| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 All authors: Department of Radiology, Abdominal Imaging Section, Tisch Hospital, New York University Medical Center, 560 First Ave., Ste. HW 207, New York, NY 10016.
Received March 27, 2002;
accepted after revision July 11, 2002.
Address correspondence to M. Macari.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. Findings of 35 CT examinations (19 patients
with small-bowel ischemia and 16 patients with intramural hemorrhage) were
analyzed by two abdominal radiologists for the degree of wall thickening,
location and length of involvement (short, 
15 cm; medium, 16-30 cm; or
long, >30 cm), presence of hemoperitoneum, and pattern of attenuation.
Patency and caliber of the superior mesenteric artery and vein were noted.
Diagnosis was confirmed by laboratory findings, clinical parameters, and
follow-up examinations, or at surgery. A Mann-Whitney U or Fisher's
exact test was used to compare the two conditions for the following features:
wall thickening, location and length of involvement, presence of
hemoperitoneum, and appearance of the target sign.
RESULTS. Among the 35 examinations, 18 abnormal segments with intramural hemorrhage and 19 abnormal segments with ischemia were identified. (Two patients with intramural hemorrhage each had two segments involved.) Mean bowel wall thickness was 11.7 mm (range, 4-25 mm) in patients with intramural hemorrhage and 4.0 mm (range, 1-9 mm) in patients with ischemia. Length of involvement was short in 14 segments with intramural hemorrhage and medium in four segments with intramural hemorrhage; none of the segments with intramural hemorrhage had long involvement. Among the segments with ischemia, length of involvement was medium in three and long in 16; none of the ischemic segments had short involvement. Fifteen (94%) of 16 segments with intramural hemorrhage and six (32%) of 19 segments with ischemia had hemoperitoneum. Seven of the 18 segments with intramural hemorrhage and nine of the 19 with ischemia had a target sign. Segments with intramural hemorrhage exhibited a higher statistically significant degree of wall thickening (p < 0.001), a shorter length of involvement (p < 0.0001), and a higher incidence of hemoperitoneum (p < 0.001) than did segments with ischemia. The two groups were not statistically different in location of involvement (p = 0.12) or in the incidence of the target sign (p = 0.18).
CONCLUSION. Although some of the CT features overlap, a short segment involvement with wall thickening of 1 cm or greater is typical of intramural hemorrhage; a long segment involvement with wall thickening of less than 1 cm is typical of ischemia.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The CT findings of small-bowel intramural hemorrhage and ischemic bowel have been previously reported [2,3,4,5,6]. CT findings of the ischemic small bowel include bowel wall thickening with or without appearance of a target sign, lack of enhancement of the bowel wall, pneumatosis, arterial embolization, or venous thrombosis [3,4,5]. In patients with intramural hemorrhage, some of the CT findings may be similar to those of the ischemic small bowel, such as homogeneous bowel wall thickening, presence of a target sign on IV contrast-enhanced studies, and highly attenuated ascites [2, 6].
Only a few reports in the literature describe the imaging findings of intramural hemorrhage after administration of IV contrast material [2, 6]. Moreover, to our knowledge, the similarities and differences in the CT appearance between these two conditions have not been previously addressed in a comparative analysis. The purpose of our study was to evaluate whether CT differences between small-bowel intramural hemorrhage and ischemia could be discerned if a pattern approach to interpreting imaging findings was applied. We retrospectively evaluated proven cases of small-bowel intramural hemorrhage and small-bowel ischemia for the degree of mural thickening, length and location of intestinal involvement, presence of hemoperitoneum, and pattern of attenuation in an attempt to determine if contrast-enhanced CT scans can be used to diagnose these two conditions.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The CT database runs on FileMaker Pro 5.5 (FileMaker, Santa Clara, CA), a program that allows users to access data by entering keywords related to anatomy and pathology. For this study, the anatomic keyword used was "small bowel," and the pathologic keyword was either "ischemia" or "hemorrhage." Therefore, the database searched for cases of small-bowel ischemia or small-bowel hemorrhage.
Sixty CT examinations with one of these two diagnoses were found to have been performed during the specified time period. Thirty-nine depicted small-bowel ischemia, and 21 depicted small-bowel intramural hemorrhage. Of these 60 cases, 15 were eliminated because oral and IV contrast materials were not used, leaving 45 patients as potential study cases. For our study, we relied on the final clinical diagnosis as opposed to the final imaging diagnosis as proof of either small-bowel hemorrhage or small-bowel ischemia. In 10 of these 45 patients, clinical follow-up data could not be obtained, and these patients were eliminated from the study.
