Keywords
adenomyosis, pregnancy, spontaneous abortion, uterine arteriovenous fistula, uterine artery embolization
1. Introduction
Uterine arteriovenous fistula (UAVF) is a rare uterine vascular malformation characterized by abnormal direct communication between uterine arteries and veins, causing high‐flow, low‐resistance shunting that bypasses the capillary network [1]. Although incidence is low (estimated at less than 1/100 000), UAVF can cause sudden, uncontrollable, massive bleeding leading to hemorrhagic shock and represents an acute critical condition in obstetrics and gynecology [1]. Based on etiology, UAVF is classified as congenital or acquired. Congenital UAVF is rare and often associated with embryonic vascular developmental abnormalities. Acquired UAVF accounts for over 80% of cases and is typically secondary to trauma, surgery, infection, or pregnancy‐related factors [2].
Pregnancy‐related acquired UAVF is a rare but life‐threatening complication, typically presenting with miscarriage or explosive postpartum vaginal bleeding. Its pathogenesis involves the interaction between physiological vascular remodeling and pathological damage during pregnancy. During normal pregnancy, uterine blood vessels undergo significant changes: spiral arteries are replaced by trophoblast cells, and vascular endothelial growth factor (VEGF) and placental growth factor (PlGF) expression increase, resulting in increased vascular permeability and thinned vessel walls [3]. During miscarriage or delivery, placental detachment and uterine contractions can damage vessel walls. If underlying vascular lesions exist, abnormal arteriovenous connections may form. Pregnancy‐related factors account for 25%–40% of acquired UAVF cases and represent the most common cause [3, 4].
Adenomyosis is a benign condition characterized by endometrial gland and stroma invasion into the myometrium. Its pathological features include myometrial structural disruption, vascular smooth muscle component defects, local inflammatory microenvironment, and abnormal angiogenesis [5]. These changes may increase the risk of pregnancy‐related UAVF: inflammatory cytokines (IL‐6, TNF‐α) secreted by ectopic endometrial tissue promote angiogenesis and matrix metalloproteinase activation, weakening vessel wall integrity [5]. However, a possible association between adenomyosis and pregnancy‐related UAVF remains underrecognized in current literature. This case report presents a hypothesis‐generating observation and summarizes key diagnostic and therapeutic considerations.
2. Case Report
A 36‐year‐old married woman with known adenomyosis presented to the emergency department with severe vaginal bleeding 1 month following spontaneous miscarriage. Her uterus was enlarged to the size of a 3‐month pregnancy; she had long‐standing severe dysmenorrhea (VAS score 7–8) and menorrhagia (7–9 days menstrual periods requiring 10–12 sanitary pads daily). She had intermittently used leuprorelin (GnRH‐a) for symptomatic management. The diagnosis of adenomyosis was based on clinical history and prior ultrasound findings at external hospitals; prepregnancy imaging records were not available for objective confirmation of extent and severity. Her obstetric history included one prior embryo miscarriage at 8 weeks in 2022. This was her second pregnancy (G2P0).
The patient underwent IVF‐ET at another hospital due to secondary infertility using an antagonist protocol. A total of 12 oocytes were retrieved, 3 high‐quality embryos were formed, and 1 blastocyst was transferred. After confirming pregnancy, she received continuous hormone support (progesterone, dydrogesterone, estradiol), antispasmodic treatment (ketotifen), and immunomodulation (tacrolimus, later adjusted to hydroxychloroquine and cyclosporine after transfer to our hospital).
During early pregnancy (6–7 weeks), she experienced recurrent vaginal bleeding with a smaller volume than menstrual flow, dark red in color, accompanied by mild lower abdominal distension. External hospital visits diagnosed threatened miscarriage; conservative treatment with progesterone and dydrogesterone was ineffective, with bleeding recurring repeatedly. No pelvic vascular ultrasound was performed. At 6 + 5 weeks, transfer to our hospital revealed a progressively expanding subchorionic hematoma on ultrasound. Despite intensified conservative treatment including tranexamic acid, vitamin K, and snake venom hemocoagulase, bleeding remained refractory.
