The effect of intravenous insulin infusion on renal blood flow in conscious sheep is partially mediated by nitric oxide but not by prostaglandins.

To test the effect of insulin on renal perfusion and the participation of NO and PG as me- diators of this response, renal blood flow (RBF) was measured in sheep (n = 8) implanted with ultrasonic flow probes around renal arteries and with a systemic arterial pressure (SAP, n = 4) telemetry device. Three protocols were performed: 1) RBF and SAP were re- corded (0800 to 1800 h) in fed and fasted sheep, with the latter receiving intravenous (i.v.) infusions (0.5 mL/min) of insulin at 2 or 6 mU/(kg∙min); 2) fasted sheep received i.v. infusions of either an inhibitor of NO synthesis (N G -nitro-l-arginine methyl ester, L-NAME) alone (0.22 mg/(kg∙min), 1000 to 1200 h) or L-NAME (1000 to 1200 h) + insulin during the second hour (6 mU/(kg∙min), 1100 to 1200 h); and 3) the same proto- col was followed as in protocol 2, substituting L-NAME with ketoprofen (0.2 mg/(kg∙min)), a cyclooxygenase inhibitor. In all protocols, plasma insulin and glucose were determined. During insulin administration, eugly- cemia was maintained and hypokalemia was prevented by infusing glucose and KCl solutions. After the onset of meals, a long-lasting 18% increase in RBF and a 48% insulin increase were observed (P < 0.05), without changes in SAP. Low- and high-dose insulin infusions increased RBF by 19 and 40%, respectively (P < 0.05). As after meals, the increases in RBF lasted longer than the insulin increase (P < 0.05). The L-NAME infusion decreased RBF by 15% (P < 0.05); when insulin was added, RBF increased to preinfusion values. Ketopro- fen decreased RBF by 9% (P < 0.05); when insulin was added, RBF increased to 13% above preinfusion values (P < 0.05). In no case was a modification in SAP or glucose noted during the RBF changes. In conclusion, insulin infusion mimics the meal-dependent increase in RBF, independent of SAP, and lasts longer than the blood insulin plateau. The RBF increase induced by insulin was only partially prevented by L-NAME. Ke- toprofen failed to prevent the insulin-dependent RBF increase. Both facts suggested that complementary va- sodilatatory agents accounted for the insulin effect on sheep renal hemodynamics.