**Background:** Gestational diabetes mellitus (GDM) affects approximately one in seven pregnancies and programs offspring for elevated risks of metabolic disorders including obesity, type 2 diabetes, and cardiovascular disease. Leptin and adiponectin are adipose tissue-secreted hormones that regulate energy balance and insulin sensitivity. GDM has been associated with increased leptin and decreased adiponectin in newborns. Docosahexaenoic acid (DHA), an n-3 long-chain polyunsaturated fatty acid with anti-inflammatory and insulin-sensitizing properties, may potentially counter these adverse programming effects. This trial tested whether DHA supplementation in GDM women affects neonatal metabolic health biomarkers.
**Methods:** This single-center, open-label RCT was conducted at Xinhua Hospital, Shanghai, between August 2017 and March 2019. Sixty-eight singleton pregnant women with de novo GDM diagnosed by 75g OGTT at 24–28 weeks (IADPSG criteria: fasting ≥5.1 mmol/L, 1-h ≥10.0 mmol/L, or 2-h ≥8.5 mmol/L) were randomized to either 500 mg/day DHA supplementation (n=30) or standard care (n=38). Inclusion criteria: Han Chinese, age 20–45 years, natural conception, no current fish oil/DHA use. Exclusion criteria: severe illness, illicit drug use, chronic hypertension, pre-gestational diabetes, fetal congenital malformations. The intervention group received six DHA capsules (83.2 mg each) daily until delivery. Both groups received standard dietary and exercise guidance, with insulin if needed. Cord blood was collected at delivery. Primary outcomes were cord serum leptin and total adiponectin. Secondary outcomes included HMW adiponectin, IGF-1, maternal glycemic control, and birth outcomes. Leptin was measured by Invitrogen ELISA, adiponectin by ALPCO ELISA, IGF-1 by Crystal Chem ELISA, and erythrocyte DHA by GC/FID. The study had 80.6% power to detect a 0.7 SD difference at α=0.05. Analysis was by intention-to-treat using t-tests and chi-square tests.
**Key Results:** Baseline characteristics were similar between groups except for higher total triglycerides at first prenatal visit in the DHA group (2.1±0.8 vs. 1.7±0.6 mmol/L, p=0.047) and more cesarean sections in the DHA group (76.7% vs. 52.6%, p=0.041). There were no significant differences in cord serum leptin (DHA: 9.0±6.4 vs. control: 8.9±5.3 ng/mL, p=0.66), total adiponectin (14.5±4.0 vs. 15.3±5.2 μg/mL, p=0.69), HMW adiponectin (7.0±2.9 vs. 8.2±4.1 μg/mL, p=0.37), or IGF-1 (28.7±10.9 vs. 29.8±10.2 ng/mL, p=0.60). Cord blood erythrocyte DHA was non-significantly higher in the DHA group (6.62±1.05% vs. 6.45±0.96%, adjusted p=0.29). Birth weight was significantly higher in the DHA group (3,494±489 vs. 3,259±420 g, p=0.04), as were birth weight z-score (0.5±1.2 vs. -0.1±1.0, p=0.02) and ponderal index (27.9±2.3 vs. 26.9±1.5 kg/m³, p=0.03). Head circumference showed a marginal increase (34.6±1.5 vs. 34.0±1.1 cm, p=0.08). Maternal fasting and 2-h postprandial glucose were similar between groups throughout follow-up. No adverse events were reported. In a parallel comparison, GDM pregnancies had lower cord total and HMW adiponectin than euglycemic pregnancies.
**Clinical Implications:** DHA supplementation at 500 mg/day in the third trimester for GDM women did not alter neonatal leptin, adiponectin, or IGF-1 levels, which is reassuring regarding potential adverse effects on adipokine profiles. The observed increase in birth weight and ponderal index suggests possible benefits to fetal growth, though the clinical significance of this finding requires further study. The higher cesarean section rate in the DHA group warrants investigation. Limitations include small sample size (powered only for ≥0.7 SD differences), open-label design, and lack of long-term follow-up. Larger trials are needed to confirm these findings and assess long-term metabolic outcomes in offspring.