Transcatheter Arterial Embolization for Palliation of Uterine Body Cancer Bleeding.

Endometrial cancer, or more broadly, uterine body cancer, is now the most common gynecologic malignancy in developed countries. Its incidence is increasing in many countries including South Korea, where the incidence of uterine body cancer has surpassed that of cervical cancer as of 2019, according to the Korea Central Cancer Registry ( 1 ). Patients commonly present with vaginal bleeding, and while this is mostly managed conservatively, the definite, mainstay of treatment for uterine body cancer is surgery including hysterectomy with bilateral salpingo-oophorectomy. However, in advanced stages or in recurrent diseases, surgery may not be applicable and other measures such as radiation therapy or chemotherapy are taken ( 2 3 ). In these cases, additional options for bleeding control need to be considered, since the tumor is remaining in the pelvic cavity, often in close relation to the numerous vessels within, all of which have the potential to cause bleeding. Transcatheter arterial embolization (TAE) is now widely used in various gynecologic and obstetric situations. The most well-known indications include symptomatic uterine fibroids and intractable postpartum hemorrhage ( 4 5 ). However, the spectrum of applicable situations is much wider, and includes other benign conditions such as adenomyosis, but also non-benign conditions such as bleeding due to advanced stage gynecologic malignancies. While cases of successful TAE of refractory bleeding related to cervical cancer have often been reported and widely studied ( 6 7 ), there have been very few reports regarding TAE for bleeding due to cancers involving the uterine body. In this study, we evaluated the safety and efficacy of TAE performed in uterine body cancer patients presenting with bleeding.

A total of six patients with varying types of uterine body cancer were included in the study ( Table 1 ). The median age of patients was 58 years (range = 29–79 years). Two patients (patient no. 2, 5) had received hysterectomy, and were included due to bleeding from recurred tumor. Five patients had received chemotherapy, except one patient (patient no. 3) who was lost to follow-up after the initial diagnosis and had not received any treatment. None of the patients had received radiotherapy before TAE. More than half of the patients ( n = 4 of 6) presented with vaginal bleeding. Two patients who had previously received hysterectomy presented with hematochezia due to fistulas between the bowel and recurrent masses within the pelvic cavity ( Fig. 1 ). Four patients presented with chronic, persistent bleeding, one patient presented with chronic, intermittent bleeding, whereas one patient (patient no. 6) presented with an acute event of massive bleeding. One patient (patient no. 4) went through two sessions of TAE, and presented with vaginal bleeding on the first session, and hemoperitoneum due to tumor bleeding on the second session. Out of six patients and seven procedures, relevant CT scans at the time of TAE were available in five procedures. In one patient who presented with acute massive bleeding (patient no. 6), evidence of active bleeding was seen on CT as contrast extravasation ( Fig. 2 ). The rest of the CT scans showed a necrotic mass with or without vascular encasement. Details of TAE including angiographic findings, embolized arteries, and embolic material are summarized in Table 1 . On angiography, active bleeding was shown as contrast extravasation in none of the patients. Instead, angiography showed hypervascular tumor stainings within the uterus or pelvic cavity. The main feeders of the tumor staining were considered culprit arteries and subsequently embolized. The uterine arteries were embolized in four patients and the internal iliac arteries were embolized in two patients. Tumor feeders from uterine arteries or internal iliac arteries were selectively embolized in sessions in which tumor feeders were identified. Tumor feeders from the superior rectal artery and omental branches were also embolized in two patients (patient no. 2, 4), respectively. In most of the sessions ( n = 6 of 7), bilateral arteries (e.g. both uterine or internal iliac arteries) were embolized. Selective embolization of the left IIA was performed in one patient (patient no. 5). The most used embolic material was gelatin sponge particles ( n = 6). Coils with gelatin sponge particles were used in one patient (patient no. 5), and PVA was used in another patient (patient no. 1). After the procedure, there were no remaining bleeding focuses, and disappearance of over 90% of the extent of tumor staining was seen in all patients, thus resulting in technical success in all patients. Bleeding control for at least one week from the procedure was seen in three patients, resulting in a clinical success rate of 50% (3/6). When a total of seven TAE were included, bleeding control for at least one week was achieved in 42.8% (3/7) of the treatments. Conservative measures were taken in patients who experienced re-bleeding. There were no major complications during the immediate follow-up. One patient (patient no. 6) showed mild fever on the day after the procedure, which subsided after conservative management. One patient (patient no. 4) went through a second session of TAE 6 months after the initial treatment due to hemoperitoneum related to tumor bleeding. In this second session, even though technical success was achieved, the patient shortly presented with recurrent bloody ascites presumably due to persistent tumor bleeding. On follow up, most patients ( n = 5 of 6) expired within 60 days (median = 46, range = 17–60 days). One patient (patient no. 5) died of hypovolemic shock 47 days after the TAE. This patient exhibited persistent hematochezia after TAE, even after having received radiotherapy to the pelvis as a palliative measure to control the bleeding. Another session of TAE was considered in this patient, but eventually was not performed due to the patient’s wishes. The others were terminal stage cancer patients at the time of TAE, and died of disease progression.

