Imaging of the gastrointestinal tract-encompassing not only the alimentary tract itself but the liver and pancreas-has changed dramatically in the last 10 years.? A decade ago, barium enema studies were the standard method of evaluating the bowel, and many people argued that the barium or double-contrast enema would become the standard method of screening for colon cancer. Also, contrast angiography was a common tool for surgical planning. If one wanted to examine the biliary tree, endoscopic retrograde cholangiopancreatography (ERCP) was the method of choice. The use of CT was just beginning, and one of the most common applications was arterial portography, in which a catheter was placed, and the patient was transported from the angiography suite to the CT scanner.

As explained by Pablo R. Ros, MD, MPH, vice chair of radiology at Brigham and Women’s Hospital and professor of radiology at Harvard Medical School, Boston, CT is now the focus of gastrointestinal radiology. The speed and spatial resolution of the newest scanners permit acquisition of images of 1-mm resolution in the early, mid, and late arterial phases after contrast injection. This modality permits a comprehensive review of tumors for both diagnosis and staging, as well as evaluation for obstruction, inflammation, infection, and trauma. The department with a multidetector array scanner can do vascular imaging and three-dimensional surgical planning and avoid angiography except for intravascular interventional procedures. Also gaining popularity is MRI, particularly as specialty coils (eg, for the rectum and esophagus) are introduced by scanner manufacturers. Use of ultrasonography also is expanding as greater value is found for Doppler and harmonic imaging methods.? Reimbursement is now available for positron emission tomography (PET) as employed in the evaluation of esophageal and colon cancer.

As a result of all these changes, barium studies are now rare at many medical centers,, as are diagnostic ERCP and diagnostic angiography. As the indications and capabilities of imaging broaden, the term “abdominal imaging” has come to replace gastrointestinal radiology. Image-guided therapy also is increasing. This article reviews some of the most common reasons for imaging the gastrointestinal tract and the methods now in use.

Pancreatic and Liver Carcinoma

CT is the standard method for the diagnosis of pancreatic cancer, which kills approximately 27,000 persons each year in the United States. Despite the accuracy of the imaging, most of these lesions are unresectable at diagnosis, and the 5-year survival rate is no better than 4%.

Both CT and MRI have their advocates as a method of diagnosing hepatocellular carcinoma (HCC). One indication for the study is potential liver transplantation, as this cancer is a contraindication to such surgery. Mark S. Peterson, MD, and colleagues of the University of Pittsburgh used triphasic contrast-enhanced CT in 430 patients referred for transplantation in whom there was no suspicion of liver cancer.1 Among these patients, 59 (14%) were found to have HCC by histologic examination of the excised liver, including 27% of those whose cirrhosis was a consequence of hepatitis B infection and 22% of those with hepatitis C. The CT scan identified 59% of these patients and 37% of the tumors. Janice Ward, MSc, and associates of St James University in London demonstrated that double-contrast MRI (gadolinium and supraparamagnetic iron oxide [SPIO]) had an 86% accuracy in identifying HCC in transplant candidates.2

Staging of Cancer

Metastases to the liver are common, especially in patients with colorectal or lung cancer. Sometimes, these lesions can be resected or ablated with chemoembolization or by other methods; in other patients, they are contraindications to treatment of the primary tumor. Thus, the radiologist may be called on to answer two questions: are hepatic metastases present? And if so, are they resectable?

More than 155,000 new cases of colon cancer are diagnosed each year in the United States. Hepatic metastases are present in as many as 15% of patients at diagnosis, and 40% of patients will eventually have hepatic-only disease. The inaccuracy of previous methods of evaluating the liver is proven by the fact that as many as 60% of patients undergoing surgery had hepatic recurrences, indicating that tumor was left behind.

Spiral CT is the standard method for hepatic evaluation in colorectal cancer, and a recent large series reported from Barcelona demonstrates its accuracy.3 Among 157 patients with metastatic colorectal cancer, spiral CT with 5-mm collimation identified 247 of the 290 lesions that were found by intraoperative ultrasonography and histopathologic examination. The size of the lesions missed by CT ranged from 0.3 to 1.5 cm with a mean of 0.7 cm. In this series, CT indicated that the lesions were resectable in 113 patients, and resection actually proved possible in 112 of them. Only 19% of the patients had hepatic recurrences during follow-up. The survival rate was 89% at 12 months and 58% at 48 months.

