RESUMEN
BACKGROUND AND OBJECTIVE: The canine model of esophageal varices with an agar constrictor has been used for studies of the endoscopic treatment of esophageal variceal bleeding, but it has limitations in both stability and successful rate. This study was designed to enhance the model's efficiency and success rate by using a novel approach with a balloon dilatation constrictor. METHODS: We used 22 adult mongrel dogs to establish the model by progressively compressing and constricting the portal vein through a rechargeable balloon dilatation constrictor in combination with side-to-side portocaval shunt and inferior vena cava (IVC) ligation to increase portal vein pressure (PVP). The rechargeable balloon dilatation constrictor was composed of a hyaline polypropylene (PP) ring, a silica gel tube with a balloon, and an injection pedestal (silica gel) in an implantable vascular access port. The effects were evaluated with pre- and post-shunt PVP measurement, weekly gastroscopy, and portocaval venography. RESULTS: The mean PVP increased significantly from a preoperative (before side-to-side portocaval shunt and IVC ligation) 12.86 +/- 0.18 cmH(2)O to 26.75 +/- 0.39 cmH(2)O after the model had been established (P < 0.05). After the model was established, gastroscopy performed to assess esophageal varix size demonstrated four varicose veins of grade I, six of grade II, eight of grade III and four of grade IV. The portocaval angiography showed that the IVC and portal vein were completely blocked, that the anastomosis stoma was unobstructed and that the blood flow through esophageal varices via splenetic and gastric veins was blocked. CONCLUSION: The novel canine model developed with a rechargeable balloon dilatation constrictor is feasible and reliable for modeling esophageal varices.