Insulin treatment of patients with gestational diabetes: does dosage play a role?

In nonpregnant patients high insulin requirements are associated with hypoglycemia and weight gain but not with improvement in glucose control. The effect of insulin requirement on maternal and neonatal outcomes in gestational diabetes mellitus (GDM) is yet unknown.

We conducted a retrospective cohort study of maternal and neonatal outcomes of pregnancy according to insulin requirements in women with GDM who were followed and delivered at the Yitzhak Shamir Medical Center between 2006 and 2016. The daily insulin dose in units per body weight was divided into quartiles and analyses were performed to compare the lowest, highest, and two middle quartiles. The primary outcome was a composite of any of the following: cesarean-section (CS), preeclampsia, macrosomia and large for gestational age (LGA) birth weight, neonatal intensive care unit admission, need for phototherapy, and neonatal hypoglycemia.

Women were divided according to their insulin requirements as follows: 79 (24.8%) women who needed <0.13 IU/kg/day of insulin (insulin-sensitive group), 160 (50%) women who needed 0.14-0.42 IU/kg/day of insulin (comparison-group), and the rest who needed >0.43 IU/kg/day of insulin (insulin resistant group). There were no differences in the composite outcome between the groups (64.6, 61.3, and 69.6% for the insulin sensitive-, comparison- and resistant- groups, respectively, p = .44). Women in the insulin-resistant group had higher fasting glucose levels in the first trimester (91, 98 and 102 mg/dL for women in the insulin sensitive-, comparison- and insulin-resistant groups, respectively; p = .01). Women in the insulin-sensitive group had significantly better glycemic control (fasting glucose levels ≤90 mg/dL and 1-hour and 2-hour postprandial glucose levels ≤140 mg/dL and ≤120 mg/dL for more than 80% of measurements) than those in the insulin-resistant group (70.3 versus 29.9%; p < .001). The rate of CS was significantly higher in the insulin-resistant group (42.3 versus 24.1%; p = .03), but the rate of LGA birth weight was surprisingly higher in the insulin-sensitive group (29.5 versus 16.7%, p = .04). After controlling for confounders, women in the insulin-sensitive group had a decreased risk for CS in relation to the comparison group (OR = 0.46, 95%CI 0.23-0.9, p = .025).

We found no association between insulin requirements and adverse composite outcome in women with GDM. However, those with higher insulin requirements have poorer glucose control and higher rates of CS than those with lower insulin requirements. Larger studies are needed to inquire short- and long-term outcomes of insulin requirements on fetal and maternal outcomes.