OBJECTIVE: The current study sought to determine whether a dietary cocktail of deficient substrates and antioxidant could reduce the incidence of diabetic embryopathy to the background rate in an in vivo rat model.

STUDY DESIGN: Sprague-Dawley rats 70 to 90 days old were assigned to one of eight groups: two control groups (groups 1 and 2) and six diabetic groups (groups 3 to 8). One group of controls (group 2) received an oral supplemental cocktail of vitamin E (⍺-tocopherol, 400 mg/day), safflower oil (arachidonic acid, 1 ml/day), and myo-inositol 0.08 mg per day. Four other diabetic groups (groups 5 to 8) received the cocktail in varying strengths. One diabetic group (group 3) received a normal diet only without insulin or cocktail, whereas another diabetic group (group 4) received insulin but no cocktail supplementation. On day 6 of gestation diabetes was induced in groups 3 to 8 with intravenous streptozotocin (65 mg/kg), and maternal glucose levels were monitored. Animals were killed on day 12; embryos were examined for size, somite number, and evidence of malformations and were biochemically evaluated for vitamin E and myo-inositol levels and superoxide dismutase activity.

RESULTS: The diabetic groups had a significantly higher (p < 0.05) mean blood glucose level than controls did. The insulin-treated group 4 had glucose levels that were comparable to those of controls. The unsupplemented diabetic group 3 had a neural tube defect rate of 23.7% versus 4.04% in controls and 3.55% in insulin-treated diabetics (p < 0.05). The rate of neural tube defects was significantly reduced to the background level in animals receiving half-strength cocktail or stronger doses (groups 6 to 8) compared with the diabetic unsupplemented controls (group 3). Diabetic animals in group 5 receiving only quarter-strength cocktail did not demonstrate a significant reduction in the malformation rate. Serum myo-inositol levels were not significantly different among the groups. However, serum levels of vitamin E were significantly higher in diabetics receiving half-strength cocktail than in nondiabetic controls, diabetics receiving no supplements, and diabetics receiving quarter-strength cocktail. Superoxide dismutase activity was also significantly increased in diabetic animals receiving supplementation versus animals not receiving the same, and the increases in vitamin E and superoxide dismutase were significantly correlated (r = 0.66, p < 0.05).

CONCLUSION: These data demonstrate that a cocktail containing deficient substrates and an antioxidant in varying strengths reduces the malformation rate to background in offspring of diabetic rats. (Am J Obstet Gynecol 1997;176:790-8.)