At 14 weeks, vaginal bleeding suddenly increased significantly, with bright red blood exceeding menstrual flow, accompanied by paroxysmal lower abdominal pain. External ultrasound suggested an inevitable miscarriage. Complete spontaneous abortion occurred at our hospital without curettage. The miscarriage process was complete with no placental residue. Post‐miscarriage bleeding gradually decreased and stopped within 1 week. Follow‐up ultrasound showed no intrauterine residual tissue.
One month post‐miscarriage, the patient suddenly developed heavy vaginal bleeding without an obvious trigger. The blood was bright red with numerous clots; five sanitary pads were soaked within 1 h. She experienced dizziness, fatigue, and palpitations. Emergency hemoglobin was 102 g/L. Intravenous oxytocin combined with tranexamic acid temporarily reduced bleeding. However, after discontinuing hemostatic drugs, massive bleeding recurred (approximately 500 mL over 24 h), with hemoglobin dropping to 91 g/L, suggesting organic vascular lesions rather than simple uterine atony.
Serum β‐hCG was measured at presentation and was negative (< 5 IU/L), effectively excluding ongoing pregnancy, retained trophoblastic tissue, and gestational trophoblastic disease (GTD).
On admission, physical examination revealed: temperature 36.5°C, pulse 110 beats/min (bpm), respiration 20 breaths/min, blood pressure 95/60 mmHg. She appeared conscious with anemic pallor. Gynecological examination showed the external genitalia stained with blood, a large amount of bright red blood and clots in the vagina (estimated 200 mL accumulated), loose cervical os allowing one finger passage with active bleeding, and a uterus enlarged to 3‐month pregnancy size (approximately 12 × 8 cm), soft, with positive tenderness.
Pelvic ultrasound revealed a 118 × 81 mm uterus with uneven myometrial echo, more prominent in the anterior wall, with multiple hypoechoic areas (largest 44 × 45 × 44 mm in the right anterior wall). Multiple irregular tubular anechoic areas were noted in the anterior myometrium (largest 10 × 8 mm). Color Doppler flow imaging demonstrated these anechoic areas filled with blood flow signals; the uneven echo area showed mosaic blood flow signals (62 × 27 × 31 mm). Pulse Doppler detected rough spectral blood flow with a resistance index of 0.52. Peak systolic velocity was not documented in the emergent ultrasound examination. Ultrasound diagnosis: Abnormal blood flow signals in the uterine myometrium, considered UAVF. Representative color Doppler ultrasound images are shown in Figure 1.
Based on pregnancy history, delayed explosive bleeding, and ultrasound findings of abnormal arteriovenous shunting, the clinical diagnosis was acquired UAVF (pregnancy‐related). After a multidisciplinary team discussion, an emergency UAE was planned. Due to limited interventional facilities at our hospital, the patient was transferred with informed consent.
At the external hospital, angiography confirmed enlarged uterine arteries with abnormal vascular clusters in the uterine body and premature venous filling, indicating an arteriovenous fistula. Super‐selective catheters were inserted into lower branches of the uterine artery, and the abnormal vascular clusters were embolized using 300–500 μm polyvinyl alcohol (PVA) particles combined with gelatin sponge particles. The bilateral uterine artery main trunks were preserved. Completion angiography confirmed fistula closure and disappearance of abnormal shunting.
Postoperatively, vaginal bleeding significantly decreased within 24 h and completely stopped by Day 3. Hemoglobin stabilized at 94 g/L without fever, increased abdominal pain, or other complications. At 1‐month follow‐up, pelvic ultrasound showed normal uterine size with no abnormal intramyometrial blood flow signals. Menstruation resumed naturally at 3 months post‐surgery with regular cycles and moderate flow. Contraception for 6–12 months was recommended to allow vascular remodeling and endometrial repair before reassessing pregnancy risk.
