| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 All authors: Department of Radiology, Hospital of the University of Pennsylvania, 3400 Spruce St., Philadelphia, PA 19104.
Received March 8, 2000;
accepted after revision May 1, 2000.
Address correspondence to W. T. Miller, Jr.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. A search of medical and radiology files revealed 29 patients with esophagitis in whom thoracic CT was performed within 1 month of the diagnosis. The CT scans were reviewed retrospectively for esophageal wall thickening, a target sign, or other abnormalities. The thickness of the esophageal wall was also measured on CT in these 29 patients and compared with a control group of 85 patients.
RESULTS. We found that 16 patients (55%) with esophagitis had abnormal findings on CT, including a thickened esophageal wall (using 5 mm as the threshold for wall thickening) in all 16 (55%) and a target sign in five (17%). The overall mean esophageal wall thickness was 4.7 mm (standard deviation [SD], 2 mm; 95% confidence interval [CI], 0.7-8.7 mm) in patients with esophagitis versus a mean wall thickness of 2.9 mm (SD, 0.8 mm; 95% CI, 1.3-4.5 mm) in controls (p <.001). Using the same 5-mm threshold for wall thickening, we found that only three (4%) of 85 controls had a thickened esophageal wall on CT.
CONCLUSION. The majority of patients with esophagitis had
abnormalities on CT, including a thickened esophageal wall (
5 mm) in 55%
and a target sign in 17%. Although barium studies and endoscopy are more
sensitive modalities for detecting this condition, the CT finding of a
relatively long segment of circumferential esophageal wall thickening, with or
without a target sign, should suggest the diagnosis of esophagitis in the
proper clinical setting.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
All 29 patients underwent helical CT (HiSpeed Advantage and HiSpeed CT/i; General Electric Medical Systems, Milwaukee, WI). In 28 patients, CT was performed with 7-mm collimation and a pitch of 1.5. In the remaining patient, we performed CT with 5-mm collimation, reconstructing the images at 2.5-cm intervals during full inspiration and end expiration. Twenty-seven of these patients underwent enhanced Ct with IV contrast material. In all 29 patients, high spatial frequency (i.e., bone) algorithms were used for image reconstruction. CT was performed without prior administration of an effervescent agent to dilate the esophagus or stomach.
The thoracic CT scans of these 29 patients with esophagitis were reviewed retrospectively by an experienced thoracic radiologist for esophageal wall thickening or an abnormal enhancement pattern (e.g., a target sign). When esophageal wall thickening was present, the radiologist noted whether the thickening was symmetric or asymmetric and diffuse or focal. Esophageal wall thickening was considered diffuse when the narrowed segment was greater than 3.5 cm in length and focal when this segment was less than 3.5 cm in length. The radiologist also noted whether this wall thickening involved the proximal, mid, or distal thirds of the thoracic esophagus.
Apart from the subjective interpretation of a thickened esophageal wall, we obtained the actual wall thickness in these 29 patients by direct measurement. In 20 patients in whom the esophageal lumen could be identified because the esophagus was not collapsed or partially collapsed, the thickness of a single esophageal wall was measured from its outer to inner borders. In the remaining eight patients in whom the lumen could not be identified because the esophagus was collapsed, this value was obtained by measuring from the outer border of one wall to the outer border of the opposing wall in the short axis of the cross section of the esophagus and dividing by two.
In 25 patients, measurements were obtained at four levels, including the aortic arch, carina, superior pulmonary vein, and gastroesophageal junction. In the remaining four patients in whom the esophagus could not be adequately identified at all these levels, measurements were obtained at three levels. Factors that prevented adequate visualization of the esophagus included motion or beamhardening artifacts, lack of mediastinal fat, and adjacent pleural effusions. A mean wall thickness was then calculated at each level for all patients with esophagitis. Finally, an overall mean esophageal wall thickness was determined by combining the measurements at the aortic arch, carina, and superior pulmonary vein and dividing by three. Measurements at the gastroesophageal junction were excluded from the determination of overall mean wall thickness because of a known problem with apparent thickening of the wall or pseudomass lesions at this level [8, 9].
