Long-term risk prediction of diabetes mellitus using the 100-g oral glucose tolerance test during pregnancy: Insights into test patterns, single abnormal values, and abnormal fasting glucose.

Introduction: Gestational diabetes mellitus (GDM) is defined as an abnormal glucose tolerance first diagnosed during pregnancy. The 100-g oral glucose tolerance test (OGTT), applied exclusively during pregnancy, is used to diagnose GDM. GDM is associated with long-term maternal complications, including the development of diabetes mellitus (DM). This study aimed to investigate the relationship between various patterns of abnormal OGTT results and the subsequent development of DM, emphasizing the predictive value of both the number and timing of abnormal glucose measurements. Methods: A retrospective cohort study was conducted, including pregnant individuals who underwent a 100-g OGTT. Participants were followed for up to 11 years after the test. Exclusion criteria encompassed preexisting DM, steroid use, and incomplete OGTT data. Two models were analyzed: the first compared pathological OGTT patterns (≥2 abnormal results) to normal OGTTs (<2 abnormal results), while the second examined DM risk based on the number and timing of the abnormal values (i.e., fasting, 1-, 2-, or 3-h postglucose ingestion). Statistical analyses utilized hazard ratios derived from multivariable Cox regression models, adjusting for relevant variables. Results: Among 7645 participants, 6.4% developed DM during follow-up. Those with a pathological OGTT were older, more obese, and had higher rates of hypertension and hyperlipidemia. During follow-up, 21.6% of people with ≥2 abnormal results developed DM compared to only 4.6% in the group with <2 abnormal results (hazard ratio [HR] 2.03, 95% confidence interval [CI] 1.85-2.23). Compared to those with all four results normal, the HRs for developing DM with one, two, three, or four abnormal results were 3.2 (95% CI 2.54-4.03), 4.89 (95% CI 3.82-6.26), 8.18 (95% CI 6.0-11.16), and 15.46 (95% CI 9.21-25.93), respectively. Among people with only one abnormal result, abnormal fasting glucose showed the highest risk HR = 7.12 (95% CI 5.10-9.94). Conclusion: The study demonstrates the predictive value of the 100-g OGTT for future DM development. Even a single abnormal result, particularly the fasting glucose, significantly increases risk, with risk escalating as additional abnormalities occur. These findings underscore the need for proactive postnatal follow-up and early interventions to mitigate long-term metabolic risks.