MR-guided focus ultrasound (MRgFUS) for symptomatic uterine fibroids: predictors of treatment success

Transcription

1 Human Reproduction, Vol.27, No.12 pp , 2012 Advanced Access publication on September 26, 2012 doi: /humrep/des333 ORIGINAL ARTICLE Gynaecology MR-guided focus ultrasound (MRgFUS) for symptomatic uterine fibroids: predictors of treatment success Ronit Machtinger 1,2, Yael Inbar 3, Shlomo Cohen-Eylon 1, Dahlia Admon 1, Aviva Alagem-Mizrachi 4, and Jaron Rabinovici 1, * 1 Department of Obstetrics and Gynecology, Sheba Medical Center, Sackler School of Medicine, Tel Aviv University, Tel Hashomer, Israel 2 Department of Obstetrics and Gynecology, Brigham and Women s Hospital, Harvard Medical School, Boston, MA, USA 3 Department of Radiology, Sheba Medical Center, Sackler School of Medicine, Tel Aviv University, Tel Hashomer, Israel 4 Department of Advanced Technologies, Sheba Medical Center, Tel Hashomer, Israel *Correspondence address. Tel: ; yaronr@tau.ac.il Submitted on August 17, 2011; resubmitted on August 1, 2012; accepted on August 10, 2012 study question: What are the factors associated with long-term success in patients with symptomatic uterine fibroids treated by magnetic resonance-guided focus ultrasound (MRgFUS) and can they be employed to create a clinically useful index that predicts longterm efficacy? summary answer: Hypo-intense fibroids on T2-weighted magnetic resonance imaging (MRI) and older age were associated with higher success rates and can be used to predict success rates on the basis of their presence or absence as pre-treatment parameters. what is known already: The signal intensity of baseline T2-weighted MRI images and non-perfused volume at the end of the treatment can be correlated with MRgFUS outcome. study design, size and duration: This was a retrospective analysis of 81 patients who were treated by MRgFUS for symptomatic uterine fibroids, in an academic affiliated center between 2003 and participants/materials, setting and methods: There was a post-treatment phone interview.6 months following MRgFUS for symptomatic uterine fibroids. main results and the role of chance: The eighty-one patients completed a successful MRgFUS treatment during this period, of whom 74 were included in the final analysis (1 was post-menopausal at treatment and 5 were lost for follow-up). The mean time for the phone interview was months (range: 6 53 months) after the MRgFUS treatment. Fifty-five patients (69%) did not need any additional alternative treatment following MRgFUS. Nineteen patients (24%) underwent other surgical interventions. Hypointense fibroids were associated with a higher chance of success than hyper-intense fibroids [odds ratio ¼ 2.96 ( ); P ¼ 0.04] for surgery in hyper-intense fibroids). Women with long-term treatment success were significantly older at the time of treatment [ (range: 37 53) years versus (range: 36 51) years, respectively; P ¼ 0.02]. limitations and reasons for caution: Retrospective non-comparative studies are suboptimal and might overemphasize favorable outcomes. wider implications of the findings: This paper can contribute to selection of suitable candidates for the MRgFUS treatment for patients with uterine fibroids and can serve as a guide for gynecologists for a better patient selection. study funding/competing interest(s): None. trial registration number: N/A. Key words: MRgFUS / uterine fibroids / long-term outcome Presented in part at the 64th Annual Meeting of the American Society for Reproductive Medicine. & The Author Published by Oxford University Press on behalf of the European Society of Human Reproduction and Embryology. All rights reserved. For Permissions, please journals.permissions@oup.com

2 3426 Machtinger et al. Introduction Uterine fibroids are the most common benign pelvic tumors of the female genital tract, with a prevalence of up to 50% in premenopausal women. In 25% of women fibroids are associated with clinically significant symptoms such as abnormal vaginal bleeding, bulk symptoms resulting in urinary or bowel discomfort, abdominal pressure or back pain (Verkauf, 1992; Stewart 2001; Donnez and Jadoul, 2002; Rabinovici et al., 2007). Treatment options include invasive surgical procedures (abdominal or laparoscopic hysterectomy or myomectomy), less invasive treatments (hysteroscopic myomectomy, uterine artery embolization), or conservative alternatives such as magnetic resonance-guided focused ultrasound (MRgFUS) as well as pharmacological options. Ideally, the choice of treatment has to be optimally tailored for each patient, depending on the type and localization of the fibroid, associated symptoms, the wishes of the