Gestational Diabetes

"A major public health problem" or a " diagnosis looking for a disease"?

Medical opinion varies widely about the condition called gestational diabetes and considerable controversy and confusion exists over its definition, its significance, the necessity of screening all pregnant women and the effectiveness of treatment.

Before 1994, gestational diabetes amongst the general population of pregnant women was rarely mentioned, let alone diagnosed.  Around that time however, it became apparent that the screening test for gestational diabetes had become widespread in New Zealand and there was a growing demand for information on the condition.

So what is it that makes gestational diabetes such a confusing and controversial matter and why has it been added to the growing list of tests for which pregnant women are routinely referred?

 Definition of gestational diabetes          

There is international agreement that gestational diabetes is diabetes (or glucose intolerance) that is first recognized in pregnancy and in New Zealand, occurs in 2 - 5% of pregnant women.  There is however disagreement over whether the diagnosis should only include those women whose glucose tolerance reverts to normal following pregnancy or should it include all women who are glucose intolerant and this is first recognized in pregnancy?

The reason for this disagreement is that in communities with a high incidence of glucose intolerance in the general population, many of those who are identified as having gestational diabetes actually have pre-existing and previously unrecognized non-insulin dependent diabetes.  As it is estimated that the proportion of undiagnosed diabetes in New Zealand is around 50%, this may explain why at National Women’s Hospital, some 40 - 50% of women diagnosed as having gestational diabetes have been found to have permanent diabetes (either insulin or non-insulin dependent) afterwards.  As the prevalence of gestational diabetes is closely linked to the prevalence of glucose intolerance in the general population, it would also appear to be a good reason to promote better general screening programs for diabetes rather than concentrating on pregnant women.

Screening Tests?          

 The initial or "routine' test is the Oral Glucose Challenge, better known as the polycose test.  There is considerable variation internationally in the criteria for the polycose test. T he polycose loading ranges from 50 - 100g, the wait before testing is 1 - 3 hours and the blood glucose concentration on which the provisional diagnosis of gestational diabetes is made also differs. The difficulty seems to be to set the blood glucose level at a level which will detect Impaired Glucose Tolerance but which not lead to unnecessarily large numbers of women going on to have the more stringent Oral Glucose Tolerance Test (OGTT)

In New Zealand women drink 50g of glucose or polycose and have their blood tested after 1 hour.  If the 1-hour blood glucose is greater than 7.8mmol/l, then the woman is referred for the Oral Glucose Tolerance test (OGTT).  Diabetic specialists consider that the polycose test blood glucose level of greater than 7.8 mmol/l is at an acceptable level in its detection of those who may have the condition and its ability to identify those who are normal.

In the OGTT test, the woman fasts overnight and her blood glucose levels are tested with her not having eaten for 12 hours. A level of greater than 5.5 mmol/l is regarded as the woman having gestational diabetes.   She is then given 75g of polycose to drink and her blood glucose level is further tested after 2 hours. Should this be above 9.0 mmol/l, then again she is regarded as having gestational diabetes.

When the blood glucose level after two hours in the OGTT is over 9.0 mmol/l as stated above, the number of women diagnosed as having gestational diabetes would be between 4 - 5%.  As the prevalence of gestational diabetes for European women is around 2% and for Maori and Pacific Island women around 5%, this would seem reasonable and not result in over-diagnoses and in many essentially normal women being subjected to unjustified intervention at considerable cost.

A criticism of both the polycose test and the OGTT is that they may not be "reproducible" - that is if the test were repeated the next day or the next week, different results would be obtained.   The OGTT has been particularly criticized in this respect and is said to be non-reproducible in at least 50% of cases. One study gave two OGTT's one to two weeks apart to 64 pregnant women who had "failed" the polycose test. Of these 64 women, 16 had one abnormal OGTT but only two tested abnormal on both OGTT's.

There is debate as to why women test abnormal one week and normal the following week when the same test is used.   However, it is known that glucose tolerance is adversely affected by low carbohydrate diet preceding the test, illness, bed rest, certain medications and anxiety.

Are the screening tests harmful?          

Health professionals associated with testing for gestational diabetes state that in general, women were pleased with the availability of the tests and did not find them too stressful.  They also state that women do not find the polycose drink unpleasant.  Anecdotal evidence however suggests that most women have not been stressed by the polycose test, but "failing" it and being referred to take the fasting OGTT has been an extremely stressful experience. Many have also reported the polycose drink to be "disgusting" and all would have preferred to eat several candy bars!

Some concern has also been expressed about the effects of polycose loading, particularly when done in the morning. The concerns relate to very complex metabolic issues but are essentially about the effects of increased insulin response from the pancreas and the consequent loss of both chromium and zinc.  It is believed that the loss of these minerals, particularly zinc, can lead to a lowering of resistance to infection. 

