| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Vascular and Interventional Radiology |
1 From the Department of Radiology, Georgetown University Hospital, 3800 Reservoir Rd NW, CG 201, Washington, DC 20007-2197. Received January 21, 2003; revision requested April 11; revision received May 28; accepted July 1. Address correspondence to J.B.S. (e-mail: spiesj@gunet.georgetown.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Contrast material–enhanced pelvic MR imaging was performed in 20 patients before UAE, at 3 months after UAE, and then yearly for up to 3 years. Two readers compared the uterine fibroid, dominant (ie, largest) fibroid, and percentage of perfusion measurements from each of these examinations by using intraclass correlations. Seventeen patients underwent contrast-enhanced MR imaging at baseline and 3 months and 3 years after treatment. Among these patients, those with complete infarction were compared with those with incomplete infarction of the dominant fibroid at 3 years to determine extents of infarction, differences in baseline characteristics, degrees of volume reduction of the uterus and fibroid, and extents of symptom change. Comparisons were performed by using t and Pearson 
2 tests. Differences in proportions, with 95% CIs, were calculated. Each follow-up MR image was also evaluated for the presence of myometrial perfusion defects and new fibroids.
RESULTS: Intraclass correlation coefficients calculated for the two readers (range, 0.974–0.995) and with the MR imaging data (range, 0.966–0.988) were high. Of the 17 patients included in the outcome analysis, the 12 with complete fibroid infarction were more likely not to have enhancing lesions at 3-year follow-up (P = .002) than were those with incomplete infarction. No significant differences in volume or symptom changes between the two groups were detected, but growth of residual perfused portions of the incompletely infarcted fibroids was seen in three patients, two of whom had recurrent symptoms. Four patients developed new fibroids, none of which has caused symptoms. There were no instances of myometrial infarction.
CONCLUSION: Although the small study population prevented the drawing of definitive conclusions, the data suggest that although incomplete fibroid infarction may not affect outcome immediately, regrowth of uninfarcted fibroid tissue may result in symptom recurrence.
© RSNA, 2004
Index terms: Arteries, therapeutic embolization, 854.1264, 854.1266 • Uterine neoplasms, 854.315 • Uterine neoplasms, MR, 854.121411, 854.121412, 854.121416, 854.12143 • Uterine neoplasms, therapy, 854.1264, 854.1266
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
During our initial study (4) of UAE for treatment of fibroids, we evaluated a number of aspects of outcome, including the magnetic resonance (MR) imaging findings, if any, detectable at 3 months after treatment that could help predict the long-term symptom or imaging outcome. Furthermore, we were interested in knowing whether estimates of the size (ie, volume) and degree of enhancement of fibroids and the uterus varied between image readers and thus affected the reliability of the observations made from these examinations. With MR images that were obtained up to 3 years after embolization being available, the purpose of our current study was to assess and report the long-term MR imaging outcomes of fibroid tumors treated with UAE.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
At 1, 2, and 3 years after UAE therapy, additional MR imaging examinations were offered without additional cost to the first 75 patients treated. Twenty of these patients underwent MR imaging at a minimum of at baseline, 3 months, and 3 years after therapy. This group represents the study population examined for the analysis of interobserver reliability. For three of these patients, contrast material–enhanced MR images were missing from the set of images obtained at 3 months after therapy. Therefore, 17 patients were included in the analysis of outcome based on contrast-enhanced MR image findings. The mean age of the 20 patients was 43.7 years (age range, 34–50 years).
Embolization Procedure
All patients were treated between July 1997 and December 1998. One interventionalist (J.B.S.) with more than 10 years experience in interventional radiology performed embolization in all patients. The procedure was technically successful in all patients who were included in this analysis. UAE was performed in the following manner: Bilateral selective uterine artery catheterization was performed by using 5-F catheters and fluoroscopic guidance. In most cases, the catheter tips were placed in the transverse portions of the uterine arteries, at the level of the cardinal ligaments. Embolization with 500–710-µm polyvinyl alcohol particles (Contour, Boston Scientific, Boston, Mass; Ivalon, Cook, Bloomington, Ind; or Trufill, Cordis, Miami, Fla) was performed, with the end point being stasis or near stasis in the artery and with no large uterine artery branches remaining patent.
