Utility of Post-Procedural Transjugular Intrahepatic Portosystemic Shunt Ultrasound in the Covered Stent Era

Introduction Current ultrasound surveillance guidelines for detection of stenosis following transjugular intrahepatic portosystemic shunt (TIPS) were based on experience with procedures performed with bare metal stents—devices which have been shown to have a higher frequency of post procedural stenosis when compared with polytetrafluoroethylene (PTFE) covered stents.^1–3 In this investigation, we aim to study the utility of surveillance ultrasound following TIPS performed with covered stents.

Materials and Methods This is an institutional review board (IRB)-approved, retrospective review of patients receiving TIPS between May 2012 and December 2016. Patients were selected for this study if they received a covered stent at the time of their TIPS procedure, had at least one ultrasound follow-up, and had follow-up in our institution (to provide consistent reporting). Study endpoints included markers of stenosis based on expert consensus (1): TIPS patency, portal vein velocity status (abnormal ≤30cm/s), portal vein flow direction, intra-stent velocity gradient >100 cm/s, and overall velocity status (abnormal = velocity <90 or ≥190 cm/s). Data were summarized using survivor functions estimated with Turnbull’s method. All analyses were performed in R version 3.3.2.⁴

Results A total of 403 ultrasounds were reviewed from TIPS procedures performed between May 2012 and December 2016 from 151 patients (mean age: 56 years, 82 male) that met criteria for this study. The median number of ultrasound follow-ups for a patient after the initial TIPS placement was 2 (range: 1–7). A patient’s last ultrasound follow-up was 284 days after the initial TIPS placement on average (range: 3–1,786 days). A total of 42 (28%) patients had at least one TIPS revision. Of these revisions, a TIPS malfunction was suggested on ultrasound for 37 (76%). The results from the ultrasound follow-ups among the 151 patients in this sample were summarized at various time points after the initial TIPS placement. Over 90% of ultrasound follow-ups indicated patent-TIPS, pulmonary vein (PV) velocity <30 cm/s, and normal flow direction. Over 85% of ultrasound follow-ups indicated a lack of stenosis. A relatively large number of ultrasound follow-ups, however, indicated at least one velocity <90 or ≥190 cm/s.

Conclusion A majority of patients receiving covered stents during TIPS placement did not have an abnormality detected by ultrasound surveillance until beyond 6 months post procedure. Our study demonstrates that in addition to identifying better markers for TIPS malfunction on surveillance ultrasound, there may not be a role for ultrasound surveillance until >6 months post TIPS procedure. A future prospective study using a less frequent ultrasound follow-up interval following TIPS would provide better data on the healthcare cost savings associated with decreasing ultrasound surveillance.