SUBMITTED 16 JUN 21 1 REVISION REQ. 22 AUG 21; REVISION RECD. 10 OCT 21 2 ACCEPTED 7 NOV 21 3 ONLINE-FIRST: JAN 2022 4 DOI: https://doi.org/10.18295/squmj.1.2022.006 5 6 Acquired Uterine Vascular Anomaly 7 Experience from a tertiary care center in Pakistan 8 Kulsoom Fatima,1 Mustafa B.A. Zubairi,2 Muhammad Azeemuddin,1 *Raza 9 Sayani1 10 11 1Department of Radiology, Aga Khan University Hospital and 2Medical College, Dow University 12 of Health Sciences, Karachi, Pakistan 13 *Corresponding Author’s e-mail: sayani_raza@yahoo.com 14 15 Abstract 16 Objective: To retrospectively review imaging findings and the outcomes of uterine artery 17 embolization (UAE) in symptomatic uterine vascular anomalies. Methods: We identified 15 18 cases of acquired uterine vascular anomaly from 2010 to 2020 who were evaluated with 19 ultrasound, computed tomography, and magnetic resonance imaging, either alone or in 20 combination. All patients had history of dilatation and curettage or uterine instrumentation. They 21 underwent angiography and embolization of the uterine arteries. Primary outcome post 22 embolization was assessed clinically and/or in combination with ultrasound. Post procedure 23 pregnancies were also recorded. Results: Non-invasive imaging was abnormal in all patients, 24 however this pre intervention imaging was unable to accurately classify the type of vascular 25 anomaly except in the case of pseudoaneurysm. Conventional angiography showed uterine artery 26 hyperemia in 6, arteriovenous malformation in 7 and pseudoaneurysm in 2 patients. The 27 technical success rate was 100% with no repeat embolization needed. Follow up ultrasound in 12 28 patients revealed resolution of abnormal findings, remaining three were normal on clinical 29 follow up. Seven patients (46.7%) had a normal pregnancy, 15.7 months after the procedure 30 mailto:sayani_raza@yahoo.com (range 4-28 months). Conclusion: UAE is a safe and effective management option for intractable 31 severe bleeding in patients with uterine vascular anomaly post instrumentation and is seen not to 32 impair future pregnancy. 33 Keywords: Uterine artery embolization; Pseudoaneurysm; Arteriovenous malformation; 34 Ultrasound; Computed tomography; Magnetic resonance imaging. 35 36 Advances in Knowledge 37  All hypervascular lesions in the uterus on ultrasound are not true arteriovenous 38 malformations. Placental subinvolution theory should be kept in mind when assessing 39 patients for suspected uterine vascular anomaly post-abortion on imaging. 40 41 Application to Patient Care 42  Pelvic artery embolization prevents hysterectomy in cases of severe vaginal bleeding. 43 Embolization of the uterine artery does not preclude successful future pregnancy 44 45 Introduction 46 Uterine vascular anomalies (UVA) are classified into two main types, a) vascular malformations, 47 and b) vascular neoplasms, according to the International Society for the Study of Vascular 48 Anomalies classification system.1 The first category includes several entities such as venous 49 malformations, arteriovenous fistulas, pseudoaneurysms, arteriovenous malformations (AVMs), 50 and rarely a combination of pseudoaneurysm and AVM. 1-3 AVMs are the most reported uterine 51 vascular anomaly although their true incidence is unknown.4,5 These may be either congenital or 52 acquired, the latter being far more common.4,6 Acquired uterine AVMs are most commonly 53 secondary to uterine trauma, such as curettage or uterine surgery, which results in abnormal 54 communication of uterine artery branches with the myometrial venous plexus and lack a true 55 nidus.4,7,9 56 57 Rare causes of uterine AVM include endometrial or cervical carcinoma, leiomyoma, uterine 58 infection, gestational trophoblastic diseases or endometriosis.9-11 Direct communication between 59 artery and vein results in arteriovenous fistulas.1,12 Pseudoaneurysms constitute another rare 60 acquired vascular anomaly. These