| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Department of Imaging, Great Ormond St. Hospital for Children, London, United
Kingdom.
2 Department of Imaging, Hammersmith Hospital, Du Cane Rd., London W12 0HS,
United Kingdom.
Received June 23, 2005;
accepted after revision September 4, 2005.
Address correspondence to J. E. Jackson
(jejackson{at}hhnt.nhs.uk).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. A retrospective review was performed of all visceral angiography referrals over a 10-year period. During that time, 522 visceral angiograms were obtained for the investigation of acute or chronic gastrointestinal bleeding. IBD was reported as a possible or probable diagnosis based on the angiographic findings in 43 patients (8.2%) (male-to-female ratio, 23:20; age range, 11-79 years; mean age, 48.6 years).
RESULTS. Active IBD was confirmed in 25 (58%) of these 43 individuals (male-to-female ratio, 15:10; mean age, 46.5 years), representing 4.8% of all patients referred for the investigation of gastrointestinal bleeding. In 14 of these individuals, a diagnosis of IBD had not been suspected previously. A diagnosis other than IBD was found to be the cause of the angiographic findings in seven subjects. In the 32 patients in whom a cause for the angiographic findings was identified, IBD was the diagnosis in 25 (78%).
CONCLUSION. In a group of patients referred for angiography to investigate acute or chronic gastrointestinal bleeding, IBD can be expected to be the cause in approximately 5%. Radiologists should be aware of the angiographic abnormalities seen in patients with IBD and be prepared to consider this diagnosis even if the results of other previous investigations have been reported as normal.
Keywords: angiography • gastrointestinal bleeding • inflammatory bowel disease
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We present a series of 43 patients in whom a diagnosis of IBD was considered on the basis of findings at visceral angiography and review the final diagnoses in this group of individuals.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
|
|
|
|
|
|
|
Several angiographic abnormalities were visualized and their presence or absence was recorded:
Increased contrast medium staining of the bowel wall—The most common angiographic feature, which was present in all patients, was one or more segments of bowel showing increased contrast medium staining compared with adjacent bowel loops (Figs. 1A, 1B, 1C, 1D, 2A, 2B, 2C, 3A, 3B, 3C, and 3D). This finding was usually, but not invariably, accompanied by an increase in caliber of the supplying arteries. The length of involved bowel ranged from segments of small bowel measuring only a few centimeters to most of the large bowel.
|
|
|
|
Abnormalities of the vasa recta—In some cases, the vasa recta within diseased segments of bowel appeared markedly abnormal, showing abrupt truncation, angulation, or both (Figs. 1C, 4A, 4B, and 4C). This finding was generally best seen on magnified views.
|
|
|
Strictured segments of bowel—In some cases, the width of a segment of bowel showing increased contrast medium staining was consistently narrowed on several separate angiographic runs, suggestive of the presence of a stricture (Fig. 2C). The presence of a stricture was reported only if there were accompanying angiographic abnormalities.
Double-layer appearance of the bowel wall—This appearance has previously been referred to as the "zoning" sign (vide infra in Discussion section) and describes a densely staining inner wall, representing a hypervascular submucosa and mucosa, and a thickened but relatively avascular outer muscle layer. This sign was seen in this series only when disease involved the colon (Figs. 5A and 5B).
|
|
Presence of hypervascular mesenteric lymph nodes—Hypervascular mesenteric lymph nodes were seen on some of the angiograms adjacent to mesenteric arterial branches (Fig. 2A).
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
A diagnosis of active Crohn's disease was established in 21 of these 25 cases, active ulcerative colitis in two cases, nonspecific colitis in one case, and nonspecific inflammatory small-bowel ulceration in one case. Contributory investigations after angiography included small-bowel enema, colonoscopy, and laparotomy. Histologic proof of the diagnosis was obtained in 19 cases (76%). In the remaining six cases, the diagnosis was based on characteristic changes on barium studies and appropriate response to subsequent medical therapy.