In a separate part of the study, patients who underwent thoracic CT at our hospital during a 4-month period from July 1999 to October 1999 were asked to complete screening questionnaires to obtain a control group without known esophageal disease. In the questionnaire, patients were asked if they had any of the following: a history of disease or surgery involving the esophagus; symptoms related to the esophagus, including heartburn, dysphagia, odynophagia, or substernal chest pain; or predisposing factors for esophageal disease, including mediastinal irradiation, immunocompromised states, prior caustic ingestion, or the use of oral medications such as tetracycline, doxycycline, potassium chloride, quinidine, or nonsteroidal antiinflammatory agents. The questionnaires identified 85 patients with no known esophageal disease in whom thoracic CT was performed. Our control group comprised these 85 patients. The average age was 54 years (range, 16-87 years). Fifty-two patients were male and 33 were female.
All 85 controls underwent helical CT. In 81 patients, CT was performed with 7-mm collimation and a pitch of 1.5. In the remaining four patients, alternative CT protocols were used. Sixty-four of these patients underwent enhanced CT with IV contrast material.
We measured esophageal wall thickness in the 85 controls at the levels of the aortic arch, carina, superior pulmonary vein, and gastroesophageal junction, using the same methodology as in the group with esophagitis. Similarly, mean esophageal wall thickness was calculated at each level for the control group, and overall mean wall thickness was also determined.
After mean values and standard deviations for esophageal wall thickness were obtained for patients with esophagitis and controls, a Student's t test was performed to assess the presence or absence of statistically significant differences in esophageal wall thickness between these groups.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
In the past, both 3 mm and 5 mm have been suggested as maximum normal values for esophageal wall thickness [1, 2, 10]. If 3 mm is used as the threshold for esophageal wall thickening, 23 (79%) of our 29 patients with esophagitis had a thickened esophageal wall, as did 36 (42%) of 85 controls, limiting the value of this finding. Instead, we used 5 mm as the threshold for esophageal wall thickening in our study. With this threshold, 16 (55%) of 29 patients with esophagitis had a thickened esophageal wall (Figs. 1A,1B,2,3), whereas only three (4%) of 85 controls had a thickened wall (Fig. 4). Thirteen (45%) of the 29 patients with esophagitis also had a subjective assessment of esophageal wall thickening in the retrospective review of CT images. All 13 of these patients had an esophageal wall thickness of 5 mm or greater.
|
|
|
|
|
In Table 2, we summarize the findings in the 29 patients with esophagitis, using 5 mm as the threshold for esophageal wall thickening. All 16 patients with a thickened esophageal wall had concentric circumferential wall thickening (Figs. 1A,1B,2,3). Twelve of these patients (75%) had diffuse wall thickening, and the remaining four (25%) had focal wall thickening that involved an esophageal segment less than 3.5 cm in length. The distribution of esophageal wall thickening included the entire thoracic esophagus in eight patients (three had unspecified esophagitis, one had reflux esophagitis, one had herpes esophagitis, one had radiation esophagitis, and two had combined radiation and reflux esophagitis); the proximal and mid esophagus in five patients (four had radiation esophagitis and one had reflux esophagitis); the mid esophagus in one patient with reflux esophagitis; and the distal esophagus in two patients (one had combined Candida and reflux esophagitis and one had unspecified esophagitis).
|
Five patients (17%) with a thickened esophageal wall also had evidence of a target sign on CT with enhancement of the mucosa and a hypodense submucosa [2, 11] (Fig. 3). None of the controls exhibited a target sign. Both esophageal wall thickening and the target sign were detected in various forms of esophagitis without apparent differences between these groups, but our sample size was too small to permit a statistical analysis.
In both patients with esophagitis and controls, mean wall thickness at the gastroesophageal junction was significantly greater than overall mean esophageal wall thickness (p < 0.001). Mean wall thickness at the gastroesophageal junction was also significantly greater in patients with esophagitis than that in controls (p < 0.001) (Table 1).
The mean interval between CT and the diagnosis of esophagitis was 5.4 days (range, 1-30 days) in 16 patients with esophageal wall thickening versus 8.5 days (range, 1-22 days) in 13 patients without esophageal wall thickening.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Although there are considerable data on the role of CT in staging esