Women with lower serum levels of vitamin D during the first trimester of pregnancy are at greater risk for developing gestational diabetes mellitus (GDM) later in pregnancy, according to Marilyn Lacroix, a master's degree candidate from the Faculty of Medicine and Health Sciences at the University of Sherbrooke in Quebec, Canada. She reported the results here at the European Association for the Study of Diabetes (EASD) 48th Annual Meeting.
The association between vitamin D level and GDM risk was independent of age, season of blood sampling, vitamin D supplementation, and adiposity of the mother.
The relationship had been shown before, but previous study results had been inconsistent because of potential confounding factors, which this study took into account.
The active metabolite, 1,25-dihydroxyvitamin D, formed from 25-hydroxyvitamin D (25OHD), is involved in calcium balance and bone metabolism, acts as a transcription factor, and functions in glucose metabolism.
Women aged 18 years or older (n = 655) who were in their sixth to 13th week of pregnancy and in good health were recruited from the Sherbrooke area, which lies at about 45 degrees north latitude and therefore gets relatively less sun exposure than more southerly locations. Women with known type 1 or type 2 diabetes, diagnosed GDM in the first trimester, multiple pregnancy, or a history of miscarriage or drug or alcohol abuse were excluded from the study.
After the presentation, an audience member asked whether prior GDM was an exclusion criterion, noting that women with a history of GDM are at greater risk of developing it again. Lacroix replied that such women were not excluded.
The researchers made anthropometric measurements and determined 25OHD levels at the time of recruitment. Between the 24th and 28th weeks of pregnancy, they performed a 75-g fasting oral glucose tolerance test (OGTT) to determine normal glucose tolerance or GDM according to criteria of the International Association of the Diabetes and Pregnancy Study Groups (fasting glucose ≥ 5.1 mmol/L; 1 hour post-OGTT glucose ≥ 10.0 mmol/L; 2-h post-OGTT glucose ≥ 8.5 mmol/L).
Predictors of GDM
Fifty-four women (8.2%) were found to have developed GDM. At baseline, the women with GDM were significantly older than the women with normal glucose tolerance (mean age, 30.4 ± 5.4 years vs 28.2 ± 4.4 years; P = .005) and had a greater waist circumference (mean, 96.0 vs 88.0 cm, respectively; P = .02). These measurements were made at equivalent gestational age (9.0 vs 8.9 weeks, respectively; P = .71).
"The mean total 25OHD levels in our cohort was about 63 nmol/L, and participants with gestational diabetes mellitus had lower levels of 25OHD compared to [women with] normal glucose tolerance," Lacroix reported. "The overall prevalence of vitamin D deficiency in our cohort was about 27%."
"We observed a significant correlation between 25OHD levels at first trimester and insulin sensitivity at the second trimester," she said. "There is an association between 25OHD levels at the first trimester and glycemic indices at the second trimester."
Lower levels of 25OHD were associated with an increased risk of incident GDM. When adjusted for age, season of blood sampling, vitamin D supplementation, and waist circumference, the risk for GDM increased by 40% for each standard deviation (SD) decrease in 25OHD level (1 SD, 18.8 nmol/L; odds ratio, 1.40/1 SD [95% confidence interval, 1.02 - 1.91]; P = .04). This result was consistent regardless of the measurement of adiposity used (waist circumference, body mass index, or percentage body fat).
At the second-trimester measurements, the women with GDM had lower insulin sensitivity than women with normal glucose tolerance, as reflected in the mean Matsuda indices of 4.9 vs 7.9, respectively ( P < .001). There was a linear correlation between the level of 25OHD and the Matsuda index ( r = 0.13; P < .001), indicating that with higher 25OHD levels, there was greater sensitivity to insulin.
The women with GDM also had lower β cell compensation as measured by the Insulin Secretion Sensitivity Index-2 (mean score, 491.1 vs 866.4, respectively; P < .001). The groups did not differ in terms of insulin secretion, as measured by the ratio of the total areas under the insulin and glucose curves ( P = .95).
The investigators did not find any significant correlation of 25OHD levels in the first trimester and insulin secretion or β cell compensation during the second trimester after adjustment for confounders.
Lacroix and colleagues concluded that lower vitamin D levels in the first trimester are associated with increased risk of developing GDM, independent of age, season of blood sampling, vitamin D supplementation, and adiposity measurements. Lower vitamin D levels are associated with insulin resistance but not with insulin secretion or β cell compensation after adjustment for confounders.
Ethnicity May Play a Role in Vitamin D Deficiency
Session moderator Anne Dornhorst, BM, BCh, from the Department of Diabetes and Endocrinology at Imperial College Healthcare NHS Trust in London, United Kingdom, commented to Medscape Medical News that vitamin D and its active metabolites are certainly involved in the expression of insulin resistance, and in pregnancy insulin insensitivity can change up to 3-fold. "People come into pregnancy with different levels of insulin resistance, but then they all will increase. And so vitamin D is very likely to be involved in this," she explained.
She said that in London women who are at higher risk for type 2 diabetes often come from the ethnic groups that are inherently vitamin D deficient, "often from lower social-economic status and groups that have probably less vitamin D in the diet and less sun exposure," including women who come from cultures in which they are covered up.
"If you take women from sub-Saharan Africa and Muslim women who are not only dark skinned but covered, you can say to yourself, is there any point measuring it? Just give them vitamin D," Dr. Dornhorst recommended.
She said also that blacks are at high risk for type 2 diabetes and gestational diabetes. Indians, too, have a very high level of vitamin D deficiency along with a high level of type 2 and gestational diabetes.
Dr. Dornhorst pointed out that a problem affecting studies of vitamin D is that the vitamin is difficult to assay; results are reported in different units; and threshold levels for deficiency, insufficiency, and vitamin D repletion vary among laboratories. "The real active metabolites are very, very difficult to measure," she said. [But studies are] going to provide when it is all unraveled quite good translational research...the importance of supplementing it and who to target to supplement."
She said a European study in progress is assessing whether vitamin D supplementation will prevent GDM.