Although 14 of these subjects had no diagnosis before angiography (Table 1), the diagnosis of IBD had been made previously in 11 patients: seven had a history of Crohn's disease, two of ulcerative colitis, and two of nonspecific colitis (Table 2). All were thought to have quiescent disease at the time of presentation because they had no other symptoms of gastrointestinal disease and were referred for visceral angiography with the explicit purpose of establishing a second diagnosis.
Of the 18 patients in whom IBD was suspected on the basis of the angiographic findings but was not confirmed, an alternative diagnosis that explained the observed angiographic abnormality was made in six patients. Three of these diagnoses were malignant tumors (cecal adenocarcinoma, small-bowel angiosarcoma, colonic lymphoma) and three were benign causes (nonsteroidal antiinflammatory drug [NSAID]-associated enteropathy in one, edematous small-bowel loops with vascular congestion of undetermined cause in two). Active contrast medium extravasation in the right colon that was thought to be from a diverticulum was seen in one patient and was embolized; increased vascularity was also seen in the distal ileum, which was thought to be due to IBD. This patient proceeded to a right hemicolectomy because of continuous bleeding, and the histology of the resected specimen showed acute inflammation due to diverticular disease. The small-bowel loops appeared macroscopically normal, and the cause of the ileal hypervascularity seen on angiography was therefore not determined, although it may have been reactive.
A cause for the reported angiographic abnormalities has not been found in any of the remaining 11 patients, and in none of them can inflammatory disease be excluded. Three have been lost to follow-up, and one refused to undergo additional examinations. In one patient, the gastrointestinal bleeding resolved without therapy, and the patient died 2 years later of unrelated causes. Two other individuals died soon after undergoing angiography of continued gastrointestinal hemorrhage, the cause of which was not found. One patient was found to have some small intestinal telangiectasia on capsule endoscopy, but this finding was not thought to be the cause of the angiographic abnormality seen in a loop of mid ileum; chronic iron deficiency anemia was thought to be due to a large rolling hiatus hernia, and this individual improved on a proton pump inhibitor. The last three patients do not appear to have undergone further imaging of the bowel loops that were reported as abnormal on angiography despite continuing anemia.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
All of the early published studies described angiographic findings in subjects with an established diagnosis of IBD. The largest series, published in 1976 by Brahme and Hildell [9], comprised 123 angiograms obtained in 116 patients with Crohn's disease. Histologic proof of the diagnosis was available in 107 cases; the remaining patients had shown characteristic abnormalities on a minimum of two consecutive barium studies. In each case, superior mesenteric angiography was performed via conventional percutaneous catheterization, with inferior mesenteric studies performed in selected cases. The angiographic features described included an increase in the number and tortuosity of arterial vessels supplying the bowel, increased opacification and thickness of the bowel wall, and early opacification of draining veins. Abnormalities were detected in more than 90% of patients, but were thought to be nonspecific in most cases.
The two angiographic signs considered to be diagnostic of Crohn's disease were the presence of mesenteric neovascularity and the "zoning" sign. Herlinger [10] first described the zoning sign in 1972, referring to a double-layer appearance of the bowel wall with a densely staining inner wall representing a hypervascular submucosa and mucosa and a thickened but relatively avascular outer muscle layer. In the series by Brahme and Hildell [9], the zoning sign, considered to occur exclusively in Crohn's disease, was identified in 35% of cases.
Angiographic findings in ulcerative colitis have also been described, albeit in much smaller numbers [4, 5, 10, 11]. Positive findings in early published studies included increased vascularity of the bowel wall and early filling of draining veins. Tortuosity and irregularity of the arterial vessels were not features. No single finding was thought to be diagnostic of ulcerative colitis, and the relative value of different features was heavily debated.
