Nephrology Dialysis Transplantation · 1995 · 32 citations · 0 references
MedicineImmunologyNephrologyDialysisHemodialysisInflammationComplement SystemCytokine-mediated PathologyDialysate CompositionEndotoxin ControlRenal DiseaseRenal InflammationCytokine-mediated Renal InflammationEndotoxin AssaysEndotoxin QuantificationInflammation In HemodialysisInflammatory Responses To HemodialysisLimulus Amebocyte Lysate AssayLimulus Amebocyte Lysate TestMonocyte Biology
Abstract We examined in vivo the release of tumour necrosis factor alpha (TNFα) and interleukin 6 (IL-6) by uraemic monocytes upon stimulation with endotoxin-contaminated bicarbonate concentrate. Twelve uraemic patients underwent 1-month-subsequent periods of standard haemodialysis (SHD) with cuprophane (CU), a high-complement-activating membrane (6 patients), or haemodiafiltration (HDF) with polyacrylonitrile (PAN), a low-complement-activating membrane (6 patients), by using a dialysate prepared with either non-sterile bicarbonate concentrate tanks (phase 1) or sterile bicarbonate concentrate bags (phase 2). TNFα and IL-6 concentrations were determined in monocyte supernatants by ELISA; endotoxin levels in bicarbonate concentrates were measured by a chromogenic limulus amoebocyte lysate (LAL) assay. A significant increase in LAL reactivity was found in bicarbonate concentrate tanks compared to sterile bags (P<0.001). Non-sterile dialysate caused a significant (P<0.001) predialytic increase in monocyte TNFα release as compared to controls and nondialysed uraemic patients. One month treatment with sterile bicarbonate significantly decreased TNFα predialytic activity in monocyte supernatants (P<0.001) to levels closer to those of non-dialysed uraemic patients. A similar decrease was observed for IL-6 production. Dialytic treatment induced a further increase in both TNFα and IL-6 production, particularly in phase 1. When uraemic patients were examined separately according to the different dialytic procedures (SHD-CU or HDF-PAN), the use of sterile dialysate (phase 2) caused a significant decrease of predialysis TNFα release in both SHD CU patients (24.1±8.4 pg/ml versus 55.3±5.7 pg/ml, P<0.001) and HDF PAN-treated patients (16.6±5.3 pg/ml versus 29.1±5.4pg/ml, P<0.005), so that the differences between the dialytic procedures were completely abolished. In conclusion, TNFα and IL-6 release may be induced by endotoxin-contaminated dialysate during haemodialysis. The use of sterile bicarbonate can ameliorate the bioincompatibility of CU membranes and probably influences the biocompatibility of PAN membranes. Therefore, regardless of the type of dialyser used, all attempts to obtain an ultrapure dialysate are important to optimize dialytic treatment, in order to attenuate the chronic monocyte activation which occurs during haemodialysis.