ORIGINAL ARTICLE

ACTA RADIOLOGICA

Uterine Fibroid Embolization Can Still Be Improved: Observations on Post-Procedure Magnetic Resonance Imaging ˚ . JAKOBSEN E. J. DORENBERG, Z. NOVAKOVIC, H.-J. SMITH, G. HAFSAHL & J. A Department of Radiology and Department of Gynecology, Rikshospitalet University Hospital, Oslo, Norway ˚ . Uterine fibroid Dorenberg EJ, Novakovic Z, Smith H-J, Hafsahl G, Jakobsen JA embolization can still be improved: observations on post-procedure magnetic resonance imaging. Acta Radiol 2005;46:547–553. Purpose: To evaluate the efficacy and completeness of uterine fibroid embolization (UFE) measured by changes in volume and signal intensity at magnetic resonance imaging (MRI), and to compare with clinical outcome. Material and Methods: 40 women with symptomatic uterine fibroids underwent bilateral uterine artery embolization. At MRI studies, including post-contrast sequences before and repeatedly after treatment, the uterus and dominant fibroids were evaluated for volume, location, and contrast enhancement. Prior to treatment, all myomas showed significant contrast enhancement. The mean uterine volume was 929 ml. Clinical examinations with emphasis on menorrhagia, pelvic pain, and urinary dysfunction were performed before and 6 and 12 months after treatment. Results: UFE was bilaterally successful in 38 patients. After UFE, MRI showed no enhancement of myomas in 30 patients. In 8 patients, post-procedural MRI revealed partially remaining vascularization of fibroids despite angiographically complete embolization of the uterine arteries. On average, uterine volume decreased by 46.2% at 12 months. There was significant improvement of symptoms in the majority of patients, but slightly less improvement in patients with partially remaining vascularization of myomas. Conclusion: UFE causes significant volume reduction of myomas and clinical improvement. MRI can reveal remaining vascularization in myomas despite angiographically complete embolization of uterine arteries. Key words: Embolization; leiomyoma; magnetic resonance imaging; therapeutic Eric J. Dorenberg, Department of Radiology, Rikshospitalet University Hospital, NO-0027 Oslo, Norway (fax. +47 23072610, e-mail. [email protected]) Accepted for publication 25 March 2005

Uterine fibroids, giving symptoms such as menorrhagia, pelvic pain, and dysuria, affect a large proportion of premenopausal women. The traditional treatment has been either drug therapy or surgery. While treatment with hormonally active drugs gives only temporary effect, both hysterectomy and myomectomy are effective, but have significant morbidity. During the past decade, embolization of the uterine arteries with microparticles has been introduced as a new, minimally invasive treatment. The aim of this study was to evaluate the efficacy and completeness of uterine fibroid embolization (UFE). Changes in uterine volume and fibroid vascularization were evaluated with MRI and compared to clinical outcome.

Material and Methods Patients In this prospective study, women referred to our institution for treatment of symptomatic uterine fibroids were offered uterine artery embolization as an alternative, experimental, treatment to hysterectomy or myomectomy. The study was approved by the regional ethics committee. From February 1999 to May 2001, 40 women (39 of Caucasian, 1 of African origin, median age 45 years, range 30– 52 years) were consecutively included after giving written informed consent. All women were over 30 years of age and had no desire of future pregnancy. Patients with contraindications to angiography or MRI were excluded from the study. Further exclusion criteria were lack of contrast enhancement DOI 10.1080/02841850510021706

