Ulcerative Colitis Imaging

Updated: Nov 22, 2015
  • Author: Ali Nawaz Khan, MBBS, FRCS, FRCP, FRCR; Chief Editor: Eugene C Lin, MD  more...
  • Print


Preferred examination

Plain abdominal radiographs (see the images below) are a useful adjunct to imaging in cases of ulcerative colitis of acute onset. In severe cases, the images may show colonic dilatation, suggesting toxic megacolon; evidence of perforation; obstruction; or ileus. Radiographic imaging has an important role in the workup of patients with suspected IBD and in the differentiation of ulcerative colitis and Crohn disease. Because of its ability to depict fine mucosal detail, double-contrast barium enema examination is a valuable technique for diagnosing ulcerative colitis and Crohn disease, even in patients with early disease. Traditionally, barium enema examination has been the mainstay of radiologic investigation for suspected ulcerative colitis. [1, 2, 3]

Double-contrast barium enema studies in a 44-year- Double-contrast barium enema studies in a 44-year-old man known to have a long history of ulcerative colitis. Images show total colitis and extensive pseudopolyposis.
Plain abdominal radiograph in a patient (same as i Plain abdominal radiograph in a patient (same as in the previous image) who presented with an acute exacerbation of his symptoms. Image shows thumbprinting in the region of the splenic flexure of the colon.
Plain abdominal radiograph obtained 2 days later i Plain abdominal radiograph obtained 2 days later in the same patient as in the previous image shows distention of the transverse colon associated with mucosal edema. The maximum transverse diameter of the transverse colon is 7.5 cm. The patient was treated for toxic megacolon.

Transabdominal ultrasonography (US) is a noninvasive modality that may be helpful in the diagnosis of IBD, but it cannot be used to distinguish between ulcerative colitis and Crohn disease. US is also a useful technique in the investigation of biliary complications of the disease. [4]

Generally, CT has a limited role in the diagnosis of uncomplicated ulcerative colitis. However, CT plays an important role in the differential diagnosis of ulcerative colitis, and it is an excellent modality in the diagnosis of complications associated with the disease. Biliary dilatation suggests primary sclerosing cholangitis. Cross-sectional imaging studies such as CT, MRI, and US are useful for showing the effects of these conditions on the wall of the bowel and for demonstrating intra-abdominal abscesses and other extraluminal findings in patients with more advanced disease. Thus, barium studies and cross-sectional imaging studies have complementary roles in the evaluation of ulcerative colitis. [5, 4]

Radionuclide studies (see the image below) are useful in cases of acute fulminant colitis when colonoscopy or barium enema examination is contraindicated. Radionuclide studies are also useful in depicting disease activity and the extent of disease and in monitoring the response to therapy. [6]

Scan obtained with technetium-99m hexamethylpropyl Scan obtained with technetium-99m hexamethylpropylamine oxime (HMPAO)–labeled WBCs in a patient with active colitis involving the transverse and descending colon.

Radiologic findings in cases of acute infective enterocolitis from infection caused by Entamoeba histolytica (amebiasis), cytomegaloviral colitis, and Isospora, Salmonella, Shigella, or Yersinia may be similar to the findings seen in cases of ulcerative colitis; this is especially true with CT scans.

Etiology and pathophysiology

Ulcerative colitis is a type of inflammatory bowel disease (IBD) that characteristically involves the large bowel (see the images below). It is a multifactorial polygenic disease; the exact etiology is unknown. Included in the etiologic theories are environmental factors, immune dysfunction, and a likely genetic predisposition. Some have suggested that children of below-average birth weight who are born to mothers with ulcerative colitis have a greater risk of developing the disease.

Double-contrast barium enema study shows changes o Double-contrast barium enema study shows changes of early disease. Note the granular mucosa.
Single-contrast enema study in a patient with tota Single-contrast enema study in a patient with total colitis shows mucosal ulcers with a variety of shapes, including collar-button ulcers, in which undermining of the ulcers occurs, and double-tracking ulcers, in which the ulcers are longitudinally orientated.
Postevacuation image obtained after a single-contr Postevacuation image obtained after a single-contrast barium enema study shows extensive mucosal ulceration resulting from Shigella colitis.

