Goodwin, Scott C. MD1; Spies, James B. MD2; Worthington-Kirsch, Robert MD3; Peterson, Eric MD, MPH4; Pron, Gaylene PhD5; Li, Shuang MS6; Myers, Evan R. MD, MPH7; for the Fibroid Registry for Outcomes Data (FIBROID) Registry Steering Committee and Core Site Investigators
Click on the links below to access all the Data Supplements for this article.
Please note that Data Supplement files may launch a viewer application outside of your web browser.
Uterine leiomyomata are a major public health care problem, occurring in at least half of American reproductive-age women.1 Leiomyomata may cause pain, abnormal bleeding, pressure, and other symptoms related to uterine enlargement and fertility problems. Symptomatic leiomyomata lead to 30–40% of all hysterectomies in the United States, and to 150,000–200,000 hysterectomies annually.2 More than 25% of women in the United States will have a hysterectomy by the time they are 60 years old. It is the second most frequently performed surgical procedure after cesarean delivery for women of reproductive age in the United States.3
Uterine artery embolization, as a less invasive alternative to hysterectomy for the treatment of symptomatic leiomyomata, was first reported in 1995.4 The procedure is also known as uterine fibroid embolization. Since then, the procedure has grown in popularity, and currently approximately 25,000 uterine artery embolizations are now performed worldwide annually. In 1999, the Cardiovascular and Interventional Radiology Research and Education Foundation (now the Society of Interventional Radiology Foundation) convened an expert panel at the Rand Corporation in Santa Monica, California, to perform a systematic review of the uterine artery embolization literature and develop an agenda for research of the procedure.5 At this meeting, four areas of recommended research were put forward, including a randomized trial, development of a disease-specific quality of life questionnaire, a cost analysis, and a prospective registry. The first three goals have been met. Three prospective randomized trials have been reported.6–8 A disease-specific quality of life instrument has been developed,9 and several cost analyses have been performed.10,11
To meet the final goal of the Rand panel’s recommendations, the Society of Interventional Radiology Foundation initiated the Fibroid Registry for Outcomes Data (FIBROID) Registry, a multi-center prospective study. Goals for the Registry included assessing the short- and long-term safety and efficacy of uterine artery embolization and determining the outcome of the procedure in a large patient cohort in general practice. The methods, 30-day outcomes, and 1-year outcomes from the FIBROID Registry have been previously reported.12–14 Herein, we report the 3-year results from the FIBROID Registry.
MATERIALS AND METHODS
The FIBROID Registry methods have been previously described in detail.12–14 Briefly, in 1999 the Cardiovascular and Interventional Radiology Research and Education Foundation (now the Society of Interventional Radiology Foundation), in collaboration with the Duke Clinical Research Institute, initiated the FIBROID Registry as a single-arm prospective multi-center longitudinal outcome study of the short- and long-term outcomes of uterine artery embolization for leiomyomata throughout the United States.
Sites had to have prior uterine embolization experience of at least 24 cases and an anticipated monthly volume of at least five cases. Sites were required to have a study coordinator. Participation in the Registry was voluntary, and there was no financial support for the sites. Each site had the protocol approved by its institutional review board. Written informed consent for the collection of data was obtained from each enrolled patient. The original consent included 2 years of patient follow-up. Before commencing 3-year follow-up, each patient reconsented to the extension.
