In this double-blind, cross-over, placebo-controlled, randomized study, possible extraintestinal effects of miglitol, an absorbable a-glucosidase inhibitor, were investigated. Sixteen healthy male volunteers underwent two 75 g oral glucose tolerance tests with concomitant administration of miglitol or placebo. Peak and post-peak areas under the curve values for blood glucose, serum insulin and serum C-peptide after miglitol were not different from those found after placebo. The post-peak AUC-ratio (AUC ( peak, 180 min) on miglitol/AUC (peak, 180 min) on placebo) was for glucose 1.15 (CI 0.94-1.40, P=0.16), for insulin 1.12 (CI 0.95-1.33, P=0.17) and for C-peptide 0.98 (CI 0.81-1.18, P=0.82). It is concluded that miglitol exerts no clinically relevant extraintestinal effects on glucose control. Keywords a-glucosidase inhibitors miglitol systemic effectsIntroduction number of patients failed to confirm these findings (unpublished observations) [8 ]. To settle the contradictory findings prior to further studies we felt it necessary In the treatment of subjects with non-insulin-dependent diabetes mellitus (NIDDM ) complex carbohydrates, to once more repeat the studies of Joubert et al. to establish reproducibility of their findings and to find dietary fibre and a-glucosidase inhibitors are recommended to delay carbohydrate absorption, thereby out whether this approach is suitable for further exploration of possible systemic effects of miglitol. aiming to reduce post-prandial blood glucose fluctuations [1]. The a-glucosidase inhibitors act in this process through a competitive inhibition of brush border enzymes that cleave oligo-and disaccharides into monosaccharides [ 1, 2 ]. These compounds comprise Methods acarbose, a nitrogen-containing tetrasaccharide, and miglitol, which is derived from 1-deoxynojirimycin andThe study was performed in 16 healthy volunteers. The mean±s.d. age of these subjects was 28±9 years and is structurally similar to glucose. In contrast to the nonabsorbable acarbose, miglitol is almost completely the body mass index was 42.2±2.3 kg m−2. The subjects did not have renal or hepatic disease, had no family absorbed from the small intestine, raising the possibility that it could exert extraintestinal effects. Joubert et al.history of diabetes and were not taking other investigational drugs or medications altering the gastrointesti-[ 3, 4 ] demonstrated in healthy control and NIDDM subjects a decrease of post-peak blood glucose values nal motility and/or absorption. Written consent to participate was obtained from all subjects after oral and after an oral load of glucose in combination with miglitol but not with acarbose. Subsequent findings by written information regarding the study. The design of the protocol was a double-blind, crossother groups were interpreted in accordance with the results of that study. Our group [ 5] showed in an acute over, randomized, placebo-controlled study and was approved by the Ethics review committee of the study in NIDDM subjects that ingestion of a testmeal tog...