| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Commentary |
1 Both authors: Department of Diagnostic Radiology, The University of Texas M. D. Anderson Cancer Center, 1515 Holcombe Blvd., Box 57, Houston, TX 77030.
Received June 30, 2005; accepted after revision July 12, 2005.
Each month the American Journal of Roentgenology will republish
online one of the 100 most-cited articles from its first century. A
corresponding commentary in the print journal by a contemporary radiologist
will provide a current perspective. For a full list of these articles, see
page 3 of the January
2006 issue of the AJR or go to
www.ajronline.org.
Keywords: contrast material • CT • liver • oncologic imaging
In Les Miserables, Victor Hugo wrote "the general life of the human race is called Progress, and so is its collective march." In medical imaging, particularly CT, the march of human progress has become an all-out sprint. Nothing makes this as clear as a comparison of the CT technology of yesterday with the MDCT technology of today.
The introduction of CT revolutionized abdominal imaging. Operator-independent cross-sectional images could be obtained, markedly reducing the need for diagnostic surgical procedures. In 1977, Stephens et al. [1] published a landmark article, "Computed Tomography of the Liver," based on the evaluation, over a 10-month period, of 1,120 patients, 109 (10%) of whom were being evaluated for suspicion of disease involving the liver. Their findings, and their discussion of the implication of those findings, foreshadowed much that was to be developed in abdominal imaging as the technology of CT rapidly evolved. In this commentary, we contrast the technology, imaging findings, and implications for the diagnosis of disease, then and today.
Technique
Stephens et al. [1] used a prototype EMI scanner that could acquire, in a 20-sec breath-hold, a single 13-mm-thick cross-sectional image with a matrix of 320 x 320. Images were typically obtained without IV contrast material. However, if no lesion was seen on the unenhanced images, patients were then given, by infusion, 300 mL of 30% diatrizoate methylglucamine. Artifacts attributed to organ motion (heart, air in the stomach) and streak artifacts from materials of high or low density were problematic (Figs. 1A, 1B, and 2). The most common problem for Stephens et al. was "excessive noise."
|
|
|
The use of IV contrast material has also markedly changed. Today, IV contrast material is routinely used unless it is specifically contraindicated. A power injector is routinely used in place of a drip infusion approach that even Stephens et al. [1] noted could mask lesions. Combined with the rapid acquisition of images that is possible today, multiphasic imaging that can be reconstructed in any conceivable plane is a reality.
Diagnosis
Many of the routine findings we note currently on day-to-day CT examinations were new at the time of the article by Stephens et al. [1]. They were among the first to note the importance of using window and level settings to improve lesion conspicuity. However, the specificity of their findings was limited by their technology. Basically, nearly all lesions, cystic or solid, of whatever type, were, as now, "of diminished radiographic density [on unenhanced CT]." One of the limitations of their infusion technique was that neoplasms seen on the unenhanced examinations became isodense and essentially disappeared once contrast material was administered (Figs. 3A and 3B).
|
|
Today, we can detect small metastases to the liver and characterize them on the basis of their enhancement pattern on multiphasic MDCT imaging. Multiphasic imaging is particularly helpful with hypervascular tumors. Scanning during the hepatic artery phase is now routinely used for the detection of hypervascular hepatic metastases such as those from renal cell carcinoma, islet cell tumor, carcinoid tumor, and thyroid cancer.
This stark contrast between then and now is also apparent in the imaging of
primary liver lesions. Stephens et al.
[1] noted that in the case of a
patient "with a huge hepatoma, massive liver enlargement only was
noted." Today, we commonly use multiphasic CT in the cirrhotic patient
to detect small hepatocellular carcinomas that are typically best seen on the
arterial phase of imaging as hypervascular lesions. Research has also
suggested that the rate of washout seen on CT correlates with an elevated
-fetoprotein level
[2].
Stephens et al. [1] were also among the first to note that the information provided by a single CT examination Indeed, today we often use the pattern of organ involvement to direct us to the correct diagnosis, and we take for granted the finding of incidental lesions such as adenomas, renal cysts, or even small renal cell carcinomas.
...is not limited to a single organ. Previously undiagnosed primary malignant lesions in the pancreas...and kidney...have been found in examinations that were done primarily to investigate the liver. Likewise, clinically unsuspected hepatic metastases have been discovered in CT investigations of other suspected conditions.
Benign Lesions
The characterization of lesions as hemangiomas is a common imaging task.
The article by Stephens et al.
[1] could only note "the
highly vascular part of a cavernous hemangioma became denser than the normal
parenchyma on a scan made immediately after infusion of a large dose."
Today, routine scanning generally allows the diagnosis to be made because of
visualization of peripheral contrast puddling. With MDCT multiphasic imaging,
this characteristic can often be used to diagnose even small hemangiomas as
part of a routine evaluation of the entire liver.
In 1977, Stephens et al. [1] noted that they could not differentiate small cysts from portal tracts. Today, small cysts can be confidently identified by directly measuring their density, and small tumors can often be detected by virtue of peripheral enhancement. Focal nodular hyperplasia, typically seen on today's studies as a focus of rapid early arterial enhancement that becomes isodense on later images, was noted in the text of the Stephens article as a nonspecific "lesion of slightly diminished density."
Diagnostic Algorithms
The Stephens article also represents the early development of diagnostic
algorithms using CT. Those authors dedicated a section of their article to the
evaluation of dilated intrahepatic bile ducts, noting segmental intrahepatic
biliary dilatation with Caroli's disease, nodal obstruction at the liver hilum
as a result of metastatic lung cancer, and extrahepatic obstruction resulting
from periampullary tumors and pancreatic head tumors. Today, we can identify
on CT not only the level of biliary obstruction, but also the cause—for
example, intrahepatic cholangiocarcinoma or pancreatic carcinoma.
New Applications
The evolution of MDCT has resulted in the development of applications that
could not even be addressed in the Stephens article. The rapid advancement of
computer technology has resulted in new applications for CT data through the
use of advanced computer postprocessing of images. Multiplanar and 3D
reconstructions of thin-section multiphase MDCT images have been used before
liver transplantation to depict the vascular anatomy of the liver and to
identify important vascular variants and vascular abnormalities such as celiac
artery stenosis. CT cholangiopancreatography, using minimum intensity
projections or volume rendering, can create detailed images of the biliary
tree that are of similar quality to those of ERCP. CT perfusion imaging, using
cine techniques of imaging a bolus of IV contrast material, has been used to
quantitatively determine mean transit time, blood flow, and blood volume,
variables that may correlate with the severity of liver disease and with the
response of neoplasms to new antiangiogenesis therapy.
Conclusion
Much has occurred in the 29 years since the original article on CT of the liver by Stephens et al. [1]. The rapid evolution of CT has resulted in marked improvements in the ability to characterize diseases that involve the liver. Still-evolving new techniques using computer postprocessing algorithms and ultrafast dynamic imaging of the liver promise new perspectives from which disease will be analyzed. Competition from the other evolving technologies of MRI, PET/CT, and sonography make it difficult to predict how imaging of the liver will evolve, but the next 29 years promise to be equally exciting.
References
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Related articles in AJR:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
