Holt et al. (1) acknowledge the cost-effectiveness of the American Diabetes Association (ADA) criteria, which are similar to our recommendation of offering all women a glucose test. They have understated the importance of the article by Landon et al. (2), which showed that active management of gestational diabetes mellitus (GDM) not only led to significant reductions in neonatal size/fat mass (which may predict future diabetes/obesity in the offspring), but also significant reductions in shoulder dystocia (1.5 vs. 4.0%), cesarean delivery (26.9 vs. 33.8%), and preeclampsia/gestational hypertension (8.6 vs. 13.6%). These findings make screening for GDM even more cost-effective.
When women who are below ADA weight criteria are excluded from screening, 10% of GDM is missed, including 16% with hyperglycemia warranting insulin therapy (3). Because 80% of low-risk pregnant women needed to be tested using ADA criteria in slimmer times, the importance of keeping systems simple to maximize implementation (as implied by the postnatal testing regimen by Holt et al.) suggests universal screening would be best during pregnancy (4). The limitations of the health economic model (e.g., it excludes benefits from a more simple implementation approach and future impact) require policy implementation to take a broader view. Although some may accept the National Institute for Health and Clinical Excellence (NICE) approach, philosophically, we, and probably many others, remain committed to supporting women to choose their screening method.
We agree that the screening timing for undiagnosed preexisting diabetes remains unresolved. The International Association of Diabetes and Pregnancy Study Groups has recently recommended screening “as early as possible” (5). The NICE recommendations advise screening at 16–18 weeks, which is not early and potentially allows the growing fetus to be exposed to significant hyperglycemia for many weeks.
We remain bemused at basing national policy on one study with a relatively small number of subjects and ignoring all the other published evidence. Our references (6) provide other evidence: up to 56% of women with diabetes postnatally would be missed using fasting glucose alone. To dismiss the importance of identifying impaired glucose tolerance with its high risk of progression (preventable in some women) could be seen as inappropriate. We do applaud the 100% follow-up of women for postnatal testing, a figure rarely achieved in any health activity. However, as the key difficulty in a glucose tolerance test is attending fasting, perhaps the optimal approach, again consistent with our philosophy of patient empowerment, is to offer women, in an informed way, the option of either test and to allow the woman to make the choice. Finally, the suggested use of A1C for screening postnatally may be inappropriate because of altered red cell dynamics and the potential for anemia during the postnatal period.
Acknowledgments
No potential conflicts of interest relevant to this article were reported.
