Poster Australasian Diabetes in Pregnancy Annual Scientific Meeting 2013

Maternal vitamin D deficiency, gestational diabetes mellitus and pregnancy outcomes: results including a multivariate logistic regression model from a multi-ethnic population of women in South-Western Sydney (#110)

Min Ling 1 , Tang Wong 1 2 , Glynis P Ross 1 , Gabriel S Gabriel 2 3 , Adedapo Oni 1 , Cathy Finneran 1 , Nikki Edghill 1 , Jane Payne 1 , Jeff R Flack 1 2
  1. Department of Diabetes and Endocrinology, Bankstown-Lidcombe Hospital, Sydney, NSW 2200
  2. University of New South Wales, Sydney, NSW 2052, Australia
  3. Ingham Institute for Applied Medical Researcj, Liverpool, NSW 2170, Australia

Background: Vitamin D deficiency has been linked to impaired glucose metabolism1,2.

Aim:  To investigate the link between vitamin D deficiency and impaired glucose metabolism by studying the relationship between serum 25-hydroxyvitamin D (25[OH]D) and clinical indicators for glycaemic control and pregnancy outcome in gestational diabetes mellitus (GDM).

Methods:  Retrospective analysis of prospectively collected data (Feb-2011 to Apr-2013) on GDM women diagnosed by ADIPS criteria, with glycated haemoglobin (HbA1c) and 25[OH]D collected shortly thereafter. Logistic regression was used to identify predictors associated with vitamin D deficiency.

Results:  There were 628 women from diverse ethnic backgrounds (31.4% South-East Asian, 23.4% European, 23.2% Middle Eastern, 15.3% Indian/Pakistani, 6.7% Other), diagnosed with GDM at (mean±SD) 24.6±6.1 gestational weeks. Vitamin D deficiency (2013 ANZBMS guideline) was present in almost half (48.2%), with mild (30-49nmol/L), moderate (12.5-29nmol/L) and severe (<12.5nmol/L) degree in 61%, 34% and 5% respectively. Compared with vitamin D sufficient women, women with vitamin D deficiency had significantly higher fasting blood glucose level (BGL) (r=-0.22, p<0.001), HbA1c (r=-0.20, p<0.001), gravida (p=0.022), parity (p=0.006), pre-pregnancy body mass index (BMI) (p<0.001), number of GDM risk factors (p=0.005), likelihood of requiring insulin therapy (p<0.001) and LGA babies (p=0.038). There were no significant differences between subgroups in women’s age, gestation at GDM diagnosis, 2-hour BGL on oGTT, pregnancy weight-gain, gestational age at delivery, caesarean delivery or oGTT/HbA1c at 6-8 week post-partum. In the multivariate logistic regression model, pre-pregnancy BMI (p<0.001) and HbA1c (p=0.003) were statistically significant predictors of vitamin D deficiency.

Conclusion:  This study identified a high prevalence of vitamin D deficiency in a multi-ethnic population of GDM women, with greater insulin requirement and more LGA babies, but no other significant differences in birth outcomes or follow-up glucose tolerance in the two subgroups. Both pre-pregnancy BMI and HbA1c were independent predictors for vitamin D deficiency by multivariate regression analysis. Further exploration of these findings is warranted.

  1. Serum 25-hydroxyvitamin D and glycated haemoglobin levels in women with gestational diabetes mellitus. SL Lau, JE Gunton, NP Athayde, K Byth and NW Cheung. MJA 4 April 2011; 194(7); 334-337.
  2. Maternal vitamin D deficiency, ethnicity and gestational diabetes. RJ Clifton-Bligh, P McElduff and A McElduff. Diabet Med 28 June 2008; 25(6); 678-684.