抄録
Background: Esophageal varices are classified into two types according to their venous anatomy at the gastroesophageal junction: the palisading type (major) and the pipeline stem type (minor). Many cases of the pipeline stem type (pipeline varices) have been reported to be resistant to or associated with a high risk during endoscopic therapy. This study aimed to clarify the prevalence and clinicopathological characteristics of pipeline varices.
Materials and methods: Among 1,141 patients who underwent endoscopic treatment for esophageal varices, 210 patients for whom detailed variceal anatomy was available from endoscopic varicealography during injection sclerotherapy or from percutaneous transhepatic portography were retrospectively analyzed. Endoscopic and hemodynamic findings were assessed. Pipeline varices were further classified into typical and subtype variants, and clinical characteristics were compared between these groups
Results: Of the 1,141 patients, 64 (5.6%) had pipeline varices; 49 (76.6%) were classified as the typical type and 15 (23.4%) as the subtype. No significant clinical differences were observed between the two types. Compared with patients with ordinary esophageal varices, those with pipeline varices showed a significantly higher prevalence of portopulmonary venous anastomosis (14.1% vs. 4.8%, p = 0.0197). In both groups, the primary site of esophageal varices was the right esophageal wall, and the left gastric vein was the main feeder in the majority of cases.
Conclusions: Pipeline varices are frequently associated with portopulmonary venous anastomoses. In the absence of a preprocedural diagnosis of such anastomoses, endoscopic treatment of pipeline varices may carry a substantial risk, including severe post-procedural bleeding or fatal systemic complications caused by the inadvertent migration of injected sclerosant into the arterial circulation.
