OBJECTIVE The objective of this study is to compare the effect of intraperitoneal versus subcutaneous insulin injection on hepatic glucose production (HGP) and systemic glucose utilization (Rd) in patients with NIDDM. METHODS Eight male volunteers with NIDDM, each of whom had a programmable-rate, implantable insulin pump, were given an injection of insulin (0.15 units/kg body wt) by intraperitoneal or subcutaneous injection on separate days in randomized order. Plasma glucose was kept constant for 5 h using the glucose clamp technique, and HGP and Rd were measured using isotope dilution. RESULTS Intraperitoneal insulin injection resulted in higher and earlier peak systemic insulin concentrations (1,469 +/- 245 vs. 454 +/- 48 pmol/l, P < 0.01). Glucose Rd doubled within 1 h after intraperitoneal injection and was greater than that attained with subcutaneous injection (3.91 +/- 0.27 vs. 2.60 +/- 0.19 mg.kg-1.min-1, P < 0.01). Intraperitoneal and subcutaneous injections suppressed HGP and plasma free fatty acid to a similar extent during the first 3 h, effects tht persisted through 5 h after subcutaneous insulin injection but waned 3-4 h after intraperitoneal injection. CONCLUSIONS In patients with NIDDM, intraperitoneal insulin injection achieves more rapid and greater peak values for stimulation of glucose Rd than subcutaneous insulin injection. With regard to HGP and lipolysis, intraperitoneal and subcutaneous injections achieve similar initial suppression but this is maintained for a more limited duration with intraperitoneal as compared with subcutaneous injection. These differences in insulin action seem directly related to the rapidity of insulin absorption with intraperitoneal injection.