3. Discussion
The following mechanistic discussion represents a hypothesis based on pathophysiological reasoning and limited indirect evidence; it is not intended to imply proven causality. This case may be interpreted through the “underlying lesion‐acute injury” dual‐hit pathogenetic model. Adenomyosis may have served as the underlying lesion, potentially creating “vascular vulnerable zones” through myometrial structural disruption, inflammatory cytokine‐mediated angiogenesis, and matrix metalloproteinase activation [5]. During pregnancy, physiological vascular remodeling may have been amplified by IVF‐ET high‐dose hormonal support, potentially further increasing vascular fragility. The natural miscarriage at 14 weeks may have caused placental detachment and uterine contraction‐related mechanical damage to potentially fragile vessels. Under high local perfusion pressure, arterial blood may have directly entered the venous system, resulting in acquired UAVF. Notably, this patient lacked traditional risk factors such as cesarean section or curettage history, raising the hypothesis that adenomyosis may represent a potential predisposing factor in the development of pregnancy‐related UAVF, though this association requires validation in larger studies.
The UAVF in this case most likely developed in association with the miscarriage process, rather than representing a preexisting lesion. The temporal relationship (onset 1 month post‐miscarriage) supports this interpretation.
3.1. Differential Diagnosis
UAVF must be distinguished from several conditions with similar presentations:
Retained products of conception (RPOC) typically present with immediate post‐miscarriage bleeding, intrauterine echogenic tissue on ultrasound, and positive or declining β‐hCG. In this case, RPOC was excluded by: (a) complete spontaneous abortion without curettage, with documented expulsion of intact products; (b) follow‐up transvaginal ultrasound showing no intrauterine residual tissue and a thin, regular endometrial line; (c) negative serum β‐hCG; and (d) the delayed onset of hemorrhage (1 month post‐miscarriage), which is atypical for RPOC‐related bleeding that usually occurs within days to 2 weeks.
Enhanced myometrial vascularity (EMV) typically shows diffuse, symmetric vascular enhancement without focal arteriovenous shunting. This case demonstrated focal mosaic blood flow with a discrete resistance index of 0.52, consistent with true fistula rather than physiological vascularity.
GTD was excluded by negative β‐hCG and absence of characteristic “snowstorm” or “bunch of grapes” ultrasound appearance; no hydropic villi were identified.
Uterine artery pseudoaneurysm typically shows a saccular aneurysmal sac with “to‐and‐fro” spectral waveform and lacks early venous filling. Angiography in this case demonstrated direct arteriovenous communication with premature venous opacification, confirming fistula rather than pseudoaneurysm.
Why MRI or CT angiography was not performed before angiography: Given the patient's hemodynamic instability (pulse 110 bpm, BP 95/60 mmHg, ongoing massive bleeding with ~500 mL blood loss in 24 h), immediate hemostasis was the clinical priority. Color Doppler ultrasound provided sufficient diagnostic confidence for UAVF, and proceeding directly to diagnostic angiography with therapeutic intent was the most efficient approach to achieve simultaneous diagnosis and treatment.
The diagnostic process was significantly delayed. Early pregnancy recurrent bleeding was repeatedly misdiagnosed as threatened miscarriage. However, UAVF bleeding is typically bright red, voluminous, continuous or recurrent, and refractory to progesterone therapy, unlike the dark red, small‐volume, intermittent bleeding of threatened miscarriage [6]. We recommend color Doppler ultrasound screening for adenomyosis patients with recurrent early pregnancy bleeding, focusing on abnormal intramyometrial flow signals rather than relying solely on empirical fetal preservation measures.
