| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 All authors: Division of Abdominal Imaging and Intervention, Massachusetts General Hospital, White 270, 55 Fruit St., Boston MA 02114.
Received February 6, 2003;
accepted after revision March 27, 2003.
Address correspondence to M. G. Harisinghani
(mharisinghani{at}partners.org).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. We performed a retrospective review of 100
consecutive abdominal and pelvic CT examinations in 61 men and 39 women (mean
age, 56 years) with clinical suspicion of renal stone disease. Two
radiologists experienced in abdominal imaging performed qualitative and
quantifiable assessment of the images. Five segments of the colon (ascending
colon, transverse colon, descending colon, sigmoid colon, and rectum) and the
terminal ileum (for 
1 ft [30 cm]) were evaluated for the presence of the
fat halo sign. If the fat halo sign was present, fat density and total
wall-thickness assessments were made. Presence or absence of clinical and
radiologic signs of inflammatory bowel disease was determined. The Student's
t test was used to evaluate the statistical significance, correlating
body weight and presence of the halo sign.
RESULTS. The fat halo sign was seen in 21 (21%) of 100 patients. Of the 21 patients with the fat halo sign, six (29%) had renal stone disease and 15 (71%) had no stone disease. The density value of the halo sign ranged from -18 to -64 H (mean, -41 H). The distribution of the fat halo sign was as follows: the terminal ileum, 4%; the ascending colon, 28%; the transverse colon, 34%; the descending colon, 36%; the sigmoid colon, 14%; and rectum, 10%. No patient with this sign had any remote, recent, or subsequently recorded history of inflammatory bowel disease. A statistically significant relationship (p < 0.001) was seen between the presence of the fat halo sign and body weight distribution, with 16 of 21 patients weighing over 200 lb (90 kg).
CONCLUSION. In the absence of clinical or radiologic evidence of inflammatory bowel disease, the presence of the fat halo sign may represent a normal finding that is possibly related to obesity.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The purpose of this study was to evaluate the presence and frequency of this fat halo sign in patients in whom the reason for undergoing abdominal CT was other than gastrointestinal signs and symptoms.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
All patients underwent abdominal and pelvic CT performed with a helical CT system (HiSpeed CT/i, General Electric Medical Systems, Milwaukee, WI) on different scanners, all of which are density-calibrated on a daily basis. Axial scans were obtained from the diaphragm to the symphysis pubis using a 5-mm slice thickness, a pitch of 1, and a reconstruction interval of 5 mm. No oral or IV contrast material was administered for the initial phase of the examination. In 74% of patients, IV contrast-enhanced CT was performed with the patients in the prone position to determine whether a suspicious calcific density was located in the genitourinary tract.
Two radiologists experienced in abdominal imaging together performed
qualitative and quantifiable assessment of the images. Images were viewed on
clinical workstations at standard soft-tissue parameters (window width, 301 H;
level, 44 H). Differences between reviewers were settled by consensus. Five
segments of the colon (ascending colon, transverse colon, descending colon,
sigmoid colon, and rectum) and the terminal ileum (for 1 ft [30 cm]) were
evaluated for the presence of the fat halo sign. Stratification was deemed
present when a discrete low-density intramural layer was present over three
fourths or greater of the circumference of the bowel wall. Hounsfield unit
measurements on the submucosal low-density layer were made on magnified views
of the section in the segment optimally showing mural stratification. A round
cursor maximally adapted to the low-density middle layer was used to obtain
average Hounsfield unit measurements. The maximal thickness of the bowel wall
was also measured. If a series was performed with the patient in the prone
position, the fat halo measurements were made on a section comparable in
location to that section previously assessed on the supine view. In each
segment of bowel, the number of sections showing the presence or absence of a
layer whose density measured less than -10 H was recorded. Sections with a
low-density layer measuring greater than -10 H were not recorded although they
conceivably could have represented a partial volume of a thin layer of fat.
Additionally, for each segment, the presence or absence of periintestinal fat
stranding, creeping fat, and local (
1 cm of bowel wall) lymphadenopathy
(> 5-mm short axis) as well as central (mesenteric root) lymphadenopathy
(> 5-mm short axis) was recorded. In the patients found to have a fat halo
sign, medical records were reviewed for any previous or present history of
gastrointestinal symptoms attributable to inflammatory bowel disease or prior
imaging and colonoscopy. Records were subsequently searched by the reference
date of the CT examination for clinical or radiologic findings compatible with
inflammatory bowel disease. The longest follow-up was 60 months (mean, 34
months); the longest period since prior clinical assessment was 10 years
(mean, 4.5 years).
Presence of the fat halo sign in relation to the patient's weight was recorded. The significance of body weight for the halo sign was determined statistically using the Student's t test.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
Luminal distention of the bowel in each segment was variable as was that of the section used for measurements. The average bowel wall thickness in the sections containing the measured fat halo sign ranged from 1.1 to 2.4 cm. The selected section invariably showed less distention than others in the same segment. The fat halo sign was not necessarily seen on continuous sections; absence was particularly apparent when a section contiguous to the segment with the fat halo was distended with gas. Periintestinal fat stranding was seen in three (3%) of 100 patients. Findings in these three patients did not show the fat halo sign. A pattern of creeping fat contiguous to the sigmoid colon was observed in one of the 100 patients with no remote or recent history of inflammatory bowel disease. In 11 of 100 patients, five of whom had findings of the halo sign, central enlarged mesenteric lymph nodes were noted. Again no remote, recent, or subsequent history or symptoms of inflammatory bowel disease were specifically detected in these patients.