are focal areas of confined defects communicating with the 61 vessel lumen through a traumatic defect, frequently post pelvic/uterine surgery or a curettage 62 procedure, and lack a true wall.2 Patients with acquired uterine vascular anomaly present with 63 acute heavy bleeding, which may be intermittent or continuous.5,8 Other symptoms such as lower 64 abdominal pain, urinary frequency or incontinence, dyspareunia, and hypotension or 65 hypovolemia secondary to blood loss have also been described.11 All suspected cases of uterine 66 AVM initially undergo ultrasound evaluation for diagnosis, supplemented by computed 67 tomography (CT) or magnetic resonance imaging (MRI) in cases where ultrasound is 68 inconclusive.8,12 Although conventional angiography is the gold standard for diagnosis, it is 69 reserved for cases that are unresponsive to conservative management and in whom a therapeutic 70 embolization is planned.13-16 71 72 The purpose of our study was to review the diagnostic accuracy of imaging in identifying uterine 73 vascular anomalies in symptomatic patients and to assess the technical success of percutaneous 74 uterine artery embolization (UAE) in the management of this patient cohort. 75 76 Methods 77 This study was a retrospective analysis conducted at the Aga Khan University Hospital (AKUH), 78 Karachi after taking approval from the institute’s Ethical Review Committee (Ethical Review 79 Number: ERC # 2020-3690-10189). We searched our Radiology database for patients 80 undergoing UAE from January 2010 to May 2020. After excluding patients with known uterine 81 tumors, retained products of conception, gestational trophoblastic disease, and post-partum 82 hemorrhage, we identified 15 cases suspected of uterine vascular anomaly clinically and on 83 imaging. 84 85 A pre-structured proforma was used to record patient demographics including age, parity, 86 pattern, and volume of vaginal bleeding, history of uterine surgery or dilatation and curettage 87 (D&C), time interval since the intervention, findings on imaging and angiography, and patient 88 outcome. The duration of hospital stays, post-procedure complication, follow-up ultrasound 89 findings, and post-embolization fertility/pregnancy were also recorded. The patient’s imaging 90 was reviewed on picture archiving and communication system (PACS), Rogan Delft View Pro-91 X, while additional data was collected from the Health Information Management Services 92 (HIMS). 93 94 The pre angiography imaging modality was chosen at the discretion of the referring physician, 95 which included ultrasonography with color Doppler Imaging, pelvic MRI, and CT, either alone 96 or a combination. The referring physician decided on embolization after consulting with the 97 interventional radiologist. In the angiographic suite of Aga Khan University Hospital, 98 interventional radiologists performed embolization procedures. Consent was taken in every case 99 to explain the benefits and risks. Under local anesthesia, the procedure was performed on a flat 100 panel monoplane digital subtraction angiography machine Axiom-Artis, Siemens. The femoral 101 artery was punctured and a 4F vascular access sheath was inserted. A 4Fr Simmons (SIM 1) 102 catheter (Cordis), or a Cobra (C1) angiographic catheter (Cordis) was advanced over a 0.035-103 inch guidewire. An angiographic run was performed after selective catheterization of the uterine 104 artery, followed by super-selective cannulation using a microcatheter (Progreat Terumo) which 105 was placed coaxially as near as possible to the feeder vessel. The embolization materials used 106 were polyvinyl alcohol particles (PVA), size 355-500 µm, gel foam, glue, and coil, either in 107 combination or isolation. In a few cases, the ovarian artery was also embolized. Clinical success 108 was defined as resolution of vaginal bleeding and/or abnormal imaging findings on post 109 embolization follow-up. 110 111 SPSS version 20 was used for statistical analysis. Quantitative data were expressed as mean ± 112 standard deviation; qualitative data were expressed using frequencies (percentages). Descriptive 113 analysis was done for all variables, including the demographic variables as well as the other 114 categorical variables, and frequencies, proportions, and percentages were reported. 