The aim of this study was to investigate the association between maternal iron/zinc serum levels and their nutritional intake in early pregnancy with gestational diabetes. The maternal serum zinc/iron levels were measured in 1,033 healthy singleton pregnant women aged 20-35 between 14 and 20 weeks of gestation, within two groups: namely, normal and gestational diabetes, and participants were followed up to 24-28 weeks of gestation. Food frequency questionnaire was used to assess nutritional intakes of iron/zinc. The main outcome was gestational diabetes screened with the 50-g glucose challenge test and diagnosed with oral glucose tolerance test at 24-28 weeks of gestation. Gestational diabetes occurred in 72 (6.96 %) of 1,033 women in study. There was a statistical relationship between early pregnancy maternal serum iron and gestational diabetes, mean (SD), 143.8 (48.7) vs. 112.5 (83.5) μg/dl, P value of <0.0001. There was no statistical significant difference in zinc levels and iron/zinc nutritional intake between groups. The results remained unchanged after using regression model for adjustment of potential risk factors with an adjusted OR of 1.006 (95 % CI 1.002 to 1.009; P = 0.001) for early pregnancy maternal serum iron to cause gestational diabetes. The receiver-operator characteristic curve identified that a maternal serum iron above 100 μg/dl in early pregnancy is the optimum cutoff value for predicting gestational diabetes, which showed a sensitivity and specificity of 80.6 and 50.7 %, respectively. In conclusion, high maternal serum iron in early pregnancy could increase the risk of gestational diabetes. Also, it could be used as a sensitive and specific predictor for gestational diabetes.