We have seen 414 women this study period for their initial evaluation. Of the women seen during this study period, 116 have developed NIDDM. The range of ages for women at the last follow-up is 15-55 years. 39.6% of the women were hypertensive, 36.8% had never smoked and 80.1% reported at least 12 years as the highest grade ever completed. A multivariate analysis revealed that diabetic type during pregnancy, insulin resistance (as assessed by the MFSIVGTT), BMI LMP for GDM pregnancy, family history of diabetes and hypertension were significant risk factors for NIDDM. Women with hypertension were 2.3 times more likely to be NIDDM (95% CI 1.40, 3.89); women with NIDDM had a history of a 1.21 higher HOMA index (95% CI 1.13, 1.29) compared to non-NIDDM women; women with NIDDM had a higher BMI LMP than non-NIDDM women (OR=1.04 95% CI 1.00-1.08); women who required insulin control during GDM pregnancy were more likely to develop NIDDM (OR=4.09, 95% CI 2.45-6.85); and women with a family history of diabetes were 2.9 times more likely (95% CI 1.56-5.49) to be NIDDM compared to non-NIDDM women. Insulin resistance. The HOMA index for insulin resistance is calculated as (insulin)/22.5 e -In glucose assuming that normal weight subjects < 35 years have an insulin resistance of 1.0. Beta-cell function can be approximated by the formula: '-cell function (%)= (20 x insulin)/ (glucose-3.5). In our AA poulation with a history of GDM, 551 values are available to calculate HOMA. The mean HOMA is 5.3, range 0.03- 106.2. The upper value is skewed due to a patient with an insulin of 153.1. Repeat insulin and glucose values are available at two separate time points for 265 women preceding NIDDM. The Pearson correlation coefficient is 0.32 for HOMA. The Pearson correlation coefficients for HOMA and '-cell function at the first time point is 0.70 and at the second time point is 0.35. It does appear that HOMA and '-cell function change over time in this cohort. These studies are revealing that African American women who present with gestational diabetes are at increased risk for NIDDM. If further studies confirm and/or identify more specific markers these can be used to identify those at risk long before the disease occurs. This poulation can be targeted for preventive interventions.
Showing the most recent 10 out of 570 publications
