Endoscopic ultrasound-guided coil embolization and thrombin injection of bleeding gastroduodenal artery pseudo-aneurysm

BACKGROUND AND OBJECTIVES: Gastroduodenal artery (GDA) aneurysms are rare but a potentially fatal condition if rupture occurs. The most common etiology is acute or chronic pancreatitis. The most common clinical presentation is gastrointestinal (GI) hemorrhage secondary to rupture of the aneurysm. Such a complication is not always related to the size of the aneurysm, and therefore, treatment should be planned as soon as a diagnosis is made. Surgical, endovascular, percutaneous, and endoscopic ultrasound (EUS)-guided interventions are used in the treatment of visceral artery pseudo-aneurysms. METHODS: A 50-year-old male had an episode of alcohol induced acute moderate severe pancreatitis 1 month back. He presented with melena, requiring six units of blood transfusions for hemodynamic stabilization. After hemodynamic resuscitation, the patient underwent upper GI endoscopy to know the etiology of massive upper GI bleeding. Upper GI endoscopy showed a bulge with overlying ulceration in the second part of the duodenum. Side viewing endoscopy showed a pulsatile bulge with overlying large ulcer. Ultrasound abdomen showed pseudo-aneurysm of size 3.8 cm × 5.6 cm arising from GDA artery. Contrast-enhanced computed tomography abdomen with angiography showed a saccular pseudo-aneurysm of size 4 cm × 6 cm in relation to GDA. EUS from duodenal bulb showed a pseudo-aneurysm of size 4.1 cm × 5.8 cm arising from GDA. Radiological or EUS-guided interventions were considered. The advantages and disadvantages of both procedures were explained. The patient selected the option of EUS-guided coil embolization. RESULTS: Under EUS and fluoroscopy guidance, five coils of 10 mm size were placed within pseudo-aneurysm through 19-gauge EUS needle. After coil embolization, contrast injection into the pseudo-aneurysm showed partial filling of pseudo-aneurysm. Review EUS 1 day after coil embolization showed high flow in the pseudo-aneurysm. Around 30% of pseudo-aneurysm was obliterated. On the 3(rd) day, 6 mL of human thrombin was (3000 IU) injected during second session of intervention in six boluses of 500 unit each. After thrombin injection, high-velocity flow was confined to the neck and periphery of pseudo-aneurysm. Further 2 mL of thrombin was injected. Immediately after thrombin injection, color Doppler EUS showed complete obliteration of pseudo-aneurysm. Two weeks later, repeat EUS showed completely obliterated pseudo-aneurysm with no flow. CONCLUSIONS: This case shows the practical problems of EUS-guided coil embolization of pseudo-aneurysms.