Ovarian Artery Embolization Supplementing Uterine Embolization for Leiomyomata Mara M. Barth, MS, and James B. Spies, MD Uterine artery embolization for leiomyomata may fail due to additional blood supply from the ovarian arteries. The potential role of supplemental ovarian artery embolization is unknown. The authors present here the results for their first six patients who were treated with ovarian artery embolization. J Vasc Interv Radiol 2003; 14:1177–1182 Abbreviations:
OAE ⫽ ovarian artery embolization, UAE ⫽ uterine artery embolization
IN the past 5 years, uterine artery embolization (UAE) has become an accepted treatment for symptomatic uterine fibroids. Initial and mid-term results have demonstrated improvements of pressure symptoms, menstrual bleeding, and reduction in fibroid size (1,2). Although these reports have shown great promise for the technique, 10%–15% of patients may not improve (1–3). In at least some of these cases, ovarian arterial supply supplementing the uterine artery supply appears to be the cause of clinical failure after UAE (4,5). There are only two reported cases of successful ovarian artery embolization (OAE) as an adjunct to UAE for fibroids with successful clinical and anatomic results (6,7). The first was a case report that described selective catheterization of the ovarian artery beyond the ovary (6). The second, from the same institution, included that case and another in which embolization was performed proximal to the ovary because of extreme vessel
From the Department of Radiology, Georgetown University Hospital, 3800 Reservoir Road, NW, Washington, DC 20007-2197. Received January 7, 2003; revision requested March 10; final revision received April 25; accepted April 29. Address correspondence to J.B.S.; E-mail: [email protected]
georgetown.edu J.B.S. is a consultant for and receives research support from Biosphere Medical and Boston Scientific. © SIR, 2003 DOI: 10.1097/01.RVI.0000085772.71254.CF
tortuosity (7). We treated a series of six patients in whom one or both ovarian arteries provided significant arterial supply to the uterine fibroids. We present the results of those procedures.
METHODS Over a 3-year period, six women between 44 and 50 years of age were treated by UAE for symptomatic fibroid tumors. Each patient gave informed consent for the procedure and, in particular, was aware that amenorrhea was a likely outcome of OAE. In each of the six cases presented, angiography revealed substantial ovarian artery supply to the fibroids. The patient demographics, presenting symptoms, volume of uterus and dominant fibroid, the embolized arteries, and type and caliber of embolic particles are listed in Table 1. Informed consent was obtained for the procedure detailing risks, benefits, and alternative treatments. Standard risks described included amenorrhea, infection, and possible uterine or ovarian injury. For the purposes of this study, continuation of normal menstrual cycles was used to evaluate continued ovarian function, and no hormonal assays were used. Our approach to UAE is as follows: bilateral 5-F femoral artery sheaths and crossover 5-F Cobra catheters are placed into the proximal anterior divisions of the hypogastric arteries. This is followed by coaxial placement of
microcatheters for embolization of the distal horizontal portion of the uterine artery, with either 355–500 m polyvinyl alcohol particles (Contour, Boston Scientific, Natick, MA) or 500 –700 m tris-acryl gelatin microsphere with or without gold for opacification and (Embospheres or Embogold; Biosphere Medical, Rockland, MA). We chose the embolic material based on the product that was used in our practice at the time that the procedure was done. The endpoint of embolization was occlusion of the fibroid supply with sluggish forward flow still present in the main uterine artery. Approach to the ovarian arteries included a midstream aortogram at the uterine artery level, and selective coaxial catheterization of the ovarian artery(ies) for embolization of the fibroid feeders. The endpoint for OAE was complete occlusion of branches to the fibroids, with the ovarian artery left patent but with near stasis. All patients had undergone preprocedure MR imaging to determine the total uterine volume, the numbers of fibroid tumors, and the volume of the dominant fibroid. For embolization, clinical symptoms were assessed, and, if possible, volume changes of fibroid tumors were documented by MR imaging. Clinical follow-up was via questionnaire and telephone consultation and assessed menstrual bleeding changes, changes in bulk-related signs and symptoms, and subsequent gynecologic interventions. Follow-up was
Patient 1 Age (y) Presenting clinical symptoms
44 Heavy menstrual bleeding, polyuria, low back pain
Uterus size (mL) Dominant fibroid (mL) Left uterine artery
518 360 Embolized with 1 mL 350– 500 m PVA
Right uterine artery
Embolized with 3 mL 355– 500 m PVA
49 47 Heavy menstrual bleeding, Cramps, low back pain, pelvic pain and pressure bloating
461 50.6 Absent
247.6 45.9 Embolized with 1.5 mL 500–700 m Embospheres* Absent
44 50 Heavy menstrual bleeding, Significant left-sided flank pelvic pain and pressure pain, heavy menstrual bleeding
50 Heavy menstrual bleeding, urinary pressure
403 205 Embolized with 4 mL 500– 700 m Embospheres*
1,097 665.1 Embolized with 2 mL 500– 700 m PVA
621 271 Embolized with 6 mL of 500–700 m Embospheres* Absent
Embolized 2 mL 500–700 m PVA
* Embogold, Biosphere Medical, Rockland, MA. Note.—PVA ⫽ Polyvinyl alcohol.