115 116 Results: 117 The mean patient age was 28.2 years (range: 20-35 years). Fourteen patients had undergone a 118 prior uterine procedure. Twelve patients had a prior D&C, One patient had a repair of a uterine 119 rupture and one patient had a C-section. The patient without prior D&C or surgery had a history 120 of medical termination of pregnancy. 121 122 The clinical features are shown in Table 1. Main presenting complain was abnormal pervaginal 123 bleeding. It was considered mild if there was only spotting, when there was continuous bleeding 124 but no clots it was labelled as moderate and severe when there was passage of clots. The mean 125 time interval of patient presentation after the intervention was 64.6 days (range: 1-365 days). 126 All patients underwent pre-embolization US except one, who underwent only MRI examination. 127 Greyscale ultrasound identified an abnormal area in the myometrium in 12 patients (85.7%) and 128 an abnormal area in the endometrium in 2 patients. On Doppler imaging, eight patients showed 129 mixed arterial/venous flow. Six of the eight were confirmed as AVM on angiography (figure 1) 130 while the other two showed only uterine hyperemia on angiography. Four patients showed focal 131 increased vascularity within the myometrium, one turned out to be an AVM on angiography 132 while the other three just showed uterine artery hyperemia. Two patients showed 133 pseudoaneurysms on Doppler that were confirmed on pre-embolization CT and angiography 134 (figure 2). Two other patients had abnormal focal vascularity on arterial phase with prominent 135 veins on pre-embolization CT suggesting AVM. Among these, one proved AVM on angiography 136 (figure 3) while the other showed uterine hyperemia only. 137 138 Six out of 15 patients underwent pre-embolization MRI. MRI findings demonstrated abnormal 139 signal intensity areas in either the myometrium or endometrium with abnormal enhancement and 140 multiple flow voids suggesting AVM. Three of these were confirmed as AVM (figure 4) on 141 angiography, while the other three showed enlarged, prominent uterine arteries. 142 The details of the angiographic findings and procedures are shown in Table 2. 143 144 The embolization procedure was technically successful in all 15 patients, and none required a 145 repeat embolization or post embolization transfusion. None of the patients had an on-table 146 procedure-related or puncture site complication. The mean duration of hospital stay was 2.73 147 days, (range: 2 - 4 days). Nine out of fifteen patients had mild bleeding at the time of discharge 148 from the hospital which resolved by the next clinic visit. Two patients had an episode of per 149 vaginal bleeding a month later which responded to conservative management. 150 151 Twelve patients underwent follow-up ultrasound examinations. The mean time of the follow-up 152 ultrasound, after embolization, was 40.6 days (range- 15 to 90 days). In five patients, the follow-153 up ultrasound was completely normal. Five patients showed persistent greyscale findings, 154 however, abnormal vascularity had resolved. One patient showed a decrease in size of the 155 abnormal area on ultrasound with persistent mild vascularity, although she was asymptomatic. In 156 another patient, both greyscale and Doppler abnormality was demonstrated on initial follow-up 157 ultrasound, but it resolved completely on repeat ultrasound two months later. Three patients did 158 not have any follow-up imaging but were clinically asymptomatic. 