Uterine fibroid embolization: where are we and where should we go?

UFE is a highly effective, minimally invasive alternative to hysterectomy and is now widely accepted for the management of fibroid-related symptoms. However, there are still some questions regarding the use of UFE as an alternative to myomectomy in women desiring future fertility and in women with adenomyosis. In early series, complete occlusion of both uterine arteries to stasis with PVA particles, often supplemented with either gelatin sponge pledgets or coils, was the standard technique used (the so-called endpoint of embolization)1-6. With the introduction of tris-acryl gelatin microspheres, the appropriate endpoint has become a subject of discussion7, 8. Limited embolization of the uterine arteries leaving patent the main arterial trunk has recently become the standard technique (Figure 1)7, 8. Color Doppler ultrasound image obtained at 7 days following bilateral limited uterine artery embolization demonstrates patent uterine arteries with devascularized fibroids. Central to the successful outcome of UFE is the appropriate selection of patients. Deciding who is an appropriate candidate involves careful evaluation of the uterine fibroids, especially their size and location. Imaging plays a particularly vital role in preprocedural planning for UFE because it allows assessment of the fibroids and of associated conditions that may imitate or exacerbate the symptom complex. A team approach with both the interventional radiologist and the gynecologist equally involved is the key to a successful UFE program. The preprocedural evaluation should include a detailed history of both gynecological and general medical conditions and a gynecological examination. The decision to treat should be based mainly on symptoms. UFE appears to be effective in treating heavy menstrual bleeding and bulk-related symptoms2-5. Given the presenting symptoms, it then becomes important to determine if the symptoms can be explained by the fibroids. Preprocedural imaging using Doppler ultrasound or magnetic resonance imaging (MRI) can adequately confirm the presence of uterine fibroids, determine their location and size, and exclude conditions that produce similar symptoms such as adenomyosis or endometriosis9-12. Purely submucosal fibroids, which are relatively uncommon, should be treated surgically, ideally using hysteroscopic resection but they may also respond to UFE6. Sonohysterography is more sensitive than transvaginal sonography for detecting the presence and number of submucosal fibroids13. Color Doppler ultrasound may also help in distinguishing an endometrial polyp with a single feeding vessel from an intracavitary submucosal fibroid, which usually has several vessels arising from the inner myometrium14. Pedunculated subserosal fibroids should not be treated by embolization2-5. When the point of attachment to the uterus is narrow, it may disintegrate after UFE and the fibroid may become free in the abdominal cavity4. This situation may require surgery because of peritoneal irritation, infection or small-bowel adhesions4. The number of uterine fibroids is not a limitation of UFE since all fibroids derive their blood supply from the same source. Finally, before embolization, it is important to determine if the fibroids are viable. Viability can be assessed using Doppler ultrasound or contrast-enhanced MRI10, 11, 15. Different degrees of vascularity can be seen but typically fibroids have marked peripheral blood flow with relatively low central flow. Degenerative fibroids with cystic or hemorrhagic necrosis or calcified fibroids should be identified, as they may not respond very well to embolization. In most patients with adenomyosis, areas of hypoechogenicity or heterogeneity of the myometrium that may be focal or diffuse are present16. The echogenic areas represent heterotopic endometrial tissue, whereas the hypoechogenic regions represent smooth muscle. In approximately 50% of patients, small myometrial cysts representing dilated cystic glands of hemorrhagic foci are seen16. When attempting to distinguish adenomyosis from fibroids, the lack of contour and ill-defined margins favor adenomyosis. Color Doppler ultrasound may demonstrate randomly scattered vessels or intratumoral signals and aid in distinguishing between adenomyosis and fibroids (Figure 2)16. The diagnosis of adenomyosis by MRI is made by demonstrating abnormality of the junctional zone of larger than 12 mm16, 17. Numerous foci of high signal intensity scattered throughout the junctional zone are easily identified on T2-weighted MRI16, 17. Color Doppler ultrasound image obtained in a patient with diffuse adenomyosis demonstrates multiple signals throughout the myometrium consistent with the diagnosis. Technical success has been described as successful embolization of both uterine arteries18. The reason is that, except in rare cases, the procedure is unlikely to be successful unless both arteries are treated4, 18. The reported technical success rates range from 84% to 100%, with most series reporting more than 95% technically successful procedures1-8, 18. Clinical success has been measured by