patient and the ability of the physician (Stewart et al., 2003; Morita et al., 2008). The demand for alternative treatments has increased during the last decade, as more women wish to retain their uterus and generally avoid invasive procedures, which are associated with higher morbidity and mortality (Behera et al., 2010). MRgFUS has been used in recent years for the non-invasive, ambulatory treatment of uterine fibroids and has been shown to be a uterus-preserving, safe and effective treatment alternative (Stewart et al., 2003, 2006; Istre, 2008). As with any evolving new procedure, important concerns are the long-term outcome and reliability of the treatments and the choice of appropriate candidates for treatment. MRgFUS has been offered as a treatment option (not as a part of clinical trials) in our hospital since The aims of this study were to follow the first cohort of patients treated in our Medical Center and to identify parameters associated with long-term success or failure of MRgFUS for symptomatic uterine fibroids. Materials and Methods The long-term outcomes of women who were treated by MRgFUS at the University Affiliated Medical Center between 2003 and 2008 were assessed by a phone interview. The medical charts, the pre-treatment MR imaging (MRI) scan and the immediate treatment outcome were retrospectively reviewed. The study was approved by the Institutional Review Board of our hospital. The inclusion criteria and the technical details for MRgFUS therapy were published previously (Stewart et al., 2006; Rabinovici et al., 2007). In short, women with no significant medical contraindications for treatment who declared that they had completed their families and had clinically symptomatic uterine fibroids were evaluated (maximal fibroid diameter 13 cm). Screening included a symptom severity score questionnaire (uterine fibroid symptom-quality of life-symptom severity score, UFS-QOL-SSS) (Spies et al., 2002), a medical and gynecological assessment and a pelvic screening MRI with gadolinium. Patients were scheduled for treatment if there was no significant medical condition that would be a contraindication for treatment, no interference with the beam pass and the fibroids could be treated according to our guidelines at this time (Stewart et al., 2006). The parameters assessed on MRI also included the volume of fibroids as well as their signal intensity and heterogeneity. The classic uterine fibroid without degeneration has clear margins and low signal intensity (hypo-intense) compared with that of the myometrium on T2-weighted images. This reflects a mixture of fibrous components and reduced blood flow. The proportion between leiomyoma cell fascicles and the extra-cellular matrix affects signal intensities on T2-weighted images. Fibroids were classified on T2-weighted MRI according to their signal intensity compared with the normal myometrium (Marret et al., 2005). Fibroids with signal intensity equal to or higher than the myometrium were classified as hyper-intense or white fibroids. These cases can be attributed to cellularity, edema, high vascularity, cystic degeneration and myxoid degeneration. Fibroids with signal intensity lower than that of the myometrium were classified as hypo-intense or black fibroids. Low signal on T2 (hypo-intense) can be attributed to tight smooth muscle or hyaline degeneration. Fibroids were further subdivided into homogenous or heterogeneous according to their appearance on the T2-weighted images. The MRgFUS system (ExAblate 2000, InSightec, Haifa, Israel) integrates a standard 1.5 Tesla MRI System (GE Medical Systems, Milwaukee) with the focused ultrasound treatment transducer in the patient table. The treatment protocol was previously discussed (Rabinovici et al., 2007). At the end of the treatment, all the patients underwent i.v injection of gadolinium. Contrast-enhanced T1-weighted MRI was performed and the volume of non-enhancement was measured. The non-perfused volume (NPV) percentage of the fibroids was calculated as the non-enhanced volume divided by the total fibroid volume. In cases of patients who underwent more than one MRgFUS treatment, the NPV in the study was the NPV achieved following the last MRgFUS treatment. Patients medical data from their first visit in the clinic and screening MRIs were retrospectively reviewed. The parameters examined included patients age, previous number of pregnancies, symptoms (abnormal bleeding or urinary symptoms), baseline UFS-QOL-SSS score, previous abdominal scars and BMI. Analyses of the MR images included number of fibroids, the total volume of the fibroids, maximal fat layer thickness, distance of the dominant fibroid