Assessment of pancreatic beta-cell function during pregnancy is important because gestational diabetes mellitus (GDM) involves both insulin resistance and impaired insulin secretion. Model-based evaluation of beta-cell function is commonly performed using a five-sample oral glucose tolerance test (OGTT), but whether fewer sampling points can provide comparable estimates in pregnancy remains uncertain.

An analysis published in Diabetology evaluated whether the three-sample OGTT protocol (0, 60, and 120 minutes) could provide comparable model-based beta-cell function estimates to the five-sample OGTT reference protocol. The study included pregnant women undergoing a 2-hour OGTT with glucose, insulin, and C-peptide measurements in early and/or mid-pregnancy, generating 152 OGTT datasets. Using the five-sample OGTT as the reference model, three beta-cell function parameters were derived: glucose sensitivity (GSENS), potentiation factor ratio (PFR), and rate sensitivity (RSENS). These parameters were recalculated using values from the three-sample OGTT and compared with the reference estimates.

For GSENS, regression analysis showed conditional R² 0.56 and marginal R² 0.75 (p < 0.0001). Bland-Altman analysis showed 94.2% of samples within limits of agreement, and subgroup analyses showed consistent GSENS estimates between the two protocols. Results for PFR were less satisfactory but considered acceptable, whereas RSENS estimates derived from the three-sample OGTT were not reliable.

These findings indicate that the three-sample OGTT may provide acceptable estimates for selected model-based beta-cell function parameters in pregnancy, although reduced sampling does not reliably capture all indices.