Meticulous Control of Diabetes During Organogenesis Prevents Congenital Lumbosacral Defects in Rats

Abstract

Lumbosacral defects occur in the offspring of the diabetic rat in a fashion analogous to those occurring in infants of diabetic human mothers. To test the hypothesis that control of diabetes during the critical period of organogenesis can affect the incidence of these defects, diabetes was produced by streptozotocin in pregnant rats on day 6 of gestation. One group of pregnant rats was treated with saline (group 2), while another (group 4) received insulin twice daily between days 6 and 13. The incidence of lumbosacral malformations in the fetuses was then compared between these groups, with a pregnant, nondiabetic control group (group 1), and with a group of rats in whom diabetes was produced by streptozotocin on day 12, after the critical period of fusion of the lumbosacral spine (group 3). Two types of lumbar and/or sacral malformations were noted: a fusion defect and an ossification defect. In the nondiabetic animals (group 1), only 1 out of 59 fetuses had an abnormality. Diabetes produced after the period of fusion of the neural tube in the lumbosacral area (group 3) had no effect on the incidence of lumbosacral malformations (0 out of 54). Diabetes treated with saline during the period of organogenesis (group 2) was associated with lumbosacral malformations in 25 of 146 fetuses, which was significantly increased (P < 0.005) when compared with the nondiabetic pregnant animals. Insulin treatment during the period of organogenesis (group 4) significantly reduced (P < 0.005) the number of lumbosacral defects (5 of 106) from that observed in the diabetic animals treated with saline. When the results of the insulintreated diabetic group were further analyzed, the incidence of malformations (5 of 58) in the animals that exhibited intermediate hyperglycemia during the period of organ differentiation [two or more serum glucose values greater than 200 mg/dl, (group 4A)] was not significantly different from that observed in the saline-treated diabetic group (P > 0.1). This occurred despite a mean glucose concentration during the period of insulin treatment that was significantly lower than the saline-treated diabetic controls (180 ± 16 mg/dl vs. 348 ± 16, P < 0.001). In contrast, the incidence of lumbosacral malformations (0 of 48) observed in the fetuses of those diabetic animals in whom insulin therapy produced normal serum glucose concentrations (122 ± 9 mg/dl) during the period of organogenesis (group 4B) was sharply reduced from that seen in the fetuses taken from the saline-treated diabetic mothers who had not received insulin (P < 0.001). These data are consistent with the thesis that (1) uncontrolled diabetes produces lumbosacral malformations only during the period of organ differentiation, (2) control of diabetes during the period of organ differentiation reduces the incidence of malformations, and (3) the “quality” of that control is quite important, because normal to near-normal serum glucose concentrations must be achieved in order to be effective.