Comment on: Au et al. Body composition is normal in term infants born to mothers with well-controlled gestational diabetes mellitus. Diabetes Care 2013;36:562-564.

The article by Au et al. (1) is a wonderful example of the research that is required to enable us to determine the optimal glucose treatment targets for women with gestational diabetes mellitus (GDM), however GDM is defined. The article by Au et al. deals with fetal outcomes. The infants born to women with tightly controlled GDM are smaller than the infants born to women with normal glucose tolerance, although none of the differences are statistically significant. If these differences were real, a simple calculation, using the average SD (that is the average SD of the GDM and normal glucose tolerant groups) to make a power calculation, suggests that to have 80% power to detect a difference in weight, the number needed is 101, and for head circumference the number is 107. This suggests that if these differences were real, this study would not have detected them as being significant. The readers should take this into consideration when interpreting these results. Previous work has suggested that lower levels of glycemia during treatment of GDM can result in infants that are too small (2).

It should be noted that the difference in weight between the infants of women with the treated GDM and those with normal glucose tolerance is greater than the difference in the two recent large, randomized controlled trials of treating GDM, in which the treated groups were compared with groups of untreated women with GDM (3,4).

More work is required to determine the optimum glucose treatment targets for women with GDM. For the moment, we should use those defined by Landon et al. (4), fasting glucose <5.3 mmol/L and a 2-h glucose <6.7 mmol/L but using clinical judgment including assessment of fetal growth (abdominal circumference) to decide on the targets for pharmacological therapy (5). Women with GDM are at risk for having large, overgrown fetuses. However, not all fetuses are affected, and affected fetuses can be detected using ultrasound. It seems reasonable to limit tighter treatment targets to women who have an overgrown fetus. This will limit pharmacological therapy to the subgroup of fetuses that may benefit from tight control.