Purpose: The aim of this study was to evaluate the feasibility of cone-beam computed tomography (CBCT)-based real-time 3-D guidance of TIPSS placement and its positioning compared to standard guiding methods.
Materials and methods: In a prospective, randomized, consecutive study design from 2015 to 2017, we included 21 patients in the CBCT guided group and 15 patients in the ultrasound (US) guided group. The prospective groups were compared in terms of success rate of intervention, portal vein puncture/procedure time, number of puncture attempts and applied dose. We furthermore retrospectively analyzed the last 23 consecutive cases with fluoroscopic guided portal vein puncture in terms of success rate, procedure time and applied dose, as it has been the standard method before US guidance.
Results: The median number of puncture attempts (CBCT: n = 2, US: n = 4, p = 0.249) and the mean puncture time (CBCT: 32 ± 45 min, US: 36 ± 45 min, p = 0.515) were not significantly different. There were furthermore no significant differences in the mean time needed for the total TIPSS procedure (CBCT: 115 ± 52 min, US: 112 ± 41 min, fluoroscopy: 110 ± 33 min, p = 0.996). The mean applied dose of the complete procedure also showed no statistically significant differences (CBCT: 563 ± 289 Gy·cm2, US: 322 ± 186 Gy·cm2, fluoroscopy: 469 ± 352 Gy·cm2, p = 0.069). There were no image guidance related complications.
Conclusion: Real-time 3-D needle guidance based on CBCT is feasible for TIPSS placement. In terms of puncture attempts, duration and dose, CBCT guidance was not inferior to the control groups and may be a valuable support for interventionists in TIPSS procedures.
Keywords: Cone-beam CT (CBCT); Portal vein mapping; Real-time needle guidance; Transjugular intrahepatic portosystemic stent shunt (TIPSS).