| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Clinical Observations |
1 Service de Radiologie A, CHU Jean Minjoz, 3 Bvd Fleming, 25030
Besançon, France.
2 Service de Chirurgie Digestive, CHU Jean Minjoz, Besançon,
France.
Received March 19, 2008;
accepted after revision June 9, 2008.
Address correspondence to E. Delabrousse
(eric.delabrousse{at}orange.fr).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CONCLUSION. In cases of SBO, although some CT features of bezoars and small-bowel feces overlap, a well-defined mass mottled with gas bubbles associated with an encapsulating wall, the newly described "floating fat-density debris" sign, and a lesion in the stomach that appears similar to the obstructing mass is typical of a small-bowel bezoar; an isolated amorphous mass mottled with gas bubbles is typical of small-bowel feces.
Keywords: bezoar • CT • small-bowel feces • small-bowel obstruction
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
However, first described in 1995 by Mayo-Smith et al. [2] in patients with SBO, the "small-bowel feces" sign, which corresponds to the presence of particulate material mixed with gas bubbles in the small bowel, may simulate a bezoar. Only a few reports in the literature describe precisely the imaging findings of small-bowel feces [2–5]. To our knowledge, the similarities and differences in the CT appearance between small-bowel bezoar and small-bowel feces have not been previously addressed in a comparative analysis. Nevertheless, this distinction is clinically important because small-bowel feces are very often present in adhesive SBO, the treatment of which is conservative, whereas impacted small-bowel bezoar requires surgery to correct the condition.
The purpose of our study was to evaluate the accuracy of CT in differentiating small-bowel bezoar from small-bowel feces in cases of SBO.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CT Protocol
All of the CT scans that we reviewed were obtained with a 64-MDCT scanner
(Brilliance 64, Philips Healthcare). The CT protocol routinely used in our
institution for evaluation of suspected SBO was performed in each patient. All
patients underwent contrast-enhanced CT. The CT para meters were as follows:
0.625-mm detector collimation, 2.0-mm slice thickness, and 1.0-mm re
construction interval. Scanning began 70–90 seconds after the start of
an IV injection of 120 mL of contrast material (iomeprol [Iomeron 400, Bracco
Imaging]) delivered at a rate of 2–4 mL/s using a power injector
(EnVision CT, MedRad). No patient was given an oral contrast agent. Images
were reconstructed with a soft-tissue algorithm.
Imaging Analysis
The 27 CT examinations were evaluated randomly and in consensus by two
abdominal radiologists blinded to the final diagnosis. To enhance the
sensitivity of CT for gas detection, all of the CT images were assessed using
an adapted window setting (level, –50 HU; width, 500 HU) to allow air
densities to be differentiated from fat densities. The length of the
intraluminal lesion was assessed using automatically generated multiplanar
reformations in every case.
The CT features that were evaluated in our study were those previously reported in patients with a small-bowel bezoar: well-defined ovoid shape (defined by an ovoid lesion distinct from the surrounding intraluminal fluid), short length (< 10 cm), encapsulating wall (defined by a high-attenuation boundary of the lesion), and a lesion in the stomach that appears similar to the obstructive lesion in the small bowel.
A newly proposed CT sign, the "floating fat-density debris" sign, was positive when images showed fat-density debris floating in the lumen of a dilated bowel, proximal to the obstructive lesion. The fat density was proven by region-of-interest (ROI) measurements ranging from –50 to –150 HU.
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
The length of the mass was short in five of eight (63%) bezoars and long in 12 of 19 (63%) small-bowel feces. Small-bowel bezoars were not found to be significantly shorter in length (p = 0.398) than small-bowel feces.
An encapsulating wall was diagnosed in three of eight (38%) bezoars; none of the small-bowel feces had an encapsulating wall. The presence of an encapsulating wall was found to be statistically more significant (p = 0.019) in bezoars than in small-bowel feces (Fig. 3).
|
Bezoars were located in the ileum in six of eight (75%) patients with SBO, and small-bowel feces were seen at the same location in 17 of 19 patients (89%) with SBO. The location of the obstructive mass in the two patients groups was not significantly different (p = 0.558).
A lesion that appeared to be similar to the obstructing small-bowel mass was diagnosed in the stomach in five of eight (63%) patients with SBO secondary to a bezoar and two of 19 (11%) patients with SBO secondary to small-bowel feces. Bezoars were found to be significantly associated with the presence of similar-appearing lesions in the stomach (p = 0.011), whereas this finding was not significantly associated with small-bowel feces (Fig. 4).
|
|
|
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Considering that the clinical symptoms of SBO resulting from a small-bowel bezoar do not differ from those of SBO resulting from other causes, bezoars are often revealed on CT scans [8]. The typical small-bowel bezoar CT findings are an intraluminal ovoid or round mottled-appearing mass with soft-tissue density containing air in its interstices and outlined by fluid or oral contrast material in the dilated small bowel at the site of the obstruction [9]. According to some authors, this appearance of obstructing small-bowel bezoars on CT scans is sufficient for the diagnosis of a small-bowel bezoar [1, 7]. However, Quiroga et al. [9] reported that this appearance may be similar to the small-bowel feces sign.
The small-bowel feces sign on CT was first described in 1995 by Mayo-Smith et al. [2]. It is defined by the presence of particulate matter mingled with gas bubbles in the lumen of dilated loops of small bowel, resembling the appearance of stool in the colon on CT scans [3]. The small-bowel feces sign is most often seen in distal small-bowel loops [3]. The prevalence of the small-bowel feces sign reported in the first studies was low (7–8%) [2, 3, 5]. Curiously, this prevalence was high (56%) in a large and more recent series of 34 patients with SBO [4]. According to Lazarus et al. [4], the small-bowel feces sign is usually seen immediately proximal to the level of the obstruction. Therefore, the small-bowel feces sign may be helpful in recognizing the exact site and cause of SBO [3, 4].
