Skip Navigation U.S. Department of Health and Human Services
Agency for Healthcare Research Quality
Archive print banner
The FIBROID Registry

This information is for reference purposes only. It was current when produced and may now be outdated. Archive material is no longer maintained, and some links may not work. Persons with disabilities having difficulty accessing this information should contact us at: Let us know the nature of the problem, the Web address of what you want, and your contact information.

Please go to for current information.



Uterine leiomyomas, commonly known as fibroids, are smooth muscle and fibrous tumors of the uterus. They are very common among women of reproductive age and may be asymptomatic. The cumulative incidence of a diagnosis of symptomatic fibroids among women ages 25 to 44 is 30 percent1 and among women ages 35 to 49, 35 percent had a previous diagnosis of fibroids and on ultrasound examination, an additional 51 percent had unsuspected fibroids.2

Leiomyomas are the most common indication for hysterectomy in the United States. Approximately 200,000 women undergo hysterectomy each year due to leiomyomas, which represents 30 to 40 percent of all hysterectomies performed annually.3 An additional 34,000 women undergo myomectomy (surgical removal of the fibroids, leaving the uterus intact) each year according to the National Inpatient Sample (NIS) on-line query system, available at In 2000, hospitalizations due to uterine fibroids were estimated to exceed 250,000 admissions. Hospital costs, costs related to outpatient diagnosis and care, and the morbidity associated with leiomyomas add up to make this a major public health concern.

Fibroids may cause a variety of symptoms, depending on their size and location. The most common symptom is heavy menstrual bleeding (menorrhagia), which may be associated with severe menstrual cramping (dysmenorrhea). Fibroids also cause a variety of other non-bleeding symptoms, including pelvic pain, pelvic pressure, back or leg pain, urinary frequency, incontinence, and constipation. In addition, fibroids may be associated with pain during sexual intercourse (dyspareunia). Fibroids also increase the risk of pregnancy-related complications, and there is a suspected association with infertility.

Surgical Options

Surgical treatment options for uterine fibroids include hysterectomy and myomectomy. Myomectomy is typically performed for women who want to maintain their fertility. Both hysterectomy (removal of the uterus with or without removal of the ovaries and fallopian tubes) and myomectomy are usually abdominal surgeries that entail 6 to 8 weeks for patient recovery. Hysterectomy has been demonstrated to be effective in controlling fibroid-related symptoms, and it results in substantial improvement of health-related quality of life.4,5 Abdominal myomectomy has been less well studied, but it appears to control fibroid-related symptoms in 80 to 90 percent of patients while preserving the uterus.6

There are less invasive variations on these procedures, such as vaginal hysterectomy, laparoscopic-assisted vaginal hysterectomy, hysteroscopic myomectomy, and laparoscopic myomectomy. These less-invasive options generally are limited to patients with relatively small uteri and those who have only one fibroid or just a few fibroids in accessible locations within the uterus. Thus, they may not be appropriate therapies for some women who have uterine fibroids.

The most comprehensive published data currently available is from the CDC's hysterectomy surveillance project, which uses samples of discharge data to estimate national trends. The most recent publication was in 2002, looking at data from 1994 to 1999. The publication is available at

In the 2002 study, the ratio of vaginal (including LAVH) to abdominal hysterectomies did not significantly change, although the proportion of "vaginal" hysterectomies which were LAVH did increase; black women were more likely to get abdominal hysterectomies, and, since 68 percent of hysterectomies in black women were for fibroids, it seems unlikely that there was a substantial shift. Although it is possible that more recent data would show a different trend nationally, more recently published data on the proportion of surgery for fibroids, which are performed laparoscopically, could not be found.

Uterine Embolization

Embolization is an established interventional radiology technique that has been used since the early 1980s to control postpartum hemorrhage. Uterine embolization for fibroids was first introduced in the United States in 1997, and it has been quickly adapted in practice both in the United States and around the world.

