Abstract
BACKGROUND: Vitamin B plays a crucial role in pregnancy, influencing maternal and fetal health. However, the impact of Vitamin B metabolism alterations on pregnancy outcomes remains unclear. This study investigates the association between Vitamin B metabolism and adverse pregnancy outcomes (APO) using a retrospective approach. METHODS: A total of 1,086 pregnant women were included, categorized into normal and APO groups. Vitamin B markers during pregnancy, including red blood cell (RBC) and plasma levels of folate (5MTHF), Vitamin B2 (VB2), and Vitamin B6 (VB6), as well as methylmalonic acid (MMA) and homocysteine (HCY), were analyzed. LASSO regression was used for feature selection, followed by logistic regression to develop a predictive model. The model's performance was evaluated using ROC curve, calibration curves, decision curve analysis, and external validation. RESULTS: Higher plasma-5MTHF, RBC-5MTHF, RBC-VB2, and RBC-VB6 levels were significantly associated with normal pregnancy outcomes, whereas elevated MMA and HCY levels correlated with an increased risk of APO (P < 0.05). In addition, maternal age, pregnancy type, pre-BMI, and passive smoking were identified as significant baseline factors. By integrating these baseline characteristics with Vitamin B-related biomarkers, the predictive model demonstrated excellent performance, with an AUC of 0.85 (95% CI: 0.818-0.883) in the training set and 0.807 (95% CI: 0.738-0.876) in the external validation set. CONCLUSION: Alterations in Vitamin B metabolism are significantly associated with pregnancy outcomes. The developed predictive model provides a valuable tool for risk assessment in clinical practice.