Fibroblast growth factor 21 (FGF21) is implicated in metabolic regulation and has been associated with obesity, impaired glucose tolerance, and insulin resistance in nonpregnant populations. The clinical significance of perinatal changes in FGF21 and whether those changes relate to postpartum metabolic outcomes in women with gestational diabetes mellitus (GDM) remain unclear. This study aimed to clarify perinatal FGF21 dynamics and their association with early postpartum metabolic risk among women diagnosed with GDM.
This investigation was a single-center retrospective observational study that leveraged prospectively collected residual serum samples. The cohort comprised 45 women with GDM and 30 control women without GDM. The analysis was longitudinal, with serum samples obtained at three perinatal timepoints for each participant: approximately 26 weeks of gestation (T1), approximately 36 weeks (T2), and within 7 days after delivery (T3).
Serum FGF21 concentrations were measured at the three predefined timepoints: T1 (~26 weeks), T2 (~36 weeks), and T3 (within 7 days postpartum). Postpartum metabolic outcomes in the GDM group were assessed using the homeostasis model assessment of insulin resistance (HOMA-IR). The abstract does not provide assay methods, measurement units, or laboratory reference ranges; those methodological details were not reported in the source abstract.
Longitudinal changes in serum FGF21 across timepoints were evaluated using the Friedman test with post hoc comparisons. Predictive performance of FGF21 for postpartum insulin resistance in the GDM group was assessed by receiver operating characteristic (ROC) analysis; bootstrap-derived 95% confidence intervals were reported for area under the curve (AUC) estimates. Additional statistical details (covariate adjustment, handling of missing data, exact definitions or cutoffs for postpartum insulin resistance beyond HOMA-IR use) were not provided in the abstract.
Across both the GDM and control groups, serum FGF21 concentrations increased significantly from T1 to T2 and remained elevated at T3 based on the Friedman test with post hoc comparisons. The pattern of rising FGF21 in late pregnancy and persistence into the immediate postpartum period was observed irrespective of GDM status.
Within the GDM group, FGF21 measured during pregnancy showed predictive ability for postpartum insulin resistance as estimated by HOMA-IR. The highest predictive performance occurred at the T1 timepoint (approximately 26 weeks), with an AUC of 0.871 and a bootstrap 95% confidence interval of 0.732–0.973. The abstract does not report specific sensitivity, specificity, or optimal cutoff values, nor does it report whether multivariable models were used to adjust for potential confounders.
Although FGF21 concentrations remained elevated at the early postpartum timepoint (T3) overall, the study observed a decline in the early postpartum period. Specifically, T3/T1 ratios were significantly lower in samples collected on postpartum Days 5–7 compared with Days 1–4, suggesting an early postpartum downward trend in circulating FGF21 within the first week after delivery.
The authors interpret these findings to suggest that perinatal dynamics of FGF21 may mirror physiological changes in insulin sensitivity during late pregnancy and the early postpartum period. In women with GDM, FGF21 measured during pregnancy—particularly around 26 weeks—may serve as a complementary biomarker to help stratify postpartum metabolic risk, as indicated by its predictive association with postpartum HOMA-IR.
This study is single-center and observational in design. The abstract notes that further validation is required before clinical application of FGF21 measurement for postpartum risk stratification. Several methodological and outcome details were not reported in the abstract, including assay specifics, precise HOMA-IR thresholds used to define postpartum insulin resistance, adjustment for confounders, and longer-term postpartum follow-up. Those omissions limit assessment of generalizability and clinical readiness.
In this cohort of 45 women with GDM and 30 controls, serum FGF21 increased from mid- to late-pregnancy and remained elevated into the early postpartum period. In women with GDM, pregnancy FGF21—notably at ~26 weeks—demonstrated the strongest ROC performance for predicting early postpartum insulin resistance (AUC 0.871; bootstrap 95% CI 0.732–0.973). Early postpartum samples suggested a decline in FGF21 over the first week after delivery. The authors conclude that perinatal FGF21 dynamics may reflect changes in insulin sensitivity and that pregnancy FGF21 has potential as a complementary indicator for postpartum metabolic risk stratification in GDM, pending further validation.
Note: Specific methodological parameters, assay methods, exact HOMA-IR cutoffs, extended follow-up outcomes, and multivariable analyses were not reported in the source abstract and therefore are not provided here.