159 160 Seven patients (46.7%) had normal pregnancies that carried to term after the procedure. The 161 mean time interval between the procedure and the pregnancy was 15.7 months (range- 4 to 28 162 months). The remaining eight did not conceive to our knowledge. 163 164 Discussion 165 Our retrospective study at a tertiary referral center reviewed the spectrum of imaging findings in 166 patients with suspected acquired vascular uterine anomalies and the outcomes of super-selective 167 UAE. We found true AVMs in seven cases including one arteriovenous fistula. Additional 168 anomalies that we found were uterine hyperemia and pseudo-aneurysm. Timmerman et al 169 studied 30 cases of suspected uterine vascular malformations out of which eight underwent 170 angiography. Their study showed true AVMs in three patients while the rest had only abnormal 171 arterial blush.17 Occasionally, the number of AVMs is purportedly higher on (conventional) 172 angiography. Hugues et al did a study on 26 cases of suspected uterine AVMs, of which about a 173 quarter showed uterine hyperemia, whilst true AVMs were observed in the remaining cases.18 174 Also, a study of iatrogenic uterine arterial injuries, which were treated by UAE, found AVMs in 175 the majority of cases (15 out of 24), either alone or in combination with pseudoaneurysm.2 176 177 All of our cases showed abnormalities on both greyscale and color Doppler ultrasound 178 examinations. Both pseudoaneurysms were accurately identified. Pseudoaneurysms appear as 179 cystic spaces on greyscale ultrasound which show swirling multidirectional flow on color 180 Doppler with varying degrees of turbulence, allowing for correct identification in most cases.1,2 181 The greyscale appearances of AVMs, on the other hand, are non-specific, ranging from subtle 182 myometrial inhomogeneity to linear, anechoic spaces in the myometrium which show color 183 filling on Doppler interrogation with a mosaic pattern.12,19 The specificity is increased by Duplex 184 US/ spectral analysis which reveals high velocity, low resistance arterial flow.17 185 186 Unfortunately, the spectral analysis findings were not available in all our cases as mostly 187 greyscale and color Doppler evaluation was done. It has been seen that low resistance abnormal 188 vascularity often persists in the myometrium at the site of trophoblast/placental implantation and 189 takes time to resolve.1,11 This has been referred to as sub-involution of placental bed and may 190 account for abnormal findings on ultrasound and MRI as seen in some of our cases which 191 subsequently demonstrated only uterine hyperemia on angiography with no AVM. Nevertheless, 192 angioembolization was justified as these patients presented with moderate to severe bleeding that 193 had been resistant to conservative management. Symptoms completely resolved following a 194 single session of embolization, with no discernable adverse effects. 195 196 Currently, digital subtraction angiography (DSA) is the gold standard for the diagnosis of uterine 197 vascular anomalies; however, its use is not justified unless it is the precursor to an embolization. 198 The management of uterine vascular anomalies depends on the clinical presentation as well as 199 the severity of the anomaly. Patients who have minimal symptoms and/or are hemodynamically 200 stable may be followed clinically and by ultrasound. Bleeding usually resolves spontaneously 201 within weeks to months in milder cases. 20 UAE is offered to patients with severe intractable or 202 recurrent bleeding. It avoids hysterectomy preserving chances of future fertility. UAE may 203 theoretically result in reduced vascular supply to the uterus; however, the presence of rich 204 collaterals prevents uterine infarction.13 Many case reports and studies have shown successful 205 pregnancy outcomes post angioembolization.13,20,21 Nearly half of our cases had pregnancies that 206 carried to term, post embolization. 207 208 Several embolic agents have been used for the treatment of UVAs. These include gel foam, PVA 209 particles, glue, coil, or a combination 13,15,16,18,22,23. In our study, PVA particles were the most 210 used embolization material, used in thirteen out of fifteen cases. It was the sole embolic agent in 211 ten cases and used in combination with gel foam in one case and combination with histoacryl 212 glue and coil in two patients. Gel foam was used as the sole embolic agent in one case. One 213 patient with a right uterine artery pseudoaneurysm was embolized with cyanoacrylate glue. The 214 reported complication rate of pelvic artery embolization is low.23 215 216 Minor complications such as puncture site pain or hematoma, fever, and transient lower limb and 217 buttock claudication are more frequent than rare severe complications such as iatrogenic rupture 218 of a pelvic artery, sloughing of perineal skin, vesicovaginal fistula, or major distal 219 ischemia.13,22,23 We did not encounter any major post-procedural complications. Only three 220 patients experienced mild abdominal pain not requiring any treatment while fever was observed 221 in two patients which resolved before discharge. 222 223 Conclusion 224 Ultrasound is the first-line imaging modality employed for the diagnosis of uterine vascular 225 anomaly. It has high specificity for pseudoaneurysms but lacks specificity for AVMs. It is 226 important to consider placental bed sub-involution whilst diagnosing AVMs on non-invasive 227 imaging in cases with a history of a recent miscarriage. In summary, UAE is a safe and effective 228 management option for intractable severe bleeding in patients with uterine vascular anomaly post 229 instrumentation and does not limit future pregnancy outcomes. 230 231 Conflict of Interest 232 The authors declare no conflicts of interest. 233 234 Funding 235 No funding was received for this study 236 237 Author Contributions 238 KF, RS and MA conceived the idea, KF and MZ collected the data, KF, MZ and RS analysed 239 and interpreted the data, KF, RS and MZ drafted the article, MA critically reviewed the 240 manuscript and supervised the study. All authors approved the final draft. 241 242 References: 243 1. Vijayakumar A, Srinivas A, Chandrashekar BM, Vijayakumar A. Uterine vascular 244 lesions. Rev Obstet Gynecol 2013; 6:69-79. https://doi.org/10.3909/riog0207 245 2. Kwon JH, Kim GS. Obstetric iatrogenic arterial injuries of the uterus: diagnosis with US 246 and treatment with transcatheter arterial embolization. Radiographics 2002; 22:35-46. 247 https://doi.org/10.1148/radiographics.22.1.g02ja0735 248 3. Mammen T, Shanthakumari H, Gopi K, Lionel J, Ayyappan AP, Kekre A. Iatrogenic 249 secondary post-partum hemorrhage: apropos of two uncommon cases. Australas Radiol 250 2006; 50:392-4. https://doi.org/10.1111/j.1440-1673.2006.01609.x 251 4. O'Brien P, Neyastani A, Buckley AR, Chang SD, Legiehn GM. Uterine arteriovenous 252 malformations: from diagnosis to treatment. J Ultrasound Med 2006; 25:1387-253 92. https://doi.org/10.7863/jum.2006.25.11.1387 254 5. Kim TH, Lee HH. Presenting features of women with uterine arteriovenous 255 malformations. Fertil Steril 2010; 94:2330.e7-10. 256 https://doi.org/10.1016/j.fertnstert.2010.03.031 257 6. Vogelzang RL, Nemcek AA Jr, Skrtic Z, Gorrell J, Lurain JR. Uterine arteriovenous 258 malformations: primary treatment with therapeutic embolization. J Vasc Interv Radiol 259 1991; 2:517-22. https://doi.org/10.1016/s1051-0443(91)72234-3 260 7. Hoffman MK, Meilstrup JW, Shackelford DP, Kaminski PF. Arteriovenous 261 malformations of the uterus: an uncommon cause of vaginal bleeding. Obstet Gynecol 262 Surv 1997; 52:736-40. https://doi.org/10.1097/00006254-199712000-00004 263 8. Cura M, Martinez N, Cura A, Dalsaso TJ, Elmerhi F. Arteriovenous malformations of the 264 uterus. Acta Radiol 2009; 50:823-9. https://doi.org/10.1080/02841850903008792 265 9. Fleming H, Ostör AG, Pickel H, Fortune DW. Arteriovenous malformations of the 266 uterus. Obstet Gynecol 1989; 73:209-14. https://pubmed.ncbi.nlm.nih.gov/2643064/ 267 10. Cockshott WP, Hendrickse JP. Persistent arteriovenous fistulae following chemotherapy 268 of malignant trophoblastic disease. Radiology 1967; 88:329-33. 