Even if the small-bowel feces sign is reported in the small bowel of patients without SBO [10], the small-bowel feces sign seems to be very indicative of SBO [3]. Catalano [5] observed that all patients with SBO in whom the small-bowel feces sign was present had an insidious onset of symptoms. Catalano stated that the small-bowel feces sign is more likely to be present if the obstruction developed progressively, resulting in sufficient time for the effects of slowed transit, decreased absorption, and increased secretion to occur [3, 10]: It is not the degree of obstruction that appears to be important in the formation of small-bowel feces, but the chronicity of the process [4].
Several similarities and differences between small-bowel bezoars and small-bowel feces on CT scans have already been reported in the literature. Ko et al. [1] noted that in contrast to the small-bowel feces sign, which appears more amorphous than defined and affects longer segments, a small-bowel bezoar is a well-defined, focal, ovoid, intraluminal mass with a mottled gas pattern at the site of the obstruction [1]. According to Ripollés et al. [7], small-bowel feces yield an elongated form of greater length and a less compact nature than bezoars. Vitellas et al. [11] reported that small-bowel feces are more amorphous and affect longer segments than bezoars. Zissin et al. [12] also reported that small-bowel feces are typically longer than bezoars. Moreover, in that series, small-bowel bezoars were at the transition zone between dilated and collapsed small-bowel loops, whereas small-bowel feces appeared to be in dilated small-bowel loops. In a recent study of 19 patients with SBO secondary to phytobezoars, Kim et al. [8] observed that, although statistical analysis between bezoars and small-bowel feces was not performed, small-bowel feces tend to be more tubular in shape without evidence of an encapsulating wall and in the distal small-bowel loops.
To our knowledge, our study is the first to address in a comparative analysis the similarities and differences in the CT appearances of small-bowel bezoars and small-bowel feces. Among the 27 examinations of our series, we observed that the eight small-bowel bezoars and 19 small-bowel feces were all identified as intraluminal mottled-appearing masses containing air bubbles and were located immediately proximal to the transition point of the SBO. These results confirm that the site and the global aspect of bezoars and small-bowel feces are quite similar on CT scans. However, using a pattern approach to the CT evaluation of the two groups, we observed several statistically significant differences between small-bowel bezoars and small-bowel feces. Contrary to previously published results, we found that a well-defined shape, although close to being statistically significant, is not seen more in bezoars than in small-bowel feces. Moreover, small-bowel bezoars did not show a significantly shorter length than did small-bowel feces. An encapsulating wall, as first described by Kim et al. [8], was diagnosed at CT in 38% of the bezoars we observed. None of small-bowel feces had an encapsulating wall. Thus, although infrequent, this CT finding seems to be very specific for bezoars.
In our series, the jejunum was not the preferential site of small-bowel bezoars and, as encountered in small-bowel feces, bezoars tend more frequently to become impacted in the ileum, which corresponds to the narrowest portion of the small bowel. Therefore, the jejunal or ileal location of a lesion should not be used for the differential diagnosis in distinguishing between small-bowel bezoars and small-bowel feces.
A lesion similar to the obstructive mass diagnosed in the stomach was assumed to be very indicative of bezoars. Actually, we found a gastric bezoar in 63% of patients with a small-bowel bezoar. Compared with the results of previous studies, this incidence is quite high. Vitellas et al. [11] reported that a concurrent gastric bezoar was found in only 17–21% of their patients with a small-bowel bezoar. The explanation for such a difference in incidence may be that bezoars may go undetected if the CT scans are read at a standard abdominal soft-tissue window setting (level, 50 HU; width, 350 HU). Modifying the window setting by reducing the level to –100 HU makes it possible to better identify bezoars in the stomach [7]. Using almost the same procedure in our study, we probably allowed increased sensitivity for detecting gastric bezoars and probably improved the detection of the small-bowel feces sign and the floating fat-density debris sign by selecting a window setting for differentiating air density and fat density. The results of our study indicate that the presence of a similar lesion in the stomach is of important value in differentiating bezoars from small-bowel feces.
The floating fat-density debris sign corresponds to the presence of fat-density debris floating in the dilated bowel loops proximal to the obstructive lesion. With visibility largely improved by the use of an adapted window setting (level, –50 HU; width, 500 HU), this sign was seen in 100% of patients with bezoars in our study. The fat density of the floating debris was confirmed by ROI measurements ranging from –50 to –150 HU. We believe that this odd density for intraluminal debris probably results from the sum of the tissue density of the debris and the very negative density of the air in which it is trapped. To our knowledge, this sign has not yet been studied in the setting of bezoars. This CT sign has a excellent sensitivity (100%) and a good specificity for bezoars and should thus be retained.
There are several limitations to our study. First, although surgical proof was obtained for the bezoars, the small-bowel feces sign remained only a CT diagnosis with surgical exclusion of the diagnosis of bezoar. A second limitation is obviously the low number of cases in our series, which reflects the infrequency of small-bowel bezoar and small-bowel feces in SBO. Finally, our study was retrospective.
In conclusion, CT may be considered the imaging technique of choice to differentiate small-bowel bezoar from small-bowel feces in cases of SBO. Although some CT findings of small-bowel bezoar and small-bowel feces encountered in our series overlap, we found that a well-defined mass mottled with gas bubbles associated with an encapsulating wall, the newly described floating fat-density debris sign, and a lesion in the stomach that appears similar to the obstructive mass is typical of small-bowel bezoar; an isolated amorphous mass mottled with gas bubbles is typical of small-bowel feces.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
