Glucagon-like peptide-1 (GLP-1) displays concentration-dependent effects on metabolism, appetite and angiogenesis; therefore, accurate determination of circulating GLP-1 concentrations is important. In this study, we compared GLP-1 concentrations in venous versus arterialized blood in both fasted and fed conditions. Venous and arterialized blood samples were obtained simultaneously from 10 young, healthy men before and 30, 60 and 120 min after ingestion of 75 g glucose. Plasma GLP-1 concentrations increased in response to glucose ingestion (time effect, P < 0.01) and to a lesser extent in venous versus arterialized plasma (time × arterialization interaction, P < 0.01). Accordingly, the plasma incremental area under the curve was lower in venous versus arterialized plasma (974 ± 88 versus 1214 ± 115 pmol l (120 min) , respectively, P = 0.049). In the postprandial state, there was a positive relationship between arterialized GLP-1 concentrations and the venous-arterialized difference in GLP-1 concentrations (r = 0.51; P < 0.01). Both arterialized and venous peak GLP-1 concentrations showed positive relationships with peak arterialized insulin concentrations (both r > 0.6, P < 0.01). Venous sampling results in lower concentrations of GLP-1 in the postprandial but not the fasted state compared with arterialized blood. This absolute difference is biologically meaningful and is magnified when GLP-1 availability is high. Therefore, sampling from arterialized blood may provide a better chance of detecting small differences in postprandial GLP-1 availability with interventions. If absolute GLP-1 concentrations are of interest, the blood sampling method should be considered carefully and reported clearly.