# 2005 Taylor & Francis

548

E. J. Dorenberg et al.

in dominant fibroids on MRI prior to angiography, and pedunculated, subserosal fibroids. Preprocedural clinics Clinical examination included vaginal ultrasound examination, histologic sample of endometrium, and cervical cytology. Pre-embolization blood samples consisted of WBC, hemoglobin, thrombocytes, INR, CRP, and creatinine not more than 7 days prior to treatment. In addition, progesterone and FSH were measured. Pregnancy tests were performed on the day of treatment. The patients’ symptoms were evaluated with an emphasis on bleeding pattern, pressure symptoms, and urinary function. Thirty-three women (83%) had heavy menstrual bleedings or menorrhagia, 31 (78%) suffered from pain related to pressure, whereas 26 women (65%) had urinary symptoms. There was no statistically significant correlation between uterine volume and type of symptoms. Preprocedural MRI Preprocedural MRI was performed using a 1.5T unit (Magnetom Vision Plus; Siemens, Erlangen, Germany). Each examination included sagittal and axial T2-weighted turbo spin-echo (TSE) images as well as sagittal contrast-enhanced fat-suppressed T1-weighted images. For contrast enhancement, 0.1 mmol/kg b.w. gadopentate dimeglumine (Magnevist; Schering, Berlin, Germany) was injected intravenously. The number of all myomas and locations of the dominant myomas were assessed. The uterine volume was measured by manually drawing the uterine area in each axial image with an electronic caliper and multiplying the sum of the areas by the distance between slices. The axial slice thickness was 5 mm with a gap of 1.25 mm, giving a distance between slices of 6.25 mm. When present, the volumes of dominant intramural, subserosal, or submucosal myomas were evaluated using the formula for a prolate ellipse (length6depth6width60.5233). The uterine volume averaged 929 cm3 (range 230– 3878 cm3, SD 762 cm3; Table 1). Eight patients had

just 1 fibroid, 10 patients 2–5 fibroids, 13 patients 6– 10 fibroids, and 9 patients more than 10 fibroids. In 32 of 40 patients, dominant fibroids were localized within the uterine wall, whereas dominant subserosal and dominant submucosal fibroids were found in 6 and 10 patients, respectively. Intramural and submucosal fibroids tended to be larger than subserosal fibroids (Table 1). Embolization procedure The embolization procedure was performed in an angiography suite (Angiostar; Siemens, Erlangen, Germany). Patients received 1.5 g cefuroxime iv (Zinacef; Glaxo Wellcome, London, UK) and 400 mg metronidazole (Flagyl; Aventis Pharma, Strasbourg, France) as a prophylactic antibiotic treatment. All treatments were carried out under epidural anesthesia. To speed up the learning process with a new procedure, all embolizations were performed by two of three interventional radiologists (IR), only one embolization procedure was done by one IR alone. They all had broad experience in embolization procedures. We used a unilateral femoral approach. After placing a 4 French introducer sheath in the right common femoral artery, a 4 French Cobra 1 catheter was advanced to the ipsilateral uterine artery. In the case of spasm (5 patients) or difficult anatomy (6 patients), a microcatheter was used (Tracker 325; Boston Scientific, Natick, Mass., USA). Correct catheter position with the tip in the horizontal part of the uterine artery, if possible distal to the cervical branches, was confirmed by angiography. Special care was taken concerning the presence of uterine-ovarian anastomoses, but the number of patients with ovarian anastomoses was not recorded. Under intermittent fluoroscopic guidance, polyvinyl alcohol particles (Contour; Boston Scientific, Natick, Mass., USA), 355–550 mm in size were injected until complete cessation of flow in the uterine artery was obtained. Embolization was strictly flow directed; in the event of spasm we used vasodilative agents or changed to microcatheter.