Histocompatibility human leukocyte antigen (HLA)–B27 is identified in most patients with ulcerative colitis, though this finding is not associated with the condition. Immune dysfunction has been postulated as a cause, although the clear evidence of this is limited. Ulcerative colitis might also be linked to diet, though diet is thought to play a secondary role. Food or bacterial antigens might exert an effect on the already damaged mucosal lining, which has increased permeability.

The severity of ulcerative colitis can be graded as follows:

  • Mild — Bleeding per rectum and fewer than 4 bowel motions per day
  • Moderate — Bleeding per rectum with more than 4 bowel motions per day
  • Severe — Bleeding per rectum, more than 4 bowel motions per day, and a systemic illness with hypoalbuminemia (< 30 g/L)

Limitations of techniques

All imaging techniques lack specificity. Mucosal ulceration depicted on barium studies is nonspecific and is encountered in a variety of colitides. In severe cases, barium enema may precipitate toxic megacolon. Generally, barium enemas may be performed safely only in mild cases. Thickening of the bowel wall, as seen on cross-sectional imaging (CT, MRI, and US), is a nonspecific finding seen in a variety of bowel conditions besides IBD. Increased radionuclide focal activity may be related to a variety of physiologic and pathologic conditions unrelated to ulcerative colitis. Motion artifacts may interfere with cross-sectional imaging. Although this is not a major problem with modern scanners, some investigators still use a hypotonic agent to decrease bowel peristalsis.

Patient resources

For patient information resources, see the Crohn Disease Center and the Esophagus, Stomach, and Intestine Center, as well as Crohn Disease and Crohn Disease FAQs.

Ulcerative Colitis: Surgical Perspective.



Plain radiographic findings

Plain radiographs frequently yield useful information that may serve as a guide to further clinical evaluation and imaging (see the images below).

Plain abdominal radiograph in a patient with known Plain abdominal radiograph in a patient with known ulcerative colitis who presented with abdominal pain, peritonism, and leukocytosis. At surgery, a perforated toxic megacolon superimposed on ulcerative colitis was confirmed.
Increased postrectal space is a known feature of u Increased postrectal space is a known feature of ulcerative colitis.
Plain abdominal radiograph on a patient with known Plain abdominal radiograph on a patient with known ulcerative colitis who presented with an acute exacerbation of his symptoms. Image shows thumbprinting in the region of the splenic flexure of the colon.
Plain abdominal radiograph in a 26-year-old with a Plain abdominal radiograph in a 26-year-old with a 10-year history of ulcerative colitis shows a long stricture/spasm of the ascending colon/cecum. Note the pseudopolyposis in the descending colon.
Lateral radiograph of the lumbar spine shows a bam Lateral radiograph of the lumbar spine shows a bamboo spine.

The intraluminal gas pattern may show pseudopolyposis and deep ulcers. The colon may appear shortened, with associated loss of colonic haustra. Toxic megacolon is a complication of ulcerative colitis. The plain abdominal radiograph may show massive colonic dilatation associated with an abnormal mucosal contour. The dilatation is most pronounced in the transverse colon. Colonic perforation is another known complication of ulcerative colitis. Perforation may occur with or without a toxic megacolon. Pneumoperitoneum associated with a colonic perforation is usually massive. Fecal residue may be absent in inflamed bowel.

Barium enema findings

The findings on a barium enema examination vary with the stage and severity of disease (see the images below). The radiographic changes may involve the whole colon; when the disease is segmental, the left colon is usually involved. Rectal sparing is rare; it is thought to occur in 5% of cases. Skip lesions are unusual. The colon may appear narrow; such narrowing is often associated with incomplete filling caused by colonic spasm and irritability.