Transvaginal ultrasound is the preferred screening modality for UAVF [6], though reported diagnostic accuracy varies considerably across studies depending on operator experience, equipment quality, and lesion characteristic [7, 8, 9]. Key diagnostic features include intramural abnormal anechoic areas, abundant mosaic blood flow signals, high‐velocity low‐resistance arterial spectra, and early venous visualization. MRI assists in evaluating lesion extent and its relationship with adenomyosis, displaying characteristic rapid enhancement patterns. Digital subtraction angiography remains the diagnostic gold standard and enables simultaneous interventional treatment [10].
Uterine artery embolization is currently the preferred treatment for acquired UAVF, with success rates of 85%–95%, offering minimally invasive uterine preservation, particularly suitable for young patients with fertility requirements [11]. This case used 300–500 μm PVA particles (permanent embolic material reaching fistula extremities for long‐term occlusion) combined with gelatin sponge (absorbable material providing immediate hemostasis), balancing acute efficacy and long‐term prognosis [10]. Super‐selective embolization preserving uterine artery trunks optimizes fertility protection compared with main trunk embolization. Bilateral embolization completely eliminates fistula flow. Recurrence risk is approximately 10%–20%, potentially higher in adenomyosis due to extensive myometrial lesions and abundant collateral circulation [12]. Re‐embolization with alternative materials or laparoscopic fistulectomy may be considered for recurrence [11].
Fertility preservation requires comprehensive management. Post‐UAE temporary amenorrhea or reduced menstrual flow may occur due to decreased uterine blood supply; this patient's menstrual resumption at 3 months suggests successful endometrial revascularization. The ovarian blood supply may be partially affected through uterine‐ovarian anastomotic branches, but regular menstrual cycle resumption in this case indicates preserved ovarian function. Subsequent pregnancy carries dual risks: adenomyosis‐related complications (miscarriage, preterm delivery, fetal growth restriction, abnormal placentation) and UAVF recurrence [13]. Preconception imaging assessment, early pregnancy vascular screening, and delivery at facilities with interventional emergency capabilities are essential. Postabortion pregnancy rates after UAE are 30%–50%, significantly superior to hysterectomy [14, 15].
3.2. Literature Review
Published cases of UAVF specifically associated with adenomyosis are extremely rare. We searched PubMed and identified no prior case reports describing this co‐occurrence. Regarding pregnancy‐related UAVF, available literature indicates that pregnancy‐related factors account for.
In total, 25%–40% of acquired UAVF cases [16], with typical presentations including miscarriage or explosive postpartum vaginal bleeding. The temporal relationship between pregnancy termination and UAVF onset varies from immediate to several months postpartum [17]. This case shares similarities with previously reported pregnancy‐related UAVF cases in terms of delayed explosive bleeding and successful UAE treatment but differs in the potential association with underlying adenomyosis.
Terminology note: “Uterine arteriovenous malformation (UAVM)” is sometimes used interchangeably in literature, particularly for congenital lesions with a vascular nidus. However, “uterine arteriovenous fistula (UAVF)” is the preferred term for acquired lesions characterized by direct arteriovenous communication without an intervening vascular nidus and is therefore used consistently throughout this report.
3.3. Limitations
This report has several important limitations. First, the coexistence of adenomyosis and UAVF in this single patient does not establish that adenomyosis contributed to UAVF development; any proposed association remains speculative and requires validation in larger, controlled studies. Second, prepregnancy imaging to confirm the extent and severity of adenomyosis was not available, limiting objective assessment of this potential predisposing factor. Third, peak systolic velocity was not documented in the emergent ultrasound examination. Fourth, angiographic and post‐embolization imaging were not available for publication, as the procedure was performed at an external facility, limiting the completeness of imaging documentation. Fifth, longer‐term follow‐up data regarding recurrence, future fertility, and pregnancy outcomes were not available, as the patient was lost to follow‐up after her 3‐month postoperative visit. Finally, while adenomyosis is associated with increased local VEGF and inflammatory cytokine expression [18], and IVF hormonal stimulation promotes vascular remodeling, no direct evidence links these factors to UAVF formation in adenomyosis patients.