Uterine embolization is a procedure that is based on angiographic techniques and is performed by interventional radiologists. An angiographic catheter is advanced from the femoral artery at the top of the leg into first one and then the other uterine artery using fluoroscopic guidance. Small plastic beads or particles are injected into each uterine artery. The flow in the vessel carries the particles into the branches feeding the fibroids, which typically are enlarged and have increased arterial flow.

The goal of the procedure is to occlude permanently the fibroid arterial supply, while sparing the normal myometrium and leaving the uterine arteries patent. Using current materials and techniques, this goal usually is achieved. As a result of the embolization, the fibroids undergo ischemic infarction and over a period of weeks and months, the fibroids shrink and are converted to scars via hyaline degeneration. Once they are successfully treated, individual fibroids cannot recur. As with any uterine-sparing therapy, including myomectomy, new fibroids may develop over time in a premenopausal woman.7

Although the data to date have been very promising, long-term (5 year) fibroid recurrences rates for uterine artery embolization (UAE) are not known.

There have been many case series reported on the outcome from this procedure showing that bleeding symptoms are improved in 82 to 98 percent of patients, and bulk symptoms are improved in 64 to 97 percent of patients.8-15 Serious complications have been rare16 and they typically occur less often than with abdominal hysterectomy.17

The FIBROID Registry

In 1999, the Society of Interventional Radiology (SIR) Foundation, the SIR research and education foundation, commissioned an evidence report from the RAND Corporation to assess the current state of knowledge on uterine embolization and more importantly to develop a research agenda focused on the procedure.18 The report identified four important research needs:

  • Development of a disease-specific quality-of-life instrument.
  • Establishment of a registry for prospective collection of outcomes data.
  • A randomized trial.
  • A comparative cost analysis.

The SIR Foundation subsequently funded the development of a disease-specific quality-of-life instrument19 (used in the registry and described below) and oversaw the development of the current study, the Fibroid Registry for Outcomes Data (FIBROID).

Spies and colleagues at Georgetown University developed and validated a disease-specific quality of life instrument for symptomatic fibroids, the Uterine Fibroid Symptom and Quality of Life Questionnaire (UFS-QOL).19 The UFS-QOL was developed from focus groups, clinician opinion, and a thorough literature review, using standard instrument development techniques. The validation study was conducted in both nonsymptomatic volunteers and women with uterine fibroids.

The patient-rated severity scores were categorized into three groups (< or = 4, > 4 but < 8, and = or > 8) based on distribution of scores and clinical meaningfulness. The UFS-QOL discriminated between mildly, moderately and severely symptomatic women, thus indicating sensitivity in the instrument. Similar results were found with the physician ratings of patient symptom severity. Test-retest reliability was also good, with intraclass correlation coefficients of 0.76 to 0.93.

The final instrument yielded a questionnaire with 37 questions, including 8 symptom questions and 29 quality of life questions. There are six quality of life subscales:

  • Concern.
  • Activities.
  • Energy/mood.
  • Control.
  • Self-consciousness.
  • Sexual function.

The questionnaire can be scored by hand and yields both a symptom and quality of life score. Each of these scores range from 0 to 100, but in an inverse manner. For the symptom scale, a lower score is better because it indicates lower symptoms, while a higher quality of life score indicates better quality of life status. These scores are the primary outcome measure of the FIBROID Registry. a

FIBROID Registry Goals

The goals of the FIBROID Registry are to:

  • Capture high-quality patient safety and effectiveness data for UAE.
  • Demonstrate patient volume and rates of acute outcome events for patients undergoing uterine artery embolization (UAE).
  • Collect and quantify longitudinal functional and clinical outcomes of patients undergoing UAE.
  • Assess and benchmark clinical practice patterns (patient selection, devices, use of the procedure across country).
  • Facilitate data collection for post-market surveillance and randomized clinical trials (using infrastructure in place for the Registry).
  • Collect and quantify use of resources for patients undergoing UAE.
  • Develop and refine standards of care for use of UAE for leiomyomas.
  • Monitor risk-adjusted procedural outcomes.
  • Provide data to support development and design of randomized clinical trials.