Double-contrast barium enema study shows changes o Double-contrast barium enema study shows changes of early disease. Note the granular mucosa.
Double-contrast barium enema studies show changes Double-contrast barium enema studies show changes of early disease. Note the granular mucosa.
Double-contrast barium enema studies in a 44-year- Double-contrast barium enema studies in a 44-year-old man known to have a long history of ulcerative colitis. Images show total colitis and extensive pseudopolyposis.
Double-contrast barium enema study shows pseudopol Double-contrast barium enema study shows pseudopolyposis of the descending colon.
Single-contrast enema study in a patient (same pat Single-contrast enema study in a patient (same patient as in the previous image) with known ulcerative colitis in remission shows a benign stricture of the sigmoid colon.
Single-contrast enema study in a patient with tota Single-contrast enema study in a patient with total colitis shows mucosal ulcers with a variety of shapes, including collar-button ulcers, in which undermining of the ulcers occurs, and double-tracking ulcers, in which the ulcers are longitudinally orientated.
Double-contrast barium enema study shows total col Double-contrast barium enema study shows total colitis. Note the granular mucosa in the cecum/ascending colon and multiple strictures in the transverse and descending colon in a patient with a more than a 20-year history of ulcerative colitis.
Single-contrast barium enema study shows burnt-out Single-contrast barium enema study shows burnt-out ulcerative colitis.
Single-contrast barium enema study in a patient wi Single-contrast barium enema study in a patient with Shigella colitis.
Postevacuation image obtained after a single-contr Postevacuation image obtained after a single-contrast barium enema study shows extensive mucosal ulceration resulting from Shigella colitis.
Double-contrast barium enema studies show granular Double-contrast barium enema studies show granular mucosa associated with Campylobacter colitis.

The earliest mucosal changes are best depicted on a good-quality double-contrast barium enema study. Before ulcers appear, mucosal edema has a fine, granular appearance when the radiographs are seen en face. When ulcers first appear, the mucosa may have a fine, stippled appearance when seen en face. When mucosal ulcers become established and confluent, the mucosa is replaced by granulation tissue; on double-contrast enema examination, the characteristic appearance is coarsely granular.

In the acute and subacute phases of the disease, the ulcers may acquire a variety of shapes: collar-button ulcers occur with undermining of the ulcers; double-tracking ulcers are longitudinally orientated and are sometimes several centimeters long.

Symmetrical thickening of haustral folds may produce the appearance of thumbprinting. Pseudopolyps are a consequence of severe mucosal disease; they appear as multiple filling defects of varying sizes. These may develop rapidly, and they tend to persist, even when ulcerative colitis is quiescent. Occasionally, mucosal bridges are formed between pseudopolyps, which may be radiologically demonstrable.

With increased severity and duration of disease, the involved colon may become narrow, shortened, and loose in terms of its normal redundancy and haustral pattern. On lateral projections, rectal narrowing is easily recognizable as increased retrorectal space. When the entire colon is involved, changes in the terminal ileum may be seen (backwater ileitis); this involves 4-25 cm of the terminal ileum. The ileocecal valve appears patulous. The mucosa is granular and is usually associated with the absence of peristalsis.

Benign strictures occur in 1-11% of patients with long-standing disease. Such strictures are predominantly found in the left colon. Carcinomas that complicate ulcerative colitis are usually annular and may be difficult to differentiate from benign strictures. Often, however, malignant strictures are eccentric, with nodular narrowing and shouldered edges. Multiple carcinomas are not rare in the setting of ulcerative colitis. About 50% of these tumors are not detected on clinical or radiologic examination; instead, they are diagnosed at colectomy or autopsy.

Degree of confidence

Although many radiographic findings are nonspecific, many other features are typical, if not pathognomonic, of ulcerative colitis.

The diagnostic signs of various colitides overlap considerably. Similar radiographic signs may be seen in cases of infective diarrhea, Crohn disease, ischemic colitis, drug-induced colitis, and amebic colitis. Infective diarrhea is occasionally the presenting feature of IBD.


Computed Tomography

With increased use of CT as a primary imaging modality for evaluating inflammatory bowel disease (IBD), radiologists must be able to recognize the features of ulcerative colitis on CT. Although barium studies remain the principal tool for diagnosing and evaluating suspected IBD, CT may aid in differentiating ulcerative colitis and Crohn disease when results of barium studies are equivocal. High-resolution thin-section imaging of both the intraluminal and extraluminal components enables radiologists to detect and stage colonic pathology. [7, 8, 9, 10, 11]

CT is valuable for the detection and characterization of ulcerative colitis. CT typically demonstrates circumferential, symmetrical wall thickening with fold enlargement. Thickening of the colon wall (mean, 7.8 mm; standard deviation, 1.9) may be present, with inhomogeneous attenuation, a target appearance of the rectum, and the proliferation of perirectal fat. The normal colonic wall has a maximal thickness of 3 mm with the lumen distended and 5 mm with the lumen collapsed. In comparison, Crohn colitis causes greater bowel wall thickening (mean, 11 mm; standard deviation, 5.1). Such thickening appears in association with homogeneous attenuation, fistula and abscess formation, and mesenteric abnormalities.