Our series included the remaining 35 patients (17 men and 18 women) who underwent contrast-enhanced CT and for whom clinical follow-up information was available. Nineteen patients had intestinal ischemia, eight men and 11 women who ranged in age from 58 to 93 years (mean age, 71 years). Sixteen patients had intramural hemorrhage, nine men and seven women who ranged in age from 52 to 85 years (mean age, 66 years).
CT Technique
All CT examinations were performed with HiSpeed Advantage or CT/i scanners
(General Electric Medical Systems, Milwaukee, WI). Helical CT data were
acquired with either a slice collimation of 7 mm with a reconstruction
interval every 6 mm or a slice collimation of 5 mm with a reconstruction
interval of 4 mm, a pitch varying between 1.5 and 2.0, 120 kV, and 200-240 mA.
Oral administration of 800 mL of barium sulfate suspension (2.1%
weight/volume) (Readi-CaT 2; E-Z-EM, Westbury, NY) or diluted (2%)
water-soluble contrast material (Gastrografin [meglumine diatrizoate];
Bristol-Meyers Squibb, Wallingford, CT) was begun approximately 1 hr before
scanning, with the patients ingesting the contrast medium in small increments.
Then, 150 mL of nonionic IV contrast material (Ultravist [iopromide], 300 mg
I/mL; Berlex Laboratories, Wayne, NJ) was administered at a rate of 2 mL/sec
via a 22-gauge catheter inserted into an antecubital vein using a power
injector (Envision CT injector; Medrad, Pittsburgh, PA).
Image Interpretation
The 35 CT examinations were evaluated randomly and in consensus by two
abdominal radiologists who had no knowledge of the final diagnosis. In two
patients with intramural hemorrhage, two separate areas of abnormal
small-bowel mural thickening were found. Therefore, 18 abnormal intestinal
segments were evaluated in patients with intramural hemorrhage and 19 abnormal
segments were evaluated in patients with intestinal ischemia. Images were
reviewed on hard copy (n = 22), or if available, images stored on
optical disks were retrieved and restored on a workstation (Advantage Windows;
General Electric Medical Systems) (n = 13) running Windows 3.1
(Microsoft; Redmond, WA).
A pattern approach was used to evaluate each abnormal segment. Features
used to define an abnormal segment were the degree of mural thickening,
location and length of involvement, presence of hemoperitoneum, and pattern of
attenuation. The degree of greatest mural thickening was measured (in
millimeters) in a segment of thickened bowel that was oriented either
longitudinally or axially in the plane of the image. The location of
involvement was defined as being predominantly jejunal, ileal, or jejunoileal.
The length of involvement was arbitrarily defined as short ( 15 cm),
medium (16-30 cm), or long (>30 cm). The pattern of attenuation was defined
as being a target appearance (alternating concentric layers of differing
attenuation), homogeneous attenuation, or decreased attenuation (lower
attenuation than that seen in the adjacent loops of bowel).
Consensus interpretation of the scans was performed for several reasons. First, an exact measurement of the length of involvement could not be obtained, given the random orientation of involved loops of small bowel. However, we used the centimeter scale on the side of each CT image to attempt to place each abnormal segment in one of the three categories. We believed that a consensus evaluation was more likely to resolve any ambiguity in the estimation of this parameter. Second, two of the authors in consensus picked the loop of bowel that showed the greatest degree of thickening and that was, at the same time, well distended. Measurement of collapsed segments was avoided. In addition, the loop of bowel needed to be oriented directly perpendicular or parallel to the z-axis so that overestimation of thickness resulting from the oblique orientation could be avoided. All measurements were calculated by a single author. Wall thickness on the hard copies was measured with a caliper and the centimeter scale on the side of the scan; wall thickness on soft copies was measured with an electronic caliper.
We also used consensus interpretation for evaluation of hemoperitoneum. For soft-copy CT scans, a single author placed a round or oval region of interest in the center of the ascites to determine attenuation. Attenuation measuring more than 30 H was considered evidence of hemorrhage. Because some CT scans were evaluated only on hard copy, a region-of-interest measurement on ascites could not be performed. In these cases, a consensus interpretation was used to determine whether the ascites was highly attenuated relative to other fluid-filled structures, such as renal cysts or the gallbladder.
Statistical Analysis
To evaluate wall thickening, we used a Mann-Whitney U test to
compare segments with ischemia or intramural hemorrhage. Fisher's exact test
was used to compare the two patient groups with respect to the location and
length of involvement and the presence of hemoperitoneum and a target
sign.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
|
|
|
In the remaining 10 patients, the diagnosis was based on imaging and clinical parameters. One patient had thrombosis of the superior mesenteric vein and was treated with anticoagulation therapy (Fig. 3A,3B,3C). In the remaining nine patients, the diagnosis of nonocclusive low-flow mesenteric ischemia was based on clinical parameters. These nine patients had acute cardiac decompensation or an elevated lactic acid level; they experienced improvement in their symptoms after appropriate resuscitation. None of these patients had an alternative condition—such as Crohn's disease or infectious enteritis—that was clinically diagnosed. Moreover, none of these patients had an elevated international normalized ratio.
|
|
|
Statistical Analysis
A total of 18 abnormal small-bowel segments were identified in 16 patients
with submucosal hemorrhage. Nineteen abnormal segments were present in the 19
patients with intestinal ischemia. Using a pattern approach to CT evaluation
of the abnormal small-bowel segments, we found several statistically
significant differences between submucosal hemorrhage and intestinal ischemia
(Table 1).
|
The mean thickness of segments with intramural hemorrhage was 11.7 mm, and the median was 11 mm (range, 4-25 mm). In segments with small-bowel ischemia, both the mean and the median thicknesses were 4.0 mm (range, 1-9). Mural thickening was found to be statistically more significant (p < 0.001) in the segments with intramural hemorrhage than in those with small-bowel ischemia (Figs. 4 and 5).
|
|
In the 18 abnormal segments with intramural hemorrhage, six (33%) were located in the jejunum, eight (44%) were in the ileum, three (17%) were in the jejunoileal region, and one (6%) was in the duodenum. Of the 19 ischemic segments, two (11%) were jejunal, seven (37%) were ileal, nine (47%) were jejunoileal, and one (5%) involved the ileocolic region. The two patient groups were not significantly different in terms of location (p = 0.12).
Fourteen (78%) of 18 segments with intramural hemorrhage showed short segment involvement, four (22%) showed medium-length involvement, and no segment showed long involvement. Of the 10 segments with intestinal ischemia, none showed short segment involvement, three (16%) showed medium-length involvement, and 16 (84%) showed long segment involvement. The abnormal segments with intramural hemorrhage were found to have a significantly shorter length of involvement (p < 0.0001) than did those segments with ischemia (Figs. 6A,6B and 7A,7B).
|
|
|
|
Fifteen (94%) of 16 patients with intramural hemorrhage showed hemoperitoneum (Fig. 8). Hemoperitoneum was present in only six (32%) of 19 patients with intestinal ischemia. A higher incidence of hemoperitoneum was found in patients with submucosal hemorrhage than in those with ischemia (p < 0.001).
|
Regarding the pattern of attenuation, the target sign was present in seven (39%) of 18 segments with intramural hemorrhage, and the attenuation was homogeneous in the remaining 11 (61%). The pattern of attenuation in the 19 ischemic segments showed a target appearance in nine (47%) and was homogeneous in nine (47%). One case of small-bowel ischemia showed decreased attenuation in the involved small bowel (5%). The two patient groups were not significantly different in terms of the pattern of attenuation of the mural thickening (p = 0.18).
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Common causes of small-bowel ischemia are poor perfusion resulting from cardiac failure or hypovolemia, arterial embolus or thrombus, and venous thrombosis [3, 10,11,12]. In recent surgical reviews of mesenteric ischemia, approximately 50% of cases were due to a superior mesenteric artery embolus, 20-25% were due to a superior mesenteric artery thrombosis, 20-25% were due to nonocclusive mesenteric ischemia, and 5% were due to mesenteric venous thrombosis [10, 11, 13].
The clinical presentation of small-bowel ischemia, irrespective of its cause, includes abdominal pain that is out of proportion to the findings at physical examination, nausea, anorexia, vomiting, and bloody diarrhea [2, 10]. Occult blood may be detected in the stool in nearly 50% of patients. In addition, the clinical presentation varies and depends on the severity of the process; the degree of collateral vascular supply; and, if ischemia is due to vascular occlusion, whether the occluded vessel is an artery or a vein [2, 10, 11]. The nonspecific symptoms and clinical evaluation often result in a delay in the diagnosis. The clinical presentation of patients presenting with spontaneous intramural hemorrhage is also nonspecific and may be similar to that of intestinal ischemia. The findings consist of nausea, vomiting, acute abdominal pain, and gastrointestinal bleeding [14].