Registry Research Questions

The FIBROID Registry was designed to answer the following key questions about uterine embolization:

  1. Is UAE a safe treatment for leiomyomas? What is the incidence of minor and serious short-term complications?
  2. Is UAE an effective treatment for leiomyomas? What is the likelihood of symptom relief?
  3. How durable is the treatment? What is the likelihood of a subsequent procedure or medical therapy to treat recurrent symptoms of leiomyomas?
  4. What is the likelihood that a woman who undergoes UAE will be able to conceive and deliver a subsequent pregnancy?
  5. Is there a difference in outcome based on device (product size, primary and secondary embolic material) used?
  6. Are there certain patient subgroups at higher risk or that have an increased likelihood of treatment benefits?
  7. What number of procedures should be recommended for training interventional radiologists and for maintaining skills?

Registry Structure

The Registry is a cooperative venture between SIR Foundation and the Duke Clinical Research Institute (DCRI). It is a voluntary registry of patients treated with uterine embolization at participating hospitals and is organized as follows:

  • The Steering Committee is responsible for all aspects of Registry organization, including final design of all data collection forms, adjudication of adverse events, communication with onsite principal investigators regarding site performance issues, and establishment of analytical and publication plans and other policies.
  • An external advisory committee was formed prior to initiation of the Registry that included representation from the Society of Interventional Radiology and the Society of Interventional Radiology Foundation, the Duke Clinical Research Institute, and the U.S. Food and Drug Administration (FDA) Center for Devices and Radiological Health (CDRH). The advisory committee consisted of members with expertise in interventional radiology, epidemiology, and gynecology. Members of the advisory committee provide input as needed.
  • A committee of sponsors provides input, primarily on specific data requests for sponsors.
  • The roles of "core" and "participating" sites and on-site investigators are described in more detail below.

All data management, site data management, data analysis, and data reporting are managed at the DCRI. SIR Foundation staff members are responsible for overall site management, including site identification, investigator recruitment, training, and monitoring, and they serve as primary liaisons with the investigators, Steering Committee, and external organizations.

SIR Foundation staff members coordinate and participate in data interpretation and preparation of manuscripts, with a writing group composed of members of the Steering Committee. The onsite investigators and research coordinators gathered the baseline and 30-day outcomes data and submitted the data to the registry. Longitudinal followup of patients beyond 30 days is conducted by DCRI.

UAE Registry Sites

Recruitment of sites was done through the SIR Foundation. Members of the SIR were surveyed to identify sites that were interested in participating in the Registry. Sites were classified as "core" or "participating" sites depending on the amount of data collected.

Core sites were sites with previous experience (at least 24 procedures) and a willingness to collect complete baseline data on patients. Participating sites had no requirement for prior experience with the procedure and had fewer baseline data requirements. No research funding support was provided to the sites participating in the Registry.


Once patient consent was obtained, baseline demographic data, reproductive and gynecologic history, and medical history were recorded. Each patient at a core site (and many at participating sites) completed the UFS-QOL questionnaire. The interventional radiologist recorded details of pelvic imaging (either ultrasound or magnetic resonance imaging [MRI]). At the time of the procedure, technical details—including data about peri-operative care and embolic material used—were recorded. These baseline and procedure data were entered on a Web-based case report form. Separately, patient contact information was also recorded for core site patients for use in long-term follow up.

Data Collection

Thirty-day followup data were obtained for each patient and were recorded via the Web-based reporting system. Recovery and adverse events were the primary focus of the 30-day followup. AEs were scored according to the SIR scale for severity.20 In addition, AEs were classified using a system of complication definitions derived from American College of Obstetricians and Gynecologists (ACOG) quality indicators, which has been used previously to classify complications of both hysterectomy and myomectomy.21 Details about adverse events were recorded, and a subcommittee of the steering committee adjudicated these. After completion of 30-day followup, longer-term followup data were gathered by DCRI via mailed questionnaires. Non-responders were contacted by telephone and given a choice of completing the questionnaire via interview or having another questionnaire set mailed to them.