For patients with very small uterine arteries or flow-restricting spasm, a coaxial .025-inch microcatheter (FasTracker 325; Boston Scientific) was inserted through the 5-F catheter, with the outer catheter retracted into the hypogastric artery. The number of cases in which a microcatheter was used was not recorded. After each artery was embolized, a final uterine arteriogram was obtained. None of the patients underwent abdominal aortography either before or after embolization, and, thus, the presence of an ovarian collateral blood supply was not evaluated. After the UAE procedure, each patient was transferred to an observation unit for overnight (23-hour) postprocedural care.
MR Imaging Protocol
MR imaging was performed with a 1.5-T superconducting unit (Magnetom Vision; Siemens Medical Systems, Iselin, NJ) and a standard phased-array torso coil. The MR imaging protocol used has been described in detail elsewhere (7); it consisted of the following sequences: transverse T2-weighted turbo spin-echo imaging targeting the pelvis (3,200/138 [repetition time msec/echo time msec], 180° flip angle), orthogonal half-Fourier acquisition single-shot turbo spin-echo imaging (HASTE; Siemens Medical Systems) targeting the uterus (64-msec echo time, 150° flip angle), transverse T1-weighted fat-saturated spoiled gradient-echo imaging targeting the pelvis (175-msec repetition time, 80° flip angle), sagittal dynamic postcontrast T1-weighted spoiled gradient-echo imaging targeting the uterus (150/4.1, 80° flip angle) at 0, 30, 60, and 90 seconds after contrast material administration, and transverse delayed T1-weighted fat-saturated spoiled gradient-echo imaging targeting the uterus (175/4.1, 80° flip angle). In the contrast-enhanced examinations, gadopentetate dimeglumine (Magnevist; Berlex Laboratories, Monteville, NJ) was administered at a dose of 0.1 mmol per kilogram of body weight.
Data Collection
Each patient was prospectively followed up clinically and with MR imaging for a minimum of 3 years; the data were collected at approximately 3 months and 1, 2, 3, and 4 years (in four cases) after the embolization. At these follow-up examinations, the uterine fibroid and dominant fibroid dimensions were measured in three planes by two independent readers (J.P.P., J.B.S.), and the volumes (in cubic centimeters) were calculated by using the formula for a prolate ellipse: length x width x depth x 0.5233 (8). Dominant fibroid was defined as the fibroid with the largest volume measured at baseline follow-up MR imaging.
All volume and perfusion measurements were calculated by two experienced radiologists (J.P.P., J.B.S.) independently. The mean value of the two observers’ measurements was used for the outcome analyses. Before obtaining the measurements, the reviewers agreed as to which fibroid was dominant in each patient. Each postprocedural MR image was also reviewed for the presence of new fibroids. The printed MR imaging report was used to obtain an additional set of measurements of the uterus and the dominant fibroid and to determine the fibroid position.
The percentage of tissue perfused in each fibroid—from 0% to 100%—was estimated in 10% intervals. To accomplish this, the transverse and sagittal contrast-enhanced MR images were reviewed and the volume of the dominant fibroid tissue that enhanced relative to the total volume of the fibroid was estimated visually (without measurement); this estimation was converted to a percentage of volume enhancing. The contrast material bolus technique used yielded a clear demarcation of enhancing and nonenhancing tissue. Zero percentage of enhancement was considered to mean 100% (or completely) infarcted. In addition, each follow-up image was assessed for the presence of new fibroids and for defects in the normal myometrial perfusion. Only the presence or absence of these findings was noted.
The symptom status, in terms of menorrhagia and pelvic pain, at each follow-up interval was compared with that at baseline by using questionnaires that were mailed to the patients. If the questionnaire was not returned, a research nurse read the questionnaire to the patient over the telephone. The questionnaire initially consisted of a five-point Likert scale for grading symptom status: much better, slightly better, no change, slightly worse, or much worse. After the initial group of 61 patients were questioned by using the five-point scale, the scale was expanded to an 11-point linear scale with a range of grades that included -5, meaning markedly worse; -3, meaning moderately worse; -1, meaning slightly worse; 0, meaning unchanged; +1, meaning slightly improved; +3, meaning moderately improved; and up to +5, meaning markedly improved. Even numbers (ie, -4, -2, +2, and +4) were included but not labeled on the numeric linear scale.