The popularity of MRI for cancer staging is growing. Some of this growth is attributable to the introduction of phased-array and multi-array torso coils that permit acquisition of thinner sections of higher quality in less time. Also contributing are new contrast agents that improve lesion detection. The SPIO agents are taken up by the reticuloendothelial system (RES) and thus are not accumulated by metastatic lesions. A drawback of these agents is the long infusion time (30 minutes) and therefore the need to schedule two sessions on the scanner to acquire the necessary unenhanced and enhanced images. However, high-field scanners are not needed, nor is patient breath-holding during image acquisition, and the imaging window is long. The other new class is the hepatocyte-specific agents, only one of which, mangafodipir trisodium (Mn DPDP) has been licensed for use in the United States. Like the RES agents, hepatocyte-specific agents are not taken up by metastases. Beyond their use in staging of cancer, mangafodipir and related agents are useful to evaluate hepatic function and the biliary tree.

As Semelka and Helmberger have noted,4 the more types of data a radiologist acquires, the less likely he or she is to overlook disease, and “MR imaging simply can acquire more and different types of data than do [ultrasonography] or CT.” The few comparative trials that have been reported show MRI to be more accurate than CT in detecting and characterizing hepatic lesions. Also of importance is the greater ability of MR to detect extrahepatic disease. In one series of patients with various cancers, spiral CT detected 65%-and fat-suppressed spoiled GRE MR 90%-of the 155 surgically proved sites of extrahepatic disease.5

Ultrasound or CT in the ED?

Debate continues about the relative merits of CT and ultrasonography in the patient with blunt abdominal trauma. On the one hand, CT permits a rapid evaluation of the entire body, and many larger emergency departments have CT scanners. On the other hand, ultrasonography is fast, portable, and inexpensive, and the study can be carried out in the hemodynamically unstable patient while the resuscitation team is working. Hospitals in Europe and Asia tend to prefer ultrasonography as the initial study, and some US centers employ a four-quadrant scan by surgeons (sometimes called focused assessment for sonographic examination of the trauma patient [FAST]).

Mich?le A. Brown, MD, and her associates at the University of California-San Diego recently described the use of ultrasonography in almost 2,700 patients with blunt abdominal trauma.6 The images were reviewed immediately by a radiology resident and a trauma surgeon and, if they were obtained after hours, by a member of the ultrasonography staff the next day. The overall accuracy in identifying significant injuries was 96%, and none of the patients with false-negative findings died as a result of treatment delay. The hospital has now changed its practice to make ultrasonography the first study in the trauma patient. A CT scan is performed only if there are positive findings at ultrasonography, there is clinical suspicion of injury despite a negative ultrasound examination, or ultrasonography is not available. However, the authors caution that ultrasonography is appropriate only if there will be a period of observation so that additional scans can be performed if the patient’s clinical condition worsens.

Examining the Bowel Wall

Two types of disease affect the bowel wall in significant numbers of patients: inflammatory disease (eg, Crohn’s disease, ulcerative colitis) and colorectal cancer. In the past, the principal method for diagnosing and evaluating these conditions was the barium enema, which provides only an indirect view of the wall.

High-resolution ultrasonography may enable a distinction between the various types of inflammatory bowel disease,7 a potential that is more important now that disease-specific treatment agents are being introduced. Assessment of regions of inflammatory activity is possible with Doppler ultrasonography,8 providing a ready means of following patients over the long course of these diseases.? Complications such as abscesses and fistulas are readily seen.

At least 76% of the 56,000 deaths every year from colon cancer could be prevented by finding and removing large adenomatous polyps.9 However, the public generally resists the most accurate techniques now available for finding such lesions, namely sigmoidoscopy and colonoscopy. A possible alternative is virtual colonoscopy or, more accurately, CT colonography. (See related story, page 44.)The principal focus of research at present is early detection of significant polyps and adenomas (defined as those 10 mm or larger) and colon cancers in symptomatic patients and those with a strong family history of colorectal cancer.?

Perhaps the largest series of patients was reported earlier this year by Judy Yee, MD, and her colleagues from the University of California-San Francisco and the San Francisco Veterans Affairs Medical Center.10? Three hundred patients underwent CT colonography followed by colonoscopy, which was the standard for diagnosis. Importantly, CT was 90% sensitive in detecting polyps 10 mm or larger and 94% sensitive in detecting adenomas of similar size. All eight of the cancers were found by CT colonography.

