| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Radiology, Section of Abdominal Imaging and Intervention,
Brigham and Women's Hospital, Harvard Medical School, 75 Francis St., Boston,
MA 02115.
2 Department of Pathology, Brigham and Women's Hospital, Harvard Medical School,
Boston, MA 02115.
Received April 4, 2003;
accepted after revision June 2, 2003.
Address correspondence to K. J. Mortelé
(kmortele{at}partners.org).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. Patients (n = 22) with a history of allogeneic bone marrow transplantation and pathologically confirmed acute gastrointestinal graft-versus-host disease underwent contrast-enhanced (n = 13) and unenhanced (n = 9) CT. CT scans were retrospectively evaluated for intestinal and extraintestinal abnormalities by two radiologists who were unaware of the biopsy results. The CT findings were correlated with the pathologic grade of disease severity using the Pearson's correlation coefficient and the two-tailed nonparametric Spearman's rank correlation coefficient.
RESULTS. CT features included small-bowel wall thickening (22/22), engorgement of the vasa recta adjacent to affected bowel segments (20/22), stranding of the mesenteric fat (16/22), large-bowel wall thickening (13/22), bowel dilatation proximal to thickened wall segments (5/22), ascites (10/22), periportal edema (8/22), mucosal enhancement (7/13), and serosal enhancement (4/13). The wall thickening had a discontinuous distribution in nine patients (41%). Bowel wall thickening was associated with high-grade graft-versus-host disease in patients in whom the distal esophagus (p = 0.015), ileum (p = 0.034), or ascending colon (p = 0.05) was involved. Increasing numbers of thickened bowel segments correlated with high-grade graft-versus-host disease (r = 0.548, p = 0.008). Increasing numbers of abnormal CT findings did not correlate with high-grade graft-versus-host disease (r = 0.117, p = 0.604).
CONCLUSION. A variety of bowel abnormalities can be seen on CT in patients with acute graft-versus-host disease. CT findings associated with high-grade graft-versus-host disease are thickening of the distal esophagus, ileum, or ascending colon, as well as increasing numbers of thickened bowel wall segments.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Because abdominal symptoms are nonspecific, imaging of the abdomen is frequently performed to evaluate patients for abscess or perforation of the bowel wall. The prognosis of patients with acute graft-versus-host disease is dependent on whether they receive early treatment with immunosuppressive medications; thus, a prompt and accurate diagnosis is essential [4, 5]. The CT appearance of acute gastrointestinal graft-versus-host disease in an adult population has been described only briefly in the literature [6, 7]. The purpose of this study was to evaluate the CT findings in adults with biopsy-confirmed gastrointestinal graft-versus-host disease and to correlate the CT findings with the pathologic results.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
At presentation with suspected acute gastrointestinal graft-versus-host disease, patients had symptoms such as nausea, vomiting, abdominal pain, anorexia, diarrhea, and intestinal bleeding. All patients underwent endoscopic examination with biopsies of the upper and lower gastrointestinal tracts to confirm the diagnosis, determine disease severity, and exclude coexisting infectious enteritis.
Imaging Technique
Abdominopelvic CT was performed using single- or multidetector CT units
(Somotom Plus 4 or Volume Zoom, Siemens Medical Systems, Erlangen, Germany).
Contrast-enhanced CT scans (slice thickness, 3 mm; reconstruction interval, 5
mm) were obtained 60–70 sec after IV administration of 100 mL of 300 mg
I/mL iopromide (Ultravist 300, Berlex Laboratories, Montvale, NJ), injected at
a rate of 3.0 mL/sec using a power injector. Opacification of the digestive
tract was achieved with oral administration of 900 mL of a contrast agent
(Readi-Cat, E-Z-EM, Westbury, NY) before the examination.
Image Analysis
CT scans were retrospectively evaluated by two radiologists who were
unaware of the biopsy results, with the final interpretations made by
consensus. The gastrointestinal tract was divided into 10 segments (distal
esophagus, stomach, duodenum, jejunum, ileum, ascending colon, transverse
colon, descending colon, sigmoid colon, and rectum) and was evaluated for the
presence of wall thickening (defined as > 3 mm) of the distal esophagus,
stomach, small intestine, or colon; bowel dilatation (defined as 
3 cm for
the small bowel and 8 cm for the colon); and fluid-filled loops of bowel,
pneumatosis, mucosal enhancement, or serosal enhancement in any location. The
presence of extraintestinal findings was also assessed; these included
engorgement of the vasa recta, stranding of the mesenteric fat, ascites,
mesenteric lymphadenopathy, gallbladder dilatation, gallbladder wall
thickening or enhancement, biliary sludge, pericholecystic fluid,
hepatomegaly, periportal edema, splenomegaly, urinary bladder wall
enhancement, and bladder wall thickening.
Histopathologic Analysis
All gastrointestinal biopsy specimens were fixed in formalin and processed
for paraffin sections in the routine fashion. H and E–stained slides
were reviewed and graded by an experienced gastrointestinal pathologist. The
histologic diagnosis of acute graft-versus-host disease was based on
previously described criteria
[8] in patients whose biopsies
showed epithelial cell apoptosis or crypt cell dropout or destruction.
