The purpose of the study was to examine whether insulin sensitivity was associated with fasting plasma ghrelin concentrations in a population-based sample of 58-year-old clinically healthy Caucasian men. The methods used were dual-energy x-ray absorptiometry (DXA) for measurement of body composition and a conventional euglycemic hyperinsulinemic clamp, measuring glucose infusion rate (GIR) that was adjusted for fat-free mass. Plasma ghrelin was measured by radioimmunoassay. The results showed that ghrelin was not associated with GIR adjusted for fat-free mass or with GIR adjusted for body mass, and body fat, or waist circumference. Plasma ghrelin correlated negatively to body fat (−0.46, P < .001) and waist circumference (−0.45, P < .001). Ghrelin was also inversely related to systolic and diastolic blood pressure (r = −.29 and r = −0.34, respectively, P < .01) and positively to high-density lipoprotein (HDL) cholesterol (0.33, P < .01), and low-density lipoprotein (LDL) particle size (0.34, P < .001), but these associations did not remain after adjustment for body fat. Plasma ghrelin was associated with current smoking independent of waist circumference. Among current smokers, circulating plasma concentrations were higher in those who had smoked during the hour preceding the blood sample than those who had smoked 2 to 12 hours ago (P = .043). The conclusion is that whole body insulin sensitivity was not associated with plasma ghrelin concentrations. Body fatness was the strongest determinant of circulating ghrelin. It was found that acute smoking may affect ghrelin levels.