| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Letters |
Gunma University Hospital
Maebashi, Gunma, Japan
We read with interest the article by Dr. Weidekamm and colleagues [1] in the February 2005 issue of the AJR, in which they illustrated their innovative application of perfusion CT to evaluate changes in liver perfusion after transjugular intrahepatic portosystemic shunt (TIPS) placement. Their data indicated, first, that in patients with cirrhosis, hepatic arterial perfusion increased, whereas portal venous and total perfusion decreased compared with that of healthy volunteers; and, second, that TIPS placement caused a statistically significant increase of the hepatic arterial and total hepatic perfusion, but the portal venous perfusion remained unchanged.
In their study, Weidekamm and colleagues [1] calculated hepatic parenchymal portal perfusion using the maximum-slope method [2, 3]. This technique is simple, thus attractive: Tissue perfusion is calculated by dividing the peak upward slope of the tissue time-attenuation curve by the peak attenuation increase of the input function. However, this algorithm depends on several assumptions [4, 5], one of which is that the time-attenuation curve of an appropriate input function should be obtained. In their study, circular regions of interest (ROIs) were drawn in the portal vein, and the data from these ROIs were used as the input function when the hepatic parenchymal portal perfusion was calculated.
Hepatofugal flow in the portal venous system is relatively common in patients with liver disease, and its prevalence in studies of patients with cirrhosis evaluated with Doppler sonography was 3-23% [6]. In such a case, the portal vein is not an appropriate input function when the hepatic parenchymal portal perfusion was calculated by this algorithm because the hepatic parenchyma may not be perfused by this portal venous flow. When there is a transhepatic shunt, either spontaneous (paraumbilical vein) or iatrogenic (TIPS), such shunts are often associated with arterioportal shunting and hepatofugal flow in intrahepatic portal veins despite brisk hepatopetal flow in the main portal vein; thus, the flow dynamics are more complex [6].
We suggest that the flow direction of the portal venous system should be checked when portal perfusion CT is applied in patients with cirrhosis and that the maximum-slope method cannot be applied when reversal of blood flow in the portal venous system is detected.
References
Medical University of Vienna
Vienna, Austria
We appreciate the interest of Dr. Tsushima and Dr. Endo in our recently published study [1]. We agree with their comment about the importance of the evaluation of the flow direction of the portal venous system before performing hepatic dynamic CT.
In our study, we analyzed the flow direction in the portal vein with Doppler sonography before and after transjugular intrahepatic portosystemic shunt (TIPS) placement and verified that all patients had hepatopetal flow. Unfortunately, this information was missing from the Subjects and Methods section. However, the blood flow in the portal venous system does not affect the arterial liver perfusion. The most important finding of our study is the increase in arterial liver perfusion in cirrhotic patients after TIPS placement.
However, we agree with Tsushima and Endo that liver cirrhosis causes regional hepatic and general hemodynamic changes, partly due to intrahepatic shunts between branches of the hepatic artery, the portal vein, and the hepatic veins [1, 2]. The arterialization of liver perfusion was indicated by significantly lower hepatic perfusion during the portal venous phase in the cirrhotic group than in the control group (72 vs 102 mL blood/min/100 mL tissue, respectively) [1].
Finally, we agree with Tsushima and Endo that in patients with hepatofugal flow in the portal vein, hepatic dynamic CT is not advisable for the evaluation of portal hepatic perfusion. Because there is no portal venous inflow into the liver, the hepatic portal venous perfusion is zero.
References
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
