| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Surgery, Mayo Clinic, 13400 E. Shea Blvd., Scottsdale, AZ
85259.
2 Department of Diagnostic Radiology, Mayo Clinic, 13400 E. Shea Blvd.,
Scottsdale, AZ 85259.
3 Division of Gastroenterology and Hepatology, Mayo Clinic, Scottsdale, AZ
85259.
Received December 10, 2001;
accepted after revision May 20, 2002.
Address correspondence to J. M. Collins.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. We retrospectively reviewed the records and radiologic studies of 13 patients with known intramural small-bowel hemorrhage.
RESULTS. The mean age at presentation was 64 years. Sixty-two percent of patients had warfarin toxicity. The diagnosis was evident on CT performed in all patients. Small-bowel obstruction was present in 85% of patients, and biliary obstruction was present in 8%. A single hematoma was present in 85% of patients, and multiple hematomas were present in 15%. The jejunum was the most common site of the hematoma (69%), followed by the ileum (38%) and duodenum (23%). The hematoma extended into the cecum in 15% of patients. The estimated average length of the hematoma was 23 cm, and the shortest segment was 8 cm. Resolution of the hematoma was seen on CT as early as 1 week after onset. Eleven patients (85%) with non-extensive hematomas were dismissed from the hospital without any short- or long-term complications (mean follow-up, 35 months). Two patients with extensive hematomas involving more than half the length of the small intestine died.
CONCLUSION. Spontaneous intramural small-bowel hematoma is rare. It occurs in patients who receive excessive anticoagulation with warfarin or who have some other risk factor for bleeding. CT characteristics include circumferential wall thickening, intramural hyperdensity, luminal narrowing, and intestinal obstruction. Early diagnosis is crucial because most patients are treated nonoperatively with a good outcome.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Because of the rarity of spontaneous intramural small-bowel hematoma, no single institution, to our knowledge, has gathered substantial experience treating patients with this condition. Furthermore, the CT and sonographic findings have been reported only rarely [17, 19,20,21,22]. The aim of our study was to review the imaging findings and outcomes of patients with spontaneous intramural small-bowel hematoma. With a total of 13 patients gathered from an experience spanning 17 years, we believe this is the largest single series in the medical literature from an institution reviewing this disease.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Two patients died during hospitalization. Both of these patients had extensive hematomas and sepsis at presentation. One extensive hematoma involved the duodenum and the entire jejunum. The other extensive hematoma stretched from the proximal jejunum to the ascending colon.
The other 11 patients (85%) had nonextensive hematomas with an average length of 23 cm. All these patients were dismissed from the hospital without short-term complications. At the last follow-up (mean, 35 months), four of these patients had died of causes unrelated to the previous spontaneous intramural small-bowel hematoma, and seven patients were alive and well. None of the patients had recurrence of intestinal tract hematoma, abdominal pain, or intestinal tract obstruction.
Imaging Findings
At presentation, CT was performed in all patients; upper gastrointestinal
tract studies with follow-through of the small bowel, in two patients; and
enteroclysis, in one patient. Follow-up CT was performed in seven patients,
and upper gastrointestinal tract studies were performed in one patient.
Diagnosis was evident on CT performed in all patients.
Figures 1A,1B,2A,2B,3A,3B,4,5,6A,6B,7,8 represent the findings observed in some patients. CT characteristics included circumferential bowel wall thickening, luminal narrowing, and intestinal tract obstruction. Intramural hyperdensity was noted in five patients. A stacked-coin appearance was also noted on upper gastrointestinal tract or enteroclysis studies in these five patients. None of our patients had pneumatosis intestinals or air in the portal circulation. Small-bowel obstruction was present in 11 patients (85%); it was partial in 64% and complete in 36% of this group. Biliary obstruction was present in only one patient with a large duodenal hematoma that obstructed the ampulla of Vater.
|
|
|
|
|
|
|
|
|
|
|
|
A single hematoma was noted in 11 patients (85%), and multiple hematomas (involving separate segments of bowel) were noted in two patients. The jejunum was most commonly involved with the hematoma (69%), followed by the ileum (38%) and the duodenum (23%). The small-bowel hematoma extended into the cecum in two patients. The small bowel was extensively involved with the hematoma in two patients. Intraperitoneal fluid was seen in seven patients.
