A Randomized Controlled Trial Investigating the Effects of a Low-Glycemic Index Diet on Pregnancy Outcomes in Gestational Diabetes Mellitus

Jimmy Chun Yu Louie, MNUTRDIET, APD; Tania P. Markovic, MBBS, PHD; Nimalie Perera, MBBS; Deborah Foote, BSC DIPNUTRDIET, APD; Peter Petocz, PHD; Glynis P. Ross, MBBS; Jennie C. Brand-Miller, PHD


Diabetes Care. 2011;34(11):2341-2346. 

In This Article


Contrary to our hypothesis, this randomized controlled trial of an LGI diet versus a conventional high-fiber diet found no differences in key pregnancy outcomes in GDM. Average infant birth weight, birth weight centile, and ponderal index were within healthy norms in both groups. One explanation for the findings is that both groups of women achieved a relatively LGI diet, with only a modest 5-point difference between groups. Irrespective of dietary assignment, all had received early nutrition counseling in a group setting. Thus, on enrollment, both groups were found to be consuming a diet with a lower GI than population norms. Compared with routine care in another Australian study,[18] both dietary interventions resulted in a lower prevalence of LGA (9 vs. 22%), macrosomia (4 vs. 21%), and emergency caesarean section (16 vs. 20%). Hence, in the setting of intensive medical management of GDM, our findings suggest that both an LGI and HF diet produce optimal pregnancy outcomes.

Our findings increase the evidence supporting the safety and efficacy of an LGI diet in GDM. Moses et al.[12] also found no significant differences in key fetal and obstetric outcomes between subjects who followed an LGI diet (GI = 48) versus a higher-GI diet (GI = 56). However, unlike in the current study, they found that a significantly higher proportion of women in the higher-GI group met the criteria to commence insulin (59 vs. 29% in the LGI group). In addition, almost one-half of the women in the higher-GI group who met the criteria for insulin commencement avoided insulin by switching to an LGI diet. Their insulin treatment protocol, however, was different from that of the current study, in which more stringent criteria were used as the basis for insulin treatment.

A recent Canadian study,[19] in which women with GDM or impaired glucose tolerance monitored their own blood glucose levels, found that those who were randomized to an LGI diet versus those assigned to the conventional diet had a greater proportion of their 2-h postprandial levels on or below the treatment target. Although there was a tendency for higher birth weight in the control group, the study was a pilot and underpowered to detect a statistically significant difference.

Another explanation for our findings is the relatively normal weight of most of our subjects (68% had a BMI <25 kg/m2). It is possible that an LGI diet may be more effective among overweight and obese gravidas with higher degrees of insulin resistance and β-cell deficiency.[20] Rhodes et al.[21] reported higher head circumference and a lower proportion of early delivery (<38 weeks' gestation) in overweight and obese nondiabetic pregnant women assigned to a low GL diet. However, there was no significant difference in birth weight, ponderal index, or pregnancy weight gain, which are more sensitive to maternal glycemic control.[22]

The lack of difference in our study may also relate to the timing and duration of the intervention. Dietary instruction began at the start of the third trimester (29 weeks' gestation) and lasted, on average, 6–7 weeks. It is likely that maternal hyperglycemia during the first and second trimester will also drive excessive fetal growth. In a post hoc analysis of women who started dietary intervention before 25 weeks of gestation (10 from the LGI group and 5 from the HF group), those in the LGI group showed a tendency to lower birth weight (LGI 3.2 ± 0.2 kg vs. HF 3.5 ± 0.1 kg; P = 0.224) and lower birth centile (LGI 45.3 ± 11.0 vs. HF 57.5 ± 12.2; P = 0.476), suggesting that an earlier intervention may be beneficial. However, apart from a small number of high-risk women who are screened early, in most countries GDM screening occurs at 26–28 weeks' gestation,[23,24] which means that any intervention in GDM will be necessarily short. A more viable test of our hypothesis would therefore be an appropriately powered study in women at high risk of developing GDM (e.g., women with a BMI >30 kg/m2 or previous GDM), starting on or before the start of the second trimester, to determine the effect of an LGI diet on both pregnancy outcome and risk of developing GDM.

The failure to achieve the target GI of ~60 in the HF group could reflect high recognition of the GI concept among Australians diagnosed with diabetes, particularly among those with higher education (in the current study, two of three subjects had a university degree). In the group education session conducted soon after diagnosis, all the women, irrespective of future dietary assignment, were encouraged to limit total carbohydrate to ~180 g per day and to consume a greater proportion as fruit and dairy products—changes which are likely to lower the GI of the overall diet. Self-monitoring of blood glucose levels was also encouraged and may have provided feedback that discouraged consumption of high glycemic foods. Finally, the use of data collected from medical record may be subject to inaccuracy, e.g., birth weights were measured and entered by different staff, therefore biasing the result toward the null hypothesis.

In conclusion, we found that both an LGI diet and a conventional high-fiber diet produced comparable pregnancy outcomes in women with GDM. Both groups achieved a relatively low GI diet and had mean birth weight, birth weight centile, and pregnancy weight gain within population norms. An LGI diet appears to be a safe alternative to the traditional pregnancy diet for women with GDM and expands the range of dietary strategies that can be offered. Further studies in overweight and obese individuals and earlier interventions in women with risk factors for GDM are warranted.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: