Umbilical Artery Thrombosis in the Third Trimester of Pregnancy: A Retrospective Analysis of 13 Cases

To editor, Umbilical artery thrombosis (UAT) is a rare obstetric complication, with a reported incidence ranging from 0.0025% to 0.045%.1 UAT is associated with significant fetal morbidity, including acute fetal compromise, growth restriction, visceral infarction, neonatal encephalopathy, fetomaternal hemorrhage, and intrauterine demise. Notably, the exact etiopathogenesis of UAT remains elusive, with hypotheses implicating thrombophilic disorders, vascular endothelial injury, and hemodynamic disturbances. This case series retrospectively analyzes 13 confirmed UAT cases, aiming to elucidate clinical patterns and inform evidence-based management strategies. A retrospective cohort study was conducted involving 30,328 pregnant women who underwent delivery at our hospital between October 2022 and October 2024. The incidence of UAT in our cohort was 0.043% (13/30, 328). The mean maternal age was 30.20 ± 4.49 years. Antenatal diagnosis was achieved in 9 cases (69.2%, 9/13) via ultrasonography between 27 and 39+1 weeks of gestation (Fig. 1A). Seven cases were identified by isolated single umbilical artery (SUA) on third-trimester ultrasound, while two cases were diagnosed following sonographic suspicion of UAT (e.g., luminal filling defects, abnormal Doppler waveforms). Clinical management diverged based on disease severity: five pregnancies underwent immediate termination, whereas four cases received expectant management (median duration: 34 days; range: 4 to 55 days) with prophylactic low-molecular-weight heparin (LMWH).Figure 1.: Imaging and histopathological features of umbilical artery thrombosis. (A) Ultrasound image of umbilical artery thrombosis (indicated by arrow). (B) Cross-section of the umbilical cord shows thrombus formation within a single umbilical artery (arrow denotes the occluded vessel). (C) Thrombus identified within the umbilical artery (hematoxylin-eosin staining, ×4). HE: Hematoxylin-eosin.Maternal comorbidities included gestational diabetes mellitus (n = 4, 30.8%), adverse obstetric history (n = 4, 30.8%), antiphospholipid syndrome (n = 3, 23.1%), fetal growth restriction (n = 2, 15.4%), and hypertensive disorders of pregnancy (n = 1, 7.7%) . Fetal surveillance abnormalities were observed in 76.9% (10/13) of cases: non-reassuring fetal heart rate patterns (n = 10), reduced fetal movements (n = 7), umbilical artery doppler velocimetry abnormalities (n = 5), middle cerebral artery pulsatility index deviations (n = 4). Gestational age at delivery ranged from 30+1 to 40+1 weeks. Cesarean delivery was performed in 12 cases (92.3%), with one vaginal delivery (7.7%). All maternal outcomes were favorable without complications (Supplementary Table 1, https://links.lww.com/MFM/A112). Of the 13 live-born neonates, seven (53.8%) were preterm (< 37 weeks) and six (46.2%) were term. Two infants (15.4%) were small for gestational age (birth weight < 10th percentile). Neonatal asphyxia was observed in 5 cases (38.5%): One case presented with severe hypoxic-ischemic encephalopathy accompanied by multiple congenital malformations, for which parental consent was obtained to withdraw life-sustaining treatment. The remaining four cases were classified as mild asphyxia, all of which achieved complete clinical recovery. UAT was histologically confirmed in all cases (Fig. 1B&C). Concurrent umbilical vein thrombosis was identified in one case (7.7%). Additional cord abnormalities included excessive coiling (n = 5, 38.5%) and velamentous/marginal cord insertion (n = 2, 15.4%). Umbilical cord thrombosis incidence differs between clinical (0.08%) and pathological (0.1%) reports. Venous thrombosis is more common than arterial, possibly due to venous anatomy. UAT pathogenesis involves multifactorial interplay (anatomical predisposition, hemodynamics, maternal/fetal hypercoagulability, infection, diabetes, hypertension).2,3 Our data align with risk patterns (hypercoiling 38.5%, aberrant insertions 15.4%, diabetes 30.8%). Autoimmune comorbidities (30.8%), including antiphospholipid syndrome (23.1%), support findings that maternal autoantibodies may damage umbilical arteries.4 Umbilical arterial thrombosis constitutes a critical determinant of perinatal morbidity and mortality. The clinical severity correlates with thrombus chronicity, occlusion magnitude, coexisting placental pathology, and residual circulatory compensation. In cases of unilateral arterial thrombosis, the patent artery may partially compensate fetal perfusion, though intrauterine growth restriction development remains probable. This compensatory mechanism becomes vulnerable during mechanical compression of the umbilical cord-occurring either spontaneously through fetal movement or during intrapartum manipulation. Acute obstruction of the residual arterial flow under these circumstances may precipitate fetal distress secondary to hypoperfusion. Persistent occlusion ultimately culminates in progressive fetal deterioration, with intrauterine demise representing the most catastrophic outcome. UAT is generally asymptomatic in most cases, with only a subset presenting clinical indicators such as reduced fetal movements or abnormal cardiotocography patterns.5 Definitive diagnosis requires histopathological confirmation; however, prenatal identification relies exclusively on ultrasonography, which carries inherent diagnostic challenges. Ultrasonographic evaluation of umbilical cord cross-sectional views permits prenatal detection of UAT. A critical diagnostic pitfall arises when unilateral arterial thrombosis mimics SUA due to absent Doppler flow. This underscores the necessity for: (1) expertise in serial image interpretation; (2) comparative analysis with prior ultrasound records; (3) meticulous evaluation of vascular spatial relationships. The characteristic “an orange held by a hand” sign described by Tanaka et al.6, wherein the thrombosed artery maintains parallel alignment with the patent vessel while encircled by umbilical veins serves as a key differentiating feature. Although SUA can be identified through fetal bladder arterial flow assessment, distinguishing congenital absence from acquired thrombosis remains problematic. When SUA is newly detected in late gestation, retrospective review of first-trimester imaging becomes essential to establish developmental chronology. UAT remains a clinically challenging condition due to the absence of standardized management protocols. Current evidence does not establish optimal timing for pregnancy termination, with cesarean delivery predominantly performed in response to fetal distress. In our cohort, abnormal cardiotocography findings were documented in 10 cases, reduced fetal movements in 7, umbilical flow abnormalities in 5, and middle cerebral artery (MCA) hemodynamic alterations in 4. These observations suggest compensatory mechanisms through the patent umbilical artery may mask typical fetal distress signs, necessitating heightened ultrasonographic vigilance-particularly for subtle reductions in fetal movement. For prenatally diagnosed UAT, cesarean delivery following fetal lung maturation represents the predominant intervention.7,8 Early gestational diagnosis poses a therapeutic dilemma: immediate termination risks iatrogenic prematurity, while expectant management carries intrauterine demise risks. This underscores the critical need for meticulous fetal surveillance incorporating biophysical parameters (movement patterns, cardiotocography), hemodynamic markers (umbilical/MCA Doppler indices), and growth metrics to optimize delivery timing. Current decision-making integrates maternal-fetal status, gestational age, and neonatal care capacity, with successful pregnancy prolongation documented in select cases.9 Given the effectiveness and safety of LMWH in thrombotic diseases, LMWH treatment during the observation period may be beneficial. Existing evidence supports the therapeutic potential of LMWH administration in UAT for optimizing perinatal outcomes.10 In our study, four cases diagnosed with UAT immediately received LMWH treatment, and during the observation period, fetal growth, umbilical blood flow, and middle cerebral artery blood flow were monitored regularly. The pregnancy was extended by 4 to 55 days, with a median of 34 days. This suggests that for umbilical arterial thrombosis pregnancies diagnosed early with poor neonatal survival potential, initiating LMWH treatment as early as possible may be an effective therapeutic approach. However, due to the small sample size, further clinical validation is needed. Cesarean section is recommended to avoid umbilical compression risks during labor. Vaginal delivery remains feasible in select cases, as evidenced by one successful instance in our cohort following cervical dilatation, though this requires cautious patient selection. Pregnancies complicated by prior adverse obstetric outcomes or maternal autoimmune disorders warrant heightened vigilance for UAT development. Timely prenatal diagnosis through ultrasonographic surveillance and tailored therapeutic interventions are critical for optimizing perinatal outcomes and mitigating late-gestation fetal demise. Prenatal ultrasound remains the first choice for detecting fetal umbilical arterial thrombosis. New single umbilical arteries discovered in the late pregnancy should raise suspicion of umbilical arterial thrombosis. Pregnancy monitoring and perinatal management of umbilical arterial thrombosis should be individualized, and decisions on observation or immediate pregnancy termination should be based on gestational age and fetal condition. Early LMWH treatment may improve pregnancy outcomes, especially when the diagnosis is made early, and the fetus has poor survival potential. Acknowledgments We gratefully acknowledge the invaluable contributions of all study participants and their families. Funding None. Author Contributions The study was conceived of and designed by Ying Jin, Shimao Zhang, Xihao Wang, Wen Xiong, Dongmei Tang and Sumei Wei. Data acquisition and collection was by Ying Jin, Shimao Zhang, Xihao Wang and Wen Xiong. Data analysis was by Shimao Zhang, Xihao Wang, Wen Xiong and Dongmei Tang. Literature search and manuscript preparation was by Ying Jin. The manuscript was edited by Ying Jin and Shimao Zhang and reviewed by Wen Xiong, Dongmei Tang and Sumei Wei. Ying Jin and Shimao Zhang contributed equally to this manuscript; Dongmei Tang and Sumei Wei contributed equally to this manuscript. Conflicts of Interest None. Data Availability The data that support the findings of this study are available from the corresponding author upon reasonable request.