August 1999, Vol. 6, No. 3
The Journal of the American Association of Gynecologic Laparoscopists
Selective Uterine Artery Embolization as Primary Treatment for Symptomatic
Leiomyomata Uteri Francis L. Hutchins, Jr., M.D., Robert Worthington-Kirsch, M.D., and Robert P. Berkowitz, M.D.
Abstract Study Objective. To analyze initial experience with uterine artery embolization for treatment of symptomatic leiomyomata. Design. Prospective, longitudinal study (Canadian Task Force classification 1t-2). Setting. Private practice, university-affiliated hospital. Patients. Three hundred five women (age 26-52 yrs). Interventions. Uterine artery embolization, performed over 2 years by a single radiologist working in collaboration with a single gynecology practice. Measurements and Main Results. Embolization was technically successful in 96% of patients. No major complications occurred. Average reduction in uterine volume was 48%. Control of menorrhagia was reported by 86% of patients at 3 months, 85% at 6 months, and 92% at 12 months after the procedure. Bulk symptoms were satisfactorily controlled in 64% of patients at 3 months, 77% at 6 months, and 92% at 12 months. Six women subsequently underwent hysterectomy and five had myomectomy. Conclusion. Uterine artery embolization appears to be a highly effective treatment for symptomatic uterine leiomyomata. Its impact on fertility and pregnancy remain to be investigated fully. (J Am AssocGynecol Laparosc6(3):279-284, 1999)
Uterine leiomyomas are the leading cause of hysterectomy, which is the second most common surgical procedure in the United States. 1 Because of the magnitude of this problem, much effort has been expended in attempting to develop more effi-
cient treatment modalities that would be associated with less morbidity and more rapid recovery than hysterectomy. Particular attention has been given to minimally invasive approaches, which until recently were limited to hysteroscopic and laparoscopic
From the Department of Obstetrics and Gynecology, Thomas Jefferson University, Philadelphia (Drs. Hutchins and Berkowitz); Hutchins Institute for Women's Health EC., Bala Cynwyd, Pennsylvania (Dr. Hutchins); and Department of Interventional Radiology, Delaware Valley Imaging, Ltd., Bala Cynwyd, Pennsylvania, and Department of Radiology, Philadelphia College of Osteopathic Medicine, Philadelphia, Pennsylvania (Dr. Worthington-Kitsch). Address reprint requests to Francis L. Hutchins, Jr., M.D., Hutchins Institute for Women's Health, One Bala Avenue, Suite 120, Bala Cynwyd, PA 19001; fax 610 668 1482. Accepted for publication May 24, 1999.
Uterine Artery EmboJization for Leiomyomata Uteri Hutchins et al
surgery. Embolotherapy has long been emergency therapy for pelvic hemorrhageY and recently was introduced as primary treatment for symptomatic fibroids 4-6
ibuprofen 600 mg orally every 4 hours. The morning after the procedure patients were converted to oral analgesia with ibuprofen 600 mg four times/day for 10 days and oxycodone 1 to 2 tablets every 4 to 6 hours as necessary, and discharged from the hospital.
Materials and Methods
Follow-up Three hundred five premenopausal women with uterine leiomyomas confirmed by ultrasound were treated by selective uterine artery embolization (UAE). All had menorrhagia and/or bulk-related symptoms such as pelvic pressure, urinary frequency, or constipation. Patients were excluded for pregnancy, active pelvic infection, severe contrast allergy, arteriovenous fistula, or undiagnosed pelvic mass.
