To examine the association between different gestational weight gain (GWG) patterns and the risk of gestational diabetes mellitus (GDM), and to identify high-risk weight gain characteristics that may inform GWG management and early prevention of GDM.

A prospective cohort study was conducted among 312 pregnant women who completed follow-up at the Obstetrics Department of the First Affiliated Hospital of Shanxi Medical College between January 2022 and December 2024. According to the Institute of Medicine (IOM) recommendations, and based on pre-pregnancy body mass index(BMI) and weight changes across the first, second, and third trimesters, latent class growth analysis (LCGA) was used to classify GWG trajectories into three patterns: steady-appropriate, moderately rapid, and persistently rapid. GDM was diagnosed using the results of a 24-28-week oral glucose tolerance test (OGTT). Multivariable logistic regression models were applied to assess the independent associations between GWG patterns and GDM, and model performance was evaluated using the Hosmer-Lemeshow goodness-of-fit test and receiver operating characteristic (ROC) curve analysis.

The overall incidence of GDM was 22.44% (70/312). Among all participants, 55.13% (172/312) showed a steady-appropriate GWG pattern, 31.09% (97/312) showed a moderately rapid pattern, and 13.78% (43/312) showed a persistently rapid pattern, with corresponding GDM incidences of 14.53% (25/172), 29.90% (29/97), and 37.21% (16/43), respectively (P<0.001). Multivariable analysis showed that compared with the steady-appropriate pattern, the moderately rapid pattern was associated with an increased risk of GDM(OR=1.68, 95% CI: 1.02-2.77, P=0.041), and the persistently rapid pattern showed a stronger association (OR=2.95, 95% CI:1.54-5.64, P=0.001). Pre-pregnancy BMI (OR=1.10, 95% CI: 1.01-1.21), family history of diabetes (OR=2.13, 95% CI: 1.13-4.00), and low physical activity level (OR=1.89, 95% CI: 1.04-3.44) were also independently associated with GDM (all P<0.05). The Hosmer-Lemeshow test indicated good model fit (P=0.537). The area under the ROC curve was 0.781 (95% CI: 0.726-0.835), with a sensitivity of 72.86% and a specificity of 73.44%.

GWG patterns are closely associated with GDM. Monitoring abnormal weight gain trends during pregnancy may have indicative value for metabolic health management.