The long-term followup schedule was 6 months, 12 months, 24 months, and 36 months after treatment. The 6- and 12-month data collections are complete, and the initial analysis of these data is included in this report. Data gathering is underway for the 24-month followup interval, but these data are not included here.

Initial data analysis was completed using summary descriptive statistics. Predictors of outcomes were assessed using univariable and multivariable analysis, using linear and logistic regression models. Predictors of time to events for multivariable risk-adjustment were examined using Cox proportional hazards regression.

Patient Enrollment

Eighty-five sites completed the process for participation in the Registry. The first patient was enrolled December 13, 2000, and, at closure of the Registry on December 31, 2002, 3,319 patients had been logged into the Registry by 72 sites. Of these patients, complete data on core variables were available for 3,005 (94.9 percent.). Twenty-six of the core, or high-volume, sites logged 2,922 (88 percent) of the patients into the Registry; of these, 2,782 (95 percent) were enrolled. Forty-six of the participating sites logged in 397 (12 percent) patients, of which 378 (97 percent) were enrolled.

Of the 3,160 enrolled patients, in-hospital data were obtained on 3,005 patients, and complete 30-day data were received on 2,729 (86 percent) cases. At the 6-month followup interval, 2,112 patients qualified for further followup. Followup data were received on 1,797 of these (85.1 percent) cases, and 12-month data were received on 1,701 (80.5 percent) cases. A small percentage of cases at each time point were lost to followup.

Summary of Registry Results to Date

Data that may address some of the key research questions listed above are now available, and the results pertinent to them are presented here.