Patients were seen in follow-up at 3 months unless they required an earlier visit. Ultrasound examination was performed at that time to evaluate uterine size. Uterine volume was calculated using the formula (length)(width)(AP thickness)(0.52), as an approximation for the volume of a prolate ellipsoid. Patients were interviewed by telephone at 3, 6, and 12 months, with several attempts made to reach every one. The following questions were asked:
1. How many days did it take to feel back to normal after embolization? 2. How has the abnormal component of your bleeding changed? (stopped completely, greatly improved, moderately improved, slightly improved, unchanged, worse) 3. How have your bulk symptoms changed? (not applicable, stopped completely, greatly improved, moderately improved, slightly improved, unchanged, worse) 4. If you had to do it over again, would you have an embolization procedure? (yes, maybe, no)
Uterine artery embolizatiom was performed according to published protocol.6 Before arteriography, a baseline ultrasound of the pelvis was performed in all patients with measurement of uterine dimensions. Prophylaxis consisted of a single dose of cefazolin 1 g intravenously; patients with a history of penicillin allergy were given clindamycin 300 mg intravenously. Patients received morphine sulfate 5 to 10 mg, midazolam 2 to 5 rag, and fentanyl 50 to 200 ~g, all intravenously, for conscious sedation. Arteriography was performed using the right femoral approach. A 5Fr diagnostic catheter (generally Levin 1; Cook Ob/Gyn, Spencer, IN) was used in almost all cases. After a mapping aortic injection to define iliac bifurcations and (ideally) uterine arteries, the left uterine artery was selectively catheterized. The catheter tip was positioned at or beyond the junction of descending and horizontal portions of the left uterine artery, and embolization was performed with 500 to 700 g polyvinyl alcohol (PVA) particles (Biodyne; Cook). The PVA was injected into each uterine artery until there was complete cessation of flow or reflux of contrast into the anterior division of the internal iliac artery. Embolization coils (Gianturco coils; Cook) were added to ensure cessation of flow, if necessary. After verifying occlusion of the left uterine artery, the right uterine artery was catheterized and embolized in the same fashion. Postoperative pain was managed with morphine sulfate by patient-controlled anesthesia (PCA) pump (dose 1 rag, delay 6 min, basal rate none), ibuprofen 800 mg orally immediately after the procedure, and hydroxyzine 75 mg intramuscularly followed by
In addition, complications of the procedure or hospitalization were discussed during the interviews. Results
Technical The technical success rate for UAE was 96%, with 13 unsuccessful attempts. In four women, one uterine artery was embolized, but the second thrombosed during catheterization, preventing embolization with PVA. At least two of these patients had an excellent clinical result. In five patients only one uterine artery was seen on arteriography, and it was embolized in all of them. Two of these women had previously undergone myomectomy, and presumably the absent uterine artery was ligated at that time. One patient had an anatomic variant in which one uterine artery was a branch of the ovarian artery, rather than of the internal iliac artery. In two patients one uterine artery proved impossible to catheterize selectively because of tortuosity. The procedure had to be abandoned in one woman because
August ] 999, Vd. 6, No. 3
The Journal of the American Association of Gynecologic Laparoscopists
TABLE 1. Analysis of 305 UAEs by Duration of Follow-up
of failure of angiography equipment. None of these patients experienced clinical benefit from UAE.
Length of Follow-up > 3 Months > 6 Months >_1 Year
No major complications occurred. One patient developed severe nausea and vomiting secondary to PCA and had a single episode of hematemesis. This was self-limited and presumably represents a MalloryWeiss tear due to retching. Two women were readmitted because of pain. One was treated with intravenous narcotics for 18 hours with good results and discharged within 24 hours. The second had severe postembolization syndrome, and hysterectomy was performed on postembolization day 12. Four women developed small hematomas at the puncture site, which were treated by observation without further incident. There were no infectious complications of UAE.
