Exercise training has been shown to be effective in preventing the development of non-insulin-dependent diabetes mellitus (NIDDM) in a model rat (Otsuka-Long-Evans-Tokushima Fatty [OLETF]). For determination of how long a preventive effect of exercise training against the development of NIDDM lasts in this model, six male OLETF rats each were assigned to training (1) for a whole experimental period, from 7 to 28 weeks of age (E-E); (2) for the first half of the period, from 7 to 15 weeks of age (E-S); and (3) for the second half of the period, from 16 to 28 weeks of age (S-E). In addition, eight male OLETF rats were given no exercise during the experimental period (S-S). At 28 weeks of age, E-E, E-S, S-E, and S-S rats, weighed averages of 514, 542, 557, and 669 g and had abdominal fat deposits of 13.9, 21.3, 38.2, and 76.0 g, respectively. At 28 weeks of age, the cumulative incidence of NIDDM in S-S was 100%, while none of the trained rats were diabetic. The glucose infusion rate (GIR) during a hyperinsulinemic euglycemic clamp test, an index of insulin sensitivity, in the E-E group was significantly greater than that in the S-S group. The values in the E-S and S-E groups were slightly, but not significantly, less than that in the E-E group. Morphologic studies on the pancreas of E-E rats and S-E rats showed minimal changes of islets, whereas sections of islets from E-S rats appeared slightly enlarged and fibrotic, although significantly less than those of islets of S-S rats. These results demonstrate that the preventive effect of exercise training against the development of NIDDM lasts for at least 3 months after the cessation of exercise in this model. (C) 1996 by W.B. Saunders Company