Table 1. Uterine volumes and volumes of dominant fibroids before, 6, and 12 months after treatment Before UFE

Uterus Dominant fibroid

Intramural Submucosal Subserosal

6 months after UFE

n

Mean*

SD*

n

Mean*

SD*

n

Mean*

SD*

40 32 10 6

929 358 217 108

762 545 239 132

35 29 8 6

633 164 86 69

744 299 145 85

24 17 5 3

439 133 110 27

425 170 157 14

*n5number of patients. Volumes are given as mean values and standard deviation (SD) in cm3. Acta Radiol 2005 (5)

12 months after UFE

Uterine Fibroid Embolization

In 26 cases, embolization was completed using gelfoam pledges in order to obtain more proximal embolization in addition to distal embolization with microparticles. A similar technique has been used by other investigators with either gelfoam pledges or coils in combination with PVA particles (5, 17). After the procedure, the patients were transferred to a postoperative observation unit for the next few hours. Postprocedural clinics At clinical examination 6 and 12 months after embolization, the patients’ symptoms were registered and new blood samples for analysis of progesterone and FSH were taken. Postprocedural MRI Repeated MR examinations, identical to the preprocedural examination, were scheduled at 1, 3, 6, and 12 months after embolization. The remaining volumes of uterus and myomas were measured and expressed as percentage of volumes prior to treatment. The relative decrease was calculated as difference in volume expressed in percentage of volume prior to treatment. Statistics All statistical analyses were performed with SPSS for Windows (SPSS Inc., Chicago, Ill., USA, version 12.0.1). For comparison of volumetric data before and after treatment we used paired samples t tests and chi-square tests for comparison of subgroups. Results Embolization procedure Bilateral embolization of the uterine arteries was achieved in 36 of 40 patients (90%) on the first attempt and in 2 patients (5%) on the second attempt. Because of dissection of the common iliac artery during catheterization, the procedure was interrupted in one patient. Two weeks later she underwent treatment without any complications.

549

In one patient, selective catheterization of the right uterine artery failed due to severe spasms. After unilateral embolization, the referring department decided to wait for the clinical response. Because of progressing symptoms she underwent a second procedure one year later and on this attempt bilateral embolization was completed without complication. The fibroids of the remaining two patients were only partially embolized. Both had many small feeders to the uterus arising from the left internal iliac artery. In one patient the grossly enlarged uterus had additional blood supply from the right ovarian artery, which was not accessible for selective catheterization and therefore not embolized. Within 2 years this patient underwent 3 embolization procedures, but her myomas continued to grow and her symptoms persisted. In the other patient an additional, preoperative embolization prior to myomectomy was performed. This second embolization was uncomplicated, but due to bleeding complications during surgery she finally underwent hysterectomy. Postprocedural MRI Even though follow-up MRI examination were scheduled for all 40 patients, only 21, 36, 35, and 24 MRI studies were completed at 1, 3, 6, and 12 months, respectively. Nevertheless, at least one postprocedural MRI was available for all patients. Uterine volume decreased on average by 37.1% within 6 months and 46.2% within 1 year (Table 2). Only the patient who underwent repeated, unsuccessful, embolization showed an increase in uterine volume from 2120 cm3 to 3430 cm3 6 months after the first embolization. Submucosal fibroids responded better to embolization than fibroids in other locations. After 6 months their volume decreased on average by 56.1%, while subserosal fibroids decreased by 43.9% and intramural fibroids by 51.3% (Table 2). Post-embolization MRI showed complete lack of contrast-enhancement in all fibroids in 30 of 40 patients. In 10 patients only partial embolization

Table 2. Remnant volume of uterus and fibroids in different locations 6 and 12 months after UFE. Volume is expressed as percentage of volume prior to treatment. 95% confidemce intervals (95% CI) and number of patients (n) are given 6 months

Fibroids

Uterus Intramural Submucosal Subserosal

12 months

Volume (%)

95% CI (%)

n

Volume (%)

95% CI (%)

n

37.1 51.3 56.1 43.9

29.6–44.5 43.7–58.9 31.9–80.2 23.8–63.9

35 29 8 6

56.2 57.5 56.7 39.6

37.5–54.8 48.1–66.9 9.1–104.2 15.0–64.3

24 17 5 3

Acta Radiol 2005 (5)

550

E. J. Dorenberg et al.

Table 3. Number and percentage of patients with different symptoms before and at clinical control 6 or 12 months after treatment Bleeding

