Fetal exposure to hyperglycemia is a major determinant of large-for-gestational-age (LGA; birth weight >90th centile for gender) neonates (1), yet targets for glycemic control beyond the first trimester in type 1 diabetes (T1D) pregnancy remain controversial. As HbA1c represents a summary measure of glycemic control, it might not adequately reflect acute glucose fluctuations or glycemic variability (GV) that contributes to excess fetal growth. Moreover, neonates born to women who attain HbA1c <6% (42 mmol/mol) in the third trimester of pregnancy have an LGA prevalence of 25% (2), with associated adverse perinatal outcomes (3). In contrast to HbA1c, continuous glucose monitoring (CGM) allows precise observation of GV. Several studies have demonstrated an association between higher GV and increased birth weight (1,4,5). The capability of GV compared with HbA1c to identify women likely to have LGA neonates is, however, unclear. We evaluated the association between various measures of GV, HbA1c, and LGA neonates in T1D pregnancy.
Twenty-one pregnant women with T1D were recruited over a 2-year period, and measurements of HbA1c and GV (EasyGV, University of Oxford, Oxford, U.K.) using CGM (Medtronic, Macquarie Park, New South Wales, Australia) were carried out at three time points: 14–18, 24–28, and 32–36 weeks' gestation. The study was approved by the local ethics committee (Northern Sydney Local Health District Human Research Ethics Committee), and all participants gave written informed consent.
The mean ± SD gestational age at delivery was 37.5 ± 1.4 weeks, and birth weight was 3,454 ± 576 g, with eight neonates born LGA. HbA1c at each time point was 6.1 ± 0.9%, 6.0 ± 0.8%, and 6.2 ± 1.0%, respectively. The LGA group had significantly higher mean glucose and GV indices than the non-LGA group at 24–28 weeks’ gestation (Fig. 1A–D) and higher HbA1c at each time point (Fig. 1E).
Indices of glycemic variability and HbA1c at 14–18, 24–28, and 32–36 weeks of gestation were compared between women with and without LGA neonates using the Mann-Whitney U test. Results presented as mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001. CONGA-4, continuous overall net glycemic action at 4 h.
Indices of glycemic variability and HbA1c at 14–18, 24–28, and 32–36 weeks of gestation were compared between women with and without LGA neonates using the Mann-Whitney U test. Results presented as mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001. CONGA-4, continuous overall net glycemic action at 4 h.
Linear regression demonstrated a significant association between birth weight centile and each of the glycemic measures at 24–28 weeks of gestation. Because the GV indices demonstrated significant colinearity, we included the J-index alone in subsequent analyses, as it was most strongly correlated with birth weight centile (r = 0.853; P < 0.0001). Using univariate linear modeling, J-index at 24–28 weeks maintained a significant independent association with birth weight centile (r2 = 0.229; P < 0.05), whereas HbA1c did not (r2 = 0.008; P = 0.713). The combination of J-index and HbA1c at this time point resulted in a greater association with birth weight centile (r2 = 0.477; P < 0.01), with mean values of 31.7% and 5.95%, respectively. Furthermore, using cutoffs of HbA1c >6% and J-index >30 identified all neonates that were born LGA (receiver operating characteristic curve analysis, data not shown).
Despite attaining close to recommended HbA1c target levels for T1D in pregnancy (HbA1c ≤6%), our cohort of women demonstrated an ∼40% incidence of LGA neonates, which concurs with the results of recently published studies. We found that the optimal combination of glycemic measures associated with LGA neonates is J-index and HbA1c measured at 24–28 weeks’ gestation. These findings were highly statistically significant, even with a small cohort size. Consequently, our results suggest that CGM (to calculate J-index) and HbA1c measured in the late second trimester may be useful clinical tools to identify women with T1D at high risk of LGA neonates; however, this hypothesis should be confirmed in a larger cohort. Our results provide an opportunity for future studies to determine whether targeting GV, as well as lowering HbA1c, could reduce fetal overgrowth in T1D pregnancy.
Article Information
Acknowledgments. The authors acknowledge Vanessa Donnelly and Carol Palmisano of the Diabetes Education Team at Royal North Shore Hospital for assistance with CGM device insertion and data upload. They also thank all of the women who participated in the study.
Funding. This study was supported by funding from the Novo Nordisk Australia Regional Diabetes Support Scheme and the Ramsay Research and Teaching Fund, Sydney, Australia. The funders had no role in the study design or in the decision to submit the manuscript for publication.
Duality of Interest. R.T.M., S.J.G., G.R.F., and S.L.H. have received research support for investigator-initiated studies from Novo Nordisk. S.J.G., G.R.F., and S.L.H have received honoraria from Novo Nordisk. S.J.G. has received honoraria from Medtronic. No other potential conflicts of interest relevant to this article were reported.
Author Contributions. R.T.M. recruited participants, collected and analyzed study data, and wrote the first draft of the manuscript. S.J.G., S.K.S., E.S.S., and G.R.F recruited participants, collected data, and interpreted the results. S.L.H. designed the study, recruited participants, collected data, and interpreted the results. All authors contributed to the manuscript and approved the final version prior to submission. R.T.M. is the guarantor of this work and, as such, had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Clinical trial reg. no. ACTRN12614001295639, www.anzctr.org.au.