  1. Is UAE a safe treatment for leiomyomas? What is the incidence of minor and serious short-term complications?
    • Overall, the incidence of complications was quite low and recovery was rapid. Patients returned to normal activities in a mean of 14 days. Of those women who worked outside the home (N = 2,404), the mean number of days lost from work (including the day of the procedure) was 10.
    • There were no deaths and only two events that resulted in permanent patient injury. These injuries were bilateral oophorectomy 3 weeks after embolization for pelvic pain in one patient and a puncture site femoral nerve injury in another.
    • Ninety of 3,005 patients reported a total of 94 adverse events during hospitalization for UAE (3 percent).
    • Of the 94 adverse events, 20 events were in-hospital major adverse events (0.66 percent of patients). These were primarily for prolonged pain or nausea (10 events) requiring hospitalization longer than 48 hours.
    • Adverse events (major or minor) occurred between hospital discharge and 30 days in 710 women (26 percent), of whom 191 (7 percent) had more than one adverse event. (Go to Table 2 for a description of major and minor events.)
    • Major events occurred in 111 of 2,729 patients (4.1 percent) within 30 days of hospital discharge. The most common major event was emergency room care or readmission for recurrent pain (65 patients, 2.4 percent).
    • During the period between discharge and 30 days after UAE, 32 (1.2 percent) of patients had a surgical intervention for an adverse event. The most common intervention was dilatation and curettage (D&C), and the primary indication was for management of a fibroid being expelled from the uterus.
    • Three hysterectomies were performed in the first 30 days after the procedure.
  2. Is UAE an effective treatment for leiomyomas? What is the likelihood of symptom relief?
    • Preliminary data only are available at this time, based on 6- and 12-month outcomes, using the UFS-QOL scores as primary measures. Of 2,112 patients eligible for 6- and 12-month followup, complete data were available on 1,797 patients at 6 months and 1,701 patients at 12 months.
    • For total HRQOL scores, 85.4 percent had at least a 10 percent improvement by 6 months, and 87.5 percent had this degree of improvement by 12 months. Further correlation of the degree of improvement necessary to be considered "clinically successful" is ongoing.
    • There were large improvements in symptom scores. The mean score improved from a baseline of 58.0 (SD 20.81) to 19.9 (SD 18.6) at 6 months and 19.2 (SD 17.9) at 12 months, both well in the range seen among normals in the original validation study.
    • Similarly, there were large improvements in HRQOL scores. The mean score improved from a baseline of 47.3 (SD 22.9) to 85.1 (SD 20.1) at 6 months and 86.7 (SD 18.2) at 12 months; again, both scores are in the range of normals in the original validation study.
    • Patient satisfaction with treatment outcome was measured by whether the patient would recommend the procedure to family or friends. At 12 months, 82 percent of women would recommend the procedure.
    • Although overall patient retention was high, and similar to that observed in other studies, data were not available for almost 20 percent of women eligible for longer-term followup. If these women were more likely to have had poor outcomes, both overall success rates and mean improvements in QOL scores would be lower.
  3. How durable is the treatment? What is the likelihood of a subsequent procedure or medical therapy to treat recurrent symptoms of leiomyomas?
    • The definition of recurrence used in the Registry is a return of initially controlled fibroid-related symptoms greater than 12 months after embolization. Determination of the recurrence rate will thus require analysis of the data from 24 and 36 months. Those that do not have symptom improvement sustained for a year were considered clinical failures. For these analyses, the denominators were 1,797 patients at 6 months and 1,701 patients at 12 months.
    • There were 49 hysterectomies in the first year of the Registry, representing 2.9 percent of patients who completed 12-month followup. Most of these surgeries were for symptoms not controlled by the embolization procedure.
    • Additional gynecologic procedures for symptoms were performed in 4.5 percent of patients within 6 months of embolization and an additional 5.0 percent of patients by 12 months and included hysterectomy, myomectomy, D&C, and repeat embolization.
    • Within 6 months after embolization, 7.4 percent of patients needed additional hormonal or medical therapy, and between 6 and 12 months, 3.5 percent of patients needed additional medical therapy.
  4. What is the likelihood that a woman who undergoes UAE and plans subsequent pregnancy will be able to conceive and deliver a subsequent intrauterine pregnancy?
    • The data are insufficient to determine the likelihood of subsequent pregnancy, since the postprocedure interval has been too short for analysis.
    • There have been 12 pregnancies during the first 12 months of the Registry. At this writing, four pregnancies are underway and there have been two miscarriages, one therapeutic abortion, and two deliveries (one planned and one unplanned cesarean section). Data are not currently available on the other pregnancies.
    • Based on the advanced mean age (37) of women who have expressed an interest in pregnancy, the high proportion of prior infertility, and the known impact of fibroids on pregnancy rates, it appears unlikely that the Registry will generate sufficient pregnancies to make meaningful estimates of embolization impact on pregnancy rates or outcomes.
    • To determine whether there is an impact of embolization on fecundity and pregnancy outcomes, or more importantly, whether myomectomy or embolization should be the preferred treatment for patients who want to become pregnant will likely require clinical trials in this patient population.
  5. Is there a difference in outcome based on device (product size, primary and secondary embolic material) used?
    • There does not appear to be any impact of UAE related to the embolic type, embolic particle type, or other technical aspects of the procedure on adverse events.
    • The devices used do not appear to affect outcomes at 6 and 12 months after embolization.
    • Whether subtle differences in the characteristics of the embolic material affect longer term outcome will require analysis of the data from the 24- and 36-month followups.
  6. Are there certain patient subgroups at higher risk or that have an increased likelihood of treatment benefits?
    • At this time, no subgroups have been identified that had a poor outcome or for whom the procedure would not be recommended. The data suggest that most patients who have symptomatic fibroids will benefit from embolization.
    • The one strong predictor of clinical failure was unilateral embolization, which was defined at the outset as a technical failure (standard treatment is bilateral embolization). Not surprisingly, the patients with a technically failed procedure were significantly more likely to be clinical failures at 12 months (hazard ratio 2.56, 95% confidence interval [CI] 1.48-4.42).
    • There were remarkably few predictors of adverse events, and none of these substantially increased the odds ratios (OR) for an event.
    • The only demographic factors that predicted a minor increase in adverse events was African-American race (OR 1.129, P = 0.02) and current or recent smoking status (OR 1.141, P = 0.04).
    • Patients who had prior fibroid-related procedures at baseline were slightly more likely to have an adverse event (OR 1.235, P = 0.0003).
    • Among procedure-related factors, only longer duration of the procedure (OR 1.0037, P = 0.009) had a minimal impact at best. The mean procedure time was 56 minutes.
    • Prophylactic antibiotics did not influence the frequency of adverse events, although only 5 percent of patients did not have prophylaxis. The small proportion of patients without prophylaxis limited the ability to predict adverse events related to this factor. Use of deep venous thrombosis (DVT) prophylaxis decreased the risk of adverse events (OR 0.76, P = 0.005).
    • Although there were factors identified that might influence the risk of failure after the procedure, the broad definition of failure and the minimal scale of the predictors limit conclusions at this time. It is anticipated that analyses of the 24- and 36-month data will provide better insight into the predictors of long-term success or failure.
  7. What number of procedures should be recommended for training interventional radiologists and for maintaining skills?
    • There were no predictors of adverse events or symptomatic outcome to date that relate to site experience, enrollment rates in the Registry, or site type.
    • Only members of the SIR (or their international equivalent) were participants in this Registry. Members are those who have either completed fellowship training in interventional radiology or have extensive practice experience in the field. Most investigators in the Registry were either subspecialty-certified or eligible for certification. By and large, these are practitioners that have considerable angiographic and embolization experience. With relatively little difficulty, most of these practitioners would be expected to learn and master the primary techniques of this procedure.
    • The same cannot be said of those physicians (whether radiologists or not) who do not have extensive training or experience in embolization and subselective catheterization techniques. It is doubtful that uniformity of outcomes noted in this study can be duplicated for those without the requisite background.
    • As with training for the procedure, the number of procedures that would ensure maintenance of skills cannot be estimated based on this dataset.