The target sign consists of an inner ring of soft tissue attenuation, representing mucosa, lamina propria, and enlarged muscularis mucosa; a middle ring of low attenuation, resulting from widening and fatty infiltration or edema of the submucosa; and an outer area of soft tissue attenuation, representing the muscularis propria. If the submucosa is infiltrated by fat, this is a sign of chronicity; it is more specific to IBD. This sign is best appreciated on the arterial phase of enhancement. It is a nonspecific sign that is also reported in cases of Crohn disease and pseudomembranous colitis.

Ulcerative colitis is typically left sided or diffuse; only rarely does it involve the right colon exclusively. In cases of ulcerative colitis, wall thickening may be diffuse and symmetrical; by contrast, in cases of Crohn disease, wall thickening is eccentric and segmental, and skip lesions are present. The proliferation of perirectal fat is a nonspecific sign that can be seen in any of the colitides. Submucosal fat deposition is present significantly more often in ulcerative colitis (61%) than in Crohn colitis (8%).

Mural thickening of the terminal ileum may be visualized in 10-25% of patients; such thickening is caused by backwash ileitis, which occurs as a result of the reflux of colonic contents into the distal ileum. Abscesses and pseudodiverticula are not features of ulcerative colitis; they occur almost exclusively in Crohn colitis.

Degree of confidence

Cross-sectional imaging plays a prominent role in the assessment of IBD of the colon. The clinical history, laboratory data, and extent of involvement are used in conjunction with the radiographic results to reach a specific diagnosis.

CT is preferred to MRI for assessing the extent of inflammatory disease in and beyond the bowel wall. An added benefit of CT over MRI is the fact that patients with abscesses or large fluid collections may undergo drainage while in the CT scanner. CT and MRI may help to distinguish between cases of ulcerative colitis with minimal wall thickening and cases of Crohn disease with marked wall thickening with concomitant skip lesions and fistula and/or abscess formation. [12]

Although many CT findings in ulcerative colitis are striking, they are not considered specific. Some features do suggest a diagnosis of IBD, and specific features may indicate a single diagnosis of ulcerative colitis with a high degree of confidence. The halo sign typically occurs in ulcerative colitis.

False positives/negatives

Bowel thickening is a nonspecific finding; it may be encountered in a variety of colitides and other pathologies of the bowel wall. The target sign is also nonspecific; it has been reported in Crohn colitis. Mucosal ulceration is difficult to detect with CT.

The target sign may be secondary to a hypoattenuating submucosal layer of fluid (nonspecific) or fat (chronic in origin and usually from IBD). Also, there is evidence that submucosal fat may be a normal variant.


Magnetic Resonance Imaging

Giovagnoni and associates used high-resolution MRI to study 16 resected rectosigmoid specimens of patients who underwent total colectomy in treatment of severe ulcerative colitis, [13] and the authors' preliminary experience suggested that MRI should be considered a new imaging modality for detecting changes of the colon wall in ulcerative colitis. The in vitro results showed that MRI could depict all layers of the colonic wall. In ulcerative colitis specimens in particular, T1-weighted spin-echo MRI showed thickening and the peculiar abnormal hyperintensity of the mucosal and submucosal layers. The in vivo results confirmed the high signal intensity of the mucosal and submucosal layers.

Magnetic resonance cholangiopancreatography (MRCP) seems to be a reliable, noninvasive imaging method for diagnosing and following up primary sclerosing cholangitis. The rapid acquisition with relaxation enhancement (RARE) sequence has been found to have the highest diagnostic accuracy of the T2-weighted sequences.

Gadolinium-based contrast agents (gadopentetate dimeglumine [Magnevist], gadobenate dimeglumine [MultiHance], gadodiamide [Omniscan], gadoversetamide [OptiMARK], gadoteridol [ProHance]) have been linked to the development of nephrogenic systemic fibrosis (NSF) or nephrogenic fibrosing dermopathy (NFD).