The baseline data and follow-up data to 30 days after the procedure, including adverse events, were collected by the site investigator and coordinator. The baseline data included demographics, reproductive history, comorbidities, and imaging findings of the uterus and leiomyomata. These results were presented in an earlier publication.12 The procedure-related data and complications up to 30 days after the procedure were reported in the same journal issue.14
The data management center, the Duke Clinical Research Institute, obtained all the follow-up data beyond the initial 30 day follow-up, and that follow-up data set provides the basis for this report. The investigators did not have input or direct contact with the patients for this portion of the data collection. At each follow-up interval (6, 12, 24, and 36 months), the patient was sent a packet that included the Uterine Fibroid Symptom and Quality of Life questionnaire and an additional set of questions regarding outcome. These items included questions regarding presence of menstrual periods, current medications or interventions for leiomyoma-related symptoms, and emergency room visits or rehospitalization for leiomyoma-related problems. As a measure of patient satisfaction, participants were also asked to state the degree to which they would recommend the procedure to their family members or friends, responding on a scale from strongly disagree to strongly agree. Finally, there were questions about subsequent pregnancy, complications of pregnancy, and pregnancy outcomes. Analyses and presentation of pregnancy outcomes will be made in a future publication.
The primary measures of outcome were the symptom and health-related quality-of-life scores from the Uterine Fibroid Symptom and Quality of Life questionnaire. The scores on the two scales from the questionnaire range from 0 to 100; higher scores on the health-related quality-of-life scale indicate a better score, while a lower score is better on the symptom scale, indicating fewer symptoms.
At the 6-month follow-up interval, eligibility was determined by the patient meeting each of the following criteria: the patient must have indicated at enrollment a willingness to complete follow-up, had complete baseline symptom and quality-of-life scores, and had provided a complete mailing address or telephone number at baseline. At each subsequent follow-up interval, the above criteria still applied, regardless of whether a response was obtained for the preceding interval. Only those who had had a hysterectomy before the preceding follow-up or who requested to withdraw at that follow-up interval were excluded from further follow-up efforts.
The follow-up questionnaire packets were initially mailed to the patients. If there was no response, the patient was telephoned by a trained interviewer and offered to complete the questionnaires by phone. The details of this process have been described in an earlier publication.12
Summary statistics included frequency and percentage for categorical variables and means, standard deviations, and medians for continuous variables. Pearson χ2 tests were used for comparing all categorical variables, and Wilcoxon rank sum tests were used for comparing all the continuous or ordinal demographic and clinical variables. Paired t tests were used for the comparison of mean scores over time. In addition, Kaplan-Meier estimates of post–uterine artery embolization interventions were made (see the Appendix at www.greenjournal.org/cgi/content/full/111/1/22/DC2).
Furthermore, we evaluated effects of baseline patient characteristics on score (Uterine Fibroid Symptom and Quality of Life and health-related quality of life) using two methods: Initially, we used delta score as the response variable, where delta was the difference in scores between the follow-up time point (36 months) and the baseline time point. For this analysis, only patients with relevant scores available for both time points were used. The delta score outcome was analyzed using a linear mixed model, with site as a random effect and all patient-related covariates as fixed effects.
Because this model does not adequately capture each patient’s score trajectory over time, we analyzed scores using a repeated-measures method that makes use of all available data. This more comprehensive method assessed changes in scores over time and the factors associated with change. Each patient could then have a minimum of one score (baseline score) and a maximum of five scores (baseline, 6 months, 12 months, 24 months, and 36 months). Because, for a single patient, the score at any time point is correlated with scores at the other time points, it is important to take into consideration this within-patient correlation of responses. For this analysis, a linear mixed model with unstructured covariance matrix was used to model the outcome. All factors that could potentially affect the outcome were first included in the model, including a dichotomous indicator for each time period and interaction effects between covariates and time, to allow for the possibility that the effect of factors on the outcome could change over time. Factors that came into play only after the procedure and would not affect baseline symptom score were included only as interaction effects with time. In both methods of analysis, variables that were not significant in the full model were eliminated sequentially, starting with the interaction effects.
The final model in the second analysis consisted of effects that were significant at any time point, such that 1) a factor that had an effect only at a later time point would be included with all preceding time points, and 2) a factor that had an effect at an earlier time point would be included only at that and all preceding time points. Because of space limitations, only the results of symptom scores from this second method of analysis are presented.