305 No. (%) of patients interviewed No. of patients interviewed who had complained of menorrhagia No. of patients interviewed who had complained of bulk symptoms
240 185 (77)
180 121 (67)
109 59 (54)
Interview Results At the time of writing, follow-up was 3 months or more for 240, 6 months or more for 180, and 1 year or more for 109 women. Seventy-seven percent (185/240) of patients were interviewed at 3 months, 67% (121/180) at 6 months, and 54% (59/109) at 12 months (Table 1). Eighty percent of patients reported return to full activity within 4 days after UAE. Average time to complete recovery was 13 days (range 0-60 days, 95% CI 11-15 days). Clinical success was defined as moderate or better improvement in abnormal bleeding. Women who replied that this was not applicable were excluded from calculations. At 3 months the clinical success rate was 87% (95% CI 81-91%, t55/179 patients); at 6 months it was 87% (95% C1 80-92%, 101/116 patients); and 1 year it was 86% (95% CI 75-93%, 50/58 patients; Table 2). Clinical success was defined as moderate or better improvement in bulk-related symptoms (urinary frequency or urgency, constipation, sensation of pressure, etc.). Patients who denied bulk-related symptoms were
not included in calculations. At 3 months the clinical success rate for control of these symptoms was 85% (95% C178-90%, 103/121 patients); at 6 months it was 83% (95% CI 74-89%, 71/86 patients); and at 1 year it was 86% (95% C172-93%, 36/42 patients; Table 3). Changes in menorrhagia and bulk-related symptoms were determined as independent variables, as responses for each one appeared to occur at similar rates regardless of the presence or absence of other symptoms .7 Patient acceptance, as determined by a "yes" response to the relevant question, was 84% at 3 months, 85% at 6 months, and 86% at 1 year (Table 4). Eleven patients indicated during at least one interview that they would not have UAE again. Three of these women stated that some of their symptoms worsened after UAE (see below) and four had no significant improvement. The remaining four women all had significant improvements in menorrhagia and/or bulk-related symptoms; however, three considered postoperative cramping excessive, and one was dissatisfied because she developed a chronic discharge (for which she never sought medical attention). It should be noted that the three
TABLE 2. UAE Results, Menorrhagia Length of Follow-up
3 months 6 months 1 year
55 37 24
Great Moderate Improvement Improvement 79 50 23
21 14 3
Slight No Clinical Not Total Improvement Change Worse Success Applicable Interviewed 13 7 6
9 7 0
2 1 2
155 101 50
6 5 1
185 121 59
Uterine Artery Embolization for Leiomyomata Uteri Hutchins et aJ
TABLE 3. UAE Results, Bulk-Related Symptoms Length of Follow-up
3 months 6 months 1 year
43 44 23
Great Moderate Slight No Clinical Not Total Improvement Improvement Improvement Change Worse Success Applicable Interviewed 41 21 10
19 6 3
5 2 2
10 10 4
3 3 0
103 71 36
64 35 17
185 121 59
patients who were dissatisfied because of postoperative pain were operated very early in our experience, before the current pain-control protocol was fully developed. Preprocedure average uterine volume was 993 ml (range 75-19640 ml). Ultrasound examinations were performed in 92 (38%) of 240 patients at follow-up of 3 months or more. Two women had slight increases in uterine volume, and their data are excluded from outcome data as failures. Of patients who did respond in this measure, the average reduction in uterine volume was 48% (range 1-90%, 95% CI 44-52%).
symptoms (one of the two patients who showed slight increase in postembolization uterine volume). The second patient experienced a 30% decrease in uterine volume but was dissatisfied with this result. Both women desired repeat arteriograms to determine if reembolization might be helpful. In both patients, the uterine arteries remained occluded and no reembolization was performed. Six patients underwent hysterectomy after UAE. One (described above) was for severe postembolization syndrome and three were for failure of UAE to relieve bulk-related symptoms. The remaining two women had the surgery performed by local gynecologists when they aborted submucosal fibroids 3 and 5 months after UAE. Had they been treated in our practice, we believe that these patients would probably not have resorted to hysterectomy. Five patients had myomectoinies after UAE. Three were for failure of UAE to control bulk-related symptoms and two were to maximize chances for conception and pregnancy.
Failure of Therapy
All patients in whom UAE failed to provide clinical success for control o f menorrhagia and/or bulkrelated symptoms reported only slight improvement or no change in symptoms. Exceptions to this were four women who experienced worsening of symptoms. Three reported worse dysmenorrhea despite improvement in menorrhagia in one of them. One woman reported worsening of menorrhagia, with no change in bulk symptoms. On questioning, it appears that she had had 2 to 3 days of very heavy bleeding before UAE, and now had 4 to 6 days of mild to moderate bleeding with each menses. One woman stated that both menorrhagia and bulk symptoms were worse. Her UAE was a technical failure, and it is thus not surprising that there was no change in her symptoms.
