Routine screening for gestational diabetes: indication of a positive effect confirmed

Gestational diabetes is already diagnosed in several thousand pregnant women in Germany each year. However, not all of these women can be sure that this diagnosis is really justified and will help them and their babies to prevent problems in pregnancy and birth. "In Germany there is a muddle of tests and diagnostic criteria for gestational diabetes," says Peter Sawicki, Director of the German Institute for Quality and Efficiency in Health Care (IQWiG). "We do not know whether these interventions show more benefit than harm in all of these women." An IQWiG working paper now shows a way out of this muddle. IQWiG researchers evaluated current studies investigating the benefit of diagnosis and treatment of gestational diabetes in women and their offspring. The working paper updates a final IQWiG report published in September 2009, which was commissioned by the German Federal Joint Committee (G-BA). External experts and IQWiG researchers have also recently published the results in the British Medical Journal.

Conclusions of the final report confirmed

A US study, which was published in October 2009 and investigated nearly 1000 women, was newly included in the working paper. This study confirmed that women benefit from treatment for gestational diabetes, therefore reinforcing one of the conclusions in the final IQWiG report. For example, by changing their diet and in specific cases by injecting insulin, pregnant women can reduce the risk of shoulder dystocia during birth. This refers to a delay in the delivery of the baby, which occurs in 1 to 2 out of 100 births. Most cases of shoulder dystocia are without consequence. However, as there is a risk of the baby suffering a lack of oxygen, midwives and doctors often quickly take countermeasures that can lead to injuries to the mother and baby. In the studies evaluated, the number of heavy babies (over 4000 g) was reduced by about half. This is in line with the fact that babies with a high birth weight have an increased risk of shoulder dystocia.

At least one prerequisite for routine screening fulfilled

The working paper also confirmed the final report's conclusion: by proving the benefit of treatment, an essential prerequisite is provided for offering all pregnant women routine screening for gestational diabetes.

However, as was the case with the final report, IQWiG's literature search again failed to identify studies directly showing that routine screening is of more benefit than harm in all pregnant women. Women would need to take part in such studies before the diagnosis of gestational diabetes was confirmed, which also means including pregnant women for whom the test would be futile, as their blood glucose levels are low. This constitutes the vast majority. However, the analysed studies only included women in whom gestational diabetes had been diagnosed by a test, i.e. the minority of pregnant women.

The standard: a 2-step diagnosis

IQWiG therefore exactly analysed the tests by which the pregnant women had been selected, as this is how the studies defined the threshold between "healthy" women vs. those "needing treatment". It was noticeable that in all relevant studies the diagnosis of gestational diabetes was made in a 2-step selection process.

The basic principle is that women drink a glass of glucose solution, followed by the taking of one or more blood samples to determine the blood glucose level ("glucose tolerance test"). However, there are different types of tests. In the studies included, nearly all pregnant women had initially undergone a "short" tolerance test lasting only an hour. Only those women (about 25%) who had an elevated blood glucose level in the first test were asked to come back for a second test, a full glucose tolerance test.

For this 2- to 3-hour test, the women were asked to fast before attending the doctor's surgery. The suspicion of gestational diabetes was not confirmed in most cases. Only the small group of pregnant women whose blood glucose levels also exceeded a threshold value in the second test started specific treatment for gestational diabetes. "We now definitely know that women selected by this 2-step strategy can profit from this treatment," says Sawicki.

Diagnosis can also have disadvantages

In the March issue of the journal "Diabetes Care", an international group of doctors nevertheless recommended dispensing with the first test and inviting all pregnant women to a 2-hour tolerance test instead. However, this proposal is problematic, as it is unclear whether women diagnosed in this way have the same benefit from treatment. In addition, the expert group set a rather low threshold for the blood glucose level at which gestational diabetes is diagnosed, so that according to the international HAPO Study, about 18 out of 100 pregnant women would suddenly be classified as "sick persons". This would affect about 100,000 women per year in Germany. "We do not have data for Germany to be able to estimate the consequences of such an intervention," says Sawicki. The diagnosis of gestational diabetes can in itself be a burden for a mother-to-be, simply by the fact that her pregnancy is classified as a high-risk pregnancy. In addition, she would be expected to check her blood glucose levels several times a day.

Solution: a comparative screening study

According to IQWiG, a comparative study of different strategies would be the most rational solution for determining the benefit-harm ratio of screening. In a high-quality study, pregnant women should also be asked about their preferences, as it has not yet been investigated which strategy they prefer. "Such a study could soon be conducted if doctors and health care funds agreed with each other," says Peter Sawicki. "Before tens of thousands of pregnant women are labelled with a potentially burdensome diagnosis, we should know exactly what we are doing."

Contact: Tel. ++49(0)221-35685-0, info@iqwig.de