Uterine artery embolization has become an alternative to hysterectomy for the treatment of fibroids. The author addresses the issues of patient selection, UAE technique, side effects, and complications, and present a review of the literature.
is a medical student at Northwestern University Medical
School, Chicago, IL
. Dr. Goodwin
is Chief of Vascular and Interventional Radiology in the
Department of Radiological Sciences, UCLA Medical Center, Los
Uterine fibroids are benign tumors that are estrogen responsive
and occur in 40% of women, making fibroids the most common tumor in
the female reproductive tract.
Patients with fibroids commonly present with dysmenorrhea and/or
menorrhagia, and may have compressive symptoms including urinary
frequency, abdominal distention, and constipation.
The definitive treatment for fibroid disease is hysterectomy.
Approximately 600,000 hysterectomies are performed annually in the
United States, and 25% to 35% of these hysterectomies are for
treatment of symptomatic fibroids.
Although this surgical treatment is relatively safe and completely
cures the disease, it is a major surgical procedure with a major
complication rate of 1% to 2% and a death rate of 0.1%.
It also guarantees infertility, since the uterus is removed
In order to preserve the uterus, numerous alternative therapies
have been developed and investigated. These include myomectomy,
endometrial resection and ablation, and medical therapy with
progestational compounds and gonadotropin-releasing hormone (GnRH)
agonists. However, each alternative treatment has associated
problems. Myomectomy is the most invasive treatment second only to
hysterectomy. It carries the risk of excessive intraoperative blood
postoperative adhesions, and decreased fertility (postmyomectomy
fertility rate is approximately 40%).
Endometrial resection is associated with hematoma formation,
ectopic pregnancies, and cyclical pelvic pain
and has a 6-month postprocedure failure rate of 24%.
Use of GnRH agonists is limited to 6 months due to its tendency to
and symptoms of a hypo-estrogenic state, such as hot flashes,
vaginal dryness, decreased libido, and mood changes.
Once discontinued, rapid regrowth of fibroids to their original
size is often noted.
Uterine artery embolization (UAE) has been developed as an
additional treatment option. Embolization in the pelvis is a
well-established technique that has been employed in a variety of
clinical situations. The first reported use of embolization in the
pelvis for obstetrical-gynecological hemorrhage appeared in the
literature in 1979 to control postpartum hemorrhage.
Since then, the technique has been used to treat postpartum,
and postabortion bleeding,
and for presurgical prophylaxis against excessive hemorrhage.
Uterine artery embolization for the treatment of fibroids is a more
recent development. Success rates for this treatment have ranged
between 85% and 94%.
UAE works by causing infarction of the fibroids because fibroid
arteries are essentially end arteries. The myometrium is spared
because of the rich collateral network in the pelvis. Figure 1
shows a computed tomography (CT) scan of a woman's pelvis the day
of embolization. Contrast administered intra-arterially during the
embolization is retained in the fibroid (along with the embolic
material) while no contrast is seen in the myometrium because it
has washed out.
A complete history and physical examination is necessary, as
well as ultrasonographic or magnetic resonance imaging (MRI)
documentation of fibroids to allow planning of optimal approach for
treatment. Malignancy must be ruled out before selection for UAE.
Rapidly enlarging fibroids must be investigated, and women older
than 40 years of age or with intermenstrual bleeding usually
undergo endometrial biopsy to rule out endometrial hyperplasia or
neoplasia. Contraindications include pregnancy, active pelvic
infection, active vasculitis, history of pelvic irradiation,
life-threatening contrast allergy, uncontrollable coagulopathies,
and renal insufficiency. In our experience, laboratory studies can
usually be limited to a hematocrit in bleeding patients. We have
not found routine cultures to be necessary.
UAE is performed under conscious sedation and involves selective
catheterization of both uterine arteries with 4Fr to 5Fr visceral
selective catheters or with co-axial microcatheters. Particulate
emboli are used, especially polyvinyl alcohol (300 to 700 µm).
Angiography is used to determine complete embolization of both
uterine arteries. Figure 2 illustrates pre- and postembolization
angiograms of the left uterine artery in a woman with a large
myomatous uterus. The procedure time is 45 to 135 minutes with 90%
of the procedures lasting from 50 to 75 minutes. There is an
average procedural exposure of 20 rads to the ovaries. Most
patients are discharged within 24 hours; however, hospitalization
for up to 48 hours is sometimes required for the management of
The majority of patients return to normal activities within 1 week.
Side effects and complications
Postembolization syndrome (symptoms of pain, fever, vomiting,
nausea, and anorexia) occurs in most women undergoing UAE. This
syndrome occurs in patients undergoing em-bolization procedures in
general and has been well documented, especially in the liver
The symptoms usually resolve in 7 days and may be difficult to
differentiate from infection. Differential features include
leukocytosis without a left shift and negative cultures. Infection
is characterized by delayed pain and fever with leukocytosis with a
left shift, particularly when associated with purulent discharge.
