| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Report |
1 Department of Radiology, Division of Abdominal Imaging, University of
Pittsburgh Medical Center, 200 Lothrop St., Pittsburgh, PA 15213.
2 Present address: Department of Radiology, University of Palermo, Italy.
Received December 13, 2000;
accepted after revision January 15, 2002.
Address correspondence to M. P. Federle.
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CONCLUSION. CT angiography with volume rendering shows the anomalous feeding artery and hepatic draining veins that are characteristic of focal nodular hyperplasia. These features may be helpful in distinguishing focal nodular hyperplasia from other lesions.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The angioarchitecture and pathophysiology of focal nodular hyperplasia have been studied extensively by pathologists and hepatologists and have been found to differ substantially from the angioarchitecture of hepatocellular carcinoma and other malignant hepatic neoplasms [1]. Focal nodular hyperplasia is supplied by an enlarged anomalous hepatic artery; its venous drainage is always into the hepatic veins [1]. Conversely, the efferent vessels in hepatocellular carcinoma are almost exclusively drained into the portal vein system [3]. Several investigators have concluded that unequivocal visualization of hepatic venous drainage from a hypervascular hepatic mass would be useful in diagnosing focal nodular hyperplasia [3,4,5].
Although multiphasic thin-section helical CT often shows blood vessels within or surrounding hypervascular liver masses, it is often difficult to recognize the origin and path of vessels on a series of transverse CT scans. The purpose of this article is to show the ability of three-dimensional (3D) multidetector CT angiography using volume rendering to display the hemodynamics and angioarchitecture of focal nodular hyperplasia, features that should prove useful in distinguishing the lesions from malignant masses.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
For the arterial phase images, we used a collimator width of 1.25 mm and a table speed of 7.5 mm per rotation (high-speed mode; pitch, 6). For the portal venous phase, we used a collimator width of 2.5 mm and a table speed of 15.0 mm per rotation.
Three-dimensional volume-rendered reconstructions were performed using a free-standing workstation (Advantage Windows; General Electric Medical Systems). All reconstructions were performed by a radiologist experienced in 3D postprocessing techniques, and each procedure required approximately 20 min. Volumes of interest were selected manually from the axial source images to include only the aorta, celiac axis, hepatic artery, draining veins, and lesion. CT angiograms were then reconstructed using the volume-rendering algorithm with lower thresholds of 70-140 H. Display parameters including width, level, opacity, and brightness were chosen subjectively. The volume-rendered images were obtained in projections selected to best depict the course of the hepatic vasculature. Standard projections performed in all cases included right anterior, oblique, inferior, posterior, and lateral projections.
The reconstructed axial images and the 3D volume-rendered reconstructions were reviewed by two radiologists, and the arterial supply and venous drainage to and from the lesion were recorded as part of the official CT report.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
|
|
|
|
|
|
In the arterial phase, one or more anomalous feeding arteries to each lesion were identified (Figs. 1A,1B,1C and 2A,2B,2C,2D). The arteries branched into a spiderlike series of vessels on the surface of the focal nodular hyperplasia (Fig. 2A,2B,2C,2D); the course of the vessels within the focal nodular hyperplasia could not be visualized on the volume-rendered images. Anomalous draining veins were identified that could be followed into or toward larger hepatic veins and not into portal vein branches (Fig. 2A,2B,2C,2D). In the patient with the smaller lesions, an anomalous draining vein could not be identified (Fig. 3).
|
For comparison, we illustrate the portal venous drainage of hepatocellular carcinoma (Fig. 4A,4B).
|
|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The relatively minor alterations of our standard biphasic CT liver protocol and the moderate amount of time necessary to perform the 3D reconstructions provided unique insights into the angioarchitecture of focal nodular hyperplasia in these four patients. Prior attempts to study the hemodynamics of focal nodular hyperplasia have been quite invasive and labor-intensive. Transcatheter angiography was more widely used in the past to detect and characterize liver tumors, but few cases of focal nodular hyperplasia were studied and were generally characterized merely as hypervascular masses supplied by an enlarged anomalous artery [7]. Miyayama et al. [8] coupled hepatic arteriography with single-level dynamic CT and showed arteries and veins in or near the central stellate scar and at the junction of the focal nodular hyperplasia and normal liver tissue. They noted that blood (contrast media) occasionally drained from the focal nodular hyperplasia nodule into surrounding hepatic sinusoids producing an enhanced capsulelike rim. Their CT technique did not allow depiction of the shape or number of focal nodular hyperplasia lesions or the pattern of feeding or draining vessels beyond the focal nodular hyperplasia.
