Background: Elevated iron stores, reflected in high serum ferritin (SF) levels have been associated with type 2 diabetes and gestational diabetes mellitus (GDM). However, data on whether elevated SF reflects inflammation or excess iron stores is conflicting.
Aim: To determine whether first trimester SF is associated with subsequent GDM in the absence of inflammation.
Methods: Information for NSW women with archived serum samples collected for first trimester Down syndrome screening in 2007 at a state laboratory was linked to birth and hospital discharge records. Maternal weight and gestational age at testing were derived from laboratory data, maternal characteristics from birth records and GDM diagnosis from hospital data. Serum was analysed for SF (µg/L) and C-reactive protein (CRP; mg/L). Women with inflammation (CRP >5) or Type 1 or 2 diabetes were excluded. SF was used to measure iron stores with SF>200 defined as high, indicating iron overload. Medians (1st, 3rd quartile range) were calculated for SF concentrations. The association between iron stores and GDM was examined using multivariate logistic regression.
Results: Of 3,073 pregnancies, 96 (3.1%) developed GDM. Median SF concentrations was 26.3 µg/L (14.8, 43.9) for all women and were significantly higher for GDM (36.5 µg/L: 18.7, 54.4) compared to non-GDM pregnancies (25.9 µg/L: 14.7, 43.5, p= 0.001). Mean CRP levels were not different between GDM and non-GDM pregnancies (11.8 vs. 10.4 mg/L, p=0.17). Eleven women had iron overload, none of whom developed GDM. Adjusting for age, gestational age, weight, smoking, parity and multiple pregnancies, the odds of developing GDM was 1.37 (95% CI:1.03, 1.83) for every µg/L increase in SF concentrations.
Conclusions: In absence of inflammation, moderately elevated iron stores are associated with GDM and may be a useful early clinical marker for risk of GDM. Confirmation of these findings using other biochemical measures of excess iron is warranted.