Ovarian Artery and UAE for Leiomyomata
Table 1 Patient Demographics, Presenting Symptoms, Tumor Volume, Vascular Supply to Fibroids, and Embolization Techniques
Table 2 Results of Fibroid Embolization in Six Patients Patient 1
Complete devascularization Complete devascularization Complete devascularization No flow distal to tuboComplete of fibroid, bilateral OAE of fibroid, unilateral OAE of fibroid, bilateral OAE ovarian branches, devascularization of unilateral OAE fibroid, unilateral OAE Follow-up (mo) 14 10 8 9 12 33 Normal menstrual Minimal symptomatic Return of regular menstrual Substantially improved Clinical improvement/ Mild residual back pain, Normal menstrual bleeding, resolved improvement f menorrhagia, resolved cycles, resolved back bleeding, resolution of patient satisfaction decreased hemorrhagia, dysuria hysterectomy and bilateral pelvic pain and pressure pain and bloating pain and pressure resolved polyuria salpingo-oophorectomy
Final result of embolization
Complete devascularization of fibroid, bilateral OAE
September 2003 JVIR
Barth and Spies
performed at 1, 3, and 6 months postprocedure.
RESULTS The results are summarized in Table 2. Nine ovarian arteries were embolized in six patients, without any complications. Three patients (patients 1, 3, and 5) underwent unilateral, and three patients (patients 2, 4, and 6) underwent bilateral OAE to achieve the final result. In four patients (patients 2–5), the right uterine artery was absent, and in one patient (patient 2), the left was absent as well. In three patients (patients 2, 4, and 6), both ovarian arteries were feeding the fibroids. In the remaining three (patients 1, 3, and 5), only one ovarian artery supplied the fibroids. In four of six patients, supply to the fibroid by the ovarian artery was identified by aortography during the initial embolization procedure, prompted by absent bilateral (patient 2) or unilateral (patients 3–5) uterine artery supply. In two cases (patients 1 and 6), additional supply was not suspected but was discovered on a routine aortogram (Fig 1). In three patients (patients 1, 4, and 6), OAE was delayed. In one of these patients (patient 4), no initial consent for OAE had been obtained. In another patient (patient 6), persistence of symptoms more than 1 year after initial embolization prompted repeat left UAE at 15 months, when survey aortography revealed bilateral ovarian artery supply to the fibroid. In the third patient (patient 1), follow-up MR imaging revealed incomplete fibroid infarction likely owing to ovarian flow (Fig 2). Follow-up MR imaging was completed in three of six patients (patients 1, 3, and 6), with complete infarction of the dominant fibroids in each after the OAE was completed. An example is given in Figure 3. The clinical and imaging follow-up ranged from 8 to 14 months. Five of six patients had substantial clinical improvement of symptoms at follow-up (patients 1– 4 and 6). Of those five patients, four (patients 2, 3, 4, and 6) had significant improvement in their preprocedure heavy menstrual bleeding, and five had significant improvements in bulk-related symptoms.
Figure 1. Arteriograms from patient 1. (a) Abdominal aortogram reveals enlarged left ovarian artery (arrows). The right ovarian artery was not opacified. (b) Right ovarian arteriogram reveals a moderately enlarged artery with flow to the margin of the uterus but not filling fibroid branches. (c) Left ovarian arteriogram demonstrates an enlarged vessel and substantial supply to fibroid vessels (arrows). (d) Left ovarian arteriogram after polyvinyl alcohol embolization with near stasis of flow. There is no spontaneous flow to the uterus, and the ovarian vessels are not filled (arrow).