In the above multivariable analyses, the factors offered to the model to assess effect on outcomes included patient baseline characteristics (age, race, body mass index, parity, any comorbidities, smoking behavior, prior medical or procedural therapy for leiomyomata, predominant presenting symptoms [heavy bleeding, pain, bulk symptoms, or other], uterine volume, leiomyoma location within the uterus [fundal, anterior body, posterior body, lateral body, cervical], size, and morphology [submucosal, intramural, or serosal], and number of leiomyomata), and procedure factors (embolic material, deep vein thrombosis prophylactic use, prophylactic antibiotic use, length of procedure, and vessel embolized). Bulk symptoms were defined as sensations of pelvic pressure or heaviness or urinary pressure. “Other” symptoms were any primary symptom other than heavy bleeding, pain, or bulk symptoms. Missing values for the continuous risk variables were imputed to median values using the nonmissing value. Missing values of the categorical variables were imputed to their most common value.
The significance of variables was assessed at the 0.05 alpha level. All analyses were conducted with SAS 8.2 (SAS Institute Inc., Cary, NC).
Details of enrollment and follow-up are presented in Figure 1. Two thousand one hundred twelve patients were eligible for long-term follow-up, with the reasons for ineligibility summarized in the figure. At 36 months after treatment, 1,916 patients remained in the study, and of these, 1,278 patients completed the survey. Of those not completed, 291 patients could not be located, 213 refused to continue participation, and 134 patients could not be followed for other reasons. Attrition between the second and third year was largely driven by the need to reconsent the patients in that interval.
The 1,278 eligible patients with available 3-year data were compared with the 834 eligible patients with unavailable 3-year data. Age, race, body mass index (BMI), pregnancy history, smoking status, comorbidity, previous procedures and medical therapy, leiomyoma size, number, and location, uterine volume, presenting symptoms, medication use, procedural details, recovery times, site information, and adverse events were all analyzed. Statistically significant results are shown in Table 1. Those patients completing follow-up were 1 year older (median age), were more likely to be white women, presented with a different mix of predominant symptoms, had a lower BMI, more use of prior medication, less antibiotic use, and a shorter time to return to work (P values all less than .05).
A summary of the changes in the symptom and total Uterine Fibroid Symptom and Quality of Life scores are presented graphically in Figure 2. The details of the score changes are included in Table 2. The symptom score improved 41.41 points (P<.001) in comparison with baseline and also was significantly better at 3 years than at 6 months or at 1 year. The mean symptom score did not change significantly between 2 and 3 years of follow-up. The change in the quality of life score was similar, with the median improving 41.47 points over baseline at 3 years (P<.001). This score was also significantly better than the score at 6 months and 1 year. It was not significantly better than the score at 2 years. For each patient the change in symptom score was determined from baseline to the last symptom score available. There were 2,002 patients for whom these data were available. In 3 years (cumulatively), 1,708 (85.3%) patients improved 10 or more points on the symptom score.
Figure 3 shows Kaplan-Meier estimates of survival function of subsequent invasive therapy after uterine artery embolization. Cumulatively, further surgical care included myomectomy (2.82%), hysterectomy (9.79%), and uterine artery embolization (1.83%). Subsequent care, including unplanned visits to the emergency room (ER) and hospitalizations, was tracked for each follow-up interval (0–6 months, 6–12 months, 12–24 months, and 24–36 months). Emergency room visits occurred in 3.78%, 1.18%, 0.83%, and 0.70% of patients, respectively. Hospitalizations occurred in 2.89%, 2.18%, 3.81%, and 4.23% of patients, respectively. These figures included all unplanned ER or hospital care visits, whether they were or were not related to uterine leiomyomata and/or the embolization.
