[A case of non-traumatic hemobilia due to warfarin therapy].

Hemobilia is a hemorrhage into the biliary tract that may follow surgical trauma, liver biopsy, aneurysms, extra- or intra-hepatic tumors of the biliary tract, gallstones, and inflammatory lesion of liver, especially helminthic or pyogenic. Sometimes, it is associated with primary liver cancer. An 84 year-old woman was admitted because of continuous right upper quadrant pain 4 days before admission. Physical examination revealed decreased skin turgor, icteric sclerae and severe tenderness on right upper quadrant abdomen. She had no hepatosplenomegaly, and no rebound tenderness. She has been taking warfarin for 3 weeks before admission because of atrial fibrillation. On admission, serum bilirubin and transaminase were elevated. The level of hemoglobin and hematocrit were 11.3 g/dL and 37.4%, respectively. HBsAg was negative, but IgG anti-HBc and anti-HBs were positive and anti-HCV was negative. Parasite skin test and stool ova count demonstrated non-specific findings. Stool occult blood was strongly positive, and prothrombin time was markedly prolonged. According to endoscopic retrograde cholangiopancreatography, common bile duct was dilated, and filled with blood clot but there was no stone in bile tree. After two weeks, serum transaminase, bilirubin, hemoglobin, hematocrit, and CA19-9 were normalized. We report a case of hemobilia, occurring in a patient with continuous warfarin use.

[Relationship between esophageal lesions observed by endoscopy and nasogastric intubation: a study of 185 cases of percutaneous endoscopic gastrostomy].

BACKGROUND/AIMS: Percutaneous endoscopic gastrostomy (PEG) is a method widely used for long-term enteral nutrition in dysphagia. Mostly, it is preceded by nasogastric intubation (NI) for short-term enteral nutrition; endoscopic findings associated with NI are encountered during PEG. The purpose of this study was to discuss such findings and to delineate a relationship between these findings, especially esophageal lesions and the duration of NI. METHODS: This study involved 185 individuals who had undergone PEG at Kyung Hee Medical Center from January 1999 to May 2002. The medical records were examined retrospectively. RESULTS: The dysfunction of the CNS comprised 98.4% of the causes of dysphagia. The duration of NI was 15.2 weeks on average, with median value of 8.7 weeks, indicating that PEG was performed relatively soon. Endoscopic findings revealed esophagitis in 63 cases, esophageal ulcers in 27 and active bleedings in another 10. The incidence of esophageal lesions was shown to be higher in subjects with duration of NI under 12 weeks than in those with duration over 12 weeks (p=0.032). CONCLUSIONS: PEG was carried out in many cases during the early stages of dysphagia, and NI-associated esophageal lesions appeared to be more prevalent within 12 weeks of NI duration. These results may be of help in deciding the timing of PEG.