269 https://doi.org/10.1148/88.2.329 270 11. Timmerman D, Van den Bosch T, Peeraer K, Debrouwere E, Van Schoubroeck D, 271 Stockx L, et al. Vascular malformations in the uterus: ultrasonographic diagnosis and 272 conservative management. Eur J Obstet Gynecol Reprod Biol 2000; 92:171-8. 273 https://doi.org/10.1016/s0301-2115(00)00443-7 274 file:///C:/Users/raza.sayani/Downloads/%20https:/doi.org/10.3909/riog0207 https://doi.org/10.1148/radiographics.22.1.g02ja0735 https://doi.org/10.1111/j.1440-1673.2006.01609.x https://doi.org/10.7863/jum.2006.25.11.1387 https://doi.org/10.1016/j.fertnstert.2010.03.031 https://doi.org/10.1016/s1051-0443(91)72234-3 https://doi.org/10.1097/00006254-199712000-00004 https://doi.org/10.1080/02841850903008792 https://doi.org/10.1148/88.2.329 https://doi.org/10.1016/s0301-2115(00)00443-7 12. Polat P, Suma S, Kantarcý M, Alper F, Levent A. Color Doppler US in the evaluation of 275 uterine vascular abnormalities. Radiographics 2002; 22:47-53. 276 https://doi.org/10.1148/radiographics.22.1.g02ja0947 277 13. Ghai S, Rajan DK, Asch MR, Muradali D, Simons ME, TerBrugge KG. Efficacy of 278 embolization in traumatic uterine vascular malformations. J Vasc Interv Radiol 2003; 279 14(11):1401-8. https://doi.org/10.1097/01.rvi.0000096761.74047.7d 280 14. Lin AC, Hung YC, Huang LC, Chiu TH, Ho M. Successful treatment of uterine 281 arteriovenous malformation with percutaneous embolization. Taiwan J Obstet Gynecol 282 2007; 46(1):60-3. https://doi.org/10.1016/S1028-4559(08)60109-6 283 15. Salazar GM, Petrozza JC, Walker TG. Transcatheter endovascular techniques for 284 management of obstetrical and gynecologic emergencies. Tech Vasc Interv Radiol 2009; 285 12(2):139-47. https://doi.org/10.1053/j.tvir.2009.08.007 286 16. Kim T, Shin JH, Kim J, Yoon HK, Ko GY, Gwon DI, et al. Management of bleeding 287 uterine arteriovenous malformation with bilateral uterine artery embolization. Yonsei 288 Med J 2014; 55(2):367-73. https://doi.org/10.3349/ymj.2014.55.2.367 289 17. Timmerman D, Wauters J, Van Calenbergh S, Van Schoubroeck D, Maleux G, Van Den 290 Bosch T, et al. Color Doppler imaging is a valuable tool for the diagnosis and 291 management of uterine vascular malformations. Ultrasound Obstet Gynecol 2003; 292 21(6):570-7. https://doi.org/10.1002/uog.159 293 18. Hugues C, Le Bras Y, Coatleven F, Brun JL, Trillaud H, Grenier N, et al. Vascular 294 uterine abnormalities: Comparison of imaging findings and clinical outcomes. Eur J 295 Radiol 2015; 84(12):2485-91. https://doi.org/10.1016/j.ejrad.2015.10.005 296 19. Müngen E. Vascular abnormalities of the uterus: have we recently over-diagnosed them? 297 Ultrasound Obstet Gynecol 2003; 21(6):529-31. https://doi.org/10.1002/uog.163 298 20. Poppe W, Van Assche FA, Wilms G, Favril A, Baert A. Pregnancy after transcatheter 299 embolization of a uterine arteriovenous malformation. Am J Obstet Gynecol 1987; 300 156(5):1179-80. https://doi.org/10.1016/0002-9378(87)90136-0 301 21. Chow TW, Nwosu EC, Gould DA, Richmond DH. Pregnancy following successful 302 embolization of a uterine vascular malformation. Br J Obstet Gynaecol 1995; 102(2):166-303 8. https://doi.org/10.1111/j.1471-0528.1995.tb09075.x 304 https://doi.org/10.1148/radiographics.22.1.g02ja0947 https://doi.org/10.1097/01.rvi.0000096761.74047.7d https://doi.org/10.1016/S1028-4559(08)60109-6 https://doi.org/10.1053/j.tvir.2009.08.007 https://doi.org/10.3349/ymj.2014.55.2.367 https://doi.org/10.1002/uog.159 https://doi.org/10.1016/j.ejrad.2015.10.005 https://doi.org/10.1002/uog.163 https://doi.org/10.1016/0002-9378(87)90136-0 https://doi.org/10.1111/j.1471-0528.1995.tb09075.x 