Association of mid-to-late pregnancy 25-hydroxyvitamin D levels with hypertensive disorders in pregnancy: a BMI-stratified prospective cohort study

BACKGROUND: Hypertensive disorders of pregnancy (HDP) are common complications that can lead to adverse maternal and fetal outcomes. Vitamin D is a fat-soluble vitamin with anti-inflammatory and vasoprotective properties that may influence the development of HDP. Body mass index (BMI) could potentially modify this association; however, studies investigating this relationship using BMI stratification remain limited. OBJECTIVE: To investigate the association between serum 25-hydroxyvitamin D levels in mid-late pregnancy and HDP, and to analyze its predictive value based on pre-pregnancy BMI stratification. METHODS: A total of 168 pregnant women from a tertiary hospital obstetric clinic between January 1, 2023, and September 30, 2024, were enrolled and stratified into normal weight group (84 cases), overweight group (50 cases), and obese group (34 cases) based on pre-pregnancy BMI. All participants underwent serum 25-hydroxyvitamin D measurement, blood pressure assessment, weight and height evaluation, and were followed up until delivery to record pregnancy outcomes. RESULTS: Comparison of baseline characteristics revealed significant differences in pre-pregnancy BMI, systolic blood pressure, diastolic blood pressure, 24-h urine protein, total cholesterol, triglycerides, and vitamin D supplement use among groups (F = 35.624, P < 0.001; χ2 = 9.873, P = 0.007), while no significant differences were observed in age, gestational week, and parity (F = 1.624, P = 0.201; χ2 = 1.873, P = 0.392). Serum 25(OH)D levels showed significant differences among groups (F = 18.934, P < 0.001). The total incidence of HDP and preeclampsia incidence demonstrated significant differences among groups (χ2 = 11.583, P = 0.003; χ2 = 8.924, P = 0.012), whereas the incidence of gestational hypertension showed no significant difference (χ2 = 4.237, P = 0.120). Multivariate logistic regression analysis indicated that decreased 25(OH)D levels (per 5 ng/mL) were associated with increased HDP risk (Wald = 7.234, P = 0.007), while obesity (Wald = 8.362, P = 0.004), overweight (Wald = 5.102, P = 0.024), and elevated systolic blood pressure (Wald = 13.362, P < 0.001) were associated with increased risk, and vitamin D supplement use was associated with reduced risk (Wald = 6.432, P = 0.011). The area under the curve (AUC) for serum 25(OH)D in predicting HDP was 0.768, with the highest AUC in the obese group (0.808). CONCLUSION: Serum 25-hydroxyvitamin D levels in mid-late pregnancy are negatively associated with HDP risk, and vitamin D supplement use was associated with lower risk. Exploratory BMI-stratified analyses suggested a numerically strongest association and predictive value in obese women. However, the small sample size and event count in this subgroup resulted in wide confidence intervals, high risk of overfitting, and unstable estimates, necessitating extreme caution in interpretation and mandating external validation in larger cohorts.