At 36 months after uterine artery embolization, 365 (28.6%) of 1,278 patients were amenorrheic (excluding those patients who had undergone hysterectomy). Three hundred thirty-four of the 365 patients had menstrual status available at 6 months. Forty-one percent of the patents with available data who were amenorrheic at 3 years were also amenorrheic at the 6-month time point. Of the 365 patients that were amenorrheic, 78.9% were aged 45 years or older, 15.6% were between the ages of 40 and 45, and 5.5% were under the age of 40 (1.6% of the total cohort became amenorrheic and were younger than 40 years). Whether uterine artery embolization precipitated amenorrhea in these patients or the patients’ periods stopped independently of uterine artery embolization could not be determined from this study. Study participants were asked if they would recommend the uterine artery embolization procedure to family members or friends, and 1,095 (85.68%) of 1,278 of patients agreed or strongly agreed with that supposition.
Multivariable analyses of predictors of improvement in symptom scores and quality of life at 3 years after embolization are presented in Tables 3 and 4. The use of prior medication, heavy bleeding as a presenting symptom, and submucosal leiomyoma location were all associated with improved symptom score and quality of life score outcomes. Lower uterine segment or cervical position of the dominant leiomyoma was a significant predictor of a worse health-related quality of life score at 36 months compared with score at baseline. Increasing leiomyoma size predicted a poorer symptom score outcome at 36 months compared with score at baseline. Demographic factors, technical differences, and the type of embolic materials used were not significant predictors of success at 3 years in this study, nor was practice setting a factor.
A multivariable analysis of effect of patient factors on symptom score at baseline, 6, 12, 24, and 36 months was performed on the 2,112 patients available for follow-up. Time since uterine artery embolization, age, BMI, leiomyoma morphology, presenting symptoms, baseline scores, unilateral embolization, embolic material, and smoking, pregnancy, and medication history were all analyzed. Statistically significant results at 3 years are presented in Table 5. Higher age, lower BMI, and low initial symptom scores were all associated with better outcomes at 3 years.
More than 600,000 hysterectomies are performed annually in the United States, and uterine leiomyomata are the leading cause.3 Health care costs for women with leiomyomata average 2.6 times those for controls,15 and the total direct cost exceeds two billion dollars.16 Over half of American women have leiomyoma disease,1 and despite this burden of disease, our understanding of the outcomes from leiomyoma treatments is limited.17
The purpose of the FIBROID Registry was to assess the procedure’s effectiveness in improving leiomyoma-related symptoms, to determine the durability of those improvements, and to assess the safety of the procedure in broad practice. Secondary goals were to assess the effect of the procedure on fertility and quality of life and to determine patient satisfaction with outcome. This paper reports the 3-year clinical outcomes data. No additional patient follow-up is planned, although analysis of fertility in the cohort is anticipated.
In the FIBROID Registry, 3-year follow-up was achieved in 1,278 patients, with patients treated in a wide variety of practice settings. These included academic centers, community hospitals, and closed-panel health maintenance organization facilities. The sites were diverse in geographic location, site experience, and patient population composition. In this regard, this and other registries provide a means of collecting outcomes data in large populations across a wide spectrum of practice and geographic settings, which is similar to postmarket surveillance for new drugs. They also provide a means of data collection that avoids the shortcomings of single-center studies and administratively collected data. This is particularly useful when considering new procedures, because procedures do not require U.S. Food and Drug Administration approval. For example, a registry on laparoscopic cholecystectomy provided information about the relationship between physician experience and safety outcomes.18 Additionally, more meaningful risk adjustment can be offered by a registry than by administrative data.19
In the original validation study of the disease-specific symptom and health-related quality of life questionnaire for leiomyomata, the mean symptom and quality of life scores for normals were 22.5 (±21.1) and 86.4 (±17.7), respectively.9 At 3 years the patient cohort in the FIBROID registry had significantly improved scores that moved into the normal range (Fig. 2). The score improvements were independent of operator experience or practice setting. The improvements were durable and consistent with those reported previously.13 The results are also consistent with those reported in prospective trials that found quality of life scores to be equivalently improved after uterine artery embolization or surgery.6 Also, 85.68% of patients would recommend the uterine artery embolization procedure to family members or friends.
Our study identified subgroups of patients that appear to have a greater likelihood of improvement. Patients presenting with heavy bleeding as their predominant complaint had improved outcomes. Smaller leiomyoma size was correlated with a better result. Although intuitively understandable and whereas some articles have reported similar correlations,20,21 others have not.22 Older patients fared better and this has been observed before,22 and those with a lower BMI had better outcomes. Patients with submucosal leiomyomata did better, and patients with lower uterine segment or cervical leiomyomata fared worse. In addition, women who were less symptomatic at baseline had better symptom scores at 3 years after treatment.
During the 3 years of the study, hysterectomy, myomectomy, or repeat uterine artery embolization were performed in 9.79%, 2.82%, and 1.83% of the patients, respectively. This is very comparable to the surgical reintervention rates of approximately 5% per year after myomectomy.23 However, the reintervention rate in this trial was lower than that reported after uterine artery embolization in two randomized controlled trials.6,7 This may reflect an underestimation of repeat intervention because of patients who were lost to follow-up.
During the course of the Registry, subsequent care, including unplanned visits to the ER and hospitalizations, was tracked for each follow-up interval (0–6 months, 6–12 months, 12–24 months, and 24–36 months). Emergency room visits occurred in 3.78%, 1.18%, 0.83%, and 0.70% of patients, respectively. Hospitalizations occurred in 2.89%, 2.18%, 3.81%, and 4.23% of patients, respectively. In comparison,10.6% of women ages 35–44 and 10.1% of women ages 45–64 were hospitalized in the United States for any reason in 2004,24 suggesting that these rates are not unusual in this age cohort. Low adverse event rates for uterine artery embolization as reported for the FIBROID registry are consistent with those reported in prospective controlled and long-term studies.25–27
At 36 months after uterine artery embolization, just over one quarter of the patients became amenorrheic, with nearly 80% of these women older than age 45. This study did not explore the causality of these observations. These numbers are in line with prior reports. The age-related result has been observed before and may in part be explained by the natural history of the onset of menopause in the over-45 age cohort. However, the question of nontarget embolization of the ovaries as a contributing factor has been raised.28 There are some data to suggest that the ovarian effect of uterine artery embolization is no different from that following the surgical alternatives,29 but additional investigation is needed. Interestingly, if uterine artery embolization can independently lead to a decrease in ovarian reserve and amenorrhea, the effect may be mitigated by youth,30 because these effects are detected less frequently in younger women. However, the lack of amenorrhea or even normal follicle-stimulating hormone levels may not reflect the effect of uterine artery embolizations on ovarian reserve. Antimüllerian hormone has been shown to drop after uterine artery embolization consistent with damage to the ovarian reserve.31
A limitation of this study is the number of patients lost to follow-up, and we believe this was in part due to the need to reconsent each patient to carry the study to 3 years. In this Registry, there were differences between the group of patients for which data were ultimately available and the originating population(s). Patients who completed the study were 1 year older, had more use of prior medication, and had less than a one point lower BMI. As discussed above, these factors could have slightly biased the study toward a more successful outcome. Other differences, as noted in the results, were not prognosticators of outcome in this study.
This study relied on patient-reported outcomes. The indication for treatment of leiomyomata is the symptoms they cause, and the severity of these symptoms has not been shown to correlate with objective measures of uterine or leiomyoma size. The use of objective measures is also limited by the lack of a disease classification system. In the absence of reliable means of classifying disease severity objectively, we chose to base this study on a validated symptom and quality of life questionnaire. Other investigators studying other leiomyoma therapies, including hysterectomy, have used similar questionnaires to assess outcomes from therapy.32,33
The goals of the FIBROID registry were to assess the procedure’s safety, effectiveness, and durability and its effect on fertility and quality of life. At 3 years after treatment, uterine embolization appears to be a safe, effective, and durable treatment in a variety of practice settings, with substantial improvement in symptoms and quality of life for the large majority of patients. These positive outcomes after uterine artery embolization are consistent with prior reports and suggest that these results are achievable when the procedure is performed in any experienced community or academic interventional radiology practice.
1. Payson M, Leppert P, Segars J. Epidemiology of myomas. Obstet Gynecol Clin North Am 2006;33:1–11.
2. Farquhar C, Steiner C. Hysterectomy rates in the United States 1990–1997. Obstet Gynecol 2002;99:229–34.
3. Keshavarz H, Hillis S, Kieke B, Marchbanks P. Hysterectomy Surveillance—United States, 1994–1999. CDC MMWR Surveill Summ 2002 51:1–8.
4. Ravina J, Herbreteau D, Ciraru-Vigneron N, Bouret JM, Houdart E, Aymard A, et al. Arterial embolization to treat uterine myomata. Lancet 1995;346:671–2.
5. Broder M, Landow W, Goodwin S, Brook R, Sherbourne C, Harris K. An agenda for research into uterine artery embolization: results of an expert panel conference. J Vasc Interv Radiol 2000;11:509–15.
6. Edwards RD, Moss JG, Lumsden MA, Wu O, Murray LS, Twaddle S, et al. Uterine-artery embolization versus surgery for symptomatic uterine fibroids. N Engl J Med 2007;356:360–70.
7. Hehenkamp WJ, Volkers NA, Donderwinkel PF, de Blok S, Birnie E, Ankum WM, et al. Uterine artery embolization versus hysterectomy in the treatment of symptomatic uterine fibroids (EMMY trial): peri- and post procedural results from a randomized controlled trial. Am J Obstet Gynecol 2005;193:1618–29.
8. Pinto I, Chimeno P, Romo A, Paúl L, Haya J, de la Cal MA, et al. Uterine fibroids: uterine artery embolization versus abdominal hysterectomy for treatment—a prospective randomized, and controlled clinical trial. Radiology 2003;226:425–31.
9. Spies J, Coyne K, Guaou Guaou N, Boyle D, Skyrnarz-Murphy K, Gonzalves S. The UFS-QOL, a new disease-specific symptom and health-related quality of life questionnaire for leiomyomata. Obstet Gynecol 2002;99:290–300.
10. Goldberg J, Bussard A, McNeil J, Diamond J. Cost and reimbursement for three fibroid treatments: abdominal hysterectomy, abdominal myomectomy, and uterine fibroid embolization. Cardiovasc Intervent Radiol 2007;30:54–8.
11. Subramanian S, Spies J. Uterine artery embolization for leiomyomata: resource use and cost estimation. J Vasc Interv Radiol 2001;12:571–4.
12. Myers E, Goodwin S, Landow W, Mauro M, Peterson E, Pron G, et al. Prospective data collection of a new procedure by a specialty society: the FIBROID Registry. Obstet Gynecol 2005;106:44–51.
13. Spies J, Myers ER, Worthington-Kirsch Mulgund J, Goodwin S, Mauro M. The FIBROID Registry: symptom and quality-of-life status 1 year after therapy. Obstet Gynecol 2005;106:1309–18.
14. Worthington-Kirsch R, Spies J, Myers E, Mulgund J, Mauro M, Pron G, et al. The Fibroid Registry for Outcomes Data (FIBROID) for Uterine Artery Embolization: short term outcomes [published erratum appears in Obstet Gynecol 2005;106:869]. Obstet Gynecol 2005;106:52–9.
15. Hartmann KE, Birnbaum H, Ben-Hamadi R, Wu EQ, Farrell MH, Spalding J, et al. Annual costs associated with diagnosis of uterine leiomyomata. Obstet Gynecol 2006;108:930–7.
16. Flynn M, Jamison M, Datta S, Myers E. Health care resource use for uterine fibroid tumors in the United States. Am J Obstet Gynecol 2006;195:955–64.
