The hemodynamics of transjugular intrahepatic portosystemic shunts: Investigations with Doppler sonography and development of an in vitro model

Rationale and Objectives. We evaluated Doppler sonography-based measurements of transjugular intrahepatic portosystemic shunt (TIPS) function and developed an in vitro model of normal TIPS hemodynamics. Methods. We reviewed retrospectively the results of all trans-TIPS manometries (N = 116) performed during a 24-month period. Portosystemic pressure gradient was compared with peak stent velocity as measured by angle-corrected Doppler sonography. A flow phantom simulating TIPS was created using 8-, 10-, and 12-mm-diameter wire-mesh stents placed in cylindrical channels with lengths ranging from 3.4 to 6.0 cm. Results. Among 50 trans-TIPS manometries with corresponding Doppler sonography performed on well-functioning shunts, measured portosystemic pressure gradient and peak velocity were not correlated (R(2) = .014). On the basis of a regression of measurements in the flow phantom, pressure loss in a stented cylindrical channel was estimated as follows: Delta p = rho . (0.145 - 0.001 R(ey) + 0.816 . L/D) . (V-mean(2)/2, where rho is the fluid density, R(ey) is the Reynolds number, L is the channel length, D is the stent diameter, and V-mean is the time-averaged velocity within the stent. Predicted and measured pressure gradients were correlated (R(2) = .91). Conclusion. Peak velocity in patients with a normally functioning TIPS does not predict the magnitude of the portosystemic pressure gradient.