This succinct review was created by

James B. Spies M.D.
Associate Professor of Interventional Radiology
Georgetown University Medcial Center
Washington, DC

Published Data

Uterine artery embolization as a primary therapy for fibroids was reported by Ravina in 1995 (1). In that initial report, 16 patients were treated. Polyvinyl alcohol particles were used as the embolic agent, injected through catheters placed selectively in the uterine arteries. With a mean follow-up of 20 months, symptoms resolved in 11 of 16. 3 patients had partial improvement, and the residual heavy bleeding subsequently controlled with progestins. There were 2 failures, one of which required hysterectomy 6 weeks after the procedure and another requiring myomectomy 6 months after the procedure.

Goodwin at UCLA subsequently reported on the results of this treatment in 11 patients (2). The embolization procedure used by Goodwin was very similar to that of Ravina, although he used a larger size polyvinyl alcohol particle (500-700 micron). They were successful in bilateral embolization in 10 patients and unilateral in one. One patient developed endometritis and pyometrium within 3 weeks of the procedure that required hysterectomy. In the 10 other patients, the dominant symptom was noticeably improved in 8. One patient was lost to follow-up and another had no improvement. The mean decrease in uterine volume was 40% and dominant fibroid volume decreased 60-65% at three-month follow-up.

Ravina's group reported a larger group of patients in February of 1997(3). 88 women underwent attempted embolization. Of these, the procedure was not successfully completed in five and three others were lost to follow-up or required LH-RH analogue for other reasons. This paper reports on the results of the remaining 80 women. 89% (60 of 67 patients) had resolution of their menorrhagia. There were 7 failures. Fibroid volume was reduced by 55% at 2 months and 69% at 6 months. One patient required a hysterectomy for severe ischemic injury.

Bradley and Reidy reported the results of this therapy in 8 patients with large fibroids (4). Menorrhagia was controlled in 4 of 5 patients presenting with that symptom, while bulk related symptoms improved in all patients. These authors reported that most of their patients experienced an intermittent non-purulent vaginal discharge, presumed to be necrotic fibroid tissue debris. One patient did spontaneously pass a substantial portion of a submucosal fibroid 6 weeks after the procedure. In addition, these authors did have one patient, aged 41, who became amenorrheic following the procedure. Serum follicle stimulating hormone was measured at 59.8 IU/L.

A larger patient series has recently been published (5). 53 patients were treated with technically successful procedures in 52. Follow-up at three months indicated marked improvement in 88% in menstrual bleeding patterns. For the 31 patients with bulk-related symptoms, 29 (94%) experienced marked improvement. The mean reduction in fibroid volume was 46% in the 32 patients in whom follow-up ultrasound was performed. Complications included extensive infarction requiring in hysterectomy 12 days post-procedure in one patient. Two patients required re-hospitalization for post-embolization syndrome. Another patient developed a self-limited episode of upper gastrointestinal hemorrhage secondary to vomiting.

A brief report of two patients in Melbourne, Australia details the experience of two patients treated with UFE (6). One patient did not have adequate control of symptoms and underwent supracervical hysterectomy 26 weeks after her embolization procedure. The pathologic specimen revealed aseptic necrosis of two of the fibroids, with hyaline change of the others.

Two recent papers report a total of 4 cases of fibroid expulsion as a consequence of UFE. One of these patients had the fibroid removed during a simple pelvic examination and had no sequelae (7). The other three patients had clinical evidence of infection which responded to oral antibiotics (8).

Scientific Abstract Presentations

Larger series of patients than those that have been published have been reported at scientific meetings. At the Society of Cardiovascular and Interventional Radiology Meeting in San Francisco, CA in March of 1998, Goodwin presented the follow-up in a group of 50 patients (out of 66 patients with 6 months follow-up) (9). 18 of 50 (36%) had complete resolution of symptoms and 28 of 50 (56%) had substantial improvement in symptoms. Average fibroid volume decrease was 39%. In the group of 66 patients, there were 4 hysterectomies, including one for a pyometrium. At the same meeting, Katz presented the preliminary results of a randomized comparison of polyvinyl alcohol particles with gelfoam pledgets (10). Although the study size was small (n=17), the initial symptomatic control was similar both groups, raising the possibility that temporary occlusive agents may be effective in treating fibroids. Amenorrhea occurred in two of the patients in this study, both treated with polyvinyl alcohol particles. These patients were aged 52 and 47.

