Hyperglycaemia in pregnancy: what's in a name?

We can all agree that maternal hyperglycaemia, below the threshold of overt diabetes, is associated with adverse pregnancy outcomes. The key finding from the landmark Hyperglycaemia and Adverse Pregnancy Outcome (HAPO) study was that there is a continuous linear relationship between maternal hyperglycaemia and increased frequency of delivery by Caesarean section, large for gestational age infants, neonatal hypoglycaemia and cord blood C-peptide (a marker of infant adiposity) 1. This continuous relationship persisted across a range of secondary outcomes, including pre-eclampsia, preterm delivery, shoulder dystocia/birth injury and neonatal care. The associations were independent of other risk factors, including maternal BMI, and did not differ between centres or countries. However, while HAPO confirms the contribution of maternal hyperglycaemia to pregnancy outcomes, there is ongoing controversy about what degree of hyperglycaemia warrants a diagnosis of ‘gestational diabetes’. A diagnosis, or definition, can be expressed in words as well as in numbers, and at least some of the problems raised by the recent studies might be resolved by changing the words. If the condition is named ‘gestational diabetes mellitus’, defined as any degree of glucose intolerance with onset or first recognition during pregnancy—as it has been for the past 30 years—this does not fit with the proposed new cut-off numbers. Four points are inconsistent: (1) any degree of glucose intolerance would still include mothers with more severe hyperglycaemia who would now be considered to have overt diabetes mellitus; (2) glucose intolerance by itself implies the need for a glucose tolerance test, which strictly is unnecessary if the pregnancy risk is simply attributable to hyperglycaemia; (3) first recognition during pregnancy serves to miss pre-existing but hitherto unrecognized Type 2 diabetes, which should have been diagnosed and treated before, or in very early pregnancy; (4) diabetes is an inappropriate term for what is simply a risk factor for pregnancy outcome—most of these mothers do not have diabetes, and there is a greater need to avoid over-medicalization. For these reasons, the HAPO study definition ‘hyperglycaemia in pregnancy’ is preferred—’maternal hyperglycaemia less severe than that in diabetes mellitus, but associated with increased risks of adverse pregnancy outcome’. In this issue of the journal, Hartling et al. evaluate the impact of different glycaemic thresholds on maternal/fetal outcomes 2. Thirty-eight prospective or retrospective cohort studies were included, in addition to the untreated controls from randomized trials. To avoid the confounding effects of treatment, they include only women who were not treated for hyperglycaemia. This provided 13 US studies, 10 from the UK, two each from Canada and Australia and 11 from other countries. They include a wide range of diagnostic criteria spanning four studies using the recent HAPO and International Association of Diabetes and Pregnancy Study Groups (IADPSG) criteria, six with National Diabetes Data Group (NDDG), eight World Health Organization (WHO) and 17 with Carpenter–Coustan (CC) criteria. The glucose thresholds from HAPO were based on a 1.5 or 2.0 increase adjusted odds ratio relative to the mean cohort glucose levels in large for gestational age, C-peptide and infant adiposity, while the IADPSG were based on an adjusted odds ratio of 1.75 3. Three studies considered to be methodologically strong, demonstrating a continuous positive relationship between maternal hyperglycaemia and maternal/infant outcomes, were highlighted. In a prospective cohort study of 3505 unselected women, Sacks et al. were unable to identify a specific glucose threshold following a 75-g oral glucose tolerance test, which identified women with macrosomic infants 4. The Toronto Tri-Hospital study (3367 women with hyperglycaemia below NDDG criteria) found independent associations with macrosomia and Caesarean delivery, but again without specific glucose thresholds 5. The HAPO study (25 505 women from 15 centres across nine countries) examined the impact of less severe maternal hyperglycaemia (fasting glucose < 5.8 mmol/l) and did not identify a clear glucose threshold for increased risk of clinically important outcomes 6. With their large sample size, the HAPO investigators were able to compare the pregnancy outcomes of women with hyperglycaemia in pregnancy based on different thresholds (odds ratio 1.5, 1.75 and 2.0) with those of women with lower glucose values. They found fewer cases of pre-eclampsia, Caesarean delivery, neonatal hypoglycaemia or neonatal hyperbilirubinemia among women with glucose values lower than any of the odds ratio criteria [with fewer cases of shoulder dystocia/birth injury only using the more stringent criteria (odds ratio 1.5, 1.75)]. There were more Caesarean deliveries among women meeting almost all diagnostic criteria (HAPO 1.5, 2.0, IADPSG 1.75, NDDG one high value, WHO, CC). This relationship persisted even when health professionals were blinded to the glucose values. For pre-eclampsia, the diagnostic criteria were less consistent; clear associations using HAPO (1.5, 2.0), IADPSG (1.75) and CC, but not for the WHO or NDDG criteria. This could be attributable to different definitions between studies, with only two studies reporting their definition for pre-eclampsia. For maternal birth trauma, maternal weight gain and post-partum glucose status, comparisons were based only on single studies. In terms of fetal outcomes, there were significantly more cases of macrosomia for women meeting most diagnostic criteria (CC, NDDG, NDDG one abnormal value, WHO), but not IADPSG. This could also be attributable to different definitions as there were significantly more cases of large for gestational age infants in mothers meeting IADPSG, NDDG and WHO criteria. There were significantly more cases of shoulder dystocia using most diagnostic thresholds, while only the HAPO odds ratio 1.5, odds ratio 2.0, IADPSG odds ratio 1.75 and NDDG one abnormal value were associated with neonatal hypoglycaemia. Data comparing the impact of different diagnostic thresholds on longer-term metabolic outcomes related to fetal programming are insufficient. The authors conclude that higher glucose thresholds did not consistently demonstrate increased risk for all outcomes. The HAPO odds ratio 2.0 represents the threshold at which the impact of maternal hyperglycaemia on birthweight meets or exceeds that of maternal BMI. While this cut-off would prevent the much-feared IADPSG ‘epidemic’, others will argue that it represents healthcare rationing rather than evidence-based rationale. Whilst a single threshold for diagnosing hyperglycaemia in pregnancy is appealing, dichotomizing what is so very clearly a continuous variable will inevitably cause controversy. In real-life clinical practice, we are faced with unprecedented numbers of pregnant women for whom hyperglycaemia, poor diet, sedentary lifestyle, social disadvantage and overweight/obesity are all combined. In this era of personalized medicine, it is time to move beyond the arbitrary definitions and consider the entire spectrum of hyperglycaemia in pregnancy, as an additional risk factor for each and every woman. None. None declared.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.