The effect of coffee total lipids (CTL), coffee non-saponifiable matter (NSM) and coffee diterpene alcohols (DTA) extracted from Coffea arabica beans on serum cholesterol in adult male Syrian hamsters was examined. The animals were fed either a commercial laboratory chow diet (study 1), containing 5% fat and low in saturated fat (1.46 g/100 g diet) and cholesterol (0.03 g/100 g diet) or a semi-synthetic diet (study 2) set in gelatine, containing 10% fat and high in saturated fat (4 g/100 g diet) and cholesterol (0.5 g/100 g diet). The coffee lipid extracts were dissolved in olive oil (concentration either 5 mg CTL, 0.5 mg NSM or 0.5 mg DTA per 250 microliters olive oil) in study 1 and in coconut oil (concentration either 20 mg CTL, 2 mg NSM or 2 mg DTA per 250 microliters) in study 2. A dose of 250 microliters of these solutions was administered daily to hamsters by gavage. Control animals received 250 microliters vehicle only. For serum lipid analysis, blood samples were obtained on days 0, 7 and 14 in study 1 and on days 0, 7, 14 and 21 in study 2. Statistical analyses of the data in Study 1 indicated a tendency for serum total cholesterol and high density lipoprotein-cholesterol to increase with administration of CTL, NSM and DTA. In contrast, in study 2 there were no significant differences in serum lipid levels between control and coffee lipid-treated groups across time. In either study, total serum cholesterol levels of the three coffee lipid groups were not significantly different from each other. These results support the concept that coffee lipids may be hypercholesterolaemic and indicate that diterpenes could be the lipid component responsible for such an effect. However, it appears that this hypercholesterolaemic effect is apparent only when the background diet is low in saturated fat and cholesterol. A high saturated fat/high cholesterol diet may mask the hypercholesterolaemic effect of coffee lipids.