One patient from this series became pregnant after UAE and delivered a healthy singleton at term without complications. Another conceived after ovulation induction and in vitro fertilization, and at time of writing was in the first trimester with a twin gestation that appeared to be progressing normally. Thirteen women spontaneously expelled fibroids in the initial weeks after embolization, s Presumably, these were pedunculated submucous fibroids that were sloughed after infarction. None required other than expectant management. Many women complained of a persistent mucoid vaginal discharge for up to 3 months after UAE. It sometimes had a purulent or bloody component. This spontaneously resolved in all patients, and may represent sloughing of small submucous fibroids. Approximately 40% of women developed a fever in the first 2 weeks after UAE. This is considered postembolization syndrome, and is commonly seen after
TABLE 4. UAE, Patient Satisfaction Length of Follow-up
3 mo 6 mo
Further Therapies Two patients had repeat arteriograms. In one woman this was for failure of UAE to change her
August 1999, Vol. 6, No. 3
The journal of the American Association of Gynecologic Laparoscopists
"toxic" as one would expect in similar circumstances in which true infection is present, and they rarely have a temperature above 102 ~ F. As with acute fibroid degeneration, we have been unable to delineate any more objective criteria for distinguishing this syndrome from true infection. We can only recommend caution when practitioners first manage patients after UAE to avoid overtreatment. The most significant challenge was postprocedure pain. At first we underestimated its magnitude. Success in treating the problem is reflected in the approximately 90% patient acceptance, yet it remains an incompletely resolved issue. Pain control is the only reason for overnight hospitalization. If pain were further decreased, UAE could be performed as a sameday outpatient procedure, with patients discharged within 4 to 6 hours. Our data indicate a high degree of subjective improvement as well as durability in both bulk symptoms and menorrhagia, 92% at 12 months after UAE. We were concerned that patients Iost to follow-up might represent data that would be dramatically different from what we were able to retrieve. To pursue this question we compared those who responded to the questionnaire with nonresponders using initial uterine volume, PVA dose, and PCA morphine total dose. We found no statistically significant differences between responders and nonresponders. Thus we conclude that data from nonresponders would be highly unlikely to change our outcome data were they available. Pregnancy after UAE remains to be more fully evaluated before the surgery can be unequivocally recommended to women planning future childbearing. The current literature 14-17suggests that it has no negative impact on fertility and pregnancy outcome. Nevertheless concern persists over its impact on pregnancy rates, placental insufficiency, and possible weakening of the uterine wall. This issue awaits the prospective evaluation of larger populations. For the present, we recommend informing women of this fact before making UAE available to them. Similarly, studies measuring changes in hemoglobin, blood loss during menses, and quantification of pain should be undertaken. We chose primarily to evaluate symptomatic response, since these symptoms are responsible for women seeking therapy. The high degree of therapeutic success coupled with the paucity of complications in this series suggests that we may be at a major turning point not only for management of leiomyomata but for gynecology
embolization or infarction of any solid organ. 9 It was not associated with increasing pain or discharge, and in all women it resolved after 3 to 5 days, usually without treatment. Some patients received conservative therapy with acetaminophen and fluids. Discussion Although uterine artery embolization has been available for over 20 years, its performance was limited to control of acute hemorrhage .2,3.10In the meantime, nongynecologic embolotherapy for treatment of neoplasms and destruction of organs such as kidneys and spleen was widely performed.n The procedure was first proposed for uterine leiomyomas, but initially only as a preoperative maneuver to reduce blood loss during myomectomy.12 In addition to its success for this purpose, it soon became apparent that UAE was.highly effective in controlling myoma-associated menorrhagia as well as causing substantial reduction in size of the fibroid uterus. It was primary treatment in a series of 187 women with fibroids. 5 The first United States series was reported in 1996. 4 Our results of an average 48% reduction in uterine volume and control of menorrhagia in 80% to 90% of patients are similar to those in these initial reports. In fact, the consistency among all three series is noteworthy. The theoretical basis for this procedure as a treatment for uterine fibroids was predicted by the fact that reduction in size of fibroids induced by gonadotropinreleasing hormone analog was mediated through reduction in uterine blood flow. 13Similarly, the effect of UAE on menorrhagia was heralded by its earlier success in controlling pelvic hemorrhage. Complications of UAE in this series were rare (0.02%) and minor. Approximately 40% of women had some form of postembolization syndrome consisting of pain, fever and leukocytosis, a normal response to this therapy. In the first patient in our series, our poor understanding of this response was largely responsible for the woman ultimately undergoing hysterectomy. No patient since then required treatment other than analgesics. The fact that these patients have a combination of signs and symptoms commonly associated with pelvic infection is a true clinical challenge. We find it helpful always to be mindful that what we are seeing is acute fibroid degeneration, a syndrome with which all gynecologists are familiar. These patients, although uncomfortable, do not appear subjectively as
Uterine Artery Embolization for Leiomyomata Uteri Hutchins et al
as a whole, since this diagnosis is such a major part of gynecologic practice. Partially palliative procedures such as m y o m a coagulation as stand-alone operations, for example, are made unnecessary by UAE. Myomectomy and hysterectomy may be dramatically reduced in favor of this low-risk, highly effective treatment, especially when future childbearing is not an issue.
Berkowitz RE Hutchins FL, Worthington-Kirsch RL: Vaginal expulsion of submucosal fibroids following uterine artery embolization: A report of three cases. J Reprod Med 44:373-376, 1999
. Hemingway AP: Complications of embolotherapy. In: Current Practice of Interventional Radiology. Edited by S Kadir. Philadelphia, BC Decker, 1991, pp 104-109 10. Rosenthal DM, Colapinto R: Angiographic arterial embolization in the management of postoperative vaginal hemorrhage. Am J Obstet Gynecol 151:227-231, 1985
1. Greenberg MD, Kazamel TIG: Medical and socioeconomic impact of uterine fibroids. Obstet Gynecol Clin North Am 22:625-636, 1995 2. Gilbert WM, Moore TR, Resnik R, et al: Angiographic embolization in the management of hemorrhagic complications of pregnancy. Am J Obstet Gynecol 166:493-497, 1992
11. Castaneda-Zuniga WR, Tadavarthy SW, eds. Interventional Radiology, 2nd ed, vol 1, section 2, Vascular Embolotherapy. Baltimore, Williams & Wilkins, 1992, pp 8-223
3. Greenwood CH, Glickman MG, Schwartz PE, et al: Obstetric and nonmalignant gynecologic bleeding: Treatment with angiographic embolization. Radiology 164:155-159, 1987
12. Ravina JH, Bouret JM, Fried D, et al: Value of preoperative embolization of uterine fibroma: Report of a multicenter series of 31 cases. Contracept Fertil Sex 23:45-49, 1995
4. Goodwin SC, Vedantham S, McLucas B, et al: Uterine artery embolization for uterine fibroids: Results of a pilot study. J Vasc Interv Radiol 8:517-526, 1997
13. Matta WHM, Stabile I, Shaw RW, et al: Doppler assessment of uterine blood flow in patients with fibroids receiving gonadotroin-releasing hormone agonist buserlin. Fertil Steril 49:1083-1085, 1988
5. Ravina JH, Aymard A, Bouret JM, et al: Embolisation arterielle particulaire: Un nouveautraitement des hemorragies des leiomyomes uterins. Presse Med 27:299-303, 1998
14. Stancato-Pasik A, Mitty HA, Richard HM III, et al: Obstetric embolotherapy: effect on menses and pregnancy. Radiology 201 (P); 179, 1996
6. Worthington-Kirsch RL, Hutchins FL, Popky GL: Uterine arterial embolization for the management of leiomyomas: Quality of life assessment and patient response. Radiology 208:625-629, 1998
15. McIvor J, Cameron EW: Pregnancy after uterine artery embolization to control hemorrhage from gestational trophoblastic tumour. Br J Radiol 69:624-629, 1996
7. Worthington-Kirsch RL, Delaney ML, Hutchins FL: Uterine artery embolization for the management of myomata in patients without complaints of menorrhagia. Presented at the 1st international symposium on embolization of uterine myomata, 10th international conference of the Society for Minimally Invasive Therapy, London, September 3, 1998
16. Poppe W, Van Assche FA, Wilms G, et al: Pregnancy after transcatheter embolization of a uterine artery malformation. Am J Obstet Gynecol 156:1179-1180, 1987 17. Wilms G, Peene P, Baert AL: Transcatheter arterial embolization in the management of gynecologic bleeding. J Biol Rhythm 73:21-25, 1990