The length of the involved segment (mean, 23 cm) could be estimated from CT performed in eight patients. The shortest involved small-bowel segments measured 8 cm and affected the duodenum and the jejunum separately. Follow-up CT scans in seven patients showed resolution of the hematoma without evidence of intestinal obstruction. Resolution of the hematoma was noted on CT as early as 1 week after onset.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Most patients present with spontaneous intramural small-bowel hematoma after having symptoms for several days [1, 2, 7,8,9, 12, 16, 23,24,25]. The hemorrhage is usually located in the submucosal layer of the bowel and originates from a small vessel that produces slow bleeding [7, 8]. In addition to intramural bleeding, intraluminal, intramesenteric, and retroperitoneal hemorrhage can occur, especially when the duodenum is involved [12, 16]. Hemorrhagic ascites can be present and is related to leakage of blood from an engorged, thickened, and inflamed bowel wall with submucosal bleeding extending into all layers [1, 2, 4,5,6, 8, 9, 24, 25].
In our study, spontaneous small-bowel hematomas most often were single lesions and most commonly involved the jejunum, followed by the ileum and the duodenum. These findings are different from those seen in traumatic small-bowel hematomas, which most commonly affect the duodenum [26]. Furthermore, traumatic hematomas of the small bowel tend to be focal.
The involved bowel appears longer in spontaneous hematomas than in traumatic hematomas, as shown in our study and in those of others [2, 4, 19]. We saw no spontaneous intramural small-bowel hematoma that was less than 8 cm in length. The average length of a nonextensive intramural hematoma was 23 cm. We find that measuring the length is useful to help distinguish spontaneous intramural small-bowel hematoma from tumor infiltration by metastasis or lymphoma. We would expect that a neoplasm that involves a length of 23 cm of small bowel would have a prominent extramural mass component, which was conspicuously absent from our series of spontaneous small-bowel hematomas.
Small-bowel hematomas can extend into the colon, as seen in two of our patients. However, isolated cases of intramural colonic hematomas without small-bowel involvement are rare, and we noted only two patients in a review of 67 published cases [1,2,3,4,5,6,7,8,9, 11, 12, 15,16,17, 19,20,21, 23,24,25, 27, 28]. We saw no spontaneous hematoma involving only the colon in any of our 13 patients.
Before CT, the diagnosis of intramural small-bowel hematoma was made occasionally on upper gastrointestinal tract series and was more often established only at exploratory laparotomy or autopsy. Several radiographic features were helpful in making the diagnosis. These included a stacked-coin appearance representing a thickening of folds with sharp demarcation and crowding of valvulae conniventes, a picket fence appearance showing spikelike projections of barium between adjacent thickened mucosal folds, abrupt proximal and distal transition points, and luminal narrowing of a rigid and nondistensible segment of intestine [2, 3, 16, 23,24,25]. However, despite abnormal findings reported in 92.5% of patients with an intramural small-bowel hematoma, findings on barium studies were not necessarily diagnostic or specific for this condition [18, 28]. Several conditions such as lymphoma, inflammatory bowel disease, bowel infarction, and tuberculosis can mimic the findings of intestinal tract hematoma on upper gastrointestinal tract series. Therefore, it is difficult to make a definite diagnosis of this disorder solely from the results of a gastrointestinal tract barium study.
Advances in cross-sectional imaging, primarily CT, have contributed to the early and accurate recognition of this condition. Unenhanced CT findings have been reported previously [17] and include homogeneous and symmetric intramural thickening with hyperdense material (30-80 H) in the bowel wall. The hyperdensity of the bowel wall can be seen during the first 10 days after the onset of symptoms and can be helpful in distinguishing this condition from other infiltrative processes that can affect the small bowel, such as malignancy and infection [17]. However, hyperdensity decreases as the hematoma ages and evolves into a hypodense area that can mimic an intramural cystic lesion [17, 22].
In our study, the diagnosis was evident from the CT scans obtained in all patients. Only two patients underwent surgical exploration, and neither of these required bowel resection. As a result, the gold standard of a histologic diagnosis was absent from our patients. The diagnosis of spontaneous intramural small-bowel hematoma is determined from the CT findings of small-bowel wall thickening and obstruction in patients who received excessive anticoagulation or have other increased risks for bleeding and who present with abdominal pain. The diagnosis is confirmed by the spontaneous resolution of the findings on a follow-up CT scan.
Complete resolution of the hematoma usually occurs within a few weeks after onset. Imaging studies in the past have documented healing of the intestine within 2 months [2, 19, 25]. In our study, resolution was noted on CT as early as 1 week after onset. Small-bowel lesions persisting for more than 2 months should raise the radiologist's suspicion for other causes.
In our opinion, during CT performed on a patient with complete bowel obstruction, it is best to avoid the use of oral contrast material. Water rather than positive oral contrast material can provide detailed images of the gastrointestinal tract. We recommend the use of IV contrast material because it is helpful in the visualization of many gastrointestinal tract disorders (i.e., ischemic or gangrenous gut, Crohn's disease, or malignancy) and does not hinder the ability of the radiologist to make the diagnosis of spontaneous intramural small-bowel hematoma. Despite this recomendation, many patients in this series received oral contrast material with CT before the diagnosis of bowel obstruction was established. Many patients did not receive IV contrast material because their clinical condition did not allow it.
The sonographic appearance of acute intramural small-bowel hematoma has been described and consists of a thickened and echogenic submucosal layer [20, 22]. However, this abnormality is not specific for intestinal tract hematomas and can be seen in a wide spectrum of disorders, including inflammatory bowel disease, pseudomembranous colitis, infectious colitis, ischemic colitis, hemorrhagic infarction, graft-versus-host disease, and mesenteric venous obstruction [22]. Therefore, the utility of sonography for diagnosing this condition is questionable.
MR imaging findings of subacute duodenal hematomas have been reported previously [28]. A T2-weighted high-signal-intensity ring sign also has been described [28]. However, no data are available on the imaging characteristics of acute intramural hemorrhage or small-bowel hematomas other than those that occur in the duodenum. Furthermore, in view of its higher cost and currently lower temporal resolution, we do not think that MR imaging should be performed as the initial diagnostic study for patients with spontaneous intramural small-bowel hematoma who present with acute abdominal pain.
In conclusion, although nontraumatic spontaneous intramural small-bowel hematoma is rare, its incidence is expected to increase as a result of a growing number of patients with hematologic malignancies receiving chemotherapy and an aging population requiring more chronic long-term anticoagulation treatment.
Prompt and early recognition of nontraumatic spontaneous intramural small-bowel hematoma by the radiologist is a crucial contribution to patient treatment that may prevent unnecessary exploratory surgery. All patients with nonextensive intramural hematomas, with an average length of 23 cm, had a good outcome with medical treatment alone and without bowel resection.
Patients with extensive intramural hematomas involving more than half the length of the small intestine may represent an important subgroup with a poor outcome. Both our patients with extensive intramural hematomas died. Therefore, if extensive intramural small-bowel hematoma is detected on CT, the radiologist should alert the ordering physician to the seriousness of the patient's condition and recommend aggressive medical treatment.
CT findings of small-bowel wall thickening and obstruction in patients presenting with abdominal pain who are undergoing anticoagulation therapy or have other increased risks for bleeding should raise suspicion of spontaneous intramural small-bowel hematoma.
Although mural hyperdensity was present in only five of our 13 patients, we consider this condition pathognomonic of spontaneous intramural small-bowel hematoma when the CT findings of mural hyperdensity occur within an area of circumferential thickening of the small bowel with intestinal obstruction in patients who are overanticoagulated with warfarin.
CT is the imaging technique of choice in the diagnosis of spontaneous intramural small-bowel hematoma and can be complemented with barium upper gastrointestinal tract studies with enteroclysis or follow-through of the small bowel.
The triad of warfarin toxicity, circumferential thickening of the small bowel, and intestinal obstruction is characteristic of spontaneous intramural small-bowel hematoma and was seen in all patients in this series who were overanticoagulated.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  ANSWER Gut, August 1, 2008; 57(8): 1082 - 1082. [Full Text] [PDF]
|
|
|
|
 |
|
 B. R. Weil, T. J. Howard, and N. J. Zyromski Spontaneous Duodenal Hematoma: A Rare Cause of Upper Gastrointestinal Tract Obstruction Arch Surg, August 1, 2008; 143(8): 794 - 796. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. T. Johnson, K. M. Horton, and E. K. Fishman Case 127: Henoch-Schonlein Purpura Radiology, December 1, 2007; 245(3): 909 - 913. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
