Few studies that have examined the effects of catecholamines on ketogenesis have considered the effects of catecholamines on hepatic portal blood flow. Since hepatic blood flow is a major determinant of hepatic ketogenesis (via modification of free fatty acid availability), interpretation of these studies is difficult. To better define the relative contributions of these variables, we studied the effects of physiological and pathophysiological plasma concentrations of epinephrine and norepinephrine on plasma ketone body concentrations and hepatic portal blood flow in controlled paired studies in young healthy male volunteers. To assess the effects of physiological catecholamine concentrations, each of eight subjects received 60-minute sequential infusions of epinephrine (10 ng/kg/min) and norepinephrine (32.5 ng/kg/min) together with a control infusion of heparin (0.4 U/kg/min) separated by 60-minute washout periods. Similar increments in plasma nonesterified fatty acid ([NEFA] to approximately 1 mmol/L) were observed during each infusion. The ketotic ratios, calculated as the ratio of plasma ketone bodies to fatty acids integrated above baseline for 90 and 120 minutes, respectively, for epinephrine and norepinephrine infusions were both significantly greater (P < .005 for each) than for the heparin control infusion. To assess the effects of pathophysiological plasma catecholamine concentrations, each of eight subjects also received sequential 60-minute infusions of epinephrine 60 ng/kg/min, norepinephrine 80 ng/kg/min (plus heparin 0.1 U/kg/min), and a separate control infusion of heparin with or without Intralipid (KabiVitrum, Alameda, CA). Whereas integrated plasma fatty acid levels were approximately twofold greater than those observed in the physiological protocol, the absolute integrated ketone body response to the pathophysiological concentration of epinephrine was significantly lower than that observed for the physiological dose of the hormone (P < .05). In contrast, the ketotic ratio for norepinephrine was significantly greater (P < .005) than for both epinephrine and the control infusion of heparin with or without Intralipid. Significant (P < .01) increases above baseline fasting levels were observed in plasma glucose and immunoreactive insulin concentrations during infusion of pathophysiological concentrations of epinephrine. Because of the technical difficulties of simultaneously measuring portal blood and sampling blood frequently, studies were repeated in six additional subjects using noninvasive image-guided flowmetry to measure percentage changes in hepatic portal blood flow during catecholamine infusion. Norepinephrine reduced hepatic portal blood flow significantly at the low-physiological concentration by 12% (P < .05) and at the pathophysiological concentration by 18% (P < .05). In summary, (1) both epinephrine and norepinephrine were associated with significant ketotic effects at physiological plasma concentrations; and (2) when infused at pathophysiological concentrations, only norepinephrine exerted a significant additional ketotic effect. Since norepinephrine has a significant simultaneous effect of reducing hepatic portal blood flow, we conclude that previous studies may have underestimated the effect of norepinephrine on hepatic ketogenesis.