TOPLINE: 

A selected panel of cardiometabolic biomarkers measured in plasma samples collected at 10-14 weeks’ gestation predicted the risk for gestational diabetes with promising accuracy in a US study of pregnant women. These early measurements matched the results from fasting plasma samples taken at 15-26 weeks’ gestation.

METHODOLOGY:

• Diet and lifestyle changes in the first trimester can effectively control maternal hyperglycemia and reduce risks for gestational diabetes and adverse neonatal outcomes; early prediction using cardiometabolic biomarkers and metabolites can enable timely intervention.
• In a US cohort of pregnant women enrolled before 13 weeks of gestation, 91 cardiometabolic biomarkers were measured in random plasma at 10-14 weeks and in fasting plasma at 15-26 weeks to improve the early prediction of gestational diabetes.
• Of the participants, 107 developed gestational diabetes and were matched to 214 control individuals who did not.
• Candidate biomarkers were categorized by clinical accessibility: group 1 (clinically accessible tests), group 2 (clinically accessible biomarkers upon request), and group 3 (targeted metabolomics requiring a specialty lab for assessment).
• At each visit, all candidate biomarkers and conventional predictors (age, race/ethnicity, pre-pregnancy BMI, family history of diabetes, and plasma glucose levels) were evaluated to build risk prediction models for gestational diabetes. 

TAKEAWAY:

• A comprehensive model comprising conventional predictors and biomarkers (A1c, insulin-like growth factor binding protein -2, leptin, summed odd- and even-chain saturated fatty acids, glycine, and aspartic acid) achieved an area under the receiver operating characteristics curve (AUROC) of 0.842 at 10-14 weeks’ gestation.
• Another full model comprising conventional predictors and biomarkers (A1c, soluble leptin receptor, summed odd-chain saturated fatty acids, polyunsaturated fatty acid 22:4n-6, glycine, and aspartic acid) achieved an AUROC of 0.829 at 15-26 weeks’ gestation.
• Among women classified as high-risk for gestational diabetes by the 10- to 14-week full model, 69.5% developed the condition, compared with 49.1% of those classified as high-risk by a model using only conventional predictors ; additionally, among individuals predicted by the full model to be at a low risk for gestational diabetes, the majority (79.6%) were control individuals who did not develop the condition.
• A decision curve analysis showed that the full model at 10-14 weeks’ gestation provided the highest net benefit, underscoring its utility in early pregnancy.

IN PRACTICE:

“As a novel contribution to the field, this study suggests the potential of using random non-fasting blood samples for predicting [gestational diabetes] in the first trimester. The full model using random blood samples at 10-14 GW [gestational weeks] represents potentially an earlier and more convenient prediction tool than the full model using fasting blood samples at 15-26 GW,” the authors wrote.

SOURCE:

The study was led by Jiaxi Yang, Global Center for Asian Women’s Health, Yong Loo Lin School of Medicine, National University of Singapore. It was published online on in BMC Medicine.

LIMITATIONS: 

External validation of the prediction models was not conducted. The sample size with available biomarker data was modest. Differences in fasting duration among participants at 10-14 weeks’ gestation may have introduced additional variation in levels of certain biomarkers. 

DISCLOSURES:

The study received support through grants from the Eunice Kennedy Shriver National Institute of Child Health and Human Development intramural funding and American Recovery and Reinvestment Act funding. The authors declared no conflicts of interest.

This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.