Long-term Safety and Efficacy of Transvenous Extrahepatic Portosystemic Shunts

To evaluate the feasibility and long-term safety and effectiveness of side-firing intravascular ultrasound (IVUS)-guided transvenous extrahepatic portosystemic shunt (TEPS) creation for patients unsuitable for transjugular intrahepatic portosystemic shunt (TIPS) creation.

Materials and Methods:

A single-center, retrospective analysis was performed on 22 patients (18 adult, 4 pediatric) who underwent 23 TEPS creations for variceal bleeding (N=18, 82%), refractory ascites (N=3, 14%), and pre-operative portal decompression (N=1, 5%). Rationale for extrahepatic shunt included portomesenteric venous occlusion (N=20, 91%) and intervening hepatic cysts/malignancy (N=2, 9%). Baseline characteristics included active cancer (N=6, 27%), treated cancer (N=4, 22%), cirrhosis (N=9, 41%), and liver transplant (N=4, 22%). Primary outcomes were technical success and TEPS patency at 1, 3, and 5 years. Secondary outcomes included variceal hemorrhage, recurrent ascites, hepatic encephalopathy, pancreatitis, hepatopulmonary/portopulmonary syndrome, and overall survival (OS). Patency and OS were calculated by Kaplan-Meier method. Univariate survival analysis was performed with log-rank tests.

Results:

Technical success was 100% and included portocaval (N=5, 22%), mesocaval (N=16, 70%; 15 superior, 1 inferior mesenteric), splenocaval (N=1, 4%), and splenorenal (N=1, 4%) shunts. Portosystemic gradient decreased from a median of 14 to 4 mmHg. Median follow up was 33.6 months. TEPS dysfunction included stenosis (N=3, 14%) and occlusion (N=1, 5%). No cases of shunt migration or bowel injury occurred. Shunts traversed pancreatic uncinate process in 2 (9%). Primary patency at 1, 3, and 5 years was 100%, 90%, and 60%. Primary assisted patency was 100%, 100%, and 71%, and secondary patency was 100%, 100%, and 83%. Variceal hemorrhage occurred in 3 patients (14%). Patients with refractory ascites did not need long-term paracentesis. Hepatic encephalopathy occurred in 8 patients (36%) including 5 without cirrhosis, managed medically without shunt reduction. Delayed pancreatitis developed in 1 patient 2 years after TEPS. One patient (5%) developed hepatopulmonary syndrome 2 months after TEPS, requiring oxygen supplementation. One patient (5%) developed portopulmonary syndrome resulting in death 7 years after TEPS. Median OS was 38.4 months (95% CI: 10.4-127.3) with poorer OS in patients with active cancer (8.9 vs 104.8 months, p=0.022).

Conclusion:

IVUS-guided TEPS creation is feasible, effective, and durable in patients not amenable to conventional intrahepatic TIPS creation.