To Name :
To Email :
From Name :
From Email :
Comments :

Examining the Evidence

Q Does treating gestational diabetes improve outcomes?

August 2005 · Vol. 17, No. 8

<huc>A</huc> Yes, in this study, but the findings may not apply to the entire US population. Treatment reduced serious perinatal morbidity and improved the mother’s quality of life.

Expert Commentary

Gestational diabetes mellitus has occasionally seemed like a name looking for a disease. Screening recommendations and diagnostic criteria have been debated and changed, and it has appeared that, regardless of intervention, outcomes are the same.

This randomized trial sheds new light on the effectiveness of diagnosis and intervention, but some issues remain unclear—a fact pointed out in an editorial accompanying the study.1 My interpretation is similar to the one outlined in that editorial.

Nonstandard diagnostic criterion

Though the results are compelling and the randomized clinical trial model lends credence to the conclusions, the diagnostic test and criterion for diagnosing gestational diabetes (75-g glucose load with a 2-hour value >140 mg/dL) are not the standard in the United States, so the results may not be applicable in the US.

A real difference, or coincidence?

More adverse perinatal outcomes were reported among the “routine care” group than the intervention group. The authors did not clarify, however, whether the 5 perinatal deaths in the routine care group could be attributed to gestational diabetes or were coincidental. Also, although the difference in birth weight was statistically significant (mean weight of 3,335 g in the intervention group versus 3,482 g for routine care; P<.001), I am unsure of the clinical importance of this difference.

Other variables listed under adverse outcomes included 5-minute Apgar scores of less than 7 and admission to the NICU, both of which can be based on highly subjective criteria. No information was offered about whether—and how—such decision-making was standardized.

Was it ethical to ignore screening?

US practitioners would not ignore the results of a gestational diabetes screening test, as in this study (neither practitioners nor patients were made aware of the diagnosis). Thus, the findings shed little light on real-world practices of US ObGyns.

We also lack information on the cost (in dollar terms and morbidity) of any false-positive results.

Stick to ACOG guidelines

Gestational diabetes is an increasing problem, compounded by the obesity epidemic. Failing to screen patients, or ignoring a positive screen, would seem ill-advised, and glucose control would seem to be a prudent way to minimize maternal and perinatal morbidity. We need to determine the appropriate screening tools and diagnostic criteria, glucose values that should prompt intervention, and the optimal form of intervention, be it through diet alone or in combination with oral hypoglycemics or insulin.

Until these questions are resolved (probably not within this decade), I suggest we continue to follow ACOG guidelines for diagnosis and management.2

Dr. Legro has received grant support from the American Heart Association, Crown, General Mills, and Pfizer and is a consultant for Abbott and Ortho-McNeil. Dr. Greenberg and Dr. Repke report no financial relationships relevant to these articles.


1. Greene MF, Solomon CG. Gestational diabetes mellitus—time to treat. N Engl J Med. 2005;352:2544-2546.

2. American College of Obstetricians and Gynecologists. ACOG Practice Bulletin #30: Gestational Diabetes. Washington, DC: ACOG; September 2001.