Dacron-covered stent-grafts in transjugular intrahepatic portosystemic shunts: initial experience.

Purpose: To retrospectively review the authors' experience with use of a Dacron-covered stent-graft in transjugular intrahepatic postosystemic shunts (TIPS).

Materials And Methods: The need for internal review board approval was waived. Informed consent was obtained from all patients. The study was compliant with the Health Insurance Portability and Accountability Act. A retrospective analysis was performed of 16 patients who received a Dacron-covered stent-graft during revision or de novo creation of TIPS. There were 13 men and three women aged 44-80 years (mean age, 61 years). Primary unassisted and assisted patency rates and secondary patency rates were estimated. The primary unassisted patency of patients who underwent de novo placement of stent-grafts (n = 10) was compared with that of patients with stent-grafts placed during shunt revision (n = 6); in all patients, stent-grafts were placed within stents. Primary unassisted patency was also compared between patients in whom the covered stent was confined to the parenchymal tract (n = 7) and those in whom the stent extended 1 cm or more into the portal vein (n = 9). Patency was estimated with the Kaplan-Meier method, and group comparisons were performed with the log-rank test.

Results: Primary unassisted patency rates following stent-graft placement at 4, 12, and 24 months (+/- standard error) were 64% +/- 14, 54% +/- 15, and 40% +/- 16, respectively. The rates for primary assisted patency were 78% +/- 12, 67% +/- 14, and 67% +/- 14 and those for secondary patency were 91% +/- 9, 81% +/- 12, and 54% +/- 23. At 12 months, primary unassisted patency with de novo stent-graft placement was 90% +/- 9, whereas that with stent-grafts placed during TIPS revision was 17% +/- 15 (P = .005). At 12 months, the primary unassisted patency in patients with stent-grafts confined to the parenchymal tract was 75% +/- 22, and that of patients with stent-grafts extending at least 1 cm into the portal vein was 40% +/- 17 (P = .21).

Conclusion: In this small series, satisfactory long-term patency was observed among patients in whom Dacron-covered stent-grafts were placed during revision or de novo creation of TIPS. More favorable outcomes were observed when the stent-graft was placed during de novo TIPS creation and when the device was confined to the parenchymal tract.