Broad spectrum antibiotics will be effective in most patients,
although approximately 1 in 200 patients will require a
hysterectomy because of infection. Approximately 5% of patients
will pass tissue vaginally in the months following embolization.
This is not a significant problem if patients are aware of the
possibility, unless the cervical os opens and the tissue becomes
stuck, leading to infection. A D&C and/or hysteroscopy may be
necessary to address this problem in a small number of women.
Crampy postembolization pain occurs frequently. Pain usually
peaks the first day following the procedure, but occasionally on
the second day, and rarely the third day postprocedure. Resolution
of pain can be expected in 1 week. Pain syndromes lasting longer
than 2 weeks are rare. The pain is probably due to the ischemia
produced by the embolization procedure. Strong analgesics and,
particularly, patient-controlled analgesia (during hospitalization)
is extremely helpful during this period. The severity of pain is
unrelated to the size of fibroid and is unpredictable. We believe
that an overnight admission is desirable, but others have advocated
embolization on an outpatient basis.
Exposure of the ovaries to radiation is a concern, especially if
the patient desires future fertility. Ravina
reported the radiation dose to be negligible, which is in contrast
to our calculation of 20 rads to the ovaries. A study of 11
patients who underwent UAE estimated the absorbed ovarian dose to
be an order of magnitude more than for barium enema or for
hystero-salpingogram, and several orders less than for radiation
treatment of Hodgkin's disease. Based on the well-established risks
of pelvic radiation for Hodgkin's disease, the study concluded that
there is no risk for acute/long-term radiation injury or to
Low-dose fluoroscopy units with pulsed fluoroscopic capability will
keep fluoroscopy dose to a minimum. Avoidance of detailed
fluorography as well as strict coning down to the catheter tip to
keep the ovaries out of the primary field will also aid in
Premature menopause has been documented in 1% to 2% of patients
and is believed to result from non-target embolization of the
ovaries via the collateral bed between the ovarian and the uterine
arteries. This is a risk that must be considered in a premenopausal
woman who desires future fertility.
Sexual dysfunction is an unusual complication following UAE. In
those women who experience strong uterine contractions during
orgasm (internal orgasm) those contractions can be lost following
UAE, although clitoral orgasm is not lost. This complication may be
related to embolization of the cervical vaginal branch resulting in
cervical ischemia or ischemia of the neural plexus adjacent to the
cervix. Other nontarget embolization complications are rare with
damage to the rectum, bladder, buttocks, and sciatic nerve
occurring at much <1 in 1000 cases.
Two deaths have been reported in patients undergoing UAE in
Europe: one patient had sepsis and multi-organ failure and the
other had pulmonary embolism.
No deaths have been reported in more than 4000 cases performed in
the United States.
Finally, complications related to angiography are rare and include
0.2% hematoma, 0.2% to 0.4% arterial thrombosis, and 0.05% false
Transcatheter uterine artery embolization for the treatment for
fibroids was first reported in the English literature in 1995 by
Ravina et al,
and soon thereafter, by investigators including Goodwin et al
and Worthington-Kirsh et al
from the United States, and, from the United Kingdom, Bradley and
Currently, worldwide technical success rates reported for UAE are
98% to 100%.
The overall success rate, defined as marked or complete resolution
of the primary fibroid-related symptoms requiring no further
surgical treatment, is 85% to 94%.
In women who are treated with UAE for menorrhagia, reported success
rates are 86% to 92%
and symptomatic relief occurs as early as the first cycle in the
majority of patients. Bulk-related symptoms (for example, urinary
incontinence and abdominal distention) improved in 85% to 96% of
patients. Six-month pelvic ultrasound follow-up demonstrated an
average size reduction in dominant uterine leiomyomata of
approximately 60% in comparison to preprocedural volume.
The ultrasound image in Figure 3 shows a reduction in maximum
uterine cross-sectional diameter from 9.5 to 4.9 cm, representing
>75% reduction in volume, at 6-month follow-up.
In the early 1990s, Ravina
first noticed unexpected fibroid shrinkage in patients who
underwent embolization for premyomectomy or hemorrhagic conditions
in the pelvis. In 1995, his group reported using UAE as a primary
treatment for fibroids in 16 patients.
At a mean follow-up of 20 months, 11 of the 16 patients reported
complete resolution of their symptoms, three had partial
improvement, and only two failed UAE and required subsequent
surgery. Ultrasound demonstrated fibroid volume reduction of 20% to
80% at 3-month follow-up. One of their patients subsequently became
pregnant. In 1998, they reported data on a subsequent 184 patients
who underwent UAE for symptomatic fibroids. The average fibroid
volume reduction was 66% at a 10-month follow-up.