The CT findings in bowel ischemia evolve as the process in the bowel wall progresses from ischemia to infarction [3, 7]. Initially, there is mural thickening of the involved bowel segment related to intramural edema [7]. As the process progresses, an element of intramural hemorrhage may be seen [3]. Stranding and hazy changes in the perienteric fat are also usually detected. During the early stages of ischemia, a target sign may be identified in the bowel wall related to the intramural edema and preserved mucosal vasculature [3]. As the bowel becomes more ischemic, a decrease in the attenuation (enhancement) of the affected bowel wall may be present, and the intestinal wall may become thin [3, 4]. However, the early CT findings of intestinal ischemia (mural thickening, perienteric stranding, and a target sign) are not specific and may be present in patients with Crohn's disease; infectious enteritis; radiation changes; and, notably, intramural hemorrhage [3, 7, 14].
CT is helpful in establishing the diagnosis of small-bowel ischemia and has been recommended as the definitive way to diagnose the condition [10]. In patients in whom small-bowel mural thickening is present in addition to mesenteric vascular occlusion or pneumatosis, the CT findings strongly suggest intestinal ischemia. A small-caliber aorta and superior mesenteric artery as well as the collapse of the inferior vena cava are helpful secondary findings that suggest hypotensive shock and low-flow nonocclusive ischemia. However, if vascular occlusion is not present or if the CT findings are indeterminate, differentiating ischemia from intramural hemorrhage or inflammatory conditions becomes more difficult.
To better characterize an abnormal loop of bowel detected on CT, we analyzed different specific parameters of the abnormal bowel. The pattern approach used in this retrospective evaluation included parameters related to the appearance of the intestinal wall and secondary findings in the mesentery and vascular structures.
In CT findings related to the bowel wall, segments with intestinal ischemia showed mild wall thickening, whereas segments with intramural hemorrhage showed more marked thickening—a mean of 4.0 mm for intestinal ischemia versus a mean of 11.7 mm for intramural hemorrhage. As stated previously, in irreversible intestinal ischemia, the wall frequently becomes thin, measuring less than 1-2 mm [3]. The target sign has been well described in the literature in patients with intestinal ischemia but reported infrequently in patients with intramural hemorrhage [3, 6, 7]. In our study, although a target sign was seen more often in patients with ischemia, no significant statistical difference in its presence between the two conditions was found. The detection of the target sign in 39% of patients with intramural hemorrhage in this study is probably due to the use of a power injection of IV contrast material at a relatively fast rate (2 mL/sec) during peak bowel wall enhancement. Scanning during peak enhancement may cause the intramural hemorrhage to continue to appear to be less attenuated than the enhancing mucosa and muscularis.
Intramural hemorrhage tends to involve a short segment of small bowel, whereas a greater length of involvement is present in patients with intestinal ischemia. In our CT study, 78% of patients with intramural hemorrhage showed short segmental involvement, but none of the patients with intestinal ischemia showed short segment involvement. Of patients with ischemia, 84% showed long segment involvement. This distribution is probably related to the pathophysiology of intestinal ischemia. In patients with low-flow ischemia, longer segments of small bowel are at risk of hypoxia. Similarly, in patients with superior mesenteric artery obstruction, the embolus usually lodges 3-10 cm from the origin of the vessel, and hence long segments of bowel will be affected [13]. It is unclear why submucosal hemorrhage tends to involve shorter segments. Previous reports of this condition have described its distribution as being segmental or even mimicking a focal mass [6, 15].
Hemoperitoneum was also shown to be significantly more likely to be present in patients with intramural hemorrhage. Hemoperitoneum was present in 94% of patients with intramural hemorrhage but in only 32% of those with intestinal ischemia. The cause of hemoperitoneum is unclear in patients with intestinal ischemia. It may be related to a secondary intramural hemorrhage and leakage of blood from the serosal surface of the bowel wall. Hemoperitoneum is known to occur in patients with closed-loop small-bowel obstruction and, when present, is considered a specific sign for intestinal ischemia in this setting [16, 17].