the degree of improvement or the frequency of resolution of symptoms4-6. In most studies these symptoms include heavy menstrual bleeding, pelvic pain and bulk-related symptoms (pressure, bloating and urinary frequency). Non-spherical PVA particles were used as the embolization agent in most early reports1-8. Success rates for treating menorrhagia, pelvic pain and bulk-related symptoms ranged from 81% to 96%, 70% to 100% and 46% to 100%, respectively1-8. In a series of 305 women, Hutchins et al. reported control of menorrhagia and bulk-related symptoms in 92% of cases at 12 months3. Three prospective studies with more than 200 patients enrolled have been recently published in the gynecological literature5, 6, 19. From a cohort of 508 patients undergoing UFE using non-spherical PVA particles in Canada, a significant improvement was reported for menorrhagia (83%), dysmenorrhea (77%) and urinary frequency (86%) at 3 months5. We reported on experience with UFE in 400 women with symptomatic fibroids with a mean clinical follow-up of 17 months6. Menstrual bleeding improved in 84% of women and pelvic pain was improved in 79%. In a series of 200 women, Spies et al. reported similar results with improvement of menorrhagia and bulk symptoms in 90% and 91% of cases, respectively, at 12 months19. The initial experience with the use of tris-acryl microspheres mirrors the results obtained with non-spherical PVA particles7, 8. Spies et al. reported significant reduction of menstrual bleeding and pelvic pain in 92% of treated patients at 3 months7. We reported complete resolution of menorrhagia in 85% of patients with a mean follow-up of 30 months8. Another measure of outcome is the effectiveness of UFE in avoiding other treatments for fibroids, as measured by subsequent medical therapies or additional surgery. For example, hysterectomy or additional hysteroscopic resection or myomectomy for clinical failure or recurrence after UFE is an important measure of safety and a key outcome measure of UFE. Spies et al. reported nine (4.5%) hysterectomies in 200 patients within 12 months of therapy19. Seven of the patients underwent hysterectomy for clinical failure after UFE. In a series of 400 women, we reported 23 (6%) clinical failures or recurrence6. Of these, nine (2%) required hysterectomy. In their ongoing clinical experience in 80 patients, Marret et al. reported a 10% recurrence rate at a mean time of 27 months20. In this study, hysteroscopic resection of submucosal fibroids was the most common intervention for recurrent fibroids20. The recurrence rate after UFE has been reported to be lower than 10%, most cases being related to regrowth of fibroids not infarcted after the initial procedure. The long-term rate of recurrence due to the growth of new fibroids is still to be determined21. Patient satisfaction with the clinical outcome of UFE has usually been measured with follow-up questionnaires and correlates well with symptomatic improvement6. We reported that 97% of patients were pleased with the outcome and would recommend UFE to others6. In their treatment of 200 consecutive patients, Spies et al. reported that patient satisfaction paralleled the symptom results and that these results remained stable during the course of follow-up19. A disease-specific quality-of-life instrument for fibroids has been developed22. High levels of satisfaction are observed after UFE even when subsequent therapies are necessary because of clinical failure or recurrence6, 22. Uterine volume reduction and fibroid shrinkage are evaluated using ultrasound or MRI after embolization. Within 3–6 months after UFE, a 25–60% reduction of uterine volume has been reported1-8. The reduction in volume of the dominant fibroid ranges between 33% and 68% at 3–12 months1-8. From the Canadian trial with a cohort of 508 patients, published median uterine and dominant fibroid volume reduction were 35% and 42%, respectively5. We evaluated follow-up ultrasound imaging of fibroids in 400 patients who underwent UFE and found a 58% and 83% median reduction of uterine and dominant fibroid volumes, respectively, after 6 months6. Fibroid location within the uterus may correlate with outcome. Spies et al. reported that smaller baseline leiomyoma size and submucosal location were more likely to result in a positive imaging outcome19. Jha et al. confirmed that submucosal location was a strong positive predictor of fibroid volume reduction9. Using three-dimensional color Doppler sonography, Fleischer et al. found that hypervascular fibroids tend to decrease in size after UFE more than isovascular or hypovascular fibroids10. In addition to volume reduction, the detection of new fibroids should be a priority since fibroid development is very common with uterus-sparing therapies20. Detailed studies are still required to determine the duration between UFE and clinical recurrence due to new fibroids and whether this interval is different from that seen after myomectomy. Immediately after UFE, fibroids tend to have a hyperechogenic central area or air bubbles (Figure 3). Subsequently they appear to decrease in echogenicity and in general become hypoechogenic (Figure 4)11, 23. This