This case presents a hypothesis‐generating observation of a possible association between adenomyosis and pregnancy‐related UAVF, though this requires validation in larger studies. For adenomyosis patients with pregnancy, prepregnancy assessment and enhanced pregnancy monitoring may be beneficial. Recurrent early pregnancy bleeding unresponsive to conservative therapy should raise suspicion for underlying vascular abnormalities, and color Doppler ultrasound evaluation may be warranted. Delayed explosive post‐miscarriage hemorrhage refractory to uterotonics is characteristic of UAVF, requiring prompt imaging confirmation. UAE is the preferred fertility‐sparing treatment; uterine artery trunks should be preserved when possible. Given the limitations of this single case, strict 6‐ to 12‐month contraception and long‐term follow‐up are recommended to monitor for recurrence.
Author Contributions
Linlin Zhang: data curation, writing – original draft. Guochao Jiang: writing – original draft, formal analysis, data curation. Xingwei Zhang: writing – review and editing. Yuance Xu: data curation, formal analysis, writing – original draft. Shuang Guo: data curation, writing – original draft. Zheyu Song: formal analysis, writing – original draft.
Funding
The authors have nothing to report.
Ethics Statement
This case report was approved by the Ethics Committee of Jilin Women And Children Health Hospital (Jilin Province Maternal and Child Health Quality Control Center). All procedures were in accordance with the 1964 Helsinki Declaration and its later amendments.
Consent
Written informed consent was obtained from the patient.
Conflicts of Interest
The authors declare no conflicts of interest.
Data Availability Statement
All data generated or analyzed during this study are included in this published article. The raw data supporting the findings of this case report are available within the article text, tables, and figures. Due to the nature of this single case report and patient privacy considerations under the Ethics Committee approval (Jilin Women And Children Health Hospital), individual patient‐level data beyond what is presented in this manuscript cannot be made publicly available. The corresponding author can be contacted for any additional inquiries regarding the data presented in this study.
References
- 1. Zhang Q., Zhang S., Zhou D., et al., “Early Pregnancy‐Related Uterine Arteriovenous Fistula Diagnosis and Treatment Chinese Expert Consensus (2022 Edition),” Chinese Journal of Practical Gynecology and Obstetrics 38, no. 3 (2022): 284–289. [Google Scholar]
- 2. Giurazza F., Corvino F., Silvestre M., et al., “Uterine Arteriovenous Malformations,” Seminars in Ultrasound, CT and MR 42, no. 1 (2021): 37–45. [DOI] [PubMed] [Google Scholar]
- 3. Osol G. and Moore L. G., “Maternal Uterine Vascular Remodeling During Pregnancy,” Microcirculation 21, no. 1 (2014): 38–47. [DOI] [PubMed] [Google Scholar]
- 4. Lalitha N., Seetha P., Shanmugasundaram R., and Rajendiran G., “Uterine Arteriovenous Malformation: Case Series and Literature Review,” Journal of Obstetrics and Gynaecology of India 66, no. 4 (2016): 282–286. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Vannuccini S. and Petraglia F., “Recent Advances in Understanding and Managing Adenomyosis,” F1000Research 8 (2019): 283. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Timmerman D., Wauters J., Van Calenbergh S., et al., “Color Doppler Imaging Is a Valuable Tool for the Diagnosis and Management of Uterine Vascular Malformations,” Ultrasound in Obstetrics & Gynecology 21, no. 6 (2003): 570–577. [DOI] [PubMed] [Google Scholar]