from the sacrum, the fibroids characteristics and texture on T2-weighted MRI (hyper-/hypo-intense), homogeneity/ heterogeneity and the NPV at the end of the treatment. All the patients were interviewed by telephone at least 6 months after the MRgFUS treatment regarding their outcomes and any persisting fibroid-associated symptoms. MRgFUS treatment success was defined as the lack of need for any further surgical intervention, such as uterine artery embolization, myomectomy or hysterectomy for persistent fibroid symptoms. Statistical analysis Statistical analysis was performed using the SAS statistical software (version 9.1, SAS Institute, Inc., Cary, NC). Categorical variables were expressed as percentage, and continuous variables were expressed as mean + SD. Comparisons of variables were performed by Mantel Haenszel s test for trend, the chi-square test or Fisher s exact test for categorical variables and by Student s t-test for continuous variables. All the tests were two-sided and a P-value of, 0.05 was considered statistically significant. Logistic regression analyses were performed, with surgical intervention following the MRgFUS treatment as a dependent variable. The independent variables were age, basic QOL score before treatment, total fibroid volume, signal intensity of the fibroid in T2-weighted MRI (contrast level), fibroid texture and patients complaints of menorrhagia. Using this logistic regression model, we compared the estimated odds ratio (OR) and its 95% confidence interval (CI) for failure of the MRgFUS treatment while controlling for the variables mentioned. In addition, a multivariate logistic regression model was constructed for the predicting characteristics for an NPV of.45%. Forty-five percent was used as the cutoff point and indicates the upper tertile of the results for this variable. The outcome of NPV.45% was adjusted for the same baseline variables for follow-up outcome of surgical intervention.

3 Predictors for success of MRgFUS treatment for uterine fibroids 3427 Results Eighty-seven women were admitted for the MRgFUS treatment of their uterine fibroids at our hospital during the study period. The treatments of two patients were cancelled before the start of sonications because of bowels obstructing the ultrasound beam. Two others were terminated very early because of patients discomfort. Another two treatments were stopped prematurely because of technical difficulties encountered during the procedure. These women were not interested in a further treatment trial. One of the patients was excluded from the final analysis as she was menopausal at the time of the treatment. As MRgFUS is recommended for premenopausal women, there was a concern that inclusion of a post-menopausal amenorrhoeic subject might have an effect on the outcomes considered, which refer to premenopausal women. Eighty pre-menopausal patients who completed a successful MRgFUS treatment for uterine fibroids were included in this study. A prolonged phone interview follow-up of.6 months (mean months; range: 6 53 months) was achieved for 74 of the patients. Five patients were lost to follow-up (three of them live abroad and came to our hospital especially for the treatment). One patient died shortly after the procedure for non-device-related reasons. Patients characteristics Patients characteristics are shown in Table I. Fibroid symptoms before treatment included menorrhagia (27.0%), abdominal pressure or urinary complaints (28.4%), combined bleeding and urinary symptoms/pressure (41.9%) and fibroid-associated back pain (2.7%). Treatment outcome Of the 80 patients, 50 (63%) achieved a successful outcome after the original procedure. Another five patients (6%) underwent a second MRgFUS.6 months following the treatment (range: 7 18 months) because of recurrent fibroid-related symptoms. In all these 55 patients (69%), improvement of symptoms persisted through follow-up. Nineteen patients out of the 80 (24%) did not experience improvement of their symptoms and underwent further surgical intervention: 13 underwent a hysterectomy, 4 women underwent a myomectomy and 2 underwent uterine artery embolization. One of the myomectomies was performed on a patient who was admitted to the emergency room because of vaginal bleeding shortly after the MRgFUS treatment. Gynecological examination demonstrated a fibroid that protruded through the uterine cervix and was easily removed. Six of these nineteen patients attempted a second MRgFUS treatment before choosing an alternative surgical option. All the patients who underwent surgical procedures did so within 24 months of the MRgFUS treatment. Information on the exact timing of surgical