In our study, most patients (79%) with intestinal ischemia were likely to have had low-flow nonocclusive disease as a cause. Surgical series have reported a high frequency of superior mesenteric artery embolus as the cause of mesenteric ischemia [11]. Although the exact reasons for this difference are unclear, it may be related to the age of our patients because the mean age of the patients with intestinal ischemia in our series was 71 years. However, this discrepancy may also be explained by a selection bias in that patients in previously reported surgical series were more likely to have undergone surgery. Nonocclusive mesenteric ischemia, which often manifests as a less severe form of the small-bowel ischemia, may be managed medically, whereas occlusive disease usually requires surgery.
The aim of our study was to determine if CT differences exist between small-bowel ischemia and intramural hemorrhage. Therefore, we did not attempt to determine the sensitivity and specificity of CT in the diagnosis of small-bowel ischemia or intramural hemorrhage. In previous studies, the reported range of sensitivity has been 37-80% [3]. This range is likely related to differences in CT technique as well as to the cause and severity of the ischemia.
There are several limitations to our study. First, surgical proof was not obtained in most of our patients. Of the 16 patients with small-bowel intramural hemorrhage, all had predisposing factors for bleeding, including anticoagulation therapy in 14 and factor VIII deficiency in two. None of these patients had an elevated lactic acid level, and all improved clinically after the correction of the coagulopathy. In the 19 patients with intestinal ischemia, eight underwent surgery; one patient had intestinal ischemia confirmed at autopsy. One patient had mesenteric vein thrombosis that was treated with anticoagulation therapy. In the remaining nine patients, the diagnosis of low-flow ischemia was based on clinical parameters, including acute cardiac decompensation, an elevated lactic acid level, and an improvement in symptoms after appropriate resuscitation.
A second limitation involves our finding that the length of involvement is an important differentiating feature between the two conditions. It is difficult to estimate the length of involvement of an abnormal small-bowel segment. Although we attempted to determine the length of involvement by placing each abnormal small-bowel segment into one of three categories of involvement (short, medium, or long), this system rendered a gross estimate, not an exact measurement.
A third limitation of our study is that there were a large number of patients with intramural hemorrhage relative to the number of patients with intestinal ischemia, which may reflect a selection bias in the way cases are entered into our CT database. If a patient with abnormal bowel wall thickening also has an initial laboratory value indicating that the international normalized ratio is elevated, that patient's case is likely to be entered into the database as "small bowel-hemorrhage." Regarding intestinal ischemia, other conditions may be considered in the initial differential diagnosis. If clinical parameters or specific CT findings such as the narrowing or occlusion of the superior mesenteric artery or vein are not present, cases of ischemia may be entered into the database as "small-bowel inflammation," not as "small-bowel ischemia." Therefore, more patients with small-bowel ischemia were likely to have undergone CT at our institution than were included in our study.
Finally, only contrast-enhanced examinations were included in this study. The diagnosis of submucosal hemorrhage may be facilitated by the use of unenhanced CT scans through recognition of the highly attenuated hemorrhage in the bowel wall [14]. However, if only an unenhanced CT examination is performed, visualization of thrombus or embolus in the superior mesenteric artery or vein cannot be identified. In patients with abdominal pain who are known to be at risk for intramural hemorrhage, an initial unenhanced CT examination may be warranted. Unenhanced CT scans may be helpful for differentiating a hemorrhage from ischemia by showing homogenous high attenuation (>50 H) in the bowel wall, which is indicative of an intramural hemorrhage [2, 5, 6]. However, because most patients with acute abdominal pain undergo CT imaging with oral and IV contrast materials, this finding is often difficult to discern.
In conclusion, although some CT findings of intramural hemorrhage and intestinal ischemia overlap (i.e., location of involvement, presence of hemoperitoneum, and the target sign), we found that short segment involvement with a thickness of 1 cm or more is typical of intramural hemorrhage and long segment involvement with thickening of less than 1 cm is typical of ischemia. Evaluation of the superior mesenteric artery and vein is important because occlusion may be seen on contrast-enhanced CT scans. Finally, a combination of imaging, medical history, findings of physical examination, and laboratory values are all important in establishing the correct diagnosis.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  R. F. Barajas Jr., B. M. Yeh, E. M. Webb, A. C. Westphalen, L. Poder, and F. V. Coakley Spectrum of CT Findings in Patients With Atrial Fibrillation and Nontraumatic Acute Abdomen Am. J. Roentgenol., August 1, 2009; 193(2): 485 - 492. [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]
|
|
|
|
|
|
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]
|
|
|
|
|
|
J. Ahualli The Target Sign: Bowel Wall Radiology, February 1, 2005; 234(2): 549 - 550. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