 Implications for mother          

This really depends on whether the mother has diabetes that will disappear after pregnancy or whether her gestational diabetes is previously unrecognized diabetes mellitus.  There would be little disagreement that an insulin-dependent woman, whether pregnant or non-pregnant, should be prescribed insulin as soon as possible or life-threatening complications will occur.

The treatment of the non-insulin dependent woman is less straightforward however.  While more severe gestational diabetes is linked to poorer fetal outcome, mild gestational diabetes has not yet been proven to be harmful to either mother or unborn baby.  Once a woman has been diagnosed as having the condition, however mild and however doubtful the benefits of treatment, she is classified as high risk, which reduces her options and choices with respect to both pregnancy and birth.  She also becomes subject to constant testing and many interventions, all of which can be both stressful and inconvenient.

It is known that women who are diagnosed as having gestational diabetes have a much higher incidence of developing diabetes mellitus than the general population.  A further OGTT is done at 6 weeks post-partum and of those women who are shown to be severely glucose intolerant at this time, they will either already have non-insulin dependent diabetes or have an 80 - 90% chance of developing it within 5 years.  A further significant figure is that of the women with Gestational Diabetes who were retested at 6 - 8 weeks at National Women's Hospital in 1994-5, 31% of them had either Impaired Glucose tolerance or Non-insulin Dependent Diabetes.

Commencing treatment earlier could reduce many health professionals believe that with the knowledge that they are at higher than normal risk of developing diabetes, these women should be targeted with information on recognition of early symptoms of diabetes and thus the number of diabetic complications.  It is also believed that these at-risk women may delay the onset of diabetes by greater attention to diet and by making appropriate lifestyle changes such as increasing exercise and stopping smoking.

 Implications for baby          

Again this depends on the type of diabetes the mother has.  The perinatal (around birth) death rate and incidence of congenital abnormality amongst babies whose mothers have untreated insulin-dependent diabetes is high.  For the mothers of these babies, it is crucial that control of blood sugar levels should be controlled as early in pregnancy as possible.  As the screening test is not performed until around 28 weeks, it is clear that these women need to have been identified by other risk factors much earlier in the pregnancy and appropriate treatment commenced.

For the babies of mothers with milder glucose intolerance, the picture is much less clear.  It is known that a larger than average baby and a number of conditions in the newborn such as low blood sugar, jaundice and low blood calcium, may be associated with maternal gestational diabetes.

The significance of these however is also the subject of debate. More than 80% of large babies will be born to mothers who do not have gestational diabetes and the latter three conditions are not considered harmful except in very specific circumstances.   There is further controversy as it is argued that there have been no reputable studies to demonstrate clearly that these conditions are linked to mild gestational diabetes or rather to other "confounding" factors such as maternal obesity and age, family history of diabetes, and previous pregnancy loss.

Treatment for mild gestational diabetes consists of attention to diet and exercise.  Controlled trials of "dietary regulation" however have not demonstrated a significant effect on any outcome with respect to reduced perinatal death or sickness.  In trials where dietary regulation has been combined with the use of insulin, there has been a decrease in the incidence of large babies, although the rates of caesarean section and perinatal deaths have not dropped.  These trials have not shown any significant reduction in the rates of neonatal jaundice or low blood calcium.

Infants born to mothers with gestational diabetes have been identified as having a higher than normal risk of developing obesity and non-insulin dependent diabetes mellitus in later life.   However, it is not clear whether this is due to the maternal gestational diabetes or to a genetically inherited disposition or other environmental factors.

"At risk" women          

In June 1996, there was a "consensus" meeting with representatives from a wide range of special interest groups, including the NZ Society for the Study of Diabetes, the Colleges of Midwives and Obstetricians and Gynecologists, various Maori and Pacific Island groups, and consumer groups such as Parents Center.  This meeting discussed many of the issues of concern and accepted there was still much research to be done into such matters as what is the level at which a mother's blood sugar poses a risk to the unborn baby and what are the most beneficial forms of treatment?

Recommendations from the meeting included continued use of the polycose and Oral Glucose Tolerance tests and that the policies and practices with regard to gestational diabetes should be continually reviewed. It was also recommended that there should be universal screening for all non-European women as the incidence of all forms of diabetes was much greater in the Maori, Pacific island, Indian and Asian communities. Screening should also be offered to European women with known risk factors which include:

It is recognized that there is considerable uncertainty about the magnitude and severity of gestational diabetes, about the effectiveness and the reference standards of the screening tests, and about the effectiveness of treatment.  It is also questioned whether the benefit of a universal screening program is of sufficient magnitude and clinical value to justify routine screening of all pregnant women.

Many clinicians advocate a selective screening approach, with an emphasis on individual risk factors and monitoring of mother and baby during pregnancy.  While there is debate over the value and benefit of universal screening, particularly given the very doubtful benefit of diagnosing mild cases of gestational diabetes, there is little debate over the value of universal screening for those women deemed to have high risk factors.

Sharron Cole