is the expected finding given that devascularization followed by necrosis occurs. The hypoechogenic appearance of infarcted fibroids is consistent with the hyaline degeneration observed in fibroids that have spontaneously necrosed. A hyperechogenic perifibroid rim resembling the ultrasound appearance of a fetal head can also be seen at 6–12 months post-UFE (Figure 5)23. Computed tomography and pathology have confirmed that this rim is caused by calcium deposits23. About 3 months after UFE, one should expect little if any flow on color Doppler ultrasound. Tranquart et al. demonstrated an absence of intrafibroid vessels as soon as 3 months after UFE12. Transvaginal ultrasound performed 5 days after uterine fibroid embolization demonstrates a large central hyperechogenic area, which is commonly seen immediately after embolization. Transvaginal ultrasound performed 4 weeks after uterine fibroid embolization shows hypoechogenic uterine fibroids. Transvaginal ultrasound performed 8 months after uterine fibroid embolization shows a hyperechogenic perifibroid rim resembling the ultrasound feature of a fetal head. The MRI appearance of uterine fibroids after embolization has been well described9, 15, 21. The signal intensity increases on T1-weighted images, indicating the presence of proteinaceous material related to hemorrhagic infarction9. In these fibroids there is no enhancement after contrast injection15, 21. In some cases, however, not all fibroids are completely infarcted after embolization and there may be some areas of residual perfusion21. Because the technical goal of UFE is to cause complete infarction of all identified fibroids, it is important to assess after embolization the frequency with which the infarction occurs21. Complete devascularization of all the fibroids is the necessary precursor of symptom improvement in the long term. This has been demonstrated when viewing the long-term imaging outcome of embolization, because complete fibroid infarction results in long-term improvement of symptoms, whereas incomplete infarction may predispose to regrowth and clinical recurrence21. In addition, the degree of gadolinium enhancement is not correlated with fibroid volume reduction15. Therefore, these data suggest that ultrasound may not be useful for the imaging follow-up of patients with recurrent symptoms21. This observation may change if a more accurate method than color Doppler ultrasound is developed to assess residual fibroid perfusion with ultrasound11, 12. Complications associated with UFE can be classified as minor or major based on their severity evaluated by the level of care required, the interventions necessary and the final outcome24. Two different systems (from the Society of Cardiovascular and Interventional Radiology (SCVIR) and the American College of Obstetrics and Gynecology (ACOG)) developed to allow standardized reporting of complication severity have been used to precisely assess complications following UFE24. From a cohort of 400 women, the periprocedural morbidity was 8.5% according to the SCVIR classification and 5% according to the ACOG classification24. Most complications were minor and occurred during the first 3 months after UFE. Five (1.25%) major complications were reported, one (0.25%) of which necessitated hysterectomy24. From the Canadian trial, the overall complication rate after UFE was 8%5. In another study, the rate of readmission for complications from UFE was 17%25. All readmissions were due to infection, of which all but one were treated conservatively and median time to readmission was 3 weeks25. Radiation exposure during UtAE is higher than with common radiological procedures but is within acceptable limits6, 26. The mean estimated absorbed ovarian dose has been reported to be 22 cGy with a mean fluoroscopy time of 22 min and a mean number of 44 angiographic exposure26. It is obvious that screening times should be kept to a minimum and that low-frequency pulsed fluoroscopy should be used whenever possible6, 26. The reported radiation doses are, however, low enough not to interfere with ovarian and future reproductive function26. After embolization, almost all patients experience a self-limited postembolization pain lasting 6–24 h1-6. Some patients will even present with a postembolization syndrome consisting of pelvic pain, nausea, vomiting, mild fever and general malaise6. Several strategies including oral, intravenous, epidural and patient-controlled analgesia have been utilized to manage the pain associated with UFE1-8. Most centers tend to hospitalize their patients for 1–2 days to provide aggressive management of pain1-6. Amenorrhea with other symptoms of menopause is a well-documented complication following UFE4, 6, 24, 27. Symptoms commonly associated with menopause including amenorrhea, vaginal dryness, hot flushes, mood swings and night sweats have all been reported after UFE4, 6, 27. While the incidence of this complication can still be considered low (less than 4%), the impact of this complication can be quite significant, particularly in patients wishing to preserve fertility options after embolization6, 24. Small embolization particles (occluding