- 7. Timor‐Tritsch I. E., Haynes M. C., Monteagudo A., Khatib N., and Kovács S., “Ultrasound Diagnosis and Management of Acquired Uterine Enhanced Myometrial Vascularity/Arteriovenous Malformations,” American Journal of Obstetrics and Gynecology 214, no. 6 (2016): 731.e1–731.e10. [DOI] [PubMed] [Google Scholar]
- 8. T. G. Tullius, Jr. , Ross J. R., Flores M., Ghaleb M., and Kupesic Plavsic S., “Use of Three‐Dimensional Power Doppler Sonography in the Diagnosis of Uterine Arteriovenous Malformation and Follow‐Up After Uterine Artery Embolization: Case Report and Brief Review of Literature,” Journal of Clinical Ultrasound 43, no. 5 (2015): 327–334. [DOI] [PubMed] [Google Scholar]
- 9. Jeerakornpassawat D., Tantipalakorn C., Sirilert S., and Tongsong T., “Sonographic Features of Uterine Arteriovenous Malformation: A Case Series,” Diagnostics 14, no. 9 (2024): 873. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Grivell R. M., Reid K. M., and Mellor A., “Uterine Arteriovenous Malformations: A Review of the Current Literature,” Obstetrical & Gynecological Survey 60, no. 11 (2005): 761–767. [DOI] [PubMed] [Google Scholar]
- 11. Hong W., Wang B.‐Y., Wu Z.‐P., Gao F., Li S.‐D., and Li X.‐C., “Systematic Retrospective Analysis of 13 Cases of Uterine Arteriovenous Fistula: Pathogeny, Diagnosis, Treatment and Follow‐Up,” Journal of Obstetrics and Gynaecology Research 46, no. 7 (2020): 1117–1127. [DOI] [PubMed] [Google Scholar]
- 12. Zhang E., Liu L., and Owen R., “Pelvic Artery Embolization in the Management of Obstetrical Hemorrhage: Predictive Factors for Clinical Outcomes,” Cardiovascular and Interventional Radiology 38, no. 6 (2015): 1477–1486. [DOI] [PubMed] [Google Scholar]
- 13. Juang C.‐M., Chou P., Yen M.‐S., Twu N.‐F., Horng H.‐C., and Hsu W.‐L., “Adenomyosis and Risk of Preterm Delivery,” BJOG 114, no. 2 (2007): 165–169. [DOI] [PubMed] [Google Scholar]
- 14. McLucas B., W. D. Voorhees, 3rd , and Elliott S., “Fertility After Uterine Artery Embolization: A Review,” Minimally Invasive Therapy & Allied Technologies 25, no. 1 (2016): 1–7. [DOI] [PubMed] [Google Scholar]
- 15. Mohan P. P., Hamblin M. H., and Vogelzang R. L., “Uterine Artery Embolization and Its Effect on Fertility,” Journal of Vascular and Interventional Radiology 24, no. 7 (2013): 925–930. [DOI] [PubMed] [Google Scholar]
- 16. Peitsidis P., Manolakos E., Tsekoura V., Kreienberg R., and Schwentner L., “Uterine Arteriovenous Malformations Induced After Diagnostic Curettage: A Systematic Review,” Archives of Gynecology and Obstetrics 284, no. 5 (2011): 1137–1151. [DOI] [PubMed] [Google Scholar]
- 17. Yu L. and Xiao X., “Diagnosis, Management and Treatment of Acquired Uterine Arteriovenous Malformation: A Case Series and Literature Review,” International Journal of Gynecology & Obstetrics 169, no. 1 (2025): 63–70. [DOI] [PubMed] [Google Scholar]
- 18. Benagiano G., Brosens I., and Habiba M., “Structural and Molecular Features of the Endomyometrium in Endometriosis and Adenomyosis,” Human Reproduction Update 20, no. 3 (2014): 386–402. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
All data generated or analyzed during this study are included in this published article. The raw data supporting the findings of this case report are available within the article text, tables, and figures. Due to the nature of this single case report and patient privacy considerations under the Ethics Committee approval (Jilin Women And Children Health Hospital), individual patient‐level data beyond what is presented in this manuscript cannot be made publicly available. The corresponding author can be contacted for any additional inquiries regarding the data presented in this study.