Efforts continue to optimize the technique. Satisfactory images can be obtained by having patients ingest multiple doses of contrast material over the 48 hours preceding CT. The contrast labels the stool, which can then be subtracted from the images, so no colon preparation is necessary. As shown by colonoscopy, this method is almost as sensitive as CT examination of a prepared colon for the detection of polyps.11 Intravenous iodinated contrast medium significantly increases the confidence of readers in their interpretations.12 Contrast use also improves the depiction of the bowel wall and the ability to detect polyps of 6 to 9 mm in a poorly prepared colon. Of particular importance, in one study,12 three large polyps were apparent only on the contrast-enhanced studies.

There is great interest in using CT colonography the same way mammography is used: for general screening of asymptomatic patients. However, Ros cautions that at present, CT colonography is not an accepted tool for this purpose.

“Before you start screening healthy people who are at low risk for a disease with something that has some risk, as CT does, you must have clear proof of its value in perhaps 50,000 patients,” he notes. “We are years away from having such data.”

Perhaps some day, however, such proof will be available. Then, as Schoepf and associates of the University of Munich and the Cleveland Clinic Foundation speculate,13 people will undergo multislice low-dose CT as a universal screening tool for colon cancer, lung cancer, and coronary artery disease in one session!


Looking back at 50 years of abdominal imaging, Harry I. Goldberg, MD, and Alexander R. Margulis, MD, noted that the field of today “bears only a vague resemblance” to the field in 1950. The dramatic changes in technology “have changed gastrointestinal radiology irrevocably.”14?

Judith Gunn Bronson, MS, is a contributing writer for Decisions in Axis Imaging News.


  1. Peterson MS, Baron RL, Marsh JW Jr, et al. Pretransplantation surveillance for possible hepatocellular carcinoma in patients with cirrhosis: epidemiology and CT-based tumor detection rate in 430 cases with surgical pathologic correlation. Radiology. 2000;217:743?749. 2. 3.
  2. Ward J, Guthrie JA, Scott DJ, et al. Hepatocellular carcinoma in the cirrhotic liver: double-contrast MR imaging for diagnosis. Radiology. 2000;216:154?162.
  3. Valls C, And?a E, S?nchez A, et al. Hepatic metastases from colorectal cancer: preoperative detection and assessment of resectability with helical CT. Radiology. 2001;218:55?60.
  4. Semelka RC, Helmberger TKG. Contrast agents for MR imaging of the liver. Radiology. 2001;218:27?38.
  5. Low RN, Semelka RC, Worawattanakul S, Alzate GD, Sigeti JS. Extrahepatic abdominal imaging in patients with malignancy: comparison of MR imaging and helical CT, with subsequent surgical correlation. Radiology. 1999;210:625?632.
  6. Brown MA, Casola G, Sirlin CB, Patel NY, Hoyt DB. Blunt abdominal trauma: screening US in 2,693 patients. Radiology. 2001;218:352?358.
  7. Valette PJ, Rioux M, Pilleul F, Saurin JC, Fouque P, Henry L. Ultrasonography of chronic inflammatory bowel disease. Eur Radiol. 2001;11:1859-1866.
  8. Spalinger J, Patriquin H, Miron M-C, et al. Doppler US in patients with Crohn disease: vessel density in the diseased bowel reflects disease activity. Radiology. 2000;217:787?791.
  9. Winawer SJ, Zauber AG, Ho MN, et al. Prevention of colorectal cancer by colonoscopic polypectomy. N Engl J Med. 1993;329:1977?1981.
  10. Yee J, Akerkar GA, Hung RK, Steinauer-Gebauer AM, Wall SD, McQuaid KR. Colorectal neoplasia: performance characteristics of CT colonography for detection in 300 patients. Radiology. 2001;219:685?692.
  11. Callstrom MR, Johnson CD, Fletcher JG, et al. CT colonography without cathartic preparation: feasibility study. Radiology. 2001;219:693?698.
  12. Morrin MM, Farrell RJ, Kruskal JB, Reynolds K, McGee JB, Raptopoulos V. Utility of intravenously administered contrast material at CT colonography. Radiology. 2000;217:765?771.
  13. Schoepf UJ, Becker CR, Obuchowski NA, et al. Multi-slice computed tomography as a screening tool for colon cancer, lung cancer and coronary artery disease.
  14. Goldberg HI, Margulis AR. Gastrointestinal radiology in the United States: an overview of the past 50 years. Radiology. 2000;216:1?7.