Brahme [6] and others [7, 9] performed microangiographic studies in resected bowel specimens to correlate in vivo findings with histologic abnormalities. Comparison was also made with operative findings. In those studies, the authors concluded that in cases of severe or long-standing disease, angiographic abnormalities correlated well with disease distribution and severity [7]. Angiography was also useful in postoperative cases in which disease recurrence was suspected [9]. The value of the technique was questioned, however, in cases of early disease in which the correlation with surgical abnormalities was poor. This was recognized to be due, at least in part, to an inability to adequately resolve the early mucosal changes with the conventional radiographic techniques available [9]. The use of mesenteric angiography to make a primary diagnosis of IBD in previously unsuspected cases was not considered.
Angiographic Appearances of IBD
Little has been reported about the angiographic appearances of IBD since
the advent of modern angiographic techniques. Digital subtraction angiography
is associated with lower spatial resolution than conventional film but greater
contrast resolution and, for this reason, might be more sensitive for the
detection of IBD given that increased staining of the bowel wall is the most
common angiographic abnormality. Despite this, there is no series, to our
knowledge, documenting the incidence of IBD in patients referred for visceral
angiography, nor is there one describing the usefulness of angiography in
diagnosing IBD in patients in whom the condition had not been suspected at the
time of arteriography despite previous imaging.
We have shown that certain angiographic features—notably, increased staining of bowel loops, early and prominent venous return, and skip lesions—are highly suggestive of IBD. Several of the angiographic features in our series correlate well with the findings first described in early mesenteric angiography series, notably increased vascularity of the bowel wall and irregularity and tortuosity of the vasa recta. Early and prominent venous drainage from diseased segments of bowel is shown more convincingly and in a greater number of cases than has been reported in the previous literature. Skip lesions were documented in eight of the patients with Crohn's disease presented in this series, an angiographic sign that has not, somewhat surprisingly, been previously described.
Lymph node enlargement is a recognized feature of active Crohn's disease and may be well depicted on CT and MRI [12], but the angiographic visualization of hypervascular mesenteric lymph nodes in patients with IBD has not been previously documented, to our knowledge. It was, however, an infrequent finding in this series, being present in only three patients subsequently confirmed as having Crohn's disease. Enlarged vascular lymph nodes were also seen in the patient with a cecal carcinoma, in the patient subsequently diagnosed as having NSAID-associated enteropathy, and in a patient in whom IBD was never confirmed.
Mesenteric neovascularity, a finding previously reported as being indicative of Crohn's disease as discussed earlier, was the one sign that was not convincingly shown in any patient in this series. Most of the angiographic features that we have discussed in this series require images of excellent quality for their clear visualization. Complete bowel paralysis is essential and was achieved in all patients by the liberal use of an intraarterial injection of hyosine butylbromide with the addition of intraarterial glucagon if necessary.
Although no single angiographic finding has been shown to be pathognomonic of IBD, the features we have described prove to be strongly predictive of active disease. In our series, IBD was diagnosed in 4.8% of subjects referred for the investigation of gastrointestinal bleeding.
The seven individuals in whom a diagnosis other than IBD was confirmed deserve further discussion. Three were diagnosed as having malignant disease at the site of angiographic abnormality, and the remaining four underwent laparotomy at which abnormal bowel, not due to Crohn's disease, was resected. These patients are discussed in more detail in the next section.
Diagnoses Other Than IBD
Malignant disease—The patient with a cecal adenocarcinoma
involving the terminal ileum gave a 6-month history of intermittent
gastrointestinal bleeding and had been on oral anticoagulants for 17 years
after placement of an aortic valve prosthesis. A recent colonoscopy had
revealed two ascending colonic polyps but no other abnormality. Angiography
showed an area of florid increased vascularity involving the cecum and
terminal ileum associated with early venous return. A minor irregularity of
cecal branches but without truncation or neovascularity was reported, and
these features, together with the fact that the colonoscopy had not shown a
significant abnormality, were interpreted as being due to IBD rather than a
neoplastic process. This case serves to stress the fact that visualization of
a large-bowel angiographic abnormality in a patient who has recently undergone
colonoscopy that showed "normal" findings does not exclude
malignancy [13], and such
patients should undergo further investigation.
In the patient with an angiosarcoma, the angiogram depicted a loop of proximal ileum supplied by the second ileal artery that appeared pulled down into the pelvis. At the apex of this loop, a circular area of increased vascularity supplied by prominent vasa recta was seen and there was early venous return from this segment. A small vascular blush was also seen in a segment of distal ileum supplied by the eighth ileal artery, and this bowel loop was noted to lie in close approximation to the abnormal proximal ileal loop. A differential diagnosis of IBD because of the involvement of two separate bowel loops, nonspecific ulceration, and a small-bowel tumor was reported. The patient proceeded to laparotomy at which an angiosarcoma involving proximal and distal ileal loops was found.
Lymphomatous infiltration of the large bowel was the cause of the angiographic findings in one individual who gave a 12-year history of relapsing Hodgkin's disease and who presented with a large gastrointestinal bleed. The angiographic abnormality, which consisted of increased vascularity of the ascending colon with wall thickening and early venous return (Figs. 5A and 5B), was reported as being likely due to an infective colitis or lymphomatous infiltration; the latter was confirmed by biopsy.
NSAID-associated enteropathy—A confident diagnosis of Crohn's disease was made in one patient who was subsequently found to have ileal strictures due to NSAID ingestion. The angiogram revealed features identical to those seen in other patients in this series with confirmed Crohn's disease—namely, several loops of hyperemic ileum, skip lesions, early and prominent venous return, and two hyper-vascular ileocolic lymph nodes. The angiographic features of this condition have not, to our knowledge, been described previously.
Vascular engorgement—Two additional patients underwent laparotomy because of persistent anemia, both of whom had undergone previous abdominal surgery. In one patient, abnormal edematous bowel was found and resected. Histology showed only vascular engorgement without evidence of Crohn's disease. In the other patient, there were multiple adhesions and congested small-bowel vessels but no evidence of IBD on histology. It is presumed that postoperative adhesions were the cause of the vascular engorgement and the abnormalities noted at angiography.
Diverticular disease—The patient with active contrast medium extravasation in the ascending colon due to diverticular disease also had diffuse increased vascularity of distal ileal loops associated with early venous return. Although it was obvious that the source of hemorrhage was the colon, a second diagnosis of small-bowel inflammatory disease was suggested but was not confirmed at subsequent laparotomy. The cause of this angiographic abnormality is therefore unclear, although it may have been reactive due to adjacent active inflammation in the right colon.
One of questions that arises from this study is why the diagnosis of IBD had not been made in as many as 14 patients before they were referred for angiography. This may be explained partly by the fact that this diagnosis may be difficult to make, particularly when there are no gastrointestinal symptoms other than chronic bleeding as was the case in many of these individuals. A criticism that could be made of this study, however, is that review of the small- and large-bowel imaging studies performed on the patients before arteriography was not possible. One of the problems with the provision of a tertiary referral service for visceral angiography is that previous imaging rarely accompanies the patient and there is a great reliance, therefore, on reports from the referring hospital that all relevant noninvasive investigations not only have been performed but also did not show any abnormality. It is likely, if review of all previous imaging had been possible, that angiography might have been obviated in some patients either because an abnormality was present that had not been recognized at the time of the investigation or because important noninvasive investigations had not been performed.
The introduction of two fairly recent investigations, MDCT and capsule endoscopy, is likely to have an effect on the role of visceral angiography in chronic gastrointestinal bleeding; neither of these techniques was routine during the time period of this study. Several articles have documented the usefulness of both these techniques for the assessment of small intestinal abnormalities and disease including IBD [12, 14-16], and it is likely that the requirement for visceral angiography in this group of patients will be reduced.
In conclusion, in a group of patients referred for angiography to investigate acute or chronic gastrointestinal bleeding, IBD can be expected to be the cause in about 5% and may be diagnosed on the basis of angiographic findings in those individuals without a preexisting diagnosis of the condition. Radiologists should be aware of the angiographic abnormalities associated with IBD and be prepared to consider this diagnosis even in individuals in whom the results of other investigations that might be expected to diagnose IBD have been reported as normal.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
