Long-term clinical outcome of patients with cirrhosis and refractory ascites treated with transjugular intrahepatic portosystemic shunt insertion.

Transjugular intrahepatic portosystemic shunt (TIPS) is indicated for the treatment of refractory ascites in cirrhosis. The long-term outcome of TIPS for refractory ascites is unknown. The aim of this study is to describe the natural history of patients with refractory ascites post-TIPS, and compare between polytetrafluoroethylene (PTFE)-covered versus bare stents. A retrospective chart review of patients who had TIPS for refractory ascites was conducted. Prospectively collected data include demographics, angiographic data, blood work, and urinary sodium excretion. There were 136 patients who received TIPS (bare = 104, covered = 32) for over 22 years. Patients with PTFE stents had lower international normalized ratio and model for end-stage liver disease score. More patients with bare stents developed shunt dysfunction (74.0% vs 24.1%, P < 0.0001) and required more TIPS revisions (1.6 ± 0.2/patient vs 0.2 ± 0.1, P < 0.0001). Urinary sodium excretion increased significantly from first month and progressed to 98 ± 9 mmol/day at 12th month post-TIPS (P < 0.001 vs baseline), concurrent with improved renal function. Most patients (77.6%) completely cleared the ascites without diuretics, but many achieved this beyond 2 years. Number of TIPS revision was predictive of complete response at 12 months (odds ratio [OR] 0.7, 95% confidence interval [CI] 0.5-0.9, P < 0.05). Age (hazard ratio [HR] = 1.05 [95% CI 1.02-1.08], P < 0.01), complete response (HR = 0.22 [95% CI 0.12-0.40], P < 0.0001) and polytetrafluoroethylene stents (HR = 0.23 [95% CI 0.05-0.97], P < 0.05) were predictive of survival. TIPS is an effective treatment for cirrhotic refractory ascites. Ascites clearance is dependent on number of TIPS revision, whereas survival is predicted by younger age, complete response, and covered stent use, although era effect likely contributed to improved survival with covered stent use.