NSF/NFD is a debilitating and sometimes fatal disease. Characteristics include red or dark patches on the skin; burning, itching, swelling, hardening, and tightening of the skin; yellow spots on the whites of the eyes; joint stiffness with trouble moving or straightening the arms, hands, legs, or feet; pain deep in the hip bones or ribs; and muscle weakness. [5, 4]

Degree of confidence

The usefulness in transmural assessment, the capability for sagittal imaging, and the lack of invasiveness are attractive features of MRI.

In a study by Shoenut and associates, MRI was comparable to endoscopy in differentiating ulcerative colitis from Crohn disease and in assessing the severity of disease. [14] In 17 of 18 patients, a correct diagnosis was made using gadolinium-enhanced, fat-suppressed, T1-weighted, spin-echo MRI. In 15 patients, a correct diagnosis was made on endoscopy. Overall, MRI was not significantly better than endoscopy in distinguishing ulcerative colitis from Crohn disease. MRI permitted correct grading of the severity of inflammatory changes in 13 of 20 patients; endoscopy did so in 11 cases. MRI and endoscopic findings were each found to be within one grade of the histologic findings in 7 patients. No significant difference was found between MRI and endoscopy in terms of the estimation of disease severity (as determined from biopsy samples). Assessments of bowel wall thickness on MRIs correlated well with the percentage of contrast enhancement. [14]

False positives/negatives

Crohn disease is important in the differential diagnosis of ulcerative colitis, though it is not always possible to differentiate these 2 conditions on the basis of imaging.

Hansmann and associates evaluated the diagnostic role of hydro-MRI in 33 patients suspected of having either Crohn disease or ulcerative colitis and concluded that hydro-MRI was a reliable modality for diagnosing Crohn disease. [15] A correct diagnosis was made with MRI in all 24 patients with active Crohn disease. However, MRI results were positive in only 2 of 5 patients with ulcerative colitis. In 5 patients, IBD was excluded. No false-positive results occurred.



Excluding infective and toxic states, Crohn disease and ulcerative colitis are the most common causes of chronic large bowel inflammation. US features are nonspecific and include bowel wall thickening, which may involve both the hypoechoic muscular coat and the echogenic mucosa. Wall thickening extends longitudinally, and there is decreased echogenicity and luminal narrowing. Localized perforation may lead to the formation of an abscess, which may be clinically silent if the patient is receiving steroid therapy.

Several criteria have been described for the detection of Crohn colitis and ulcerative colitis. A water-filled colon allows better definition of the large bowel mucosa. With Crohn colitis, the whole bowel wall is thickened; all layers are involved with edema, fibrosis, inflammation, and lymphangiectasis. On US, the wall is clearly thickened, hypoechoic, and homogeneous initially, but it may become inhomogeneous because of fat deposition. The layered anatomy of the bowel wall and the haustra are lost. The bowel becomes rigid, with diminished compressibility and peristalsis. The diameter of the wall is approximately 13 mm.

In ulcerative colitis, the wall thickness is not as great, averaging 7.8 mm. Early disease is confined to mucosa; wall stratification is preserved, but haustra are lost. In Crohn disease and ulcerative colitis, the presence of Doppler parietal flow throughout the affected thickened segment indicates an acute condition. Similarly, an abnormally high mean portal velocity of 30-48 cm/s (normal, 15 ± 7 cm/sec) and an abnormally low resistive index of 0.58-0.78 (normal, 0.908-0.026) are detected in the superior mesenteric artery (SMA). These values, which are determined on the basis of Doppler examination, are known to revert to normal with successful therapy.

Differentiation between ulcerative colitis and Crohn disease is important from a management viewpoint. The latter tends to be transmural disease; the former tends to be a superficial inflammatory process of the mucosa.

Endosonography (EUS) was expected to be effective in discriminating cases of otherwise indeterminate colitis. Efforts to demonstrate these changes with EUS have been largely disappointing. EUS plays a limited role in discriminating ulcerative colitis from Crohn disease. On a more positive note, EUS evaluation of perirectal and perianal complications of Crohn disease has been demonstrated to be superior to fistulography and CT and equal or superior to MRI. Because accurate anatomic information is required to guide surgical therapy of these lesions, EUS may emerge as a powerful imaging tool in the management of perianorectal Crohn disease. [16]

In a study by Bru and associates, hydrocolonic US demonstrated 100% sensitivity in identifying patients with active IBD, and it demonstrated greater overall accuracy (87%) than scintigraphy (77%) in the assessment of disease extension. In addition, a hydrocolonic US activity index was strongly correlated with clinical and endoscopic activity indexes. The authors prospectively compared the usefulness of hydrocolonic US and technetium-99m (99m Tc) hexamethylpropylamine oxime (HMPAO)–labeled WBC scintigraphy in 68 patients with active IBD (34 patients had ulcerative colitis; 34 had Crohn disease), in 12 with inactive IBD, and in 10 control subjects. [17] With these criteria, The authors concluded that their study provided precise US criteria for the definition of bowel involvement by active IBD.

Sigirci and associates concluded in their study that duplex Doppler US of the inferior mesenteric artery (IMA) and superior mesenteric artery (SMA) may be used to evaluate inflammatory disease of the large bowel, to assess the extent of disease, and to document the response to therapy. They evaluated Doppler US blood-flow parameters and spectral patterns in the IMA and the SMA in patients with active and inactive (remission-phase) ulcerative colitis. [18, 4]

Degree of confidence

Visualization of the upper abdomen and the bowel, in particular, is frequently compromised by the presence of gas within the bowel. To a certain extent, this may be overcome by performing US examinations in the morning, when bowel gas is at a minimum. Water may be used to distend the colon and thereby provide an acoustic window, but many people swallow air when drinking, which may cause further artifacts.

US measurements of bowel wall thickness offer a simple, noninvasive technique for monitoring IBD in children. Thickened and hypervascularized bowel wall are characteristic findings in cases of IBD. A combination of B-mode and power Doppler US offers an additional noninvasive procedure for the determination of activity in patients with IBD.

False positives/negatives

Most bowel pathologies cause thickening of the bowel wall. On bowel examination, a target sign is usually seen. With this sign, a hypoechoic wall surrounds the echogenic mucosa, intraluminal mucous, air, and other bowel contents. This target may be round or oval (pseudo-kidney sign).

Under normal physiologic conditions, the hypoechoic bowel wall measures less than 4 mm. With few exceptions, this 4-mm rule may be applied to the whole of the bowel. Under pathologic conditions, the bowel wall thickens symmetrically in target fashion. It may be thickened asymmetrically when the echogenic lumen is displaced to one side on a true cross-section; this is the atypical target sign.

Most bowel pathology, whether inflammatory or neoplastic, may be inferred from the aforementioned signs. As with most imaging findings, these signs are nonspecific and suggest only a differential diagnosis. US findings of mural thickening of the bowel with a paucity of luminal content may be helpful in the detection of IBDs. However, the findings are nonspecific, and the differential diagnosis must be related to the specific segment of the bowel involved.


Nuclear Imaging

Until recently, gallium-67 (67 Ga) was the only radionuclide used in the scanning of inflammatory bowel disease (IBD). However, recent developments in leukocyte labeling have largely replaced the use of gallium because of better dosimetry and resolution and a lower radiation burden to the patient. Jones and associates performed67 Ga citrate scanning in 9 patients with ulcerative colitis, [19] and in all patients, there was good correlation between the regional uptake of gallium and the extent and activity of disease. In 2 patients, scans were positive during an acute exacerbation and reverted to normal or near normal during clinical remission. In 1 patient in whom the colon was resected because of toxic dilatation, good correlation was found between pathologic and scintigraphic results.

During an acute attack of ulcerative colitis, when colonoscopy or barium enema may be contraindicated, gallium scanning may provide a noninvasive means of assessing the extent of colonic involvement. It may also be an alternative means of following the clinical course of the disease.

With the availability of indium-labeled WBCs, radionuclide-imaging studies have a definite role in the diagnosis and staging of IBD. The indium-111 (111 In) WBC study is particularly helpful in evaluating recurrent disease in patients with severe intercurrent diseases and in screening patients, without the need for barium examinations. [20]

Indium-111 oxine and111 In tropolone are neutral, nonpolar, lipophilic chelates that penetrate the cell membrane of leukocytes. Once inside the leukocytes, the111 In ion dissociates from the chelate and forms a relatively stable bond with cytoplasmic and nuclear proteins. There are a number of differences in the 2 chelating agents; one of the most important is that111 In oxine cannot be used to label leukocytes in the presence of plasma because of the higher affinity of111 In for transferrin than for oxine. Indium-111 tropolone is a stronger chelating agent; its use does not require the removal of plasma before cell labeling.

Both types of111 In label all types of blood cells indiscriminately; therefore, the leukocytes must first be separated from the other cells. The result is a mixed population of labeled leukocytes, including neutrophils, monocytes, and lymphocytes; this is acceptable for most imaging. Labeling efficiencies on the order of 80-90% may be achieved with either agent. If the labeling efficiency is less than 40%, the cells should not be reinjected.

Indium-111–labeled leukocytes are normally distributed in the spleen, liver, and bone marrow and transiently in the lungs. The accumulation of activity at sites other than these sites is suggestive of infection or inflammation.

WBCs may be labeled with99m Tc. Technetium-99m HMPAO is a neutral, nonpolar, lipophilic agent that was originally developed for cerebral perfusion imaging (see the image below). The compound is unstable and breaks down into a hydrophilic secondary complex. In its lipophilic form,99m Tc HMPAO can cross the cell membrane of leukocytes. Once within the cell, its structure is altered to the hydrophilic form, and99m Tc is trapped inside. [6]

Scan obtained with technetium-99m hexamethylpropyl Scan obtained with technetium-99m hexamethylpropylamine oxime (HMPAO)–labeled WBCs in a patient with active colitis involving the transverse and descending colon.

Technetium-99m HMPAO leukocyte labeling may be performed in the presence of plasma and appears to have significant selectivity for granulocytes. As with111 In-labeled leukocytes,99m Tc-labeled leukocytes are localized in the spleen, liver, and bone marrow and transiently in the lungs.

False-positive diagnosis of bowel inflammation is possible with99m Tc-labeled leukocytes. In adults, normal localization occurs in the bowel about 3-4 hours after administration; in children, normal localization occurs earlier. This localization is associated with sloughing of leukocytes into the lumen of the GI tract. In addition, activity is seen in the urinary tract and occasionally within the gallbladder. Fasting for 2-4 hours before imaging may reduce hepatobiliary excretion and subsequent bowel activity, especially in children. The 1-hour images demonstrate lung activity but no activity in the GI system.

Studies have indicated that imaging with99m Tc-labeled leukocytes may be more sensitive than imaging with111 In-labeled leukocytes for detecting and determining the extent of IBD, especially in the small bowel.

Fluorodeoxyglucose positron tomography has been used in the diagnosis of pediatric IBD. This technique appears to provide adequate information for patients suspected of having IBD. [21]

Degree of confidence

Radionuclide-labeled WBC scanning using111 In is an accurate, quantitative, and noninvasive method for assessing the degree of bowel inflammation in cases of IBD; it is also useful in assessing response to therapy. Fecal collection or total body counts of radioactivity may be used.

The use of99m Tc HMPAO WBC scanning has several advantages over the111 In scanning, including a reduction in the radiation dose and better image resolution. The99m Tc HMPAO WBC technique may be used in combination with single-photon emission CT (SPECT), which allows visualization of the entire bowel as distinct from overlying structures.99m Tc HMPAO WBC SPECT has now been computerized and automated to permit measurement of segmental and total bowel uptake. This method could potentially become the criterion standard for an objective assessment of the response of disease activity in cases of Crohn disease and ulcerative colitis. [22]

Technetium-99m HMPAO-labeled leukocyte scintigraphy has been found to correlate well with colonoscopy in the assessment of both the extension and severity of ulcerative colitis. Use of 2-hour scintigraphic imaging decreases the number of false-positive results. [23]

Arndt and associates concluded that imaging performed 1 hour after the injection of99m Tc-labeled WBCs may reliably replace111 In-granulocyte scintigraphy for patients with IBD because the radiopharmaceutical is available on a daily basis. When bowel segments were compared, concordance was found for 102 (91.8%) of 111 segments with 1-hour WBC images and 3-hour granulocyte images. For 5 of 5 WBC-positive and granulocyte-negative segments, it could be proven that the WBC result was caused by active disease.99m Tc-labeled WBC scintigraphy detected active disease in 4 more patients than did scanning with111 In-labeled granulocytes (11 and 7 patients, respectively). Tc-labeled WBCs was superior in the assessment of active disease, especially in small-bowel segments.