Fukukura et al. [3] studied 29 surgically resected focal nodular hyperplasia lesions and three autopsy specimens with angiography; in the autopsy cases, they studied the injection of colored gelatin into the hepatic artery and portal vein. They concluded that focal nodular hyperplasia is supplied by an anomalous enlarged hepatic artery and is neither supplied nor drained by the portal vein. Rather, venous drainage of focal nodular hyperplasia was exclusively into the hepatic vein branches [3]. Conversely, several investigators have studied the vascular supply and drainage of hepatocellular carcinoma in great detail and have found that hepatocellular carcinoma is supplied almost exclusively by the hepatic artery and is drained almost always into the portal vein [4, 9, 10]. Okuda et al. [11] showed some drainage into hepatic veins in only 1.8% of hepatocellular carcinoma.
Wanless et al. [1], who have written most extensively on the pathophysiology of hypervascular liver masses, have conducted detailed studies on the morphometry of these lesions. They report that focal nodular hyperplasia is supplied by an anomalous artery, larger than expected for its hepatic location, that branches into a spiderlike structure on or within the focal nodular hyperplasia lesion. They further report that the artery is not accompanied by a portal vein branch and that focal nodular hyperplasia drains directly to hepatic veins or first through enlarged sinusoids near its periphery [1].
These and other investigators [3] have concluded that the visualization of these vascular features should prove useful in distinguishing focal nodular hyperplasia from other hepatic masses including hepatocellular carcinoma. We believe the CT angiographic techniques that we have reported accomplish this goal and do so noninvasively and relatively inexpensively.
The CT angiograms provided by volume-rendered 3D modeling showed the size, number, and angioarchitecture of focal nodular hyperplasia in each of our four patients. The images showed the characteristic enlarged anomalous feeding arteries, the spiderlike branching on the surface of the focal nodular hyperplasia, and the hepatic venous drainage (Figs. 1A,1B,1C and 2A,2B,2C,2D). The portal vein supplied neither inflow nor drainage of the focal nodular hyperplasia in our cases.
Another potential benefit of the 3D display of hepatic vasculature provided by CT angiography is in planning for embolization or ligation of vessels for the rare symptomatic cases of focal nodular hyperplasia [12].
Our study has some limitations. The detection of draining veins was not possible for the smallest of the lesions shown. From our knowledge, based on our prior surgical and gross pathologic studies, the depiction of hepatic venous drainage (for focal nodular hyperplasia) or portal venous drainage (for hepatocellular carcinoma) should be a reliable criterion. Nevertheless, our study size is small, and a study of a larger group of patients may be warranted. We conclude that 3D CT angiography can provide useful information in cases of suspected focal nodular hyperplasia that may allow a more confident diagnosis and treatment in some cases.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  I. R. Kamel, E. Liapi, and E. K. Fishman Focal nodular hyperplasia: lesion evaluation using 16-MDCT and 3D CT angiography. Am. J. Roentgenol., June 1, 2006; 186(6): 1587 - 1596. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. R. Kamel, L. P. Lawler, and E. K. Fishman Comprehensive Analysis of Hypervascular Liver Lesions Using 16-MDCT and Advanced Image Processing Am. J. Roentgenol., August 1, 2004; 183(2): 443 - 452. [Full Text] [PDF]
|
|
|
|
 |
|
 M. J. Kim, H. K. Lim, S. H. Kim, D. Choi, W. J. Lee, S. J. Lee, and J. H. Lim Evaluation of Hepatic Focal Nodular Hyperplasia With Contrast-Enhanced Gray Scale Harmonic Sonography: Initial Experience J. Ultrasound Med., February 1, 2004; 23(2): 297 - 305. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
