ABSTRACT
Fistulas between the arteries and the gastrointestinal tract are rare but can be fatal. We present a case of an ilioenteric fistula between the left external iliac artery and sigmoid colon caused by radiotherapy for cervical cancer, which was treated with endovascular management using a stent graft. A 38-year-old woman underwent concurrent chemoradiotherapy for cervical cancer recurrence. Approximately 9 months later, the patient suddenly developed hematochezia. On her first visit to the emergency room of our hospital, computed tomography (CT) images did not reveal extravasation of contrast media. However, 8 hours later, she revisited the emergency room because of massive hematochezia with a blood pressure of 40/20 mmHg and a heart rate of 150 beats per minute. At that time, CT images showed the presence of contrast media in almost the entire colon. The patient was referred to the angiography room at our hospital for emergency angiography. Inferior mesenteric arteriography did not reveal any source of bleeding. Pelvic arteriography showed contrast media extravasation from the left external iliac artery to the sigmoid colon; this was diagnosed as an ilioenteric fistula and treated with a stent graft. When the bleeding focus is not detected on visceral angiography despite massive arterial bleeding, pelvic arteriography is recommended, especially in patients with a history of pelvic surgery or radiotherapy.
ABSTRACT
PURPOSE: We evaluated the detection rate and degree of motion artifact of the modified CAIPIRINHA-VIBE (mC-VIBE) without view-sharing and compare them with the CAIPIRINHA-Dixon-TWIST-VIBE (CDT-VIBE) with view-sharing on multi-arterial gadoxetic acid-enhanced liver MRI in the assessment of hepatocellular carcinoma (HCC). MATERIAL AND METHODS: We retrospectively identified 114 pathological-proven hepatic tumors in 114 patients with risk of HCC who underwent multi-arterial gadoxetic acid-enhanced MRI between June 2016 and June 2018. All patients underwent triple arterial phase imaging using the mC-VIBE without view-sharing (54 patients; 49 HCCs and 5 non-HCCs) or the CDT-VIBE with view-sharing (60 patients; 55 HCCs and 5 non-HCCs). We compared the detection rate of two sequences for HCC, with reference to LI-RADS.V.2017. We also compared the mean motion scores and proportions of transient severe motion (TSM) in two sequences. RESULT: For the examination using the mC-VIBE, the HCC-detection rate was significantly higher, compared with that using CDT-VIBE (93.9% [46/49] vs 80.0% [44/55], respectively; p = 0.047). For the examination with the mC-VIBE, mean motion scores were significantly lower compared with those of CDT-VIBE for all multi-arterial phases (1.21, 1.19, and 1.15 vs. 1.82, 1.85, and 1.84, respectively; p < 0.001 for all three comparisons). The proportion of TSM in the CDT-VIBE was significantly higher than that in the mC-VIBE (15.0% [9/60] vs 0.0% [0/54], respectively; p = 0.003). CONCLUSION: In multi-arterial phase gadoxetic acid-enhanced MRI, the mC-VIBE sequence without view-sharing has slightly higher HCC-detection rate and fewer motion artifacts compared with CDT-VIBE with view-sharing. KEY POINTS: ⢠Multi-arterial phase using the mC-VIBE without view-sharing can overcome motion artifacts, resulting in providing optimal arterial phase imaging. ⢠The HCC-detection rate is slightly higher with the mC-VIBE vs. CAIPIRINHA-Dixon-TWIST-VIBE with view-sharing (CDT-VIBE). ⢠View-sharing of CDT-VIBE in the multi-arterial phase is associated with increased frequency of TSM.