Before UFE (n533)* Cured After UFE Improved Unchanged / worse

Urinary dysfunction

Pelvic pain

n

%

n

%

n

%

25 18 7 0

75.8 72.0 28.0 0.0

22 18 3 1

66.7 81.8 13.6 4.5

26 13 10 3

78.9 50.0 38.5 11.5

*This table only lists patients for whom clinical follow-up data are available (n533).

was achieved, including the 2 cases of technically unsuccessful embolization. Thus, in 8 of 38 patients there was remaining circulation in myomas despite angiographically complete embolization of uterine arteries. The mean decrease in uterine volume after 6 months was higher in patients without contrast enhancement in fibroids on follow-up MRI, compared to patients with signs of incomplete embolization on MRI. This difference was statistically significant (42.7% vs. 27.7%; Pv0.05). Postprocedural clinics Clinical control data after 6 or 12 months were available for 33 patients (83%), in 25 with bleeding symptoms, 26 with pressure-related pain, and in 22 patients with urinary dysfunction prior to treatment (Table 3). Clinical success was obtained in all patients with menorrhagia. Seven patients (28.0%) observed marked reduction, and 18 (72.0%) reported complete cessation of symptoms. One of the 22 patients with urinary dysfunction (4.5%) did not respond to treatment. Eighteen patients (81.8%) had normalization of urinary function, the remaining 3 (13.6%) experienced decreased symptoms. In the group with pressure-related pain, the clinical success rate was lower. At control, 13 patients (50.0%) were pain-free, 10 (38.5%) had less pain than they had prior to embolization, while 3 patients (11.5%) had no change in pain. There was a tendency towards better clinical outcome in the group of patients with complete embolization of all fibroids at MRI as compared to

those with incomplete embolization. This difference was statistically significant for patients suffering from pelvic pain (Pv0.05; Table 4). Complications As mentioned above, there was one periprocedural complication due to dissection of the common iliac artery. This patient recovered without sequelae. A total of four patients (10%) underwent hysterectomy after UFE. About 1 week after UFE, a patient with a solitary, submucosal, myoma was readmitted owing to vaginal discharge of necrotic material. In the following days she had increasing pain and fever. MRI showed uterine volume reduction from 1130 cm3 to 490 cm3 and lack of contrast enhancement in her myomas. However, as there were small air bubbles in the myoma, hysterectomy was performed because of infected necrosis of fibroid tissue causing sepsis (Fig. 1). The second patient had more than 55% reduction of uterine volume at 6 months (Fig. 2) and complete cessation of bleeding problems after UFE, but underwent surgery because of remaining diffuse pelvic pressure. The third patient was a woman with pelvic pain and urinary dysfunction prior to treatment. Despite angiographically complete embolization, MRI revealed remaining contrast enhancement in part of her myomas. After 6 months her uterine volume had reduced from 580 cm3 to 460 cm3, but there was no clinical improvement, and the patient chose to undergo hysterectomy. The remaining case was the woman who underwent repeated embolization prior to surgery.

Table 4. Number and percentage of patients with pelvic pain at clinical control 6 or 12 months after UFE, subgrouped by findings on MRI 6 months after UFE (Pv0.05) Pelvic pain after UFE (n526)

Complete embolization at MRI (n517) Incomplete embolization at MRI (n59)

Acta Radiol 2005 (5)

Cured (n513; 50%)

Improved (n510; 38.5%)

Unchanged (n53; 11.5%)

12 (70.6%) 1 (11.1%)

4 (23.5%) 6 (66.7%)

1 (5.9%) 2 (22.2%)

Uterine Fibroid Embolization

551

Fig. 1. Patient with septic fever 2 weeks after embolization. Sagittal T2-weighted image (A) and sagittal contrast-enhanced, fat-suppressed T1-weighted image (B) are presented. The large fibroid shows no evidence of contrast-enhancement (B). It is surrounded by fluid and there are air bubbles indicating infection.