17. Myers E, Barber M, Gustilo-Ashby T, Couchman G, Matchar D, McCrory D. Management of uterine leiomyomata: what do we really know? Obstet Gynecol 2002;100:8–17.
18. Orlando R 3rd, Russell R, Lynch J, Mattie A. Laparoscopic cholecystectomy. A statewide experience. The Connecticut Laparoscopic Cholecystectomy Registry. Arch Surg 1993;128:494–8.
19. Myers ER, Steege JF. Risk adjustment for complications of hysterectomy: limitations of routinely collected administrative data. Am J Obstet Gynecol 1999;181:567–75.
20. Marret H, Cottier JP, Alonso AM, Giraudeau B, Body G, Herbreteau D. Predictive factors for fibroids recurrence after uterine artery embolization. BJOG 2005;112:461–5.
21. Spies J, Bruno J, Czeyda-Pommersheim Magee S, Ascher S, Jha R. Long-term outcome of uterine artery embolization of leiomyomas. Obstet Gynecol 2005;106:933–9.
22. Goodwin S, McLucas B, Lee M, Chen G, Perrella R, Vedantham S, et al. Uterine artery embolization for the treatment of uterine leiomyomata: midterm results. J Vasc Interv Radiol 1999;10:1159–65.
23. Reed SD, Newton KM, Thompson LB, McCrummen BA, Warolin AK. The incidence of repeat uterine surgery following myomectomy. J Womens Health (Larchmt) 2006;15:1046–52.
24. Kozak LJ, DeFrances CJ, Hall MJ. National hospital discharge survey: 2004 annual summary with detailed diagnosis and procedure data. Vital Health Stat 13 2006:1–209.
25. Goodwin SC, Bradley LD, Lipman JC, Stewart EA, Nosher JL, Sterling KM, et al. Uterine artery embolization versus myomectomy: a multicenter comparative study. Fertil Steril 2006;85:14–21.
26. Spies J, Cooper J, Worthington-Kirsch Lipman J, Mills B, Benenati J. Outcome from uterine embolization and hysterectomy for leiomyomas: results of a multi-center study. Am J Obstet Gynecol 2004;191:22–31.
27. Spies J, Spector A, Roth A, Baker C, Mauro L, Murphy-Skrynarz K. Complications after uterine artery embolization for leiomyomas. Obstet Gynecol 2002;100:873–80.
28. Ryu R, Chrisman H, Omary R, Miljkovic S, Nemcek AA Jr, Saker MB, et al. The vascular impact of uterine artery embolization: Prospective sonographic assessment of ovarian arterial circulation. J Vasc Interv Radiol 2001;12:1071–4.
29. Hovsepian DM, Ratts VS, Rodriguez M, Huang JS, Aubuchon MG, Pilgram TK. A prospective comparison of the impact of uterine artery embolization, myomectomy, and hysterectomy on ovarian function. J Vasc Interv Radiol 2006;17:1111–5.
30. Ahmad A, Qadan L, Hassan N, Najarian K. Uterine artery embolization treatment of uterine fibroids: effect on ovarian function in younger women. J Vasc Interv Radiol 2002;13:1017–20.
31. Hehenkamp WJ, Volkers NA, Broekmans FJ, de Jong FH, Themmen AP, Birnie E, et al. Loss of ovarian reserve after uterine artery embolization: a randomized comparison with hysterectomy. Hum Reprod 2007;22:1996–2005.
32. Carlson K, Miller B, Fowler F. The Maine Women’s Health Study; I. Outcomes of hysterectomy. Obstet Gynecol 1994;83:556–65.
33. Carlson K, Miller B, Fowler F. The Maine Women’s Health Study: II. Outcomes of nonsurgical management of leiomyomas, abnormal bleeding, and chronic pelvic pain. Obstet Gynecol 1994;83:566–72.
© 2008 by The American College of Obstetricians and Gynecologists.