intervention was available for 13 of the women: these women underwent surgery between 1 week and 24 months following treatment (mean months). Figure 1 shows the timescale for patients follow-up. Adverse events No serious adverse events were reported. There were three cases of first-degree burns and two cases of fever.388c. One of the patients with fever also had abdominal pain and leukocytosis and was hospitalized for i.v. antibiotic treatment. Two patients reported back pain and another two reported leg pain. Those adverse events were self-limited. Factors associated with treatment success We found an association between the appearance of the fibroid on T2-weighted MRI and treatment outcome (Table II). Hypo-intense fibroids were associated with a higher chance of success than hyperintense fibroids [OR ¼ 2.96 ( ); P ¼ 0.04 for surgery in hyper-intense fibroids). Women with long-term treatment success were significantly older at the time of treatment [ (range: 37 53) years versus (range: 36 51) years, respectively; P ¼ 0.02]. There was a gradual increase in the likelihood of long-term success with increasing age (Table II). Interestingly, 56% (5/9) of our younger patients (aged 36 40) were found on T2-weighted MRI to have hyper-intense fibroids, compared Table I Patients characteristics (n 5 74). Mean + SD Range... Patient s age Gravidity Parity BMI (kg/m 2 ) SSS-QOL before treatment Average fat layer (cm) Distance from fibroid to sacrum (cm) No. of fibroids Total fibroid volume (ml) NPV (%) NPV, non-perfused volume; SSS-QOL, system severity score-quality of life. Figure 1 Timescale for patients follow-up. In cases that a surgical intervention was performed (n ¼ 19), the follow-up period represents the time from the MRgFUS treatment to the surgical procedure and not the interval time from the treatment to the phone interview.

4 3428 Machtinger et al. Table II Univariate and multivariate analyses of parameters associated with surgery. Needed surgical Univariate analysis OR P-value Multivariable analysis OR P-value treatment (%) (95% CI) (95% CI)... Age a (n ¼ 9) ( ) ( ) (n ¼ 27) ( ) ( ) (n ¼ 38) SSS-QOL-before treatment ( ) 2.81 ( ) Total volume (Per 100 ml increase) 1.10 ( ) ( ) 0.81 Contrast level Black (hypo-intense) White (hyperintense) ( ) ( ) 0.18 Fibroid texture Homogenous mixed (Homogenous/heterogenous) ( ) ( ) 0.96 Heterogenous ( ) ( ) 0.34 Menorrhagia No Yes ( ) ( ) 0.22 C ¼ a The small number of patients younger than 40 years is due to the fact that this treatment was permitted only recently for patients who desired further fertility. OR (95% CI), odds ratio (95% confidence intervals); SSS-QOL, symptoms severity score-quality of life. Table III Correlation between the following risk factors [(1) Iso/hyper-intense fibroid appearance on MRI; (2) heterogeneous fibroid appearance on MRI] and surgical intervention. Number of risk factors n % of surgical intervention /31 16% 1 6/29 21% 2 8/14 57% Total 19/74 P trend with 37% (10/27) and only 24% (9/38) among the age groups of and years, respectively. Younger patients in our cohort also tended to have higher rates of heterogeneous fibroids (6/9, 67%) compared with patients in the age group of (10/27, 37%) and older women (ages 46 53, 15/38, 39%). The number of fibroids, total fibroid volume, average fat layer, BMI, gravidity and the distance from the fibroid to the sacrum did not significantly influence the long-term treatment outcome and were therefore not further analyzed (Table II). Patients with longer-term treatment success had a significantly higher mean NPV ratio (%) at the end of the treatment ( versus ) compared with those who needed additional interventions (P ¼ 0.04). While patients with an NPV ratio greater than 45% had only a 15% risk of needing to undergo an alternative treatment, Table IV The correlation between the number of risk factors a, subsequent need for hysterectomy and patients age in women undergoing MRgFUS. n % of surgical P interventions... Age 36 40, total 5/9 (56%) b No risk factors 0/3 0% risk factors 5/6 83% b Age 41 45, total 8/27 (30%) No risk factors 2/12 17% risk factors 3/10 30% 2 risk factors 3/5 60% Age 46 53, total 6/38 (16%) No risk factors 3/16 19% risk factors 2/18 11% 2 risk factors 1/4 25% MRgFUS, magnetic resonance-guided focus ultrasound. a Risk factors are: (1) iso/hyper-intense fibroid appearance on MRI; (2) heterogeneous fibroid appearance on MRI. b Risk of surgical intervention decreased with elevated patients age, P trend patients with an NPV ratio between 10 and 20% had a 40% risk of other interventions [OR ¼ 5.33 ( )]. We achieved higher NPV ratios in women with homogenous and hypo-intense fibroids.

