These pages are best viewed with Netscape version 3.0 or higher or Internet Explorer version 3.0 or higher. When viewed with other browsers, some characters or attributes may not be rendered correctly.

Fourth International Workshop-Conference on Gestational Diabetes Mellitus

Overview and commentary on first session

Jeremy J.N. Oats, MBBS, DM, FRCOG, FRACOG

The first three workshop-conferences on gestational diabetes mellitus concentrated particularly on defining gestational diabetes mellitus (GDM) and making recommendations on screening and diagnosis. With the opening of the fourth workshop-conference, these issues have become even more controversial, hence Dr. Gabbe's conclusion in his summation of the previous three meetings and expectations for this meeting that establishing internationally acceptable criteria and methodology for the diagnosis of GDM remain the key priorities. Until these are established, selection of the appropriate degree of hyperglycemia to initiate management strategies will also remain problematic. This remained a constant theme throughout the conference and features prominently in the summaries and recommendations that arose from the meeting.

Carpenter and Coustan's review of the diagnosis of GDM further highlighted the problems with the currently used criteria: they are not based on the likelihood of adverse perinatal outcome, but either on the potential development of diabetes postpregnancy or on an adaptation of nonpregnancy criteria. Furthermore, changes in methodology have led to mathematical adjustments to the original O'Sullivan and Mahan criteria, which has resulted in a significant rise in the diagnostic thresholds.

Universal, compared with indicated, testing for GDM in women from both high- and low-prevalence societies was examined by a number of groups. McCance et al. from Belfast, Northern Ireland, concluded that there was little value in routine screening in low-prevalence areas, whereas Humphreys from Cairns, Australia, found no difference in perinatal outcome between those who had a positive glucose challenge test and a subsequent positive glucose tolerance test (GTT) and those who had a subsequent negative GTT, except that the intervention rates were higher in those with GDM. This raises the dilemma of whether the morbidity associated with the pregnancies of those who are designated as having GDM is a result of their management rather than from maternal and fetal hyperglycemia. To further confuse the issue, later diagnosis of GDM by routine screening at 26–28 weeks compared with screening at 32 weeks in those with perceived risk factors was shown by Firth and colleagues from Dublin to be associated with doubling of macrosomia, a higher cesarean section rate, preeclampsia, prematurity, and admission to special-care nursery. Again this raises the question: is the adverse outcome due primarily to the associated risk factors that initiated the diagnostic test or is it the result of the maternal and fetal hyperglycemia?

There are still no reliable short cuts for screening. Two groups, Maahs et al. from Albuquerque, New Mexico, and Reichelt et al. from Porto Alegre, Brazil, both reported that fasting blood glucose testing was not a useful screening tool. Carr stated that the new generation of portable blood glucose monitors were sufficiently accurate to permit their use in screening programs, particularly in regions without ready access to biochemistry laboratories.

The question of the impact of lesser degrees of hyperglycemia than those currently required for the diagnosis of GDM was addressed by several groups. From the perspective of pregnancy outcome, the preliminary review of the Toronto Tri-Hospital study showed that untreated borderline GDM patients had twice the rate of macrosomia and a 50% increased likelihood of being delivered by cesarean section compared with those who have normal glucose tolerance. In addition, there were increased incidences of preeclampsia, need for phototherapy, and longer hospital stay for both mother and baby. For the treated GDM patients, the birth weights were normalized, but the cesarean section rate remained much higher. Using one abnormal value of the oral GTT, Jang and colleagues from Korea reported that 27% of their subjects developed GDM by 32–34 weeks; the only predictive factor for this deterioration was a lower 1-h post–glucose load insulin level. Sixty percent of women with one abnormal value had insulin secretions and responses below the normal curve for pregnancy, and 73% remained below normal postpartum. This compared with 75% of those with GDM being below normal during pregnancy and 80% remaining abnormal 3–6 months postpartum. These studies support the contention of those who believe that the currently used criteria are too restrictive.

The etiology and pathogenesis of GDM still remains enigmatic. There is no evidence that it is a disease of autoimmune origin (studies by Kühl, Carvalheiro, Alevizaki, and Cypryk). The persistence of abnormal insulin sensitivity postpartum in women with GDM was reported by Berkus et al. The possibility that it is the pregnancy itself that causes long-term damage to insulin metabolism was canvassed by Kjos and colleagues; they concluded that subsequent pregnancies in women with prior GDM triples the risk of subsequent emergence of diabetes. They also reported that exposure to hormonal contraception or hormone replacement therapy does not increase the risk of diabetes. This is a clinically important finding, because when these women become postmenopausal, they are at increased risk of cardiovascular disease and should therefore be targeted for consideration of hormone replacement therapy. Dornhorst and Rossi called for urgent implementation of lifestyle intervention programs to reduce the development of diabetes in this high-risk group with previous GDM. By delaying type 2 diabetes for 6 years, the risk of significant sight-threatening retinopathy would be reduced by 65%.

Finally, Dr. King cautioned that from the epidemiologist's viewpoint, the pregnant state may only be implicated as a causal factor if the glucose intolerance reverts to normal after the pregnancy. Furthermore, the estimated frequency of the condition can be influenced by the degree of screening, which is particularly important in communities where type 2 diabetes develops at a relatively early age.

Copyright © 1998 American Diabetes Association
Last updated: 8/98
For ADA Related Issues contact CustomerService@diabetes.org

For Technical Issues contact webmaster@diabetes.org