One patient (patient 5) had no improvement in heavy bleeding or bulkrelated symptoms for 1 year after embolization and subsequently under-
went hysterectomy with bilateral salpingo-oophorectomy. She had undergone left UAE and right OAE with complete devascularization of the fi-
Ovarian Artery and UAE for Leiomyomata
Figure 2. Series of T1-weighted sagittal contrast-enhanced MR images of patient 1. (a) Preembolization image reveals normally enhanced uterus and a single enhancing dominant fibroid (arrows). (b) Image from an MR imaging study 3 months after embolization. Only the posterior portion of the fibroid does not enhance, indicating that only that portion is infarcted (arrows). (c) Image from an MR imaging study obtained 2 months after ovarian embolization reveals that the entire fibroid is infarcted.
broid tumor. At surgery, the uterine specimen revealed a decreased size of the dominant fibroid compared with the preembolization MR imaging volume; however, the fibroid was not completely infarcted. Bilaterally, the ovaries revealed no evidence of infarction. This patient had no identifiable right uterine or left ovarian artery.
DISCUSSION Ovarian arterial flow supplementing the blood supply to uterine fibroids has been described in case reports (4,6,8) and may predispose to the clinical failure of UAE. The frequency of additional ovarian supply is not known. Only two case reports have described unilateral OAE for fibroids with successful angiographic and clinical results (6,7). In the six cases presented here, we attempt to demonstrate the technical feasibility of the procedure and determine the clinical outcome. In particular, there are two key questions: First, does OAE result in a successful infarction of the fibroids and thus symptom relief? Second, what is the impact of OAE on clinical ovarian function? Our imaging and clinical results are
mixed. One patient had no relief and went on to subsequent hysterectomy because pathology studies revealed incomplete infarction of the dominant fibroid. However, the remaining patients had significant improvement in bleeding symptoms and relief of both menorrhagia and bulk-related symptoms. Of the patients with imaging follow-up, the decrease in both uterine volume and volume of the dominant fibroid was substantial, ranging from 43% to 78%. More important, however, there was complete infarction of the fibroids in all cases. All six patients continued to menstruate after the procedure. There are technical challenges to OAE, and the appropriate method for embolization is not known. Distal catheterization of the ovarian artery is technically difficult because of the considerable tortuosity of the vessel’s course. In two previous reports, this type of distal passage of the catheter was possible, allowing injection of particles downstream from the ovary (6,7). The approach that we have taken in our patients is to place a microcatheter well past the origin of the ovarian artery but proximal to the typical tortuous portion of the ovarian artery
above the ovary. In our experience, the attempt to advance the catheter further very often results in severe spasm, interfering with a free-flow embolization. We have used particulate embolic material (Embospheres, Embogold, or Contour polyvinyl alcohol) in all our cases, with the hope that the high flow to the fibroids will carry the particles beyond the ovary. We intentionally left some sluggish forward flow at the end of the embolization, with complete occlusion of identified fibroidfeeding branches. Others have used a similar approach, embolizing proximal to the ovary, resulting in amenorrhea (7). All our patients continued regular menses after OAE, but we believe that our data are insufficient to determine the potential impact of the procedure on ovarian function. Cross-collateralization between uterine and ovarian arteries is well documented (4,7,9 –13). Ryu et al (12) concluded that UAE could result in decreased ovarian function, presumably due to incidental ovarian collateral embolization. More recently, Razavi et al (13) noted that in three of five women who experienced menopause after UAE, bilateral ovarian artery to uterine
Barth and Spies
Figure 3. Arteriograms and MR imaging studies of patient 3. (a) Preprocedure bilateral arteriogram reveals that the right side of the uterus is completely supplied by the ovarian artery. (b) Arteriographic images after embolization of the right ovarian and left uterine arteries. (c) Sagittal contrast-enhanced T1-weighted MR image before embolization reveals an enhancing dominant fibroid. (d) Contrast-enhanced sagittal MR image 3 months after embolization demonstrates complete infarction of the fibroid with reduced volume.
