Gestational diabetes mellitus (GDM) is an increasingly common condition during pregnancy, affecting 6–10% of all pregnancies in New Zealand and a disproportionate number of Māori and Pacific women.[[1,2]] Broadly identified as hyperglycaemia during pregnancy, GDM associates with a number of maternal and foetal complications including macrosomia, pre-eclampsia, caesarean delivery and neonatal hyperglycaemia.[[3]] It is extremely important to minimise GDM risk, including ensuring that all pregnant women are educated around the need for a nutritious, non-processed diet,[[4]] with optimal gestational weight gain.[[5]] Since 2015, national Diabetes in Pregnancy guidelines promote the need for universal screening for GDM at 24–28 weeks’ gestation following screening for undiagnosed type 2 diabetes using a glycated haemoglobin (HbA1c) test as a part of antenatal booking bloods.[[6]]

In New Zealand, the HbA1c test should be performed before 20 weeks gestation, and a result of ≤40 mmol/mol is considered normal.[[6]] In contrast, women with an elevated HbA1c of 41–49 mmol/mol are deemed to be at higher risk for GDM and are recommended to be given lifestyle advice, and have the 2-hour 75g glucose tolerance test (GTT) at 24–28 weeks rather than the usual 50g glucose challenge test (but not to commence GDM management until a positive GTT test result is obtained).[[6]] While recent New Zealand data indicate that more than 90% of pregnant women undertake an HbA1c test, up to 20% of all women (30% of Māori women and 18% of Pacific women) do not go on to have any testing for GDM.[[1]] Reasons for this vary,[[7]] though it does suggest HbA1c alone may be important to support glycaemic management in some women; indeed, it is already current practice in several New Zealand regions for women with booking HbA1c 41–49 to commence GDM intervention. It is currently unclear whether this results in improved maternal and foetal outcomes.[[8,9]]

Despite international literature suggesting that an elevated HbA1c test result associates with an increased risk of GDM,[[10]] data evaluating this association in a New Zealand context are limited.[[11]] Thus, this study aims to explore the relationship between elevated HbA1c and GDM risk in a cohort of New Zealand women to determine if those women with an elevated HbA1c should be considered for early clinical management.

Clinical data (maternal age and ethnicity) were obtained and combined for a cohort of women who birthed in the Waikato region (including hospitals and birthing centres) during January–December 2018 (n=4,140) and from a second cohort who birthed in the Northland region (Whangārei, Bay of Islands, Rawene and Kaitaia hospitals) during January 2020–September 2021 (n=3,671). NHI-matched HbA1c data were obtained from local laboratories and were deemed to be related to pregnancy if the date of the test was between 4 and 40 weeks gestation (or the date of delivery, whichever came first). Gestational age at the time of each HbA1c test was recorded. Where multiple HbA1c values were provided for any one patient, the first test result was used for analysis unless an elevated HbA1c of 41–49 mmol/mol was recorded later in pregnancy, in which case the first elevated test result was used. Women were classified as having a normoglycaemic or elevated HbA1c during pregnancy (≤40 vs 41–49 mmol/mol, respectively). Women without a pregnancy-related HbA1c measurement, and those with an HbA1c of ≥50 mmol/mol (indicative of type 2 diabetes) were excluded from the study.

Clinical information was also obtained from Te Whatu Ora (formerly Waikato and Northland District Health Boards) Diabetes in Pregnancy units for women diagnosed with GDM. Women in this dataset were matched by NHI to the study population above, and women were also classified as with or without GDM.

For analysis, women were grouped by maternal age (≤20, 21–30, 31–40 and ≥41 years), gestational age at HbA1c test (0–10, 11–20, 21–30 and 31–40 weeks), ethnicity (New Zealand European, Māori, Pacific, Asian, Other) and whether the gestational age at the time of the HbA1c test was ≤20 weeks (yes/no). A logistic regression was undertaken to determine which factors impacted the likelihood of a GDM diagnosis for all women, and separately for those with an HbA1c “before” compared to “after” 20 weeks’ gestation. In the logistic regression, the outcome variable was the absence/presence of GDM. The independent predictors included HbA1c test, ethnicity, gestational age and maternal age. Data were analysed in Stata with p<0.05 accepted as significant.

In total, 5,084 women were included in the study and the majority of women had their first HbA1c test prior to 20 weeks (88.9%). This was lower for Māori (82.8%) and Pacific (79.2%) women compared to NZ Europeans (93.2%; p<0.05) or Asian women (91.3%; p<0.05).

Overall, 88 women (1.7%) had an elevated HbA1c of 41–49 mmol/mol and 324 women had a diagnosis of GDM (6.4%). Fifty-five of the 88 women with an elevated HbA1c (62.5%) went on to be diagnosed with GDM (including 57.1% of Māori (24/42), 66.7% of Pacific (4/6), 65.2% (15/23) of Asian and 55.6% (5/9) of NZ European women. Of the remaining 33 women with an elevated HbA1c, 17 (51.5%) completed a GTT and did not have GDM, five (15.2%) returned a negative glucose challenge test result and had no GTT and 11 women (eight Māori, two NZ European, one Pacific and one Asian) had no evaluation for GDM. Based on those with GTT data only, the likelihood of an elevated HbA1c resulting in a diagnosis of GDM was 76.3% overall and 71.4%, 64.7% 80.0% and 68.2%, respectively, for NZ European, Māori, Pacific and Asian women. Logistic regressions showed that after adjustment, women with an HbA1c result of 41–49 mmol/mol were more likely to have GDM with an odds ratio (OR) of 23.60 at HbA1c ≤20 weeks (p< 0.001) and an overall (anytime) OR of 16.67 (p< 0.001) (Table 1). Pacific and Asian women were more likely to have a higher risk of GDM compared to NZ European women (OR: 2.556, CI: 1.42–4.618, p<0.01 and OR: 4.67, CI: 3.45–6.32, p<0.001 respectively) as did Māori women when the HbA1c was completed before 20 weeks (OR: 1.53, CI: 1.05–2.23, p<0.05). Similarly, women of older maternal age (>30 years) were more likely to have an increased risk of GDM compared to those aged 21–30 (Table 1; all p<0.05).

With logistic regressions undertaken separately for each ethnic group, NZ European women (n=2,500) with an elevated HbA1c were more likely to have a higher probability of GDM diagnosis than those with an HbA1c of <40 mmol/mol (OR: 19.90, CI: 4.70–84.40, p<0.001). Māori wāhine (n=1,533) with an elevated HbA1c also had an increased risk of GDM (OR: 18.99, CI: 9.28–38.86, p<0.001). The odds ratios of Pacific (n=154) and Asian women (n=666) are 12.94 (p<0.05) and 11.62 (p<0.001) respectively.

We acknowledge that this was a small study and that the results may have been skewed by the omission of women for whom we had no HbA1c or GTT results. Low sensitivity of an early HbA1c to predict GDM at 24–28 weeks has been previously described,[[11,12]] but our results show that the majority of women with an elevated HbA1c during pregnancy who are subsequently screened do develop GDM. Although this is a relatively small group of women (1.7% in our cohort), our study suggests that all women with an elevated booking HbA1c should be encouraged to complete a GTT and should be considered for early targeted management. These results are timely given that the national Diabetes in Pregnancy Guideline is currently being reviewed and a well-designed randomised, controlled trial is needed to inform whether those with an elevated HbA1c in pregnancy should be treated.

View Table 1.