[Comparison of biocompatibility of hemophane, cellulose diacetate and acrilonitile membranes in hemodialysis].

INTRODUCTION: The membranes used in haemodialysis (HD) may be manufactured from cellulose, modified cellulose or synthetic polymers. Such membranes, when in contact with blood, activate the complement system, which entails changes in leukocyte (L) and platelet (P) counts. The magnitude of complement activation depends on the type of membrane used, and represents the biocompatibility index. Acrylonitrile (AN69) is synthetic membrane of high biocompatibility standards, whereas haemophane (HP) and cellulosediacetate (CD) membranes are modified cellulose-based membranes. The biocompatibly profiles of HP, CD and AN69, characterised by changes in L and P counts and activation of complement components (C3a and C5b-9), has been studied in order to assess the biocompatibility features of HP, CD and AN69 membranes. MATERIALS AND METHODS: We compared the generation of C3a des Arg and C5b-9 and the changes in L and P counts in groups of patients on HD who used HP (3 patients), CD (3 patients) and AN69 membrane (3 patients). Concentrations of C3a des Arg and C5b-9, and L and P counts were measured in the first HD session of the week, just before the start of HD (0 min) and after 15, 120 and 240 min. RESULTS: For HP, L count at 15 min had declined to 50% of the pretreatment level (p < 0.05), compared with 59.3% (p < 0.05) for CD and with 98.7% (p = ns) for AN69, returning to the pretreatment level at 240 min. Statistically significant difference was found comparing L counts for HP vs AN69 (p < 0.001) and CD vs AN69 (p < 0.001). The mean plasma concentrations of C3a des Arg were 1246 +/- 832 ng/ml for HP, 1148 +/- 774 ng/ml for CD and 639 +/- 217 ng/ml for AN69 and significant difference was found comparing HP vs AN69, CD vs AN69 (p < 0.05). The maximal values of C3a des Arg occurred at 15 min in HP and CD (p < 0.005), whilst for AN69 the concentrations showed no statistically significant differences. The plasma concentrations of C5b-9 in patients on HP and CD were the highest at 15 min (p < 0.005), while those on AN69 did not show any statistically significant rise. DISCUSSION: The fall in L counts was considerably marked for HP and CD membranes, with the lowest level at 15 min following blood exposure, thereafter returning to pretreatment level, whilst in AN69 leukopenia was not noted. The loss of L from the circulation is attributed to their pulmonary sequestration since complement activation and release of C5a fraction into the circulation results in their binding to the neutrophils and upregulation of the adhesion-promoting integrins of leukocytes, resulting in their binding to endothelial cells. The mean plasma concentrations of C3ades Arg, measured in the course of clinical use of HP and CD were significantly higher compared with synthetic membrane AN69, reaching the maximal level at 15 min. The complement activating potential of a biomaterial is one of its important biocompatibility features. The C3a complement activation curve is characteristic of cellulose-based materials and is governed by interaction between blood and hydroxyl groups on the membrane surface and by the density of hydroxyl groups available for the binding of C3b. The highest concentrations of C5b-9 in modified cellulose membranes were achieved by 15 min following blood exposure and met the activation curve for C3a. The capacity of AN69 membrane to adsorb anaphylatoxins may be a contributory factor for the failure to measure any increase in complement fraction concentration during HD for AN69. CONCLUSION: There are significant differences between HP, CD and AN69 membranes concerning complement activation magnitude and leukopenia rate. Modified cellulose membranes HP and CD induce higher C3a concentrations and superior leukopenia rate compared with AN69. Synthetic membrane AN69 proved of higher biocompatibility, with lower efficiency of complement activation.