Histologic severity of the graft-versus-host disease was graded (grade range,
I–IV), also using previously described criteria
[9]. All cases were evaluated
for viral cytopathic changes or other evidence of infection.
Statistical Analysis
CT findings were compared with the results of histopathologic examinations.
Data were analyzed using statistical software (SPSS 10.0, Statistical Package
for the Social Sciences, Chicago, IL). Associations between the CT findings
and histopathologically determined disease grade were assessed using the
Pearson's correlation coefficient. A p value of less than 0.05 was
considered statistically significant. Correlation was calculated using the
two-tailed nonparametric Spearman's rank correlation coefficient.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
|
|
|
Extraintestinal abdominal CT findings are summarized in Table 2. The most consistent finding (20/22 patients or 91%) was engorgement of the vasa recta, which was more pronounced adjacent to the thickened bowel wall segments (Fig. 5). Stranding of the mesenteric fat was present in 16 patients (73%) and was more prominent near areas of abnormal bowel wall thickening. Nine patients (41%) had one or more abnormalities affecting the gallbladder–biliary tract.
|
|
Histopathologic Findings
Diagnostic biopsies were obtained from the colon in 15 patients, from the
stomach in three patients, and from more than one site in four patients
(stomach and colon in one patient; duodenum and colon in one patient; duodenum
and stomach in one patient; and ileum, stomach, and colon in one patient).
Histologic analysis showed varying degrees of epithelial cell apoptosis, crypt
cell dropout and destruction, and mucosal inflammation (Fig.
6A,
6B). Grade I graft-versus-host
disease was diagnosed in six patients (32%), grade II in eight patients (27%),
grade III in two (9%), and grade IV in six (32%). Areas of bowel wall
thickening showed submucosal edema with varying degrees of lymphocytic
infiltration. Results of immunohistochemical stains for cytomegalovirus were
negative in all patients.
|
|
CT–Pathologic Correlation
Pathologically high-grade graft-versus-host disease (grade III–IV)
was significantly associated with the CT finding of bowel wall thickening
involving the distal esophagus (p = 0.015), ileum (p =
0.034), or ascending colon (p = 0.05). Increasing numbers of
thickened bowel locations observed on CT scans also correlated with high-grade
disease (r = 0.548, p = 0.008). However, increasing total
numbers of abnormal CT findings did not correlate with high-grade disease
(r = 0.117, p = 0.604).
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Symptoms in patients with suspected gastrointestinal graft-versus-host disease are nonspecific, and imaging is often performed to evaluate for an abdominal source of sepsis [11, 12]. Differentiation of gastrointestinal graft-versus-host disease from infectious enterocolitis is crucial because the former is treated with immunosuppressive agents, a treatment that is contraindicated in infectious enterocolitis. To our knowledge, the CT appearance of acute gastrointestinal graft-versus-host disease has been only briefly described in the radiology literature, and no large series has been performed in an adult population [6, 7, 13, 14]. Given the high sensitivity of CT for accurately showing the bowel wall and adjacent structures [7], accurate identification of gastrointestinal graft-versus-host disease may obviate further diagnostic testing and expedite therapeutic intervention.
In our series, bowel wall thickening was the most common finding, present in all patients with biopsy-confirmed acute gastrointestinal graft-versus-host disease. On CT scans, all patients had small-bowel involvement, and 59% of patients also had wall thickening of the large bowel. The areas of thickening corresponded histologically with submucosal edema and scattered inflammatory cells. The wall thickening was discontinuous in 41% of the patients, with normal bowel segments separating focally affected portions. We do not believe that this discontinuous pattern of bowel involvement has been previously described in studies of acute gastrointestinal graft-versus-host disease [6, 7, 13–20]. Bowel-loop dilatation, when present, was universally found proximal to a thickened bowel wall segment. The specificity of these findings for diagnosis of graft-versus-host disease as opposed to other causes of enteritis, however, requires further investigation.
The finding of an abnormally enhanced bowel mucosa was noted in 54% of the patients whose scans were obtained with IV contrast medium. This finding is in agreement with that of a prior study of acute graft-versus-host disease in children [14]. The reported frequency of mucosal enhancement in our series may be artificially low because the intraluminal oral contrast medium may have obscured the depiction of an enhancing layer of bowel wall mucosa [14]. Enhancement of the bowel serosa was observed in 31% of patients whose scans were obtained with IV contrast medium. We believe that this CT finding has not been described in previous reports of acute gastrointestinal graft-versus-host disease [6, 7, 13, 14].
Among the extraintestinal findings for acute graft-versus-host disease in our series, engorgement of the vasa recta adjacent to thickened bowel wall segments was the most prevalent finding, observed in 91% of our patients. This CT finding is consistent with the results of a prior color Doppler sonographic study that described increased arterial perfusion of the bowel wall and increased superior mesenteric artery flow in most patients with acute graft-versus-host disease [21]. As reported in earlier studies [6, 7, 13, 14], stranding of the mesenteric fat was observed in 73% of the patients in our series; stranding was more pronounced near regions of thickened bowel wall. However, mesenteric lymphadenopathy was not identified on the scan of any patient, suggesting that the presence of lymphadenopathy should raise the suspicion of other inflammatory or infectious processes in a patient with suspected acute gastrointestinal graft-versus-host disease.
Of note, 41% of the patients in our study showed one or more abnormalities of the gallbladder–biliary tract, including gallbladder wall thickening, pericholecystic fluid, biliary sludge, and enhancement of the gallbladder wall. A previous study concerning the CT appearance of acute gastrointestinal graft-versus-host disease in children described a high frequency of gallbladder wall enhancement and dilatation [14]. Acalculous cholecystitis is a feared complication in recipients of bone marrow transplants, with a high mortality rate in patients undergoing operative interventions such as cholecystectomy [22, 23]. Considering the significant percentage of patients in our series with abnormalities of the gallbladder–biliary tract, acute gastrointestinal graft-versus-host disease should be considered as the potential cause of acalculous cholecystitis in bone marrow transplant recipients.
The histopathologic severity of acute gastrointestinal graft-versus-host disease is determined by assessing the involvement of the intestinal tract. Grade I graft-versus-host disease is characterized as mild disease; grade II, as moderate; and grades III–IV, as severe [24]. Grading is important in evaluating the response to prophylaxis or treatment and in determining prognosis. Patients with moderate to severe graft-versus-host disease have a significantly higher mortality rate than those with the mild form of the disease [25]. Overwhelming sepsis is the primary cause of death in patients with acute gastrointestinal graft-versus-host disease [26]. In our study, histopathologic grades III and IV graft-versus-host disease were significantly associated with the CT finding of wall thickening involving the distal esophagus, ileum, or ascending colon. Similarly, increasing numbers of thickened bowel locations observed on CT scans correlated with high-grade graft-versus-host disease. The presence of these CT findings, therefore, increases the likelihood that severe acute graft-versus-host disease is present and that the patient with such findings has a poor prognosis. However, we found that increasing numbers of abnormal findings on abdominal CT did not correlate with pathologically confirmed high-grade disease.
Our study has three important limitations. It was a nonrandomized retrospective study. We had no control group of patients who were believed to have graft-versus-host disease but whose pathologic results for the disease were negative; thus, our ability to draw conclusions regarding the specificity of the CT findings for acute gastrointestinal graft-versus-host disease was limited. Finally, the diagnosis of graft-versus-host disease was presumptive in the nine patients who showed only CT abnormalities of the small bowel. However, even in patients such as these, the presumptive diagnosis of graft-versus-host disease is relatively solid if small-bowel abnormalities are seen on CT, results of cultures are negative, and biopsies of other sites (stomach, duodenum, or colon) reveal graft-versus-host disease, with no findings suggestive of viral infection present.
In summary, the CT appearance of acute gastrointestinal graft-versus-host disease in adults includes bowel wall thickening with or without proximal dilatation, engorgement of the vasa recta, mesenteric fat stranding, mucosal and serosal enhancement, gallbladder–biliary tract abnormalities, and ascites. The CT findings associated with high grade graft-versus-host disease were thickening of the distal esophagus, ileum, or ascending colon, as well as increasing numbers of thickened bowel wall segments.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  D. Ketelsen, W. Vogel, W. Bethge, M. Werner, K. Dietz, C. D. Claussen, and M. Horger Enlargement of the Common Bile Duct in Patients With Acute Graft-Versus-Host Disease: What Does It Mean? Am. J. Roentgenol., September 1, 2009; 193(3): W181 - W185. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. L. Irani, C. S. Cutler, E. E. Whang, T. E. Clancy, S. Russell, R. S. Swanson, S. W. Ashley, M. J. Zinner, and C. P. Raut Severe Acute Gastrointestinal Graft-vs-Host Disease: An Emerging Surgical Dilemma in Contemporary Cancer Care Arch Surg, November 1, 2008; 143(11): 1041 - 1045. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Stelljes, S. Hermann, J. Albring, G. Kohler, M. Loffler, C. Franzius, C. Poremba, V. Schlosser, S. Volkmann, C. Opitz, et al. Clinical molecular imaging in intestinal graft-versus-host disease: mapping of disease activity, prediction, and monitoring of treatment efficiency by positron emission tomography Blood, March 1, 2008; 111(5): 2909 - 2918. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Macari, A. J. Megibow, and E. J. Balthazar A Pattern Approach to the Abnormal Small Bowel: Observations at MDCT and CT Enterography Am. J. Roentgenol., May 1, 2007; 188(5): 1344 - 1355. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. M. Erturk, K. J. Mortele, C. A. Binkert, J. N. Glickman, M.-R. Oliva, P. R. Ros, and S. G. Silverman CT features of hepatic venoocclusive disease and hepatic graft-versus-host disease in patients after hematopoietic stem cell transplantation. Am. J. Roentgenol., June 1, 2006; 186(6): 1497 - 1501. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