All subsequently questioned patients received only the new (expanded) questionnaire, which was also used for all subsequent inquiries to the initial group of patients. For statistical analysis, the responses graded by using the earlier symptom scale were converted to new scale grades: For example, much worse was converted to -5, slightly worse was converted to -1, slightly improved was converted to +1, and much improved was converted to +5.
Statistical Analysis Methods
The data analysis was divided into two parts. The first part of the analysis was performed to determine the reliability of the measurements and estimates of perfusion of the fibroids. The second part of the analysis was an assessment of patient outcome based on the status of the fibroids as determined at follow-up MR imaging performed 3 months after UAE.
Intraclass correlation coefficients were calculated as measures of interobserver reliability (appropriate for continuous measurements). These measures are interpreted much the same way as 
coefficients, with values close to 1 indicating a high degree of similarity between measurements of the same object performed by different observers as compared with the total variability among objects. The same type of analysis was performed to compare the volumes of the uterus and the dominant fibroid measured by each reader with the volumes calculated from measurements provided in the official MR imaging report. CIs for intraclass correlation coefficients were determined when the reader 1 volume measurements were compared with the reader 2 measurements and the volumes determined by each reader were compared with the volumes cited in the official MR imaging reports. In addition, interobserver SDs were calculated.
The primary comparisons of outcome measures at 3 years were performed between two groups of patients. The first group consisted of the patients in whom complete infarction (ie, 0% perfusion) of the dominant fibroid was noted at 3 months. The second group consisted of the patients who were noted as having some residual perfusion (greater than 0%) of the dominant fibroid at 3 months. The baseline characteristics of these two groups were compared. The outcome measures compared at 3 years were (a) proportion of patients with complete infarction of the dominant fibroids, (b) proportion of patients in whom the dominant fibroids disappeared, (c) mean percentage of reduction of dominant fibroid volume, (d) mean percentage of reduction of uterus volume, and (e) mean score for change in menorrhagia and pelvic pain status. Differences in proportions were calculated, and 95% CIs and P values were calculated by using a normal approximation. Differences in mean values, 95% CIs, and associated P values were calculated. The P values were calculated by using the two-independent-sample t test. The Pearson 2 test was used to determine whether there was evidence that the two groups were different with regard to fibroid location.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
|
Group Comparisons
The comparison between patients with completely infarcted dominant fibroids at 3 months and those with incompletely infarcted dominant fibroids at 3 months revealed no significant difference in dominant fibroid (P = .525) or uterus (P = .336) volumes at baseline between the two groups. Comparison of the two groups at 3 years after UAE (Table 2) revealed that 100% (n = 12) of the dominant fibroids in the complete infarction group, as compared with only 40% (n = 2) of the dominant fibroids in the incomplete infarction group, were still completely infarcted at 3 years (Fig 1). These data represent a 60% difference (95% CI: 17%, 100%), which is statistically significant (P = .002). The mean percentages of reduction in dominant fibroid and uterus volumes at 3 years were similar between the groups, and neither percentage of reduction was found to be statistically significant. Differences in mean score for change in menorrhagia and pain status at 3 years similarly did not reach statistical significance. Additional analyses revealed no significant differences in symptom outcomes based on baseline fibroid volume (P = .5), baseline uterine volume (P = .2), or fibroid location (2 = 1.71, df = 3, P = .635).
|
|
|
|
|
Of three patients who had unequivocal incomplete infarction of their dominant fibroid 3 months after treatment, none had complete infarction at 3 years. The calculated total and perfused volumes of these fibroids are presented in Table 3. In all three of these patients, the volume of perfused tissue at 3 years was greater than that at 3 months, indicating regrowth of viable fibroid tissue. The percentage of perfused tissue increased in each patient as well. In two of these three patients (patients 2 and 3 in Table 3), fibroid regrowth occurred despite a continued reduction in the total volume of the fibroid. This was evident in the volumes measured at 1 and 2 years as well (data not shown). Both of these patients had recurrent symptoms between 3 and 4 years after UAE. Patient 2 underwent hysterectomy after the 3-year MR imaging follow-up examination, and patient 3 underwent repeat embolization approximately 4
years after the initial embolization. The third patient (patient 1 in Table 3) had no recurrent symptoms 4 years after therapy.
|
|
|
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
However, in some cases, the fibroid may not be completely infarcted after embolization, and, thus, there may be residual areas of perfusion. An ovarian arterial blood supply to fibroids has been shown to result in areas of residual perfusion (12,13). We have also observed that spasm may result in incomplete occlusion of fibroid vessels, and this may result in residual areas of fibroid perfusion (14).
Because the technical goal of fibroid embolization is infarction, it is important to assess the frequency and completeness with which the infarction occurs. In a previously published study (12), we assessed the effectiveness of tris-acryl gelatin microspheres by using a system of evaluation that is the basis (with slight modification) of the analyses performed in the present study. In the incompletely infarcted fibroids in this study, there was a clear demarcation between the perfused and nonperfused areas on MR images. This demarcation allowed us to monitor changes in the perfused and nonperfused portions of the tumors over time. We used the degree of perfusion of the dominant fibroid at MR imaging at 3 months after UAE as the variable of interest for determining complete versus incomplete infarction in an objective imaging assessment of outcome. Although the present study involved a small number of patients and therefore was not conclusive, we believe that the postembolization perfusion of fibroids seen at MR imaging may a useful predictor of the longer term imaging appearance of these tumors. In all cases in which there was elimination of perfusion at 3 months, the complete infarction persisted at 3 years and will presumably do so for the longer term, although this remains to be seen.
Perhaps of greater interest is what occurs if the fibroids are not completely infarcted. Both our analysis results and the presented images reveal that viable fibroid tissue can grow, even as the infarcted portion and the overall fibroid volume continue to shrink. Two of the three patients who had fibroid regrowth in our study had recurrent symptoms. These findings point to an important consideration: The volume reduction that occurs after embolization may mask residual viable tissue that is growing. This possibility suggests that volume measurements, which have been a regular feature in most reported series (1,3–5,15), may be misleading. This possibility, in turn, suggests that using the percentage of volume reduction as a means of comparing images, techniques, and outcome may be less useful than assessing the perfusion outcome of the fibroids. The results of our previous work (16) also suggest that degree of volume reduction is not a substantial predictor of outcome, and we discourage the use of volume reduction as a measure of technical success or indicator of the clinical outcome of UAE.
Our study findings also suggest that ultrasonography (US) may not be useful in predicting outcome or effective in examining patients who have recurrent symptoms. This observation may change if an accurate means of assessing the perfusion of the entire fibroid with US is developed. Without this information, one cannot know whether the fibroid is completely or incompletely infarcted or is likely to be the cause of persisting or recurrent symptoms. Thus, we again propose that contrast-enhanced MR imaging is the most definitive tool for assessing UAE outcome, particularly in patients with recurrent symptoms.
The development of new fibroids was not surprising, given that this is common with the other established uterine-sparing therapies. This fact prompts an obvious question: Is the interval after UAE to the recurrence of symptoms caused by new fibroids different from that seen after myomectomy, or is there a smaller long-term need for additional therapy after UAE? A limitation of our study was that the patient population was too small for us to determine recurrence rates.
The results of this analysis must be interpreted with some caution. Small differences in outcome could have easily been missed. The width of CIs is directly associated with the sample size, and the small number of patients in this study was not sufficient for estimating these differences with a high degree of precision. The P values and CIs for the menorrhagia and pelvic pain improvement values also should be interpreted with caution. Not only were the comparisons of these values made with an even fewer number of cases, but also the number of cases that were available for comparison was limited in part because of missing data.
The data from this small-population study suggest that the appearance of complete infarction at contrast-enhanced MR imaging after embolization is indicative of favorable long-term imaging and clinical outcomes. Although new fibroids may develop, we have not yet seen a fibroid that was successfully infarcted after embolization have recurrent growth. The data also suggest that fibroid symptoms may recur and fibroid tissue may regrow despite substantial volume reductions, and in cases in which these outcomes are suspected, contrast-enhanced MR imaging is our preferred method of evaluation.
|
FOOTNOTES |
|---|
Author contributions: Guarantors of integrity of entire study, J.P.P., J.B.S.; study concepts and design, J.P.P., J.B.S.; literature research, J.P.P.; clinical studies, J.P.P., N.G.G., J.B.S.; data acquisition, N.G.G., R.C.J., S.M.A.; data analysis/interpretation, J.P.P., J.B.S.; statistical analysis, J.B.S.; manuscript preparation, editing, and final version approval, J.P.P., J.B.S.; manuscript definition of intellectual content, J.B.S.; manuscript revision/review, all authors
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  G. Tropeano, S. Amoroso, and G. Scambia Non-surgical management of uterine fibroids Hum. Reprod. Update, May 1, 2008; 14(3): 259 - 274. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Recaldini, G. Carrafiello, D. Lagana, S. Cuffari, V. Bergamini, F. Ghezzi, and C. Fugazzola Percutaneous Sonographically Guided Radiofrequency Ablation of Medium-Sized Fibroids: Feasibility Study Am. J. Roentgenol., December 1, 2007; 189(6): 1303 - 1306. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. M. White, F. Banovac, S. Yousefi, R. S. Slack, and J. B. Spies Uterine Fibroid Embolization: The Utility of Aortography in Detecting Ovarian Artery Collateral Supply Radiology, July 1, 2007; 244(1): 291 - 298. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. M. Fennessy, C. M. Tempany, N. J. McDannold, M. J. So, G. Hesley, B. Gostout, H. S. Kim, G. A. Holland, D. A. Sarti, K. Hynynen, et al. Uterine Leiomyomas: MR Imaging-guided Focused Ultrasound Surgery--Results of Different Treatment Protocols Radiology, June 1, 2007; 243(3): 885 - 893. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Hald, N.-E. Klow, E. Qvigstad, and O. Istre Laparoscopic Occlusion Compared With Embolization of Uterine Vessels: A Randomized Controlled Trial Obstet. Gynecol., January 1, 2007; 109(1): 20 - 27. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. L. Spielmann, C. Keogh, B. B. Forster, M. L. Martin, and L. S. Machan Comparison of MRI and sonography in the preliminary evaluation for fibroid embolization. Am. J. Roentgenol., December 1, 2006; 187(6): 1499 - 1504. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. J. Kroencke, C. Scheurig, C. Kluner, M. Taupitz, J. Schnorr, and B. Hamm Uterine Fibroids: Contrast-enhanced MR Angiography to Predict Ovarian Artery Supply--Initial Experience Radiology, October 1, 2006; 241(1): 181 - 189. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. B. Spies Adding to our understanding of uterine fibroid embolization. Am. J. Roentgenol., March 1, 2006; 186(3): 846 - 847. [Full Text] [PDF]
|
|
|
|
|
|
T. Katsumori, T. Kasahara, and K. Akazawa Long-term outcomes of uterine artery embolization using gelatin sponge particles alone for symptomatic fibroids. Am. J. Roentgenol., March 1, 2006; 186(3): 848 - 854. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. B. Spies, J. Bruno, F. Czeyda-Pommersheim, S. T. Magee, S. A. Ascher, and R. C. Jha Long-Term Outcome of Uterine Artery Embolization of Leiomyomata Obstet. Gynecol., November 1, 2005; 106(5): 933 - 939. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J.-P. Pelage, J. Cazejust, E. Pluot, O. L. Dref, A. Laurent, J. B. Spies, S. Chagnon, and P. Lacombe Uterine Fibroid Vascularization and Clinical Relevance to Uterine Fibroid Embolization RadioGraphics, October 1, 2005; 25(suppl_1): S99 - S117. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Kitamura, S. M. Ascher, C. Cooper, S. J. Allison, R. C. Jha, P. A. Flick, and J. B. Spies Imaging Manifestations of Complications Associated with Uterine Artery Embolization RadioGraphics, October 1, 2005; 25(suppl_1): S119 - S132. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Ghai, D. K. Rajan, M. S. Benjamin, M. R. Asch, and S. Ghai Uterine Artery Embolization for Leiomyomas: Pre- and Postprocedural Evaluation with US RadioGraphics, September 1, 2005; 25(5): 1159 - 1172. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. B. Spies, D. K. Rajan, and M. R. Asch Invited Commentary * Authors' Response RadioGraphics, September 1, 2005; 25(5): 1173 - 1176. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS |
