Obstetric-Gynecologic Imaging
Case Report
MRI Detection of Uterine Necrosis After Uterine Artery Embolization for Fibroids
Audio Available | 
Share
 |
In the last decade, selective embolization of the uterine arteries has been described as a safe and effective alternative to hysterectomy and myomectomy in the treatment of symptomatic uterine fibroids. Clinical success rates have been reported to be 80–95% of patients treated [1]. Potential complications are rare and include postembolization syndrome, infection, hematoma, premature ovarian failure, vesicouterine fistula, and fatal septicemia. We present an additional complication: a pathologically proven case illustrating the MRI appearance of uterine necrosis 1 day after embolization of the uterine arteries for fibroids.
| Case Report |  |
|---|
A 51-year-old premenopausal woman, gravida 5 para 4, was observed for many years for uterine fibroids, menorrhagia, anemia, and dysmenorrhea. Pelvic and office sonographic examinations showed a large uterine fibroid with an estimated size approximating 16- to 18-week pregnancy. Her symptoms were refractory to medical management, and she was advised to undergo total abdominal hysterectomy.
Four days before her scheduled surgery, the patient suffered an acute episode of pelvic hemorrhage, similar to prior episodes of dysmenorrhea and menorrhagia resulting from her fibroid disease. At an outside hospital, her hemoglobin was found to be 7.6 g/dL and she received 2 U of packed RBCs. The patient was then transferred to our institution where she opted to undergo uterine artery embolization after a discussion of her treatment options.
Bilateral superselective catheterization of the uterine arteries was performed in the usual manner. The uterus showed normal flow, vessels, and a moderately vascular mass consistent with a uterine fibroid (Fig. 1A). Although the fibroid was not markedly hypervascular, it can be difficult to accurately assess fibroid viability on the basis of angiographic data alone. The decision was made to proceed with uterine artery embolization on the basis of the clinical scenario. Embolization was completed with 2 mL of 500- to 700-μm-diameter Embospheres (embolic particles, Biosphere Medical) used on the right side and 4 mL on the left. Embolization proceeded to the usual end point of marked slowing of flow and occlusion of branches. No intraprocedural complications were noted.
 View larger version (62K) | Fig. 1A. —51-year-old woman with large submucosal fibroid and diffuse uterine necrosis. Diagnostic arteriogram of uterus shows moderately vascular mass consistent with fibroid. |
Two hours after the procedure, the patient complained of a small amount of vaginal bleeding and typical postembolization pain that was controlled with narcotic medication. She had persistent complaints of pelvic pain, and MRI was performed after the procedure. After MRI, her pain was controlled with oral medication, and she was discharged the following day.
An MRI examination was performed using a 1.5-T scanner (Symphony, Siemens Medical Solutions) and a standard phased-array torso coil. The following pulse sequences were used: axial T1-weighted images (TR/TE, 360/11; flip angle, 90°), axial T2-weighted half-Fourier acquisition single-shot turbo spin-echo images (1,100/60; flip angle, 150°), and axial and coronal T2-weighted turbo spin-echo images (4,370/117; flip angle, 180°). Sagittal, axial, and coronal dynamic spoiled gradient echo T1-weighted images with fat suppression (128/204; flip angle, 70°), both before and after the IV administration of 20 mL of gadopentetate dimeglumine (Magnevist, Berlex), were also obtained.
On T1-weighted images, the uterus was homogeneously dark in signal intensity. On T2-weighted images, a large leiomyoma was seen measuring 6.3 × 5.2 cm (Fig. 1B). Areas of higher signal intensity on T2-weighted images might have been related to hemorrhage. No thickening of the junctional zone was noted. The endometrial stripe was widened.
 View larger version (50K) | Fig. 1B. —51-year-old woman with large submucosal fibroid and diffuse uterine necrosis. Sagittal T2-weighted turbo spin-echo image shows large submucosal mass (arrow) within uterus. Endometrial stripe partially surrounds mass (arrowhead). No junctional zone thickening is seen. |
Contrast-enhanced images revealed nonenhancement of the entire uterus with the exception of a few capsular vessels (Figs. 1C and 1D). Findings were suggestive of acute global hypoperfusion of the uterus and diffuse uterine necrosis.
 View larger version (54K) | Fig. 1C. —51-year-old woman with large submucosal fibroid and diffuse uterine necrosis. Sagittal T1-weighted gradient-echo image after IV administration of contrast material reveals marked hypoperfusion within entire uterus, suggestive of uterine necrosis. Vagina and cervix are spared because of separate internal iliac artery branches and rich vascular supply (arrow). |
 View larger version (49K) | Fig. 1D. —51-year-old woman with large submucosal fibroid and diffuse uterine necrosis. Axial T1-weighted gradient-echo image again shows lack of uterine perfusion as well as some capsular collateral vessels (arrow). |
Four days after uterine artery embolization, the patient returned to her gynecologist for evaluation. She reported severe pain. The patient was afebrile with stable vital signs. Because of the MRI findings suggesting minimal uterine perfusion, along with the clinical symptoms, a total abdominal hysterectomy and bilateral salpingo-oophorectomy were performed 4 days after uterine artery embolization. The patient tolerated this procedure well and was discharged on day 3 after the operation.
On gross inspection, the bivalved uterus was the size of a uterus in the 14th week of pregnancy and had a myometrial wall thickness of 5.5 cm. It contained a 5.5 × 5.5 × 4 cm submucosal mass with a tan–white–pink whorled cut surface. Microscopic examination revealed extensive necrosis of the endometrium and myometrium with a single necrotic leiomyoma (Figs. 1E and 1F).
 View larger version (157K) | Fig. 1E. —51-year-old woman with large submucosal fibroid and diffuse uterine necrosis. Photomicrograph of necrotic leiomyoma shows extensive areas of necrotic leiomyoma (arrow) with small viable portion of leiomyoma (arrowhead). (H and E, ×200) |
 View larger version (174K) | Fig. 1F. —51-year-old woman with large submucosal fibroid and diffuse uterine necrosis. Photomicrograph of necrotic myometrium shows myometrium with extensive necrosis and scattered acute inflammatory cells. (H and E, ×400) |
| Discussion | |
|---|
Uterine artery embolization is a widely used treatment in which both uterine arteries are embolized to arrest blood supply, thereby initiating ischemia, subsequently resulting in the infarction of uterine fibroids. Arteriographic evidence suggests that no specific artery supplies blood flow to the fibroid; rather, an organized network of peripheral vessels originates from each uterine artery. Diffuse necrosis of the uterine fibroid is an expected finding after successful uterine artery embolization; however, necrosis of the normal myometrium is a rare complication. Though the literature has described previous ischemic-related complications, to our knowledge this is the first reported case of MRI-visualized diffuse uterine necrosis after uterine artery embolization for fibroids.
Some plausible theories exist that may explain this rare outcome. First, it is possible that the leiomyoma was already nonviable before the uterine artery embolization, thereby rendering it incapable of taking up the embolic particles administered. The fact that the angiogram showed a moderately vascular, rather than hypervascular, mass may support this hypothesis. In this case, MRI was not performed before the procedure. Thus, the procedure may have led to the delivery of all embolic particles to the viable uterus instead of the desired target. The collateral blood supply to the remaining uterus may have been limited; thus, blood flow to the normal myometrium may have become compromised by the procedure, which occluded blood flow to the uterine arteries. It has been postulated that the endometrium and inner myometrium have fewer available collateral blood vessels and are therefore more susceptible to infarction and subsequent sepsis [2].
However, one commonly held theory cannot explain the outcome in this case. Uterine artery embolization is accomplished with particulate material, most commonly of 300–500 μm in diameter. When smaller particles are used, they reach farther into smaller vessels inside the myometrium causing a larger area of ischemia and infarction [1]. In this case study of uterine necrosis, embolic particle size is not thought to play a significant role because the more typical larger, safer particles were used for the procedure.
To our knowledge, this is the first reported case illustrating the MRI appearance of pathologically confirmed uterine necrosis after uterine artery embolization for fibroids. Uterine necrosis after cesarean delivery with its MRI appearance has been previously reported in the literature, as has uterine necrosis after arterial embolization for postpartum hemorrhage [3, 4]. Godfrey and Zbella [5] describe a case they believed to be diffuse uterine necrosis 2 months after uterine artery embolization. The case is likely the result of uterine infection after uterine artery embolization as suggested by their clinical data as well as CT findings. Similar critiques of this report have been published in the literature and support the belief that Godfrey and Zbella describe a case of pelvic infection after uterine artery embolization rather than uterine necrosis [6, 7].
MRI was critical in diagnosing the unexpected outcome of uterine necrosis after uterine artery embolization. Although global hypoperfusion has been described immediately after uterine artery embolization for fibroids [8], in this case such a dramatic hypoperfusion of the uterus was present that necrosis was strongly suspected. The patient was carefully observed and operated on expeditiously with an excellent outcome. As mentioned previously, one limitation of our study was the absence of MRI before the procedure. Because fibroid viability cannot accurately be established by angiographic findings alone, MRI would have been helpful in assessing the presence of fibroid necrosis. Though some institutions have made it protocol to perform MRI before and after uterine artery embolization for fibroids, no universal imaging standard has been adopted. In our case, MRI before the procedure could have been used to assess the vascularity and viability of the fibroid, thereby predicting the efficacy of uterine artery embolization. Furthermore, MRI has a high accuracy in distinguishing between fibroids and other uterine diseases such as adenomyosis and other potential lesions in the pelvis. Finally, as in the case we describe, MRI immediately after uterine artery embolization can detect potential complications such as necrosis after the procedures.
Uterine artery embolization is a safe and effective alternative to hysterectomy in the treatment of symptomatic uterine fibroids. In this report, we present a unique complication of uterine necrosis with its MRI appearance after uterine artery embolization for fibroids. We believe MRI can play a pivotal role in detecting this potential complication.
Address correspondence to H. Gabriel.
| References | |
|---|