Endovascular treatment is considered an effective therapeutic option for a wide range of gynecologic and obstetric conditions. Ever since the first reports of TAE for bleeding control of gynecologic malignancies in the 1970s ( 8 9 ), more studies have been conducted over the past 40 years, implementing different treatment methods and technologies ( 6 7 10 11 ), and showing the increasing role of TAE. These studies mainly dealt with cervical cancer bleeding, and the separate evaluation of TAE on bleeding control for cancers involving the uterine body has rarely been done, albeit its increasing incidence. We investigated seven cases of TAE performed in uterine body cancer patients presenting with bleeding and assessed the efficacy and safety of the procedure on bleeding control. TAE showed a 100% rate of technical success and a 50% rate of clinical success. Put differently, TAE was able to provide immediate hemorrhagic control in all cases whereas half of the patients once again experienced some level of bleeding within a week. This is in-keeping with the results of previous studies on TAE for gynecologic malignancy bleeding, which showed that recurrent hemorrhage after TAE was relatively common. One study ( 12 ) demonstrated that recurrent bleeding was particularly common when absorbable gelatin sponge was used, which was also the case in our study. Other studies with uterine cervical cancers ( 6 7 ) also showed moderate rates of recurrent bleeding in the first few months after the procedure, which, however, were relatively less severe or more manageable by additional TAE or conservative measures. In our study, only one patient showed major bleeding which was directly associated with death, while in the others, re-bleeding was managed conservatively. Angiography showed hypervascular tumor stainings without contrast extravasation or pseudoaneurym formation. This seems to be related to the fact that most patients presented with persistent tumor oozing bleeding instead of an acute event of massive bleeding. Considering that technical success was achieved in all of these cases, it may be useful to empirically embolize hypervascular tumor stainings for the control of tumor oozing, even when active bleeding and its culprit artery is not identified. Further, oozing bleeding and hypervascular tumor staining without contrast extravasation or pseudoaneurysm may be suggested as a characteristic of uterine body cancer, although further validation with larger subjects is needed. It may be worthy to note that two cases of TAE performed in hysterectomy state patients with recurrent masses were also included in this study. In these patients, tumor bleeding presented not as vaginal bleeding but as hematochezia. On cross-sectional images, fistulas between the recurred mass and the bowel were seen, thus accounting for the patients’ symptoms of hematochezia. Immediate bleeding control was also achieved by embolization of feeding vessels from uterine and internal iliac arteries. Thus, uterine body cancer may present with other ways of bleeding than vaginal bleeding, and even in recurrent masses arising after the initial surgery, TAE may be considered as an option for bleeding control. Five out of six patients expired within 60 days in our study due to disease progression, and the true longer-term impact of TAE in these patients’ general outcome and survival is unknown. Nonetheless, TAE was able to provide a high level of immediate bleeding control, and the procedure itself was not associated with any major complications. Thus, we believe that TAE can be considered as an effective way of hemorrhagic control in critical periods in the disease course of advanced and terminal stage cancer patients. This study is limited in its small number of included patients, being a single centered study, the retrospective nature, and the relatively short period of follow-up. However, TAE of uterine body cancers has never been solely investigated, and we believe that this study not only demonstrates the efficacy and potential of TAE of these types of cancers, but also serve as a pilot for future studies. Also, the one-week period of bleeding control which we defined as clinical success may be rather short in terms of cost-effectiveness. However, in advanced stage diseases where the life-expectancy ranges from days to weeks, this short period of palliation may also be of help. In conclusion, TAE can be considered an effective and safe way of bleeding control in inoperable, advanced stage disease in uterine body cancer patients.

The hospital institutional review board approved this retrospective study, and the requirement for informed consent was waived (IRB No. S2022-0644-0001). The data regarding patients who went through TAE due to tumor bleeding from uterine body cancer at our institution from June, 2001 to March, 2020 were retrospectively reviewed. Patients with uterine cervical cancer were not included in the study, and a total of six patients were identified. Tumor bleeding presented as vaginal bleeding or hematochezia, with varying degrees of bleeding from intermittent episodes of chronic bleeding, to acute, massive bleeding. TAE was conducted in our institution by one of three expert interventional radiologists with 7–20 years of experience. After puncturing the femoral artery, internal iliac arteriograms were obtained, selective catheterization of the feeding arteries was performed, and the culprit arteries were embolized. Embolic materials included polyvinyl alcohol (PVA; Contour; Boston Scientific, Cork, Ireland), gelatin sponge particles (Gelfoam; Pharmacia & Upjohn, Kalamazoo, MI, USA), and microcoils (Tornado or microNester; Cook Medical, Bloomington, IN, USA), and were chosen depending on the operator’s preference. Post-embolization angiography was performed immediately after the embolization procedure to evaluate if the target artery was completely excluded. According to the medical records of the patients, data regarding patient age, cancer type (pathology), The International Federation of Gynecology and Obstetrics (FIGO) stage, treatment history regarding the cancer, degree of bleeding, cross-sectional imaging, and details of TAE were obtained. Data regarding immediate follow-up including complications, long-term follow up and patient survival were also collected when available. Technical success rates and clinical success rates were calculated. Post-embolization complications were classified according to the Society of Interventional Radiology Classification System. Technical success was defined as disappearance of contrast extravasation, or successful embolization of the target vessels leading to more than 90% disappearance of hypervascular tumor staining. Clinical success was defined as the absence of rebleeding within one week of TAE. Relevant CT scans were defined as CT scans obtained within 10 days prior to TAE.