At 30-month follow-up, 93% of patients reported substantial
improvement (with improvement in menorrhagia in 91% of patients)
while 7% did not improve. One patient underwent hysterectomy 8 days
after UAE due to fibroid necrosis. Seven pregnancies, one
unsuccessful, followed the UAE procedure.
Goodwin et al
studied 11 patients who underwent bilateral uterine artery
embolization in 1996 and had a mean follow-up of 5.8 months. All
patients underwent technically successful embolization. Eight of
the nine patients who completed the follow-up questionnaire
reported noticeable symptomatic improvement, including three women
with complete resolution of symptoms. One women had no clinical
response and another developed endometritis and pyometria 3 weeks
after the procedure necessitating hysterectomy. Ultrasound
confirmed that uterine volumes decreased an average of 40% and
dominant fibroid size decreased 60% to 65%.
These promising results led to a subsequent study of 59 patients
who underwent bilateral UAE with a mean follow-up of 16.3 months.
Fifty-six patients presented with bleeding and 47 presented with
pain. Of all patients who underwent bilateral UAE, 81% had moderate
or better improvement in their symptoms. Ninety-two percent had
reductions in their uterine and dominant fibroid volumes with an
average decrease of 42.8% and 48.8%, respectively. One patient had
permanent amenorrhea following uterine artery embolization.
Six patients eventually elected to have hysterectomy after
having undergone UAE. Three of the six patients were diagnosed with
adenomyosis by postsurgical histopathology, which suggests that
adenomyosis is a condition that predisposes a patient to fail UAE
therapy. Of these three patients, one also had cervical squamous
carcinoma and another had chronic hematometra concurrent with
adenomyosis. A fourth patient had an infectious complication that
necessitated hysterectomy. A fifth patient developed renal cell
carcinoma and underwent hysterectomy for reasons unrelated to her
fibroids. A sixth patient was lost to follow-up.
In 1998, Worthington-Kirsch et al
studied 53 patients who underwent UAE and had excellent clinical
outcomes. Menorrhagia secondary to fibroids was controlled in 88%
of patients and bulk-related symptoms were controlled in 94% of
patients. Complications in this study were low and included severe
abdominal pain in two patients.
In a Canadian multicenter trial of 24 premenopausal women in
1998, Pron et al
reported successful UAE with no procedural complications. The
majority of patients were admitted for management of postprocedural
pain (average <48 hours). Many experienced persistent pelvic
pain that lasted up to 1 week. Postprocedural vaginal spotting and
vaginal discharge for several days were common. Postdischarge, five
women returned to the Emergency Department, three for pelvic pain
and two for menorrhagia. Generally, the majority of women reported
being able to return to work, or their usual activities, 1 week
after the UAE. Several women (5 of 24) required longer recovery
periods of up to 3 weeks.
Walker et al
studied 91 patients following UAE; 80% of the patients reported
that the procedure was completely successful. One patient required
hysterectomy due to an E coli-infected tubo-ovarian abscess.
Another patient had a perforated myometrium secondary to attempted
expulsion of an infected fibroid. Another patient required a blood
transfusion 4 weeks postprocedure for anemia. Finally, one patient
suffered transient amenorrhea and another two had permanent
Spies et al
studied 50 patients with a mean follow up of 12.3 months. Using
MRI, a median decrease in uterine and dominant uterine volume of
48% and 78% was achieved, respectively. Menstrual bleeding was
improved by 89%, with 81% of patients moderately to markedly
improved. Pelvic pain and pressure was improved in 96% of patients,
with moderate to marked improvement in 79%. One 50-year-old patient
had permanent amenorrhea after one normal menstrual cycle
postprocedure. In this patient, it was difficult to differentiate
between normal menopause and procedure-induced amenorrhea.
In recent years, UAE has become increasingly popular. In the
United States, the number of UAE procedures performed increased
from 50 in 1996 to more than 4,000 in 1999
and a proliferation of literature in the lay press has indicated
strong patient demand for this nonsurgical treatment.
The advantages of UAE compared with hysterectomy include
avoidance of surgical risks, shorter hospitalization, and the
potential for maintaining fertility. However, one must remember
that, unlike hysterectomy, UAE is palliative rather than
UAE patients can expect excellent short-term and mid-term
results with regards to menorrhagia, pelvic pain, bulk-related
symptoms, and reductions in uterine volume. Longer term results are
not known. In the near future, randomized controlled studies
comparing UAE with myomectomy, as well as studies of UAE regarding
preservation of fertility, will further delineate its role in the
treatment of uterine fibroids. AR