Fetal umbilical artery thrombosis: prenatal diagnosis, treatment and follow-up

BACKGROUND: To analyze the ultrasound imaging and clinical characteristics of fetuses with umbilical artery thrombosis (UAT), explore the potential causes of UAT and construct a prognostic prediction model to guide clinical practice. METHODS: This was a retrospective cohort study of fetal UAT cases examined at two academic tertiary referral care centers from 2014 to 2020. The basic information of the participants was obtained by interview during follow-up, and data on clinical treatment, delivery conditions, diagnosis and confirmation were obtained through medical records. Probable causes of thrombosis were explored by comparative analysis of the UAT group to the control group and by further regression analysis. Multivariable logistic regression models were used to evaluate risk factors for adverse pregnancy outcomes. Receiver operating characteristic (ROC) curves were constructed to evaluate the diagnostic value of the prognostic prediction model. RESULTS: Thirty fetuses with UAT were included in this study. UAT occurred mostly in the third trimester of pregnancy, and there was an obvious predominance of right UAT. An abnormal pregnancy history (53.3%) was the most common comorbidity, followed by gestational diabetes mellitus (GDM) (20.0%). GDM and umbilical cord (UC) abnormalities were found to be independent risk factors for the development of UAT. After comprehensive decision-making, over two-thirds of the patients with UAT received urgent treatment, and less than one-third received expectant management. Surprisingly, there were no significant differences in fetal outcomes between the urgent treatment and expectant management groups. Multivariate logistic regression analysis showed that gestational age (GA) at clinical diagnosis and UC abnormalities were independent risk factors for adverse pregnancy outcomes (OR 0.781, p = 0.042; OR 16.779, p = 0.023, respectively). Based on this, we constructed a comprehensive prognostic prediction model. The area under the ROC curve (AUC) was 0.877 (95% CI 0.698–0.970; p < 0.001), which suggested that the combination of GA and UC abnormalities was a better predictor for fetal outcomes in our setting. CONCLUSION: In summary, maternal GDM and fetal UC abnormalities are independent risk factors for UAT. UAT is more frequently observed on the right side. Moreover, poor clinical outcomes for fetuses with UAT are ascribed mainly to GA and UC abnormalities, which should be comprehensively evaluated to choose the appropriate treatment.