Prior exercise alters the difference between arterialised and venous glycaemia: implications for blood sampling procedures.

Oral glucose tolerance and insulin sensitivity are common measures, but are determined using various blood sampling methods, employed under many different experimental conditions. This study established whether measures of oral glucose tolerance and oral glucose-derived insulin sensitivity (insulin sensitivity indices; ISI) differ when calculated from venous v. arterialised blood. Critically, we also established whether any differences between sampling methods are consistent across distinct metabolic conditions (after rest v. after exercise). A total of ten healthy men completed two trials in a randomised order, each consisting of a 120-min oral glucose tolerance test (OGTT), either at rest or post-exercise. Blood was sampled simultaneously from a heated hand (arterialised) and an antecubital vein of the contralateral arm (venous). Under both conditions, glucose time-averaged AUC was greater from arterialised compared with venous plasma but importantly, this difference was larger after rest relative to after exercise (0·99 (sd 0·46) v. 0·56 (sd 0·24) mmol/l, respectively; P<0·01). OGTT-derived ISIMatsuda and ISICederholm were lower when calculated from arterialised relative to venous plasma and the arterialised-venous difference was greater after rest v. after exercise (ISIMatsuda: 1·97 (sd 0·81) v. 1·35 (sd 0·57) arbitrary units (au), respectively; ISICederholm : 14·76 (sd 7·83) v. 8·70 (sd 3·95) au, respectively; both P<0·01). Venous blood provides lower postprandial glucose concentrations and higher estimates of insulin sensitivity, compared with arterialised blood. Most importantly, these differences between blood sampling methods are not consistent after rest v. post-exercise, preventing standardised venous-to-arterialised corrections from being readily applied.