Table 2.

Randomized controlled trials that sought associations between iron supplementation in pregnancy and GDM.

First Author, Year of Publication	Details	Key Results	Comments
Chan, 2009 [27]	Women with baseline hemoglobin concentrations of between 8 and 14 g/dL attending Hong Kong University Queen Mary Hospital, Hong Kong were randomly allocated to receive 60 mg of elemental iron supplement (as 300 mg ferrous sulphate) daily (n = 565) or placebo (n = 599) from their pregnancy hospital booking visit. There was no difference in baseline hemoglobin or ferritin levels, or between the ratio of ferritin to transferrin between the groups. Iron supplementation (or placebo) was administered for a further 16 weeks.	In the comparison of those that supplemented with iron against those that took the placebo, there was no significant difference in the incidence of GDM at 28 weeks’ gestation (~11% in both groups; odds ratio for GDM 1.04 (0.7–1.53), p = 0.9).	There was a relatively low general compliance and involvement in the follow-up. ~40% were lost to follow-up. Maternal ferritin (p = 0.003) and hemoglobin (p < 0.001) concentrations were higher in the iron supplement group at delivery. Offspring birth weight was increased ~100 g in iron-supplemented pregnancies (p = 0.001).
Ouladsahebmadarek, 2011 [28]	A double-blind-randomized clinical trial of 960 women attending Al Zahra Hospital in Isfahan, Iran, with singleton pregnancies in the first trimester who had not supplemented with iron in the preceding month and who had hemoglobin concentrations >12 g/dL. The supplemental group (n = 480) took 30 mg elemental iron/d plus an unspecified multivitamin preparation from 13 weeks of pregnancy until delivery. The control group (n = 480) took a placebo plus the multivitamin preparation.	In the comparison of those that supplemented with iron against those that took placebo, there was no significant difference in the incidence of GDM (0.5% vs. 0.8%, respectively; relative risk 0.67 (0.11–3.97), p = 0.7).	85% of those that supplemented with iron and 78% of those who took placebo completed the course. GDM was recorded by a questionnaire, not from hospital notes or by performing oral glucose tolerance tests.
Kinnunen, 2016 [29]	Women from Tampere and five neighboring municipalities in Southern Finland were randomly allocated to receive 100 mg iron supplement daily (n = 1336; “routine” group) throughout pregnancy (regardless of their hemoglobin level) or no supplementation (n = 1358) (unless they were diagnosed with anemia (“selective” group), in which case they were supplemented with 100 mg/d (as two doses of 50 mg iron) just until their hemoglobin level increased to 11 g/dL).	In the comparison of those that routinely supplemented with iron against those that either did not or only did so selectively, there was no difference in the incidence of a composite variable related to glucose intolerance (GDM, glycosuria, and large-for-gestational-age baby): 11.0% v. 13.0% (p = 0.1), respectively.	This was a reanalysis of an original trial where GDM was not a planned outcome and therefore was not assessed systematically in all participants. Instead, GDM, glycosuria, and/or large-for-gestational-age records were abstracted from hospital records and combined into a composite variable. Few participants were overweight or obese.