At the Society of Minimally Invasive Therapy (SMIT) Annual Conference in London in September of 1998, Dr. Ravina's group at the Hopital Lariboisiere in Paris reported on the results on 184 patients treated over a six year period, from 1991 to 1997 (11). The average patient age was 43, with a range of 21 to 54. 80% of the patients had three or fewer identifiable fibroids and the average fibroid size was 5.8 cm on ultrasound imaging.

Of the 184 patients, there were 8 immediate technical failures and 19 patients were lost to follow-up. 157 patients were available for follow-up evaluation, with a average duration of follow-up of 30 months. Of the 157, 146 were judged asymptomatic with stable results. 11 patients had no response to treatment. This represents a 93% success rate of those successfully treated and available for follow-up. If the 8 technical failures are included, there was an 89% "cure" rate. 88 % of patients were adequately treated without surgery.

Complications were infrequent. Fever occurred in 11% of patients, compatible with the post-embolization syndrome. 6 patients passed a portion of their fibroids in the weeks after the procedure. Pyometria occurred in 3 patients, treated with antibiotics. One patient developed severe necrosis of a pedunculated subserosal fibroid which resulted in a small bowel infection. As a result, they no longer treat pedunculated fibroids.

87% of the fibroids had greater than 50% reduction in volume. 10% of the fibroids could no longer be identified. 7% of patients has volume reduction of 0-24%.

The longer-term follow-up has not yet revealed any recurrences. The fibroids degenerate and scar over a period of weeks to months. Pathologic correlation reveals hyaline sclerosis of the fibroids with preservation of the normal myometrium.

5 patients had amenorrhea that was judged caused by the procedure. These ages were 30, 39, 40, 47, and 48. There were 7 pregnancies in the follow-up group. These included 1 miscarriage, 1 premature delivery, 3 normal deliveries, and 2 ongoing pregnancies.

Also at the SMIT meeting, Dr. Robert Worthngton-Kirsch reviewed the results in patients based on presenting symptoms (12). The results of 126 patients were presented, whose symptoms included menorrhagia alone (46 patients), both menorrhagia and bulk-related symptoms (60 patients), and only bulk-related symptoms (20 patients). For patients with menorrhagia, the average uterine volume reduction was 41%, with control of menorrhagia in 91% of patients at 6 months. For those with both bulk-related symptoms and menorrhagia, the average volume reduction was 48% and symptoms were controlled in 81% at 6 months. For those with bulk-related symptoms alone, the average volume reduction was 45% with control of symptoms at 80% at 6 months. Dr. Kirsch noted that many patients who felt that they only had menorrhagia alone before the procedure in retrospect noted bulk-related symptoms that had resolved post-procedure. Similarly, many with bulk-related symptoms alone noted significantly improved bleeding patterns post-procedure.

At the Royal Surrey County Hospital in Guildford, England, 109 patients have been treated (13). 84% of patients treated for menorrhagia were significantly better, while 94% of patients with bulk-related symptoms were improved or had their symptoms resolve. Most patients were treated with 300 to 500 micron size polyvinyl particles. There was one technical failure. Fibroid volume decreased an average 65%. 2 patients required hysterectomy as a result of infections that developed in the uterus or adnexae. One of these patients developed intra-cavitary pus within a fibroid that resulted in uterine rupture as the uterus tried to expel the mass. One patient had an episode of vaginal bleeding 1 month after the procedure that required transfusion. One patient passed an 8 cm fibroid transvaginally. 2 patients had menopause precipitated by the procedure (ages 54 and 55). A third patient in her mid-forties developed temporary amenorrhea that resolved after three months.

Spies et al (14) summarized the initial results at Georgetown University in Washington DC at the RSNA meeting in November of 1998. Fifty patients were treated over a 16 month period. 34 patients were available for three-month follow-up. Mean clinical follow-up was 30.2 weeks. Complications occurred in four patients during the follow-up period. Two were self-limited, while hysteroscopy was required in one patient and dilatation and curettage in another. All were treated without permanent sequelae.

Menstrual bleeding was improved in 91% of patients at three months and 83% at 6 months. At three months post-procedure, pelvic pain and pressure was improved in 94% of patients; at six months these symptoms were improved in 85%. At initial imaging follow-up (mean 18.7 weeks), mean uterine volume decreased 34.5% and the mean fibroid volume decreased 53.5%. Imaging at one year showed a mean uterine volume reduction of 55.6% and average fibroid volume decrease of 65%.

At the same RSNA meeting in November of 1998, Le Dref et al from Paris (15), France presented results in 81 patients treated with UAE. These authors concentrated on patients with menorrhagia and noted normalization of menstruation in 68 of 76 patients. There were eight failures. Results were stable with a mean follow-up of 20 months. 2 patients experienced transient amennorrhea and 4 developed permanent amenorrhea as a result of therapy. No recurrences were noted in the follow-up period.

In a related subject, Nikolic and Spies (16) reported on the radiation dose associated with UAE. In 20 patients, the mean estimated ovarian dose was 22.34 cGy and the mean skin dose was 162.32 cGy. This is an order of magnitude (10 to 30 times) larger than typical diagnostic radiographic studies, but is 10 to 30 times less than radiotherapy for Hodgkin's Disease of the pelvis. Studies on Hodgkin's patients have not shown any increase in infertility or genetic defects and thus an effect from UAE is extremely unlikely.

At the Annual SCVIR Meeting in Orlando, Florida in March of 1999, Spies (17) presented an update on the results of patients treated at Georgetown University Medical Center. In the Georgetown study, 61 women with a minimum of three months follow-up after fibroid embolization were reported. This includes the group of patients previously reported at the 1998 RSNA. Menorrhagia improved in 89 % (48 of the 54 with this symptom), with moderate to marked improvement in 81%. Pelvic pain or pressure improved in 96% (44 of 46 patients) with 79% moderately to markedly improved. Mean clinical follow-up was 8.7 months. 95% of patients were satisfied to some degree, with 82% were moderately to very satisfied with the outcome of treatment.

Imaging follow-up was available in 54 patients at the initial 3 month follow-up (mean 4.4 months) and revealed a 34% median decrease uterine volume and 50% median reduction in the volume of the dominant fibroids. At one year post procedure (mean 12.3 months), MRI imaging revealed median uterine volume reduction of 50% and median fibroid volume decrease of 78%.

At the same meeting, Dr. Gaylene Pron from the University of Toronto reported the initial results of a Canadian multi-center study (18). 109 patients have been enrolled in the prospective study to date. Pain management for this group of patients has been more difficult in this group of patients than in other reported groups and typically 2 days of hospitalization has been required. One patient had successful embolization of only one uterine artery and her symptoms did not improve. One patient developed endometritis and pyometra 3 weeks after the procedure and hysterectomy was required. Fibroid shrinkage was similar to that reported in other series.

Conclusion

The results from the limited published series and those presented at scientific meetings are similar. It appears from this initial experience that this treatment controls both menorrhagia and symptoms caused by the bulk of these fibroids in 85 to 90% of patients. Patients have tolerated the procedure well and patient satisfaction is high. While severe ischemic injury to the uterus has been feared, it appears that this occurs in only 1 to 2 % of patients. Late infection of the endometrium, spontaneous passage of fibroid tissue, and amenorrhea have also been reported by several investigators. The true incidence of all these complications requires the completion and publication of larger studies.

Pregnancies have been reported in a number of patients, but the pregnancy rate is not known. The large majority of patients treated to date do not wish additional children. Further, the number of patients seeking pregnancy is not known and there are currently no reports of pregnancy rates in the literature. It is likely that a registry or large multi-center study will be needed to answer this question.

There are a number of additional questions that require study. First, it is not known what the recurrence rate will be or whether new fibroids will grow after therapy. The effect on ovarian function has been a question, given the sporadic reports of amenorrhea after treatment. It is not known whether ovarian infarction occasionally occurs to affect function or whether merely decreasing uterine flow is sufficient to affect ovarian function. Further, it is not clear whether ovarian function is affected in only a few patients or whether it is more common and just not apparent clinically.

Despite the unknowns, it appears that the initial experience with the procedure suggests that it is effective and safe in the short term and may represent an alternative to surgical therapy for this very common medical condition.

References: