would include C5a generated by contact with cellulosic membranes, endotoxin or derived fragments crossing the membrane from the dialysate and acetate ions present in the dialysate. The acute consequences of this interaction would reproduce some aspects of the “acute phase” response such as fever and increase in circulating acute phase proteins, and the chronic consequences would lead to the “shrinking man syndrome”, amyloidosis and premature ageing. The purpose of this communication will be to reexamine and reformulate the hypothesis 5 years on. In this review, the biological properties of IL-1 and tumor necrosis factor will be considered with particular attention to their relevance in patients undergoing regular maintenance haemodialysis (HD) or continuous ambulatory peritoneal dialysis (CAPD). In addition, how the HD or CAPD procedure and materials employed in both procedures can lead to monocyte/macrophage stimulation and the induction of IL-1, TNF, IL-6 or other monkines is considered. Evidence is now accumulating that the local inflammatory response is initiated by the dialysis membrane-monocyte interaction in HD and by the instillation of exchange fluid during CAPD. The consequence of these actions are the production of IL-1. Long-term production of IL-1 may account for the chronicity of atute phase changes observed in ESRD patients on HD and CAPD. The biological properties of IL-1 and TNF are so similar (Table 2), that apart from the failure of TNF to act as a growth factor for T or B lymphocytes, they are biologically virtually identical. Indeed, the necessity for them to act
View full abstract
