Diagnostic criterial predict adverse pregnancy outcomes in gestational diabetes.

Taking a blood test during pregnancy

There remains some uncertainty about the best diagnostic criteria for gestational diabetes mellitus (GDM).  This systematic review set out to test the ability of different glucose levels after a 75g oral glucose tolerance test (OGTT) to predict adverse pregnancy outcomes.

Clinical question:

In women with gestational diabetes, what is the risk of adverse outcomes?

The reviewers were interested in comparing two different diagnostic criteria for gestational diabetes: the World Health Organization (WHO) and the more recent International A Diabetes Prevention Group (IADPSG) criteria.

Current practice in the UK diagnoses GDM when fasting blood glucose is > 6.1mmol/l, or 2 hours after OGTT ≥ 7.8mmol/l or above.  This is similar to the WHO criteria (fasting ≥7 mmol/l, 2 hours after OGTT ≥7.8 mmol/l).

The IADPSG criteria are ≥ 5.1 mmol/l fasting, or 1 hour after OGTT ≥ 10.0 mmol/l, or 2 hours ≥ 8.5 mmol/l.

The evidence:

The review included cohort studies that examined the association between WHO and IADPSG diagnostic criteria with perinatal and maternal outcomes.  Eight studies, comprising 44,829 participants, were found and included in the review.  Individual data was extracted from these and pooled in a meta-analysis.

The reviewers found that women diagnosed as having GDM by both sets of criteria were more likely than those who were diagnosed as not having GDM to experience macrosomia (birth weight >4,000g) or deliver infants who were large for their gestational age. They were also more likely to have pre-eclampsia and a caesarian section.

There was an observed increase in risk of perinatal mortality but it wasn’t statistically significant, due to lack of data.  More research is needed here.

However, in all cases, the predictive power of the criteria was clinically modest.

Appraisal hints:

Users of this research should consider:

  • The review included retrospective studies, which are harder to control for bias.  However, all of the studies applied the OGTT to all participants, reducing the chances of selection bias.
  • Whether it makes sense to combine the data from these different studies.  There was considerable variation in settings, populations, baseline incidence of GDM.  In some studies there were minor differences in how the test was delivered and outcomes defined.
  • The reviewers noted significant statistical heterogeneity among the three studies that reported the IADPSG criteria.
  • The analysis was restricted to untreated women.  That is, women in the studies who met the local criteria and were subsequently treated, were not included in the analysis. This in turn suggests that the study population had a much narrower range of glycaemia than you would be expect in a real-life setting.  The results may therefore underestimate the ability of the criteria to predict outcomes.
  • Not all of the studies provided data for all of the planned analyses.  This led to some imbalance in the numbers of subjects in the comparisons.
  • This review needs to be supplemented by new, prospective studies to provide greater precision and reliability of the effectiveness of these diagnostic test criteria, particularly the newer IADPSG set.


Wendland EM, Torloni MR, Falavigna M, Trujillo J, Dode MA, Campos MA, Duncan BB, Schmidt MI.  Gestational diabetes and pregnancy outcomes – a systematic review of the World Health Organization (WHO) and the International Association of Diabetes in Pregnancy Study Groups (IADPSG) diagnostic criteria.  BMC Pregnancy Childbirth. 2012 Mar 31;12(1):23. [Epub ahead of print]

Relevant UK guideline:

National Institute for Health and Clinical Excellence (NICE). Diabetes in pregnancy. Clinical guideline 63. London: NICE; 2008.

Share this post: Share on Facebook Tweet this on Twitter Share on LinkedIn Share on Google+ Share via email


I am an information scientist with an interest in making knowledge from systematic research more accessible to people who need it. This means you. I've been attempting this in the area of Evidence-Based Health Care since 1995. So far the results have been mixed. For some reason we expected busy clinicians to search databases and appraise papers instead of seeing patients. We also expected publishers to make the research freely available to the people who paid for it.. Ha! Hence The National Elf service.

More posts - Website

Follow me here –