Fig. 2. Sagittal, contrast-enhanced, fat-supressed T1-weighted images prior to embolization (A) and 6 months after (B). Despite marked volume reduction of both uterus and fibroids, the patient decided to undergo hysterectomy due to remaining pelvic pain.

Discussion In 2000, ANDERSEN et al. (1) published the first Scandinavian article on UFE. Like other authors (14, 16), they used ultrasound to assess uterine volume and size of dominant fibroids. MRI provides more reliable information on size, number, and exact location of myomas than ultrasound (3). OMARY et al. (11) showed that MRI significantly alters the diagnosis and treatment plans in the evaluation of women with presumed symptomatic fibroids. In our own material, MRI showed contrast enhancement in parts of fibroid tissue in 10 patients, despite volume reduction of fibroids and uterus in 9 of them. This finding on MRI would probably not have been detected by ultrasonography. Even though the difference in clinical response between the patients with signs of complete embolization on MRI, compared to those with partially infarcted

fibroids, was statistically significant only for patients with pelvic pain, there was a greater reduction of uterine volume in the first group after 6 months (43% vs. 27%; Pv0.05). KATSUMORI et al. (7) described similar findings in their report from 2001 (60% vs. 35%). Additionally, partially remaining circulation might be of interest in long-term follow-up, possibly being associated with a higher rate of recurrent fibroids. We did not routinely perform abdominal aortography in order to detect blood supply from the ovarian arteries. BINKERT et al. (2) demonstrated ovarian collateral arteries in 25% of patients, and their detection influenced treatment in 6% of cases. Such collaterals might explain the incomplete embolization in some of our patients, where the embolization was complete as judged by angiography of the uterine arteries. However, a publication by PELAGE et al. (13), including 294 cases of Acta Radiol 2005 (5)

552

E. J. Dorenberg et al.

UFE and reviewing the relevance of the ovarian artery to UFE, supports the notion that abdominal aortography should not be routinely performed during UFE. They recommend the use of abdominal aortography only in women with prior tubo-ovarian pathology or surgery, or in cases of large fundal fibroids. In other patients without blood supply to the uterus from the ovarian arteries, the degree of embolization may have been misinterpreted on angiography. There seems to be a potential for ameliorating the end-point of UFE and the use of other imaging modalities might be of interest. In accordance with earlier published data, the overall clinical response was very good; all patients with menorrhagia, 95% of patients with urinary dysfunction, and 88.5% with pressure-related pain experienced symptom relief or complete cure. In our material, the case of hysterectomy due to acute infection was the only major complication (2.5%) according to the categories defined by the Society of Interventional Radiology (SIR) (4). Infection of large, submucosal myomas requiring surgery has been reported by others (6, 12). While some cases require hysterectomy, others can be treated successfully with hysteroscopically assisted myomectomy. The elevated risk for infectious complications in submucosal myomas is known, but not regarded as a contraindication. In a survey of complications after hysterectomy for benign disorders in Norway, OMA (10) reported an overall complication rate of 29.2% in 315 patients, varying from 11% to 42%, depending on the surgical technique. In an analysis of postoperative hospitalization, morbidity and readmission after hysterectomy in Denmark, MØLLER et al. (9) estimated the complication rate at a minimum of 18%. The mortality rate in this large material including over 10,000 patients was 0.06%. Using the SIR guidelines for classification of complications, SPIES et al. reported 10% of minor complications and 1.25% of major complications in 400 women treated with UFE (15). Similar values are published in other large series (8, 16), supporting the opinion that incidence of complications is significantly lower by UFE than by surgical treatment. In conclusion, UFE is an effective, minimally invasive, method of treating uterine fibroids and is capable of producing significant volume reduction of both myomas and uterus. Despite angiographically complete embolization of uterine arteries, MRI may reveal only partial infarction of Acta Radiol 2005 (5)