5 Predictors for success of MRgFUS treatment for uterine fibroids 3429 A statistical model for predicting treatment success We found the following poor prognostic factors ( risk factors ) to be associated with treatment failure: (i) hyper-intense fibroids (ii) heterogeneous appearance of fibroids on T2-weighted MRI. On the basis of these parameters, we sought to establish a statistical model with which to determine a patient s chance of long-term success (Table III). We found a significant correlation between the number of risk factors and the need for further surgical treatments (P trend 0.009). While the absence of any established risk factors carried a 16% risk of the need for alternative treatments, one risk factor or a combination of two risk factors increased this risk to 21 and 57%, respectively (Table III). There was a negative correlation between the patient s age and the need for alternative therapy (Table IV). However, while the number of poor prognostic factors was highly correlated with the need for further intervention in the younger group (,40 years) and in the group of patients aged 41 45, we did not find such a correlation for women who were 46 years or older. In the younger age group (36 40 years old), all patients who had no poor prognostic factor did not undergo further therapy. In contrast, patients in this age group who had 1 or 2 poor prognostic factors had an 83% chance of the need for further treatment (P ¼ 0.048). In the age group of years, the presence of two poor prognostic factors was also associated with a higher chance of further interventions (P trend ¼ 0.09). Older patients (46 years or older) had lower chances of needing additional interventions irrespective of the presence or absence of the prognostic factors (Table IV). Factors associated with NPV >45% at the end of the treatment In multivariate regression analysis of parameters associated with NPV.45% (Table V), we found that hyper-intense fibroids and heterogeneous or mixed fibroid texture are associated with significantly lower rates of NPV.45% (OR ¼ 0.14; 95% CI and OR ¼ 0.05; 95% CI ). Discussion This study examined the long-term outcome of MRgFUS for uterine fibroids and the parameters that might predict the success of the procedure. Hypo-intense appearance of the fibroids on MRI before the treatment and older age at study entry predicted higher rates of treatment success. In view of the previously reported benefits of the MRgFUS treatment (Stewart et al., 2003; O Sullivan et al., 2009, Rabinovici et al., 2010), the long-term success rate of 69% for MRgFUS reported in our study compares favorably with the success rates of alternative surgical and non-surgical uterus-preserving procedures. Symptom recurrence requiring alternative treatments also occurs after uterine artery embolization: 17.2% after a median follow-up of 30 months (Marret et al., 2005). In the much larger comparative HOPEFUL study, up to 23% (95% CI: 19 27%) of women who underwent uterine artery embolization required other treatments (mean follow-up of 4.6 years) after adjusting for differential time of follow-up (Hirst et al., 2008). Thus, over time, uterus-preserving fibroid treatments may necessitate alternative or additional treatments and patients who choose such treatments must be informed about this possibility. Our study demonstrated a negative correlation between the patient s age and the need