vessels of less than 500 µm) administered within the uterine arteries can potentially make their way into the ovarian arterial circulation through patent uterine-to-ovarian anastomoses, increasing the risk of reduced ovarian perfusion and subsequent ischemia4, 28. This theory is supported by the demonstration of angiographically visible anastomoses between these two arterial beds in up to 10% of patients undergoing UFE28. In addition, several reports have described the presence of particles in the ovarian or tubal arterial vasculature after UFE6, 29. Particles occluding arteries smaller than 500 µm in diameter are therefore not currently recommended for UFE8. Ovarian ischemia may also happen after aggressive embolization of both uterine arteries when the ovaries are supplied by the uterine arteries8, 30. Using ovarian Doppler flow measurements, Ryu et al. demonstrated that more than 50% of patients have decreased ovarian arterial flow after embolization of both uterine arteries to stasis30. Nevertheless, the rate of amenorrhea is mainly dependent on the age of the patient at the time of treatment27. Chrisman et al. reported a 14% incidence of ovarian failure mainly in women over the age of 45 years27. Spies et al. reported that patients older than 45 years of age are at an increased risk of experiencing significant increases in follicle-stimulating hormone levels when compared to baseline31. One of the potentially more serious complications of UFE is the occurrence of an infection after embolization. Several studies have reported cases of pelvic sepsis after UtAE4, 6, 32. However, when several of the largest published series are considered in aggregate, the overall rate of significant infection after embolization remains low and can be estimated at less than 1%2-6. It has been suggested that submucosal fibroids, pedunculated subserosal fibroids or large uterine fibroids may be associated with increased risk of infection after embolization4, 6. The severity of this particular complication was made clear by the publication of the first death due to infection reported in a 51-year-old patient who underwent UtAE to treat abnormal bleeding attributed to submucosal fibroids33. After an immediate postprocedural period highlighted by a urinary tract infection, the patient returned to the hospital 1 week later with abdominal pain, diarrhea, vomiting and fever. Despite antibiotics, the infection required a total abdominal hysterectomy and bilateral salpingo-oophorectomy. Blood cultures ultimately were positive for Escherichia coli. Two weeks later the patient died due to multiorgan failure33. It is often difficult to know exactly how to manage patients presenting with signs that might indicate the presence of a uterine infection after embolization4, 6, 24. The diagnosis is made even more difficult by the fact that mild fever is often seen during the normal postprocedural recovery period6. A patient presenting with increasing pelvic pain, high fever, vaginal discharge and leukocytosis a few weeks after UtAE should be immediately admitted for appropriate testing with imaging evaluation and treatment6. Bilateral occlusion of the uterine arteries during UtAE clearly increases the risk of global uterine ischemia and subsequent infarction in patients undergoing this procedure4, 6-8. In fact, it is not unreasonable to assume that uterine ischemia occurs in almost all patients undergoing this procedure and that this ischemia contributes to the postprocedural pain that is commonly experienced by patients after UFE. However, it is the rare patient in whom this transient ischemia worsens to the point where the uterus becomes globally infarcted34, 35. The typical presentation of uterine ischemia consists of long-standing pelvic pain, which persists for several weeks associated with fever and elevated white blood cell count6. Ultimately, these patients may require a hysterectomy for pain relief5. In most cases, however, imaging studies have been helpful in confirming myometrial perfusion and absence of myometrial ischemia in most patients after UtAE36. While the reported risk of uterine necrosis is far less than 1%, steps such as avoiding complete stasis during embolization or using large embolization particles may reduce this risk still further8. The presence of a brown or red-brown vaginal discharge after UFE is potentially a sign of impending transcervical passage of an embolized fibroid4, 6, 24. This event has been both well described and frequently reported6, 24, 37. This has been reported to occur both a few weeks after the embolization procedure and after a period of time as long as 4 years4, 6, 24, 37. Typically, patients experiencing passage of a fibroid report symptoms including vaginal discharge, hemorrhage and crampy pelvic pain24. Patients at an increased risk for expulsion include those with submucosal fibroids and those with intramural fibroids that have significant contact with the endometrial cavity37. Transcervical fibroid passage often occurs without incident6. In rare cases, retention of fibroid fragments within the endometrial cavity can potentially increase the risk of infection after embolization and