fibroids in about one-fifth of cases. This finding has to be further explored in order to improve the UFE technique. References 1. Andersen PE, Lund N, Justesen P, Munk T, Elle B, Floridon C. Uterine artery embolization of symptomatic uterine fibroids. Initial success and short-term results. Acta Radiol 2001;42:234–8. 2. Binkert CA, Andrews RT, Kaufman JA. Utility of nonselective abdominal aortography in demonstrating ovarian artery collaterals in patients undergoing uterine artery embolization for fibroids. J Vasc Interv Radiol 2001;12:841–5. 3. Dudiak CM, Turner DA, Patel SK, Archie JT, Silver B, Norusis M. Uterine leiomyomas in the infertile patient: preoperative localization with MR imaging versus US and hysterosalpingography. Radiology 1988;167:627–30. 4. Goodwin SC, Bonilla SM, Sacks D, Reed RA, Spies JB, Landow WJ, et al. Reporting standards for uterine artery embolization for the treatment of uterine leiomyomata. J Vasc Interv Radiol 2001;12:1011–20. 5. Goodwin SC, McLucas B, Lee M, Chen G, Perella R, Vedantham S, et al. Uterine artery embolization for the treatment of uterine leiomyomata. Midterm results. J Vasc Interv Radiol 1999;10:1159–65. 6. Goodwin SC, Vedantham S, McLucas B, Forno AE, Perrella R. Preliminary experience with uterine artery embolization for uterine fibroids. J Vasc Interv Radiol 1997;8:517–26. 7. Katsumori T, Nakajima K, Tokuhiro M. Gadoliniumenhanced MR imaging in the evaluation of uterine fibroids treated with uterine artery embolization. Am J Roentgenol 2001;177:303–7. 8. McLucas B, Adler L, Perrella R. Uterine fibroid embolization: nonsurgical treatment for symptomatic fibroids. J Am Coll Surg 2001;192:95–105. 9. Moller C, Kehlet H, Utzon J, Ottesen B. Hysterectomy in Denmark. An analysis of postoperative hospitalisation, morbidity and readmission. Dan Med Bull 2002;49:353–7. 10. Oma J, Urdal K. [Gynecological laparoscopy in Norway], Tidsskr Nor Laegeforen 2003;123:328–9. In Norwegian. 11. Omary RA, Vasireddy S, Chrisman HB, Ryu RK, Pereles FS, Carr JC, et al. The effect of pelvic MR imaging on the diagnosis and treatment of women with presumed symptomatic uterine fibroids. J Vasc Interv Radiol 2002;13:1149–53. 12. Pelage JP, Le Dref O, Soyer P, Kardache M, Dahan H, Abitbol M, et al. Fibroid-related menorrhagia: treatment with superselective embolization of the uterine arteries and midterm follow-up [see comment]. Radiology 2000;215:428–31. 13. Pelage JP, Walker WJ, Le Dref O, Rymer R. Ovarian artery: angiographic appearance, embolization and relevance to uterine fibroid embolization. Cardiovasc Intervent Radiol 2003;26:227–33. 14. Spies JB, Ascher SA, Roth AR, Kim J, Levy EB, Gomez-Jorge J. Uterine artery embolization for leiomyomata. Obstet Gynecol 2001;98:29–34.

Uterine Fibroid Embolization 15. Spies JB, Spector A, Roth AR, Baker CM, Mauro L, Murphy-Skrynarz K. Complications after uterine artery embolization for leiomyomas. Obstet Gynecol 2002; 100:80. 16. Walker WJ, Pelage JP. Uterine artery embolization for symptomatic fibroids: clinical results in 400 women with

553

imaging follow up [see comment]. Br J Obstet Gynaecol 2002;109:1262–72. 17. Worthington-Kirsch RL, Popky GL, Hutchins FL. Uterine arterial embolization for the management of leiomyomas: quality of life assessment and clinical response. Radiology 1998;208:625–9.

Acta Radiol 2005 (5)