Tolia and associates compared111 In-labeled WBC scanning and colonoscopy in 19 children and adolescents with chronic IBD; they compared the degree to which the 2 diagnostic modalities correlated with histologic diagnosis. [24] The data suggested that the usefulness of scans is limited to patients in whom standard diagnostic procedures are contraindicated. Additionally, the authors suggested that the visual diagnostic impression must be confirmed with colonoscopy and histologic diagnosis. The sensitivity of111 In scanning was 18%, its specificity was 62.5%, and its accuracy for colonic disease was only 37%. In comparison, the sensitivity and specificity of colonoscopy were 100% and 57%, respectively; its accuracy was 84%.

Peters and associates compared99m Tc-labeled WBCs with111 In-labeled WBCs and found that the 2 agents provided similar clinical information in 27 of 30 patients. [25] Of the 3 in whom the information differed, 1 received111 In-labeled granulocytes, which were considered to be functionally suboptimal. In the other 2, each of whom had IBD, the scans showed different distributions of abnormal bowel activity. The authors concluded that, with respect to granulocyte kinetics and clinical data, leukocytes labeled with99m Tc HMPAO were comparable to granulocytes labeled with111 In tropolonate.

Poitras and associates performed111 In-labeled WBC scanning in 10 patients with ulcerative colitis and in 39 patients with Crohn disease involving the small intestine (25 cases), the colon (17 cases), or both, [26] and they concluded that the results of radionuclide scanning were not superior to those of conventional techniques used to detect and localize IBD and that radionuclide methods cannot replace these techniques. In two thirds of the patients,111 In leukocyte scanning of the abdomen enabled accurate evaluation of the extent of disease (sensitivity, 68%). Radionuclide uptake by the gut was seen in 84% of the patients with active inflammation.

False positives/negatives

111 In-labeled WBCs accumulate nonspecifically at sites of inflammation or infection. The differential diagnosis for accumulation of111 In-labeled WBCs involving the abdomen most commonly includes Crohn disease, ulcerative colitis, diverticulitis, ischemic or infarcted bowel, phlegmon, and wound infection.

Factors that may lead to false-positive results include the swallowing of leukocytes, the use of enteric tubes, and the presence of a GI bleed. False-positive results may also occur as a result of the physiologic uptake of99m Tc-labeled WBCs; biliary and urinary excretion; and delayed normal bowel excretion.

Saverymuttu and associates reported 2 cases of colonic carcinoma that appeared as suspected IBD and in which111 In-labeled granulocytes were localized. [27] Histology revealed an acute inflammatory infiltrate in these tumors. These results indicate that111 In-labeled granulocyte scanning is of limited usefulness in the diagnosis of IBD.



Tsuchiya and associates performed selective superior and inferior mesenteric angiography in 25 patients with ulcerative colitis. [28] Vascular changes in large vessels, such as increased diameter of the IMA, were found to correlate better with the severity or activity of illness. However, in some cases, microcirculatory changes of small vessels of the intestinal wall (eg, capillary blush or loss of normal tapering of vasa recta) remained in remission of this disease. The authors assumed that this microcirculatory disturbance plays an important role in the pathophysiology of ulcerative colitis.

Hiramatsu and associates performed angiography in 11 patients with ulcerative colitis. [29] They injected prednisolone directly into the superior and IMAs in 10 patients. Hypervascularity of the affected regions was a common finding, but early venous filling was noted in only 5 of 11 cases. Good results were obtained with intra-arterial steroid injections.

Angiography is a valuable tool in the diagnosis and therapy of hemorrhage from the large bowel resulting from ulcerative colitis and other causes. [30]

Degree of confidence

Brahme and Hildell evaluated the diagnostic yield of angiography in 116 patients with Crohn disease [31] and concluded that angiography was a diagnostic adjunct to barium studies and that it revealed the presence of lesions, as well as their extent. Angiographic abnormalities were found in over 90%. Crohn disease could be distinguished from ulcerative colitis in only 30% of the cases. The authors suggested that this feature was particularly important in cases of suspected postoperative recurrences of Crohn disease and that angiography was a potential aid in the differential diagnosis of doubtful cases of inflammatory and malignant bowel disease.

Angiographic findings are nonspecific; both false-positive and false-negative results may occur.