artery anastomoses were present. Three studies also concluded that proximal OAE in combination with UAE could significantly increase the risk of iatrogenic ovarian dysfunction (6,7,13). These same studies suggested that in a premenopausal woman, UAE should be performed after angiography to determine ovarian artery collateralization (13) and that OAE should only be per-
formed unilaterally and distal to the ovary (6,7). Three of our patients were treated with bilateral OAE and retained ovarian supply after the procedure. Additionally, each of these cases resumed normal menses for the duration of follow-up, indicating continued ovarian function. We recognize that resumption of menses is only one indicator of preserved ovarian function. Definitive
proof would require identification of a dominant ovarian follicle. Previous studies suggest that routine evaluation of ovarian arterial supply to fibroids was not warranted because of the rarity of such collateralization and the additional risk and radiation exposure to the patient that may be associated with the additional imaging (4,8). On the other hand, failure to treat significant collateral supply can result in UAE treatment failure. One angiographic technique commonly used before UAE includes nonselective pelvic arteriography with the catheter placement above the aortic bifurcation. This is followed by subselective catheterization of the uterine arteries. Although this technique is unlikely to show the ovarian arteries, abdominal aortography at the renal artery level should demonstrate enlarged ovarian arteries that provide collateral supply (11). However, supply from the ovarian arteries may join the uterine artery, and therefore aortography before embolization will not necessarily determine whether there is significant separate ovarian flow not treated during the UAE. If the aortogram (and possible selective ovarian arteriogram) is obtained after embolization, its significance in terms of residual fibroid supply from the ovarian arteries will be much easier to determine. For this reason, we do routine aortography after embolization to assess the ovarian arterial supply. This is a limited study of a few patients, far too few to draw even preliminary conclusions. However, we are aware that many practitioners are performing occasional OAE procedures, and we believe that it is important to begin to assess the outcome of these interventions. Our initial results show that OAE is technically feasible, with infarction of the leiomyomatous mass as detected by MR imaging in most cases. Although most of our patients improved, the clinical data presented here are also too limited for conclusions. Perhaps the most interesting point is that in each patient treated, including the three with bilateral OAE, regular menses continued, demonstrating somewhat to our surprise that amenorrhea is not the inevitable result of the procedure.
Ovarian Artery and UAE for Leiomyomata
References 1. Spies JB, Scialli AR, Jha RC, et al. Initial results from uterine fibroid embolization for symptomatic leiomyomata. J Vasc Interv Radiol 1999; 10:1149 – 1157. 2. Goodwin SC, McLunas B, Lee M, et al. Uterine artery embolization for the treatment of uterine leiomyata—midterm results. J Vasc Interv Radiol 1999; 10:1159 –1165. 3. Worthington-Kirsh RL, Pospky GL and Hutchins Jr FL. Uterine arterial embolization for the management of leiomyomas: quality of life assessment and clinical response. Radiology 1998; 208:625– 629. 4. Nikolic B, Spies JB, Abbara S, Goodwin S. Ovarian artery supply of uterine fibroids as a cause of treatment failure after uterine artery embolization: a case report. J Vasc Interv Radiol 1999; 10:1167–1170.
5. Matson M, Nicholson A, Belli AM. Anastomoses of the ovarian and uterine arteries: a potential pitfall and cause of failure of uterine embolization. Cardiovasc Intervent Radiol 2000; 23:393–396. 6. Andrews RT, Bromley PJ, Pfister ME. Successful embolization of collaterals from the ovarian artery during uterine artery embolization for fibroids: a case report. J Vasc Interv Radiol 2000; 11: 607– 610. 7. Binkert CA, Andrews RT, Kaufman JA. Utility of nonselective abdominal aortography in demonstrating ovarian artery collaterals in patients undergoing uterine artery embolization for fibroids. J Vasc Interv Radiol 2001; 12: 841– 845. 8. Pelage JP, Le Dref O, Jacob D, Rymer R. Ovarian artery supply of uterine fibroids (letter). J Vasc Interv Radiol 200; 11:535.
9. Karlsson S, Jonsson K. Angiography of the ovarian artery in adnexal lesions. Acta Radiol Diagn 1980; 21:739 –746. 10. Karlsson S, Persson PH. Angiography in uterine and adnexal tumors. Acta Radiol Diagn 1980; 21:11–20. 11. Frates RE. Selective angiography of the ovarian artery. Radiology 1969; 92: 1014 –1019. 12. Ryu RK, Chrisman HB, Omary RA, et al. The vascular impact of uterine artery embolization: prospective sonographic assessment of ovarian arterial circulation. J Vasc Interv Radiol 2001; 12:1071–1074. 13. Razavi MK, Wolanske KA, Hwang GL, Sze DY, Kee ST, Dake MD. Angiographic classification of ovarian artery-to-uterine artery anastomoses: initial observations in uterine fibroid embolization. Radiology 2002; 224:707–712.