Our previous studies have suggested that intraperitoneal (IP) insulin delivery may be associated with alterations of reverse cholesterol transport (RCT). We thus studied two parameters of RCT. We performed RCT studies in 10 C-peptide-negative type I diabetic patients who were randomized into two groups. The experimental (A) and control (B) groups were studied at -3 and 0 months before and +3 and +6 months after IP pump subcutaneous (SC) insulin use. The first step in RCT was estimated by measuring patient serum-mediated 3H-cholesterol efflux from cultured fibroblasts. Cholesteryl ester transport protein (CETP) activity was assessed by a solid-phase assay. No changes in glucose control occurred during the study. Total low-density lipoprotein (LDL) and high-density lipoprotein (HDL) cholesterol, apoprotein B, and apoprotein A-I remained unchanged during the study. Cholesterol efflux from group A increased from baseline after 3 months of IP insulin by 9.5% +/- 3.4% (mean +/- SE, P < .05), whereas in group B patients it decreased negligibly by 1.5% +/- 2.9% (P = .045 for changes between groups). CETP activity increased from baseline by 25.3% +/- 7.7% (P < .05) in group A after 3 months of IP insulin, whereas in group B it changed little, -1.5% +/- 7.9%, with modest differences between groups (P = .16). These data indicate that (1) serum from patients treated long-term with IP insulin delivery may enhance cholesterol efflux from fibroblasts and CETP activity, and (2) these effects appear independent from glucose control, implying a direct effect by IP insulin.