Of the 21 patients with findings of the halo sign, four (19%) weighed less than 150 lb (68 kg), five (24%) weighed between 150 and 200 lb (68 and 90 kg), seven (33%) weighed between 200 and 250 lb (90 and 113 kg), four (19%) weighed between 250 and 300 lb (113 and 135 kg), and one (5%) weighed more than 300 lb (> 135 kg). A statistically significant relationship (p < 0.001) was seen between the presence of the fat halo sign and body weight.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Patients referred specifically for renal calculus protocol provided a favorable cohort to examine the validity of this sign as a possible normal finding although an increased incidence of renal stone disease may occur in patients with Crohn's disease [8]. All studies were performed without oral contrast material so that any possible artifact introduced by an intraluminal contrast agent was eliminated. Similarly, all initial studies lacked IV contrast material so that no confounding enhancement factors were introduced. Because some of these patients required an enhanced scan, which we obtained with the patient in the prone position, scanning the patient in this position afforded us a chance to see the influence of both mural enhancement and luminal distention on the halo sign when observed in segments corresponding to those scanned with the patient in the supine position. Although precise matching of the same section with such reorientation was impossible, there was a consistent segmental correlation of the persistence of the sectional fat halo sign when comparing nondistended colonic segments. In general, the sign became less apparent or disappeared as the lumen became more distended (Figs. 3A, and 3B). Presumably, this disappearance is a consequence of distention of the wall obscuring the thin layer of fat. The effects of additional luminal distention seemed more influential than IV enhancement on the consistency of an observed fat halo sign. The disappearance does, however, reinforce the observation that this sign is best appreciated when the lumen is partially collapsed. In fact, the constellation of four signs (i.e., increased prevalence largely in the collapsed state, gravity-induced luminal distensibility and absence of narrowing when the colon is distended with gas, disappearance of the halo sign with additional distention, and a thin caliber of the fatty layer) should be considered as strongly favoring a "normal" occurrence rather than inflammatory bowel disease in the colon. The presence of a normal haustral pattern also provided reassurance of normality.
|
|
On the other hand, distinguishing this thin fatty layer in the terminal ileum from findings in patients with Crohn's disease is most challenging because many of the previously mentioned colonic adjunctive findings are not available. The small intestine is often less gas-filled and is less subject to gas redistribution by gravity maneuvers. Its normal tubular configuration prevents using more subtle anatomic observations. In equivocal cases, the periintestinal fat can be closely inspected for mesenteric stranding and lymph node enlargement. In suspicious cases, an equivocal interpretation may be submitted, and the referring physician should be encouraged to question the patient specifically for present or past symptoms of inflammatory bowel disease before considering further imaging.
CT density measurements averaging -10 H or less are currently considered reliable for indicating the presence of fat. However, in our study, the specific presence of this CT finding lacks conventional documentation because pathologic confirmation is lacking. However, its increasing incidence in patients with heavy body weight suggests a causal relationship.
The advent of improved technology, including a capacity for thinner slices, may account for this observation not being previously reported. This normal fat layer is unlikely due to intramural edema because this would have higher density measurements. Coexistence of wall edema superimposed on an abnormal fat layer could produce fat measurements in the range we observed. However, the presence of edema would imply more acute disease activity, presumably with symptoms, a finding lacking in all these patients. Lack of lower fat density measurements may be further explained by the partial volume effect of 5-mm-thick slices, particularly with the thinness of this fat layer. Infrequently, we did observe a very thin central layer measuring more than -10 H, which we considered a manifestation of a partial volume phenomenon. Finally, although unlikely, some or all the patients with the fat halo sign could have had previously un-diagnosed episodes of inflammatory bowel disease and either no longer remember the diagnosis from their youth or did not report it to their present physician. Particularly suspect would be the five of 21 patients with enlarged mesenteric lymph nodes, for whom there was no alternative explanation. Further suspect are the patients found to have renal stone disease because patients with Crohn's disease are known to have a higher prevalence of renal stone disease than would be found in the healthy population [8]. Besides the lack of any historic hint of inflammatory bowel disease, the absence of any other CT findings associated with inflammatory bowel disease would argue that the findings in at least some of these patients with the halo sign were in fact normal variants. Finally, the predominance of healthy men with the halo sign is inconsistent with the known higher incidence of women affected with Crohn's disease [9].
In summary, a subset of patients examined on CT for renal stone disease has been observed to have a bowel wall fat halo sign scattered among segments of the terminal ileum, colon, or both. In the absence of historic symptoms or other radiographic signs typical of inflammatory bowel disease, we suggest that the bowel wall fat halo sign is a normal finding, perhaps related to obesity.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  M. d'Almeida, J. Jose, J. Oneto, and R. Restrepo Bowel Wall Thickening in Children: CT Findings RadioGraphics, May 1, 2008; 28(3): 727 - 746. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. J. Scholz, S. C. Behr, and C. D. Scheirey Intramural Fat in the Duodenum and Proximal Small Intestine in Patients with Celiac Disease Am. J. Roentgenol., October 1, 2007; 189(4): 786 - 790. [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]
|
|
|
|
 |
|
 J. Ahualli The Fat Halo Sign Radiology, March 1, 2007; 242(3): 945 - 946. [Full Text] [PDF]
|
|
|
|
|
|
J. M. Pereira, C. B. Sirlin, P. S. Pinto, and G. Casola CT and MR Imaging of Extrahepatic Fatty Masses of the Abdomen and Pelvis: Techniques, Diagnosis, Differential Diagnosis, and Pitfalls RadioGraphics, January 1, 2005; 25(1): 69 - 85. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
