A 48-year-old Indian male with alcoholic liver cirrhosis was admitted after being found unresponsive. He was hypotensive and had hematochezia. Esophagogastroduodenoscopy (EGD) showed small esophageal varices and a clean-based duodenal ulcer. He continued to have hematochezia and anemia despite blood transfusions. Colonoscopy was normal. Repeat EGD did not reveal any source of recent bleed. Twelve days after admission, his hematochezia ceased. He refused further investigation and was discharged two days later. He presented one week after discharge with hematochezia. EGD showed non-bleeding Grade 1 esophageal varices and a clean-based duodenal ulcer. Colonoscopy was normal. Abdominal computed tomography (CT) showed liver cirrhosis with mild ascites, paraumbilical varices, and splenomegaly. He had multiple episodes of hematochezia, requiring repeated blood transfusions. Capsule endoscopy identified the bleeding site in the jejunum. Concurrently, CT angiography showed paraumbilical varices inseparable from a loop of small bowel, which had herniated through an umbilical hernia. The lumen of this loop of small bowel opacified in the delayed phase, which suggested variceal bleeding into the small bowel. Portal vein thrombosis was present. As he had severe coagulopathy and extensive paraumbilical varices, surgery was of high risk. He was not suitable for transjugular intrahepatic porto-systemic shunt as he had portal vein thrombosis. Percutaneous paraumbilical embolization via caput medusa was performed on day 9 of hospitalization. Following the embolization, the hematochezia stopped. However, he defaulted subsequent follow-up. A 48-year-old Indian male with alcoholic liver cirrhosis was admitted after being found unresponsive. He was hypotensive and had hematochezia. Esophagogastroduodenoscopy (EGD) showed small esophageal varices and a clean-based duodenal ulcer. He continued to have hematochezia and anemia Repeat EGD did not reveal any source of recent bleed. Twelve days after admission, his hematochezia ceased. He refused further investigation and was discharged two days later. He presented one week after discharge with hematochezia. EGD showed non Abbreviated Grade 1 esophageal varices and a clean-based duodenal ulcer. Colonoscopy was normal. Abdominal computed tomography (CT) showed liver cirrhosis with mild ascites, paraumbilical varices, and splenomegaly. He had multiple episodes of hematochezia, requiring repeated blood transfusions. Capsule endoscopy identified the bleeding site in the jejunum. Concurrently, CT angiography showed paraumbilical varices The lumen of this loop of small bowel opacified in the delayed phase, which suggested variceal bleeding into the small bowel. Portal vein thrombosis was present. As he had severe coagulopathy and extensive paraumbilical varices, surgery was of high risk. He was not suitable for transjugular intrahepatic porto-systemic shunt as he had portal vein thrombosis. Percutaneous paraumbilical embolization via caput medusa was performed on day 9 of hospitalization. Following the embolization, the hematochezia stopped However, he defaulted subsequent follow-up.