for alternative therapy after MRgFUS (Tables II and IV). Women with a successful treatment outcome were significantly older than those who ultimately chose an alternative treatment. Several factors could explain these findings. Younger women (36 40 years) had a significantly higher incidence of hyperintense fibroids on MRI (59%) than the two groups of older women (41 45 and years, with 37 and 26% incidence of hyperintensity, respectively). Since the degree of intensity correlates positively with vascularity and negatively with the ability to achieve a high ratio of NPV, it seems that younger women are more difficult to treat effectively than older women. In addition, the greater remaining life span of younger women means that there is more time left for lesion and symptom recurrence. Furthermore, younger women have a generally higher risk of fibroid recurrence after any uterus-preserving treatment. In a study that examined the recurrence rate of fibroids after abdominal myomectomy, younger women (40 44 years) had a significantly higher chance of subsequent surgery than women older than 50 years (Cox proportional hazard ratio ¼ 0.5, 95% CI for the older women) (Reed et al., 2006). These findings are probably due to increased and prolonged ovarian hormonal activity that might lead to growth of uterine fibroids in younger compared with older women. On the other hand, women who undergo a uterus-preserving fibroid treatment close to menopause are more likely to benefit in the long term from the onset of ovarian hormonal quiescence. Past studies have shown that the intensity of the MR signal of the uterus may be associated with the degree of blood flow and the fluid content of the tissue (Huang et al., 2004; Lenard et al., 2008; Yoon et al., 2008). Our finding of a strong association between MR hypo-intensity of the fibroid and the long-term success of the procedure is built on the previous findings of a positive correlation between the ability to ablate the tissue by MRgFUS and the hypo-intensity and homogeneity of the fibroid on MR. Thus, increased fibroid intensity and heterogeneity was associated with a lower NPV ratio at the end of the treatment. Funaki et al. (2007a,b) examined the relationship between signal intensity of baseline T2-weighted MR images and the therapeutic effects of MRgFUS. They reported that fibroids with image intensity equal to or higher than the myometrium ( white fibroids) were less responsive to the MRgFUS treatment and showed significantly lower peak temperatures during sonication. In another study, the same researchers (Funaki et al., 2009) evaluated the clinical outcome of patients with fibroids 24 months after MRgFUS and the need for re-interventions. They found that the rate of re-interventions depended on the fibroid vascularity as determined by MRI hyper-intensity. Lenard et al. (2008) also reported that fibroids with low signal intensity on pretreatment T2-weighted MR images were more likely to shrink than those with high signal intensity. Our data demonstrate that the chance of long-term success is positively associated with the amount of ablated volume of the fibroid after MRgFUS, expressed as the NPV ratio. These findings are in agreement with the previous findings of positive correlations between NPV and shorter-term treatment outcome (Funaki et al., 2007a,b; Stewart et al., 2007; Morita et al., 2008; LeBlang et al., 2010).

