Transcatheter embolization of uterine arteries as a treatment of uterine fibroids or as a treatment of postpartum uterine hemorrhage may be considered medically necessary.

One repeat transcatheter embolization of uterine arteries to treat persistent symptoms of uterine fibroids after an initial uterine artery embolization may be considered medically necessary (see Policy Guidelines section).

Transcatheter embolization for the management of all other indications, including cervical ectopic pregnancy and uterine arteriovenous malformation is considered investigational.  

Patient selection criteria for treatment of uterine fibroids.

Initial Procedure

There are no specific criteria for uterine artery embolization regarding the size, location, or multiplicity of fibroid tumors. The American College of Obstetricians and Gynecologists has suggested the following general criteria for treatment of fibroid tumors:

•       Asymptomatic fibroids of such size that they are palpable abdominally and are a concern to the patient; OR

•       Excessive uterine bleeding as evidenced by either profuse bleeding lasting more than 8 days, or anemia due to acute or chronic blood loss; OR

•       Pelvic discomfort caused by myomata, either acute severe pain, chronic lower abdominal pain, or low back pressure or bladder pressure with urinary frequency not due to urinary tract infection.

Repeat Procedure

One repeat uterine artery embolization may be performed when there is documentation of continued symptoms such as bleeding or pain. Repeat procedures may be most appropriate when there are persistent symptoms in combination with findings on imaging of an incomplete initial procedure, as evidenced by continued blood flow to the treated regions. Limited data from case series suggest a high rate of success following repeat procedures for this purpose, with most patients reporting relief of symptoms

Uterine leiomyomata (ie, fibroids) are extremely common benign tumors that can be located primarily within the uterine cavity (submucosal fibroids) or on the serosal surface of the uterus. Treatment for uterine fibroids is usually sought when they are associated with menorrhagia, pelvic pain, urinary symptoms (ie, frequency), or are suspected to be the cause of infertility. Treatment options include medical therapy with gonadotropin agonists or gestagen suppression or various types of surgical therapy. Hysterectomy is considered the definitive surgical treatment for those who no longer wish to maintain fertility. Various types of myomectomy, which describes removal of the fibroid with retention of the uterus, have also been described. Hysteroscopic myomectomy involves removal of submucosal fibroids using either a resectoscope or a laser. Subserosal fibroids can be removed via an open abdominal or laparoscopic approach. Laparoscopic laser coagulation of uterine fibroids is a unique approach in that the fibroid is not physically removed, but instead multiple (up to 75) laparoscopic laser punctures of the uterine fibroids are performed in an effort to devascularize the fibroid and induce atrophy. 

There is interest in techniques to directly devascularize the uterine fibroid by interrupting the uterine arteries. One technique, uterine artery embolization (UAE) involves selective catheterization of the uterine arteries with injection of embolization material. UAE has also been used to control bleeding in other situations such as severe postpartum hemorrhage, treatment of cervical ectopic pregnancy, and treatment of bleeding uterine arteriovenous malformation

The policy is based in part on a 2002 TEC Assessment¹ and was updated regularly with searches of the MEDLINE database. The most recent literature review was performed through June 8, 2015. Following is a summary of the key literature to date on uterine artery embolization (UAE) therapy. 

UAE for Treatment of Uterine Fibroids Initial UAE Procedure

A number of randomized controlled trials (RCTs) evaluating UAE for treatment of uterine fibroids have been published. In addition, there have been several systematic reviews of these RCTs. A 2014 Cochrane review included 7 RCTs comparing UAE with other surgical interventions in women with symptomatic uterine fibroids.² Four of the RCTs excluded women who desired pregnancy in the future. The comparison intervention was hysterectomy in 3 trials, hysterectomy or myomectomy in 2 trials, and myomectomy in 2 trials. The review’s primary outcomes were patient satisfaction and live birth rates (the latter analysis limited to studies where the comparison intervention was a uterine-sparing procedure). Pooled analyses did not find statistically significant differences in patient satisfaction with UAE or other interventions after 2 years (6 trials; odds ratio [OR], 094; 95% confidence interval [CI], 0.59 to 1.48) or 5 years (2 trials; OR=0.90; 95% CI, 0.45 to 1.80). Only 1 study reported live birth rate so meta-analysis was not possible UAE was associated with a higher rate of minor complications at 1 year (6 trials; OR=1.99; 95% CI, 1.41 to 2.81), and there was not a statistically significant difference between groups in the rate of major complications. Moreover, the UAE group was significantly less likely to require a blood transfusion than the surgery group (2 trials; OR=0.07; 95% CI, 0.01 to 0.52). The rate of further surgical interventions within 2 years was significantly increased in the UAE group (6 trials; OR=3.72; 95% CI, 2.28 to 6.04).

In 2011, van der Kooij et al published a meta-analysis of RCTs comparing UAE and surgery for treating symptomatic uterine fibroids and presented up to 5-year follow-up data.³ The investigators identified 11 articles reporting on 5 RCTs. The overall intraprocedural and early postprocedural complication rates were similar with the 2 procedures. However, hospital length of stay, need for blood transfusion, and febrile morbidity were significantly lower in the UAE group compared with the surgery group. At 12 months, a pooled analysis of 2 studies found a significantly higher reintervention rate in the UAE group compared with the surgery group (OR=5.78; 95% CI, 2.14 to 15.58). Pooled analyses of quality-of-life (QOL) variables at 12 months did not find significant differences between groups. Results were similar after 5 years. The reintervention rate was significantly higher at 5 years, according to a pooled analysis of 2 trials (OR=5.41; 95% CI, 2.48 to 11.81). 

A 2013 systematic review by Martin et al focused on complication and reintervention rates following UAE for symptomatic uterine fibroids.⁴ Eight RCTs comparing UAE with another intervention were included.

Among the 350 UAE cases in the RCTs, the most common complications were discharge and fever (4%), postembolization syndrome (2.9%), pain (2.9%), and groin complications (2.9%). Six trials also provided data on 346 cases of surgery for uterine fibroids. The most common complications following surgery were urinary stress incontinence (3.8%), pressure symptoms (2.9%), and menorrhagia (2.6%). Three trials presented reintervention data and, in a meta-analysis of these studies, there was a significantly higher risk of reintervention after UAE compared with surgery, but a wide confidence interval indicating imprecision of the risk estimate (OR=6.04; 95% CI, 2.0 to 18.1). 

Representative trials are described next.

The Randomized Trial of Embolization versus Surgical Treatment for Fibroids (REST) multicenter trial assigned patients in a 2:1 ratio to undergo UAE (n=106) or surgery (43 hysterectomies and 8 myomectomies).⁵ The embolization group had lower postoperative pain (3.0 vs 4.6, respectively) and faster recovery (eg, 1-day vs 5-day median hospitalization, respectively). Of 7 identified pregnancies in the UAE group, 2 resulted in successful live births. Five-year follow-up data from the REST trial were published in 2011.⁶ A total of 144 of 157 (92%) randomized patients were included in the 5-year analysis. QOL and symptom scores were similar in the 2 groups at 5 years. For example, the mean symptom score was 4.5 in the UAE group and 4.8 in the surgery group (scores ranged from 15 [markedly worse] to 5 [markedly better]). By the 5-year follow-up, 27 of 106 (25%) in the UAE group and 2 of 51 (4%) in the surgery group had received an additional intervention for continued or recurrent symptoms. The total rate of further intervention for symptoms or adverse events over the 5-year period was 32% in the UAE group and 4% after surgery. In the UAE group, there were 3 technical failures of the procedure, 8 repeat UAEs, and 18 hysterectomies. Note that 1 woman had both a repeat UAE and a hysterectomy, and 2 women were not embolized after randomization and subsequently underwent surgery. 

The Embolization versus hysterectomy (EMMY) trial from the Netherlands included 177 women with uterine fibroids and heavy menstrual bleeding who were scheduled to undergo hysterectomy. They were randomized to receive UAE (n=88) or hysterectomy (n=89).^7,8 By the 2-year follow-up, 19 of the 81 (23%) women who actually received UAE had undergone a hysterectomy. An analysis of health-related QOL outcomes at 2 years found similar improvement in both groups overall. The Defecation Distress Inventory (DDI) score improved significantly in only the UAE group starting at 6 months. A report of 5-year outcome data from the EMMY trial was published in 2010.⁹ A total of 70 of the 89 (79%) patients originally randomized to the hysterectomy group and 75 of 88 (85%) in the UAE group completed questionnaires at 60 months. In an intention-to-treat analysis, 23 of 81 (28.4%) women who had received UAE underwent hysterectomy during the 5 years. Including the hysterectomies, 58 of 81 (71.6%) women in the UAE group no longer had menorrhagia. There were no significant differences between groups in health-related

QOL at 5 years, as assessed by mental and physical components of the 36-Item Short-Form Health

Survey (SF-36). For example, the mean difference in change scores at 60 months on the Physical Component Summary of the SF-36 was 1.26 (95% CI, 2.16 to 4.70). Within the hysterectomy group, there was a statistically significantly worse physical health score at 5 years (mean, 6.87) compared with 2 years (mean, 7.26; p=0.01). The UAE group did not have a significant change in the mean physical health score, which was 6.87 at 5 years and 5.80 at 2 years (p=0.34). There was also not a statistically significant difference in the rate of reported urinary incontinence. Similar to the 2-year finding, the DDI significantly improved over time in the UAE group but not in the group assigned to initial hysterectomy. There was not, however, a statistically significant difference between groups in defecation function at 5 years. The authors did not discuss their level of statistical power to detect between-group differences. 

In 2012, findings of the Fibroids of the Uterus: Myomectomy versus Embolization (FUME) trial from the U.K. were published.¹⁰ The investigators randomized women with symptomatic fibroids to UAE (n=82) or myomectomy (n=81). Mean hospital stay was significantly shorter after UAE than surgery (2 vs 4 days, respectively; p<0.001). There were no significant differences in minor or major complications. A total of 120 of 163 (74%) were available for the analysis of QOL, the primary outcome measure. There were no significant differences between groups in change in QOL scores from baseline to 1 year. Nine patients (11%) in the UAE group required additional intervention, and 3 patients (4%) in the myomectomy group later underwent hysterectomy.

An RCT by Hald et al in Norway evaluated clinical outcomes in 66 premenopausal women (mean age, 43 years) with symptomatic uterine fibroids who were randomized to treatment with either laparoscopic occlusion of uterine arteries or UAE.¹¹ Women who wanted to bear children in the future, had a large uterus, had undergone multiple previous open abdominal surgeries, and who had bleeding disorders were excluded. The primary outcome was reduction in blood loss at 6 months postintervention, as measured by a pictorial blood loss assessment chart. Fifty-eight women underwent treatment, 29 with UAE and 29 with laparoscopic occlusion. The proportion of women who had a reduction in blood loss after 6 months did not differ between the treatment groups (52% after UAE, 53% after laparoscopy, respectively; p=0.96). An additional publication reported on follow-up data at a median of 48 months after treatment (range, 8-73 months).¹² The cumulative clinical failure and recurrence rate was significantly lower for patients in the UAE group (17% [n=5]) compared with the laparoscopy group (48% [n=17]; p=0.02). Moreover, fewer patients in the UAE group (7% [n=2]) had a hysterectomy than in the laparoscopy group (28% [n=8]; p=0.41). The authors concluded that UAE is superior to laparoscopic occlusion of uterine arteries for treatment of uterine fibroids.

Section Summary

There is evidence from RCTs that the net health outcome after UAE for uterine fibroids is reasonable compared with surgery. Meta-analyses of RCTs have found similar levels of QOL after 5 years among women receiving UAE or surgery, although there were more reinterventions in the UAE group.

Repeat UAE to Treat Recurrent or Persistent Symptoms

No RCTs focusing on repeat UAE were identified; there are published case series. In 2009, McLucas et al published a study in which the charts of 1058 women who had undergone initial bilateral UAE at several U.S. centers were reviewed.¹³ Forty-two (4%) patients had documentation of persistent symptoms, and they were offered a second bilateral UAE. Thirty-nine patients had repeat procedures, and 34 of these (87%) completed a follow-up questionnaire at least 6 months postembolization. Before the second UAE procedure, 27 of the 34 (79%) women reported severe bleeding, and only 2 (6%) women reported severe bleeding after the procedure. Similarly, the number of women with severe pain decreased from 20 (59%) to 2 (6%), and with severe pressure decreased from 18 (53%) to 2 (6%). A total of 4 women experienced severe levels of 1 or more symptoms after the second UAE. Prospective comparative studies are needed to confirm the findings. 

In 2006, Yousefi et al reported on 24 patients who underwent repeat embolization for recurrent or persistent symptoms 6 to 66 months after the initial procedure.¹⁴ The most common symptoms were pressure and/or bulk symptoms (n=15), recurrent heavy bleeding (n=12), and pelvic pain or cramping (n=7). Follow-up data were available on 21 of 24 (87.5%) after the second UAE; 19 (90%) reported symptom control.

Section Summary

There is a lack of RCTs on repeat UAE for treatment of symptoms associated with uterine fibroids. However, there are data from case series showing a high rate of success after a second UAE for recurrent or persistent symptoms.

Comparison of Products Used in UAE

A 2013 meta-analysis by Das et al identified 10 studies comparing the efficacy of 1 embolic agent used in

UAE to another intervention or comparing 2 embolic agents with one another.¹⁵ Five of the studies were RCTs and 5 were controlled trials that were not randomized. Embosphere microspheres were used in all of the RCTs. In a pooled analysis of data from 2 studies comparing Embosphere with spherical polyvinyl alcohol (PVA) for treatment of uterine fibroids, there was not a statistically significantly between-group difference in outcomes (uterine volume reduction and dominant fibroid volume reduction). Data from other studies were not pooled, but a qualitative analysis of study findings did not suggest that any agent was clearly superior or inferior to any other agent. 

As a recent example, a 2014 RCT by Shlansky-Goldberg et al randomized 60 women with symptomatic uterine fibroids to UAE using either Embosphere microspheres (n=30) or spherical PVA (n=30).¹⁶ The study was funded by the manufacturer of the spherical PVA product. The primary study outcome was complete or near-complete infarction of the dominant tumor (an infarction rate of at least 91%), as measured by magnetic resonance imaging 24 hours after UAE. The primary end point was achieved in 29 of 29 (100%) cases in the spherical PVA group and 28 of 30 (93%) cases in the Embosphere group; the difference between groups was not statistically significant. One patient was excluded from the analysis. Similarly, at 3 months, there was not a statistically significant difference in the rate of complete or nearcomplete infarction rates between groups. 

Section Summary

The available evidence from multiple RCTs does not suggest that any embolization agent is clearly superior to any other agent.

UAE for Treatment of Postpartum Uterine Hemorrhage 

No RCTs or other comparative studies evaluating UAE for treating postpartum hemorrhage were identified. In 2012, Rath et al published a systematic review of literature on second-line treatment of this condition.¹⁷ The authors stated that there is a lack of RCTs on arterial embolization for postpartum hemorrhage and the emergent and multifactorial nature of this condition, as well as potential ethical issues, makes it difficult to conduct a randomized trial. The review summarized the success rate of arterial embolization for postpartum hemorrhage reported in case series as ranging from 70% to 100%, and from 60% to 83% in placenta accreta. The authors noted that UAE should not be performed in patients with excessive bleeding or when the patient is hemodynamically unstable.

Representative larger case series are described next.

In 2013, Kim et al evaluated data on 60 women who underwent UAE for treatment of postpartum hemorrhage at a single center in Korea.¹⁸ The clinical success rate (which was not explicitly defined) was reported as 96%. Eleven patients treated with UAE experienced transient fever after the procedure, and there was 1 case of ovarian failure. During the time of data collection for this study, another 61 patients at the same center underwent cesarean hysterectomy for treatment of postpartum hemorrhage; the success rate associated with that procedure was 93%. 

In 2011, Ganguli et al published data on 66 women who underwent UAE for the treatment of postpartum hemorrhage.¹⁹ The clinical success rate, defined as obviation of hysterectomy, was 95%. Three of 66 (5%) women had a subsequent hysterectomy. In addition to the 3 clinical failures, there were 3 (5%) major complications after UAE: 1 case of lower-extremity deep vein thrombosis, 1 case of postprocedural pancreatitis, and 1 admission for intravenous antibiotic treatment for presumed endometritis. Nine pregnancies after UAE were identified; there were 2 spontaneous abortions and 7 viable gestations. 

In 2009, Kirby et al published a retrospective analysis of data from 43 women who underwent UAE for primary postpartum hemorrhage.²⁰ In this study, clinical success was defined as cessation of bleeding without need for repeat embolization, laparotomy or hysterectomy and without mortality. Eight of 43 (19%) of women had a hysterectomy before UAE. Of the remaining 35 women, clinical success was achieved in 29 women (83%). Considering the sample as a whole, the clinical success rate was 29 of 43 (67%). Complications among women who had a UAE without a previous hysterectomy included 1 case of a groin hematoma, 1 inadvertent perforation of the left obturator artery during UAE, 1 bleeding necrotic fibroid tumor, and 1 case of symptoms consistent with endometritis.

Section Summary

There is a lack of RCTs on repeat UAE for treatment of postpartum hemorrhage. Conducting RCTs may be particularly difficult in this emergent and clinically complex situation. Data from case series have shown a high rate of success after UAE for postpartum hemorrhage.

UAE for Treatment of Cervical Ectopic Pregnancy

No RCTs or other comparative studies evaluating UAE for treating cervical ectopic pregnancy were identified. The published literature consisted of case series with small numbers of patients. Sample sizes ranged from 2 to 20 patients, and most studies had fewer than 10 patients. The largest prospective series was conducted in China by Xiaolin et al.²¹ Patients underwent UAE and in conjunction with methotrexate injections before, during, and after the UAE procedure. Median follow-up was 12 months (range, 1-50 months). Two of 20 patients (10%) had recurrent vaginal bleeding; the other 18 had no significant bleeding after UAE. Five patients (25%) had an additional curettage procedure due to bleeding and/or high levels of beta (b)-hCG. The uterus was preserved in all patients, and normal menses resumed after

2 to 4 months. Eight of 16 (50%) women who attempted another pregnancy achieved a normal pregnancy within 1 year. There were 2 miscarriages and 6 live births at term. 

Section Summary

There is a lack of RCTs in UAE for treating cervical ectopic pregnancy and only a few small case series are available. 

UAE for Treatment of Bleeding Uterine Arteriovenous Malformation 

No RCTs or other comparative studies evaluating UAE for treating uterine arteriovenous malformation were identified. The published literature consists of case reports and a small case series. The series, published in 2014 by Kim et al in Korea, retrospectively reviewed data from a single center on 19 patients who underwent UAE as first-line treatment of bleeding uterine arteriovenous malformation.²² All patients presented with intermittent or progressive vaginal bleeding after gynecologic procedures or obstetric events. The UAE procedures were bilateral, and a variety of embolization agents were used. A total of 17 of 19 patients (89.5%) had immediate clinical success following the UAE. Clinical success was defined as cessation of bleeding without symptom recurrence and resolution of the uterine arteriovenous malformation on postoperative imaging studies. 

Section Summary

The available limited noncomparative evidence is insufficient to determine the effect of UAE on health outcomes in patients with bleeding uterine arteriovenous malformation. 

Impact of UAE Treatment on Fertility and Pregnancy Outcomes 

Several systematic reviews have been published. In 2014, Doumouchtis et al identified 17 studies with a total of 675 participants reporting on fertility outcomes after UAE for postpartum hemorrhage.²³ To be included in the review, studies needed to report on a minimum of 5 cases. None of the studies identified were RCTs. A total of 168 of the 675 patients (25%) wanted a pregnancy following UAE, and 126 of the 168 (75%) women who desired pregnancy achieved conception. There were a total of 136 term live births and 30 cases of pregnancy loss (ectopic pregnancy, miscarriage, elective abortion). Previously, in 2013, Mohan et al identified 21 studies reporting pregnancy outcomes and/or pregnancy complications after UAE for treatment of either postpartum hemorrhage or uterine fibroids.²⁴ The authors reported that the cumulative pregnancy and miscarriage rates among women trying to conceive following UAE for uterine fibroids was 59% and 28%, respectively, and the cumulative live birth rate was 65%. The term delivery rate was 61%. In the studies on UAE for postpartum hemorrhage, the cumulative pregnancy rate, based on a small number of pregnancies, was 87.2%. Rates of miscarriage and live births were not reported following UAE for postpartum hemorrhage. Most of the studies included in the systematic review were observational and had no or inadequate controls. There was only 1 RCT. This study, by Mara et al in Czech Republic, randomized 121 women with uterine fibroids who desired future pregnancies to UAE or myomectomy.²⁵ Participants were followed-up for a mean of 25 months; they were advised to wait for at least 6 months after the procedure before they attempted pregnancy. At final follow-up, 13 of 26 women (50%) in the UAE group who tried to conceive became pregnant compared with 31 of 40 (76%) in the myomectomy group; the difference between groups was not statistically significant. Among women in the UAE group who became pregnant, the abortion rate was 64% and the delivery rate was 19%. In the myomectomy group, the abortion rate was 23%, and the delivery rate was 48%. 

Section Summary

Reviews of fertility and pregnancy outcomes after UAE suggest that successful pregnancy is possible after UAE for treatment of uterine fibroids or postpartum hemorrhage. One review found higher rates of miscarriage and postpartum hemorrhage after UAE for treatment of uterine fibroids compared with treatment of fibroids with myomectomy. There are limited data on pregnancy outcomes in women who became pregnant following UAE for treatment of postpartum hemorrhage.

Ongoing and Unpublished Clinical Trials

Some currently unpublished trials that might influence this policy are listed in Table 1.

Table 1. Summary of Key Trials 

         NCT No.                Trial Name                                                  

                                                                                                                                            Enrollment             Date

NA: not available; NCT: national clinical trial.

Summary of Evidence

The available evidence suggests that the net health outcome after uterine artery embolization (UAE) for uterine fibroids is reasonable compared with surgery. Although there is a lack of controlled studies on repeat UAE, case series have found a high rate of success after a second UAE for recurrent or persistent symptoms. Evidence from randomized and nonrandomized trials does not suggest that any embolization agent used to treat uterine fibroids is superior to any other agent. There are no controlled studies evaluating UAE for other indications, including postpartum hemorrhage, cervical ectopic pregnancy, and uterine arteriovenous malformation. However, there are case series showing benefit for UAE in the treatment of postpartum hemorrhage and there was strong support from clinical reviewers.