Evaluation of portosystemic collaterals by SPECT imaging after endoscopic variceal sclerotherapy: usefulness for predicting recurrence.

Bleeding from esophageal varices is a major cause of morbidity and mortality in cirrhotic patients. Identification of patients at high risk for bleeding is particularly important. The aim of this study was to determine whether detection of portosystemic collaterals by SPECT could predict the outcome of endoscopic injection sclerotherapy of esophageal varices and be useful for selecting appropriate therapy. METHODS: Sixty-two patients with liver cirrhosis who were considered at high risk of bleeding were treated with endoscopic injection sclerotherapy. Endoscopy was performed every 3 mo after therapy or until bleeding occurred. Before and within 2 wk after therapy, tomographic images of intra-abdominal blood pool were constructed by SPECT. RESULTS: Before therapy, the following portosystemic collateral routes were observed: coronary veins in 53 (85.5%) of 62 patients, short gastric veins in 8 patients (12.9%), splenorenal shunts in 10 patients (16.1%), and paraumbilical veins in 6 patients (9.7%). Patients positive for imaging of coronary veins were divided into 3 groups on the basis of changes in images after therapy: complete responders (n = 17), whose coronary vein images disappeared completely; partial responders (n = 18), whose images became smaller; and nonresponders (n = 18), whose images did not change significantly before or after therapy. The rates of recurrence after endoscopic injection sclerotherapy until 6 mo in complete responders (4/17, 23.5%) and partial responders (7/18, 38.9%) were significantly less (P < 0.05) than that in nonresponders (11/13, 84.6%). The rate of recurrence of esophageal varices until 6 mo in nonresponders treated with additional submucosal injection sclerotherapy (1/5, 20.0%) was significantly less (P < 0.05) than that in nonresponders without additional submucosal injection sclerotherapy (11/13, 84.6%). CONCLUSION: Abdominal blood-pool SPECT, a noninvasive method, is useful for evaluating the therapeutic effectiveness of endoscopic sclerotherapy, for predicting the recurrence of varices, and for selecting appropriate management after sclerotherapy.