Is low-dose streptozotocin in rats an adequate model for gestational diabetes mellitus?

OBJECTIVE: To examine the use of streptozotocin (SZ) in rats as a model for gestational diabetes mellitus (GDM). METHODS: We studied various doses of SZ, either as a single administration (30, 35, 40, or 50 mg/kg, intraperitoneally) on day 1 of pregnancy or as 2 low doses (30 and 20 or 30 and 30 mg/kg) administered 2 days before mating and on day 1 of pregnancy. We examined the effect on maternal and fetal glucose and insulin concentrations and on fetal weight on day 20 of pregnancy. In a second series of experiments, we studied two groups (SZ 30/20 and SZ 35) with fetal hyperinsulinemia on day 20 of pregnancy. We performed an intravenous glucose tolerance test (IVGTT) on day 20, and in separate groups we reassessed fetal weight and insulin concentrations at term (day 22). RESULTS: There was considerable variability in glucose concentrations with most SZ doses. Lower doses of SZ (30, 30/20, and 35 mg/kg) did not significantly increase maternal and fetal glucose levels, in contrast to higher doses of SZ (30/30 and 50 mg/kg). Fetuses were smaller on day 20 with all doses except SZ 30 and SZ 30/20; fetal insulin concentrations were elevated with SZ 30, 30/20, and 35. The IVGTT showed glucose intolerance in SZ 35 and SZ 30/20, but the insulin response was unaffected in either group. Fetuses were smaller on day 22 in both these SZ groups, whereas fetal insulin levels at term were not different compared with controls. CONCLUSION: Low-dose SZ is not a good model for GDM because of the high variability in glucose levels, the normal insulin response to a glucose load, the absence of fetal macrosomia, and the inconsistent effect on fetal insulin concentrations. Copyright (C) 2003 by the Society for Gynecologic Investigation.