6 3430 Machtinger et al. Table V Univariate and multivariate analyses of parameters associated with NPV >45%. NPV >45% (%) Univariate analysis OR P-value Multivariable analysis OR P-value (95% CI) (95% CI)... Age ( ) ( ) ( ) ( ) SSS-QOL-before treatment ( ) ( ) Total volume (Per 100 ml increase) 0.75 ( ) ( ) 0.27 Contrast level Black (hypo-intense) White (hyperintense) ( ) 0.15 ( ) Fibroid texture Homogenous mixed (Homogenous/heterogenous) ( ) ( ) 0.03 Heterogenous ( ) ( ) Menorrhagia No Yes ( ) 1.83 ( ) C ¼ NPV, non-perfused volume; OR (95%CI), odds ratio (95% confidence intervals); SSS-QOL, system severity score-quality of life. We attempted to establish a model that could help both the physician and the patient determine the likelihood of long-term satisfaction after the MRgFUS treatment (Table III). Unsurprisingly, we found a significant correlation between the presence and number of risk factors (hyper-intensity, heterogeneity) and the long-term need for further surgical treatments (P for trend ¼ 0.009). While the presence and number of poor prognostic factors was highly correlated with the need for further interventions in the younger group (,40 years), this correlation disappeared in older patients (46 years and older) (Table IV). These findings demonstrated statistical significance, despite the small number of women in the younger group. This retrospective study has some limitations. Retrospective noncomparative studies are suboptimal and generally tend to overemphasize favorable outcomes. Six out of the 86 premenopausal women in our study who did not undergo a full MRgFUS procedure were not included in the analysis as we did not classify them as treatment failures. In the case of an intention to treat analysis, these cases would have been included as treatment failures. If this had been the case, the calculated success rates of the procedure would have been 64 and not 69%. Another possible constraint of our study was the performance of the follow-up by a phone questionnaire. While this did not allow us an objective evaluation of blood tests and imaging, it provided us with information about the primary outcome of this study, i.e. further surgical intervention. In summary, our study demonstrates that the MRgFUS treatment is a safe and (in the long term) an effective therapy option for women with symptomatic fibroids who wish to preserve their uterus and desire to avoid more invasive procedures. Treatment success and long-term satisfaction as judged by the need for additional treatments were positively associated with greater patient age and low signal intensity (hypo-intense) and homogeneity of the fibroid on T2-weighted MRI before the treatment. These factors should guide physicians in the selection of patients for MRgFUS and assist in pretreatment patient counseling. In view of recent technological improvements in the MRgFUS technology that yield higher NPV ratios in hyper-intense fibroids (Inbar et al., 2010), we expect the future to hold even higher success rates for this ambulatory, non-invasive fibroid treatment. Acknowledgements We thank Ms Valentina Boyko, MSc, for her assistance in the statistical analysis and Ms Lili Cohen, RN, for her help with the data collection. Authors roles R.M. participated in the study design, interviewed the patients and wrote the manuscript. Y.I. participated in the study design as well as edited the manuscript. D.A. edited the manuscript. S.C-A. edited the manuscript. A.A-M. participated in collecting and analyzing the data. J.R. was responsible for the experimental design and editing and finalizing the manuscript. Funding No external funding was either sought or obtained for this study.

7 Predictors for success of MRgFUS treatment for uterine fibroids 3431 Conflict of interest Y.I and J.R. serve as clinical trial investigators for InSightec. R.M, D.A., S.C.A. and A.A-M. have nothing to disclose. References Behera MA, Leong M, Johnson L, Brown H. Eligibility and accessibility of magnetic resonance-guided focused ultrasound (MRgFUS) for the treatment of uterine leiomyomas. Fertil Steril 2010;94: Donnez J, Jadoul P. What are the implications of myomas on fertility? A need for a debate? Hum Reprod 2002;17: Funaki K, Sawada K, Maeda F, Nagai S. Subjective effect of magnetic resonance-guided focused ultrasound surgery for uterine fibroids. J Obstet Gynaecol Res 2007a;33: Funaki K, Fukunishi H, Funaki T, Sawada K, Kaji Y, Maruo T. Magnetic resonance-guided focused ultrasound surgery for uterine fibroids: relationship between the therapeutic effects and signal intensity of pre-existing T2-weighted MR images. Am J Obstet Gynecol 2007b;196:1 6. Funaki K, Fukunishi H, Sawada K. Clinical outcomes of magnetic resonance-guided focused ultrasound surgery for uterine myomas: 24-month follow-up. Ultrasound Obstet Gynecol 2009;34: Hirst A, Dutton S, Wu O, Briggs A, Edwards C, Waldenmaier L, Maresh M, Nicholson A, McPherson K. A multi-centre retrospective cohort study comparing the efficacy, safety and cost-effectiveness of hysterectomy and uterine artery embolisation for the treatment of symptomatic uterine fibroids. The HOPEFUL study. Health Technol Assess 2008;12: iii. Huang J, Holt RG, Cleveland RO, Roy RA. Experimental validation of a tractable numerical model for focused ultrasound heating in flow-through tissue phantoms. J Acoust Soc Am 2004;116: Inbar Y, Siton A, Khaitovitch B, Admon D, Rabinovici J. An enhanced system for MRgFUS treatment of uterine fibroids. In: Proceedings of the Current and Future Applications of MR-Guided Focused Ultrasound. Washington, USA: 2nd International Symposium, Istre O. Management of symptomatic fibroids: conservative surgical treatment modalities other than abdominal or laparoscopic myomectomy. Best Pract Res Clin Obstet Gynaecol 2008;22: LeBlang SD, Hoctor K, Steinberg FL. Leiomyoma shrinkage after MRI-guided focused ultrasound treatment: report of 80 patients. Am J Roentgenol 2010;194: Lénárd ZM, McDannold NJ, Fennessy FM, Stewart EA, Jolesz FA, Hynynen K, Tempany CM. Uterine leiomyomas: MR imaging-guided focused ultrasound surgery imaging predictors of success. Radiology 2008;249: Marret H, Cottier JP, Alonso AM, Giraudeau B, Body G, Herbreteau D. Predictive factors for fibroids recurrence after uterine artery embolisation. Br J Obstet Gynaecol 2005;112: Morita Y, Ito N, Hikida H, Takeuchi S, Nakamura K, Ohashi H. Non-invasive magnetic resonance imaging-guided focused ultrasound treatment for uterine fibroids early experience. Eur J Obstet Gynecol Reprod Biol 2008;139: O Sullivan AK, Thompson D, Chu P, Lee DW, Stewart EA, Weinstein MC. Cost-effectiveness of magnetic resonance guided focused ultrasound for the treatment of uterine fibroids. Int J Technol Assess Health Care 2009; 25: Rabinovici J, Inbar Y, Revel A, Zalel Y, Gomori JM, Itzchak Y, Schiff E, Yagel S. Clinical improvement and shrinkage of uterine fibroids after thermal ablation by magnetic resonance-guided focused ultrasound surgery. Ultrasound Obstet Gynecol 2007;30: Rabinovici J, David M, Fukunishi H, Morita Y, Gostout BS, Stewart EA, MRgFUS Study Group. Pregnancy outcome after magnetic resonanceguided focused ultrasound surgery (MRgFUS) for conservative treatment of uterine fibroid. Fertil Steril 2010;93: Reed SD, Newton KM, Thompson LB, McCrummen BA, Warolin AK. The incidence of repeat uterine surgery following myomectomy. J Womens Health (Larchmt) 2006;15: Spies JB, Coyne K, Guaou Guaou N, Boyle D, Skyrnarz-Murphy K, Gonzalves SM. The UFS-QOL, a new disease-specific symptom and health-related quality of life questionnaire for leiomyomata. Obstet Gyneco 2002;99: Stewart EA. Uterine fibroids. Lancet 2001;357: Stewart EA, Gedroyc WM, Tempany CM, Quade BJ, Inbar Y, Ehrenstein T, Shushan A, Hindley JT, Goldin RD, David M et al. Focused ultrasound treatment of uterine fibroid tumors: safety and feasibility of a noninvasive thermoablative technique. Am J Obstet Gynecol 2003;189: Stewart EA, Rabinovici J, Tempany CM, Inbar Y, Regan L, Gostout B, Hesley G, Kim HS, Hengst S, Gedroyc WM. Clinical outcomes of focused ultrasound surgery for the treatment of uterine fibroids. Fertil Steril 2006;85: Stewart EA, Gostout B, Rabinovici J, Kim HS, Regan L, Tempany CM. Sustained relief of leiomyoma symptoms by using focused ultrasound surgery. Obstet Gynecol 2007;110: Verkauf BS. Myomectomy for fertility enhancement and preservation. Fertil Steril 1992;58:1 15. Yoon SW, Lee C, Cha SH, Yu JS, Na YJ, Kim KA, Jung SG, Kim SJ, Sang-Wook Y, Chan L. Patient selection guidelines in MR-guided focused ultrasound surgery of uterine fibroids: a pictorial guide to relevant findings in screening pelvic MRI. Eur Radiol 2008;18: