Long-term diabetogenic effect of single pregnancy in women with previous gestational diabetes mellitus

Abstract

Background: Pregnancy is associated with marked insulin resistance that seems to have little, if any, impact on the long-term risk of non-insulin-dependent diabetes mellitus (NIDDM) in the general population. The aim of this study was to test whether pregnancy would alter the risk of NIDDM among women with a high prevalence of pancreatic beta-cell dysfunction, as indicated by a history of gestational diabetes mellitus.

Methods: The cohort consisted of 666 Latino women with gestational diabetes attending a high-risk family planning clinic. They were followed up for up to 7.5 years, during which time they were weighed and underwent an oral glucose-tolerance test annually. The effect of an additional pregnancy, and of other risk factors for diabetes, was examined.

Findings: 87 (13%) of the women completed an additional pregnancy. 80 of those women did not have NIDDM immediately after the additional pregnancy and their subsequent annual incidence rate of NIDDM was 30.9% (95% CI 12.7-49.1), more than 2.5 times the annual incidence rate of NIDDM in the cohort overall (11.9%; 95% CI 10.0-13.8). Proportional hazards regression analysis using the presence or absence of an additional pregnancy as a time-dependent variable confirmed that an additional pregnancy increased the rate ratio of NIDDM to 3.34 (95% CI 1.80-6.19), compared with women without an additional pregnancy after adjustment for other potential diabetes risk factors during the index pregnancy (antepartum oral glucose tolerance, highest fasting glucose, gestational age at diagnosis of gestational diabetes) and during follow-up (postpartum body mass index [BMI], and glucose tolerance, weight change, breast feeding, and months of contraceptive use). Weight gain also was independently associated with an increased risk of NIDDM; the rate ratio was 1.95 (95% CI 1.63-2.33) for each 10 lb (4.5 kg) gained during follow-up after adjustment for the additional pregnancy and the other potential risk factors.

Interpretation: The study showed that a single pregnancy, independent of the well-known effect of weight gain, accelerated the development of NIDDM in a group of women with a high prevalence of pancreatic beta-cell dysfunction. This finding implies that episodes of insulin resistance may contribute to the decline in beta-cell function that leads to NIDDM in many high-risk individuals.