may require hysteroscopic extraction6. It is inevitable that interventionalists will at some point perform UFE on a patient with a leiomyosarcoma instead of the more common benign uterine fibroid. Leiomyosarcomas of the uterus are rare tumors, with an incidence of less than 0.2% of uterine fibroids38. The difficulty in distinguishing between a fibroid and a leiomyosarcoma is that there are no clinical or imaging features that clearly allow differentiation between these two entities39. Common et al. reported a case where UtAE was successfully performed, but continued growth of the fibroid prompted hysterectomy 6 months after embolization38. These cases support the use of follow-up imaging after UFE because failure to respond to embolization would warrant of a diagnosis and a subsequent for have occurred following two from and two due to infection in approximately A careful of the two cases of infection that early diagnosis and appropriate management would have such a et al. reported the results of a clinical trial in patients with fibroids to two one in which the of UFE or hysterectomy was given and one in which hysterectomy was the The overall clinical success of UFE was The hospital for patients treated with UFE was days than for those who underwent hysterectomy. Of the women who underwent UFE, minor in contrast to of those who underwent hysterectomy major A cohort UFE to hysterectomy has been recently For UFE patients there were significant in blood and menorrhagia compared to 12 months a larger of hysterectomy patients improved pelvic was no between the two in the of patients with improvement in urinary symptoms or pelvic no between the was found in of quality-of-life studies UFE to myomectomy have been In their of of patients undergoing UFE and et al. found that overall symptoms improved in 92% UFE patients and 90% myomectomy patients, respectively, and that of UFE patients were with the of their procedure compared to of myomectomy However, rates myomectomy patients were lower than in UFE patients In their of consecutive patients who underwent abdominal myomectomy or UFE, et al. reported clinical success rates for myomectomy UFE of 92% for menorrhagia for pelvic pain and 91% for bulk-related symptoms found and recovery for patients treated with UFE days and 8 that appears to be with UFE in treatment of menorrhagia, and surgery may be a for symptoms related to a of fibroids. The of UFE on and fertility has to be Patients who have UFE have become and successful 6. The of women treated with UtAE for different of or gynecological hemorrhage to and successfully is well and long-term follow-up is However, in these cases the embolization agent has usually been gelatin which not produce as a as particles and therefore may the uterus ultrasound and MRI observation of the uterus following embolization demonstrates of the normal myometrium and an normal appearance of the on 3–6 MRI The may be due to the supply in the which for the complete occlusion of the uterine vessels by embolization. most have UFE for women who no have a more approach and now embolization to patients who future particularly if hysterectomy or or multiple myomectomy is the of of and of women attempting to after should be from the results of large prospective which are not currently From prospective and rates are When it has been suggested that the rate of growth and hemorrhage is higher after UFE because of in uterine blood flow after Nevertheless, in fertility rates and outcome following UFE, it should be into that women undergoing UFE are older and have larger and fibroids than those treated with The published on fertility after embolization is still whereas the on following myomectomy is prospective including embolization and myomectomy in women in future may the It appears that after evaluation and interventional UFE can be to women who future if the options are myomectomy or hysterectomy. the reported of adenomyosis has been frequently are case series reporting the use of arterial embolization in patients with adenomyosis with or without uterine UtAE is an effective procedure in the however, it is associated with a high rate of clinical with up to of embolized women ultimately however, be an in women with diffuse adenomyosis in future fertility since no uterus-sparing treatment is UFE is both a and effective procedure to to patients with symptomatic uterine fibroids. UFE has been described as a alternative to hysterectomy and multiple myomectomy. Clinical success rates for control of heavy menstrual bleeding, pelvic pain and bulk-related symptoms have been reported to be of patients with a low rate of The risk of major uterine infection infarction and ovarian with of these complications potentially without additional surgery. In addition, patient selection with preprocedural imaging evaluation may also reduce the of failures or complications by adenomyosis or pedunculated fibroids. postprocedural follow-up both the gynecologist and radiologist is the key to successful UFE imaging is necessary to fibroid shrinkage or expulsion and clinical studies are still to UFE to other therapies such as single myomectomy and to assess outcome after