22. Sidhu HK, Prasad G, Jain V, Kalra J, Gupta V, Khandelwal N. Pelvic artery embolization 305 in the management of obstetric hemorrhage. Acta Obstet Gynecol Scand 2010; 306 89(8):1096-9. https://doi.org/10.3109/00016349.2010.481015 307 23. Grönvall M, Tikkanen M, Metsätähti M, Loukovaara M, Paavonen J, Stefanovic V. 308 Pelvic arterial embolization in severe obstetric hemorrhage. Acta Obstet Gynecol Scand 309 2014; 93(7):716-9. https://doi.org/10.1111/aogs.12376 310 311 Table 1. Baseline characteristics of the study population 312 Age Years (mean) 28.2, range (20-35) Parity 0 3 1-2 7 >3 5 History of D&C 12 Time since D&C, days (mean) 42.6, range (1-60) History of uterine surgery 2 History of MTP 1 Amount of bleeding Moderate 6 (40%) Severe 9 (60%) Pattern of bleeding Intermittent 9 (60%) Continuous 6 (40%) D&C: Dilatation and curettage, MTP: Medical Termination of Pregnancy 313 314 Table 2. Details of the Embolization procedure 315 Case Angiographic Finding Vessels embolized Embolization material used Complications Duration of hospital stay Post embolization pregnancy 1 Hyperemia B/L Uterine Artery PVA None 2 No 2 Hyperemia B/L Uterine Artery PVA None 3 No 3 Hyperemia B/L Uterine Artery PVA + gel foam None 2 Yes 4 Hyperemia U/L Uterine Artery PVA Fever 3 No 5 Hyperemia B/L Uterine Artery PVA None 4 Yes https://doi.org/10.3109/00016349.2010.481015 https://doi.org/10.1111/aogs.12376 6 Hyperemia B/L Uterine Artery PVA None 4 No 7 AVM B/L Uterine Artery PVA None 2 No 8 AVM B/L Uterine Artery PVA None 4 Yes 9 AVM B/L uterine artery PVA None 2 No 10 AVM B/L uterine artery Gel foam None 3 Yes 11 AVM B/L uterine artery PVA Fever 4 Yes 12 AVM B/L uterine artery PVA None 2 No 13 AVM B/L uterine artery+ ovarian PVA + coil + glue None 2 No 14 Pseudoaneurysm U/L uterine artery Glue None 2 Yes 15 Pseudoaneurysm B/L uterine + ovarian PVA + coil + glue None 2 Yes AVM: arteriovenous malformation, B/L: bilateral, U/L: unilateral, PVA: polyvinyl alcohol 316 317 318 Figure 1: A 25-year-old with moderate vaginal bleeding. (A) Doppler ultrasound shows a 319 mosaic color pattern in the myometrium on the right side extending into endometrium indicating 320 both arterial and venous flow (arrow), (B) digital subtraction angiography (DSA) image shows 321 the AVM supplied by right uterine artery, (C) DSA image shows the early draining vein (black 322 arrow), (D) post embolization shows resolution of the AVM. 323 324 Figure 2: A 29-year-old with moderate vaginal bleeding post uterine surgery. (A) Greyscale 325 transvaginal ultrasound image shows an irregular anechoic area in the myometrium (arrow) with 326 the turbulent arterial flow on spectral analysis, (B) Doppler image shows heterogeneous color 327 filling in the pseudoaneurysm, (C) and (D) digital subtraction angiographic images, left uterine 328 artery (long white arrow) and left ovarian artery (black arrow) supplying the pseudo-aneurysm. 329 330 331 Figure 3: A 25-year-old with continuous vaginal bleeding. Transvaginal ultrasound Doppler 332 images (A) show abnormal vascularity in the uterine myometrium, (B) Coronal contrast-333 enhanced CT image confirms abnormal myometrial vascularity and dilated draining gonadal vein 334 (white arrow), (C) DSA image shows an abnormal bunch of vessels supplied by right uterine 335 artery, (D) post embolization image showing complete resolution of the AVM 336 337 338 339 Figure 4: A 31-year-old with severe vaginal bleeding. (A) T2 weighted sagittal MRI shows a 340 heterogeneous bulging mass with serpentine signal voids involving the lower uterine cavity and 341 anterior myometrium (white arrow). No myometrium is seen between the urinary bladder and 342 this mass. (B) Coronal T2 weighted image shows multiple serpentine signal voids (white arrow). 343 (C) Digital subtraction angiographic image shows dilated tortuous right uterine artery supplying 344 AVM (D) post embolization image showing complete resolution of the AVM. 345