Future Directions

As with most studies, the initial analysis of this dataset raises questions which may be answered as further data are gathered in this study. Furthermore, subsequent analyses on the existing data may further clarify our understanding of outcome. Among issues raised are the following:

  • The longer term outcome from this therapy is a key question as data continue to accrue. Determining the benefit of a treatment in terms of reduced added costs, morbidity, and time lost for treatment is apparent, but this is true only if the durability of the treatment is known.
  • There is considerable opportunity for additional analysis of this dataset to investigate whether specific subgroups fare better or worse after therapy. For example, it would be useful to determine if prior myomectomy (rather than any prior intervention) leads to more frequent adverse events or predisposes a patient to early recurrence of symptoms.
  • Additional analysis of the subscale scores of the UFS-QOL is also of interest to determine if they correlate with any baseline characteristics or if they relate to specific outcomes.
  • Because fairly detailed anatomic data were collected at baseline, additional analysis of the influences of anatomic factors related to the fibroids or uterus on either baseline or outcome scores may be useful. This will become increasingly interesting when 24- and 36-month data are available. It is presumed that there will be recurrences in symptoms, and understanding whether baseline anatomy influences recurrence may help to define patient selection criteria. Additional detailed analysis might also determine if certain combinations of fibroid size and location affect adverse events, such as fibroid passage.
  • The analysis regarding anatomic factors of the fibroids may provide insights into how a classification system of disease severity based on imaging might be developed and validated.
  • The gathering of fertility data and pregnancy outcomes has only begun, and while definitive conclusions may not be possible, this may yet represent the largest cohort of patients pregnant after embolization. Followup data should provide further insights into potential problems and complications that will be very helpful in designing future comparative studies.

An estimation of resource use—including materials, duration of hospitalization, duration of the procedure, and recovery—has yet to be undertaken. Cost estimates may be possible, but additional resources would be required.

a The FIBROID Registry has been supported through unrestricted grants provided by Biosphere Medical, Inc., Boston Scientific/MEDI-TECH, Cordis (a Johnson & Johnson company), and Cook, Inc., as well as by the Society of Interventional Radiology Foundation. Analysis of the initial data that form the basis of this report was supported jointly by the Agency for Healthcare Research and Quality and the FDA's Office of Women's Health (Contract 03R00027101D).

Return to Contents
Proceed to Next Section


The information on this page is archived and provided for reference purposes only.


AHRQ Advancing Excellence in Health Care