In vivo effects of phosphodiesterase III inhibitors on glucose metabolism and insulin sensitivity.

BACKGROUND Milrinone is a widely-used phosphodiesterase III (PDE3)-selective inhibitor. Recently, it was shown that milrinone increased hepatocyte glucose production, glycerol release, insulin secretion and fatty acid oxidation in vitro. Considering these results, it is important to know whether milrinone can induce insulin resistance in vivo. METHODS In order to investigate the effects of a selective PDE3 inhibitor on insulin secretion, glucose and lipolysis levels and dose-response relationship, varying doses of milrinone were injected into the right internal jugular vein in conscious rats, and blood glucose, plasma free fatty acid (FFA) and insulin levels were observed. The effects of milrinone on glucose metabolism and insulin sensitivity were assessed by using a hyperinsulinaemic-euglycemic clamping. RESULTS Chronically catheterized nonstressed rats were administered varying doses of milrinone (1, 5, 25 micromol/kg) and were compared with controls not treated with milrinone. After dosing, plasma FFA levels in the 3 milrinone groups significantly increased compared with before dosing and the controls. The percentages of elevation of FFA by the different milrinone doses were very similar, 50%, 52%, 55% for 1, 5 and 25 micromol/kg, respectively, at 2 min after dosing, suggesting that lipolysis is very sensitive to the effect of milrinone. However, the effect of milrinone on glucose concentration was detectable only in 25 micromol/kg group, and the plasma insulin levels were significantly elevated in the 5 and 25 micromol/kg milrinone groups, indicating that there was a dose-response relationship between milrinone and insulin and glucose levels. During hyperinsulinemic clamping, there were a significant increase in plasma FFA (173.1 +/- 15.2 to 633.8 +/- 87.3 microEq/L, p < 0.01) and a significant decrease in glucose infusion rates (GIR) to about 21%, and a slight increase in plasma insulin after milrinone treatment. CONCLUSIONS These data suggest that milrinone impaired the abilities of insulin to suppress lipolysis and insulin-mediated glucose utilization in peripheral tissue, despite having a slight increase in insulin secretion. Therefore, we conclude that milrinone administration induces an acute insulin resistance in vivo, and that may limit the therapeutic value of milrinone for human diabetic subjects.