Immunogenicity of long-term intraperitoneal insulin administration with implantable programmable pumps. Metabolic consequences.

OBJECTIVE: To assess immunogenicity of intraperitoneal insulin infusion via implanted pumps by two methods and to evaluate the possible influence of an increased antibody level on metabolic and clinical parameters. RESEARCH DESIGN AND METHODS: We studied insulin antibody levels in 17 type I diabetic patients before and until 24 months after implantation of a programmable pump delivering insulin intraperitoneally. Antibody levels were determined by radioimmunoassay (RIA) and enzyme-linked immunosorbent assay (ELISA). They were correlated with HbA1c, insulin requirements, free insulin, and the incidence of hypoglycemia. RESULTS: Insulin antibodies increased as soon as the 3rd month after implantation. This increase was sustained throughout the study period (month 0, 25.4 +/- 16.2%; month 3, 41.2 +/- 23.5%; month 12, 45.9 +/- 26%; month 24, 48.7 +/- 25%). The data was correlated with the two assay methods (RIA and ELISA). Postimplantation level was correlated with preimplantation level, which could indicate a predictive value of the latter . No correlation was observed with any metabolic parameters, particularly the number of hypoglycemic episodes. CONCLUSIONS: Our results indicate that intraperitoneal insulin administration by implantable programmable pumps leads to an increase of insulin antibodies, which are probably high-affinity antibodies (recognized by both RIA and ELISA). This increase in insulin immunogenicity did not induce significant metabolic consequences, which is reassuring for the future of programmable insulin pumps.