Russell Kelly Paul MD, Presenter: Nothing to Disclose

Rajendra M. Vazirani MD, Abstract Co-Author: Nothing to Disclose

Katherine Ann Grum MD, Abstract Co-Author: Nothing to Disclose

John Craig McDermott MD, Abstract Co-Author: Nothing to Disclose

Myron M. Wojtowycz MD, Abstract Co-Author: Nothing to Disclose

Jaon Fine PhD, Abstract Co-Author: Nothing to Disclose

Mark Alan Kliewer MD, Abstract Co-Author: Nothing to Disclose

et al, Abstract Co-Author: Nothing to Disclose

To evaluate the efficacy of a wide range of Doppler and clinical parameters for the identification of early dysfunction of TIPS.

An analysis was performed on ultrasounds and portograms performed at the UW, Madison from 1993 – 2003. 1139 Doppler examinations were performed on 189 patients with TIPS Doppler examinations were compared with the results of portovenograms. A gradient greater than 12mm Hg established TIPS dysfunction. Sonograms and portograms were contemporaneous within 40 days. Sonographic parameters assessed: days from most recent intervention, velocities within the proximal, middle, and distal third of the TIPS, portal vein velocities, direction in the left portal branch, variation in portal vein, differences in velocities from the most recent baseline, and change in shunt velocities compared to the first ultrasound performed. These values were also analyzed in combination with clinical parameters, including ascites, MELD score, and INR. Logistic regression and multivariate regression analysis were perfomed as well as ROC curves.

The most sensitive measure for TIPS dysfunction is the change in velocities within the hepatic vein end of the TIPS. The sensitivity and a specificity were 0.66 and 0.64. The time from last intervention also represented a strong predictor of TIPS dysfunction with a sensitivity of 0.53 and a specificity of 0.76. The most specific measure of TIPS dysfunction was reversal in the left portal vein: specificity of 083. However, the sensitivity was 0.33. Clinical parameters did not improve the predictive value of the Doppler study. There was no correlation between MELD score and TIPS dysfunction. Multivariate analysis demonstrated no combination of clinical or Doppler factors to be significantly more predictive than the single parameter model.

Of the parameters studied, the most sensitive measure for TIPS dysfunction is change of velocities in the hepatic venous end and the days from last intervention. Portal vein variability was not a good measure of TIPS function. The ultrasound parameters and TIPS dysfunction were independent of MELD SCORE and other clinical parameters.

Doppler ultrasound predicts TIPS dysfunction.

Paul, R, Vazirani, R, Grum, K, McDermott, J, Wojtowycz, M, Fine, J, Kliewer, M, et al, , Sonographic Evaluation of TIPS: Systematic Analysis of Doppler and Clinical Factors in a Large Cohort.  Radiological Society of North America 2006 Scientific Assembly and Annual Meeting, November 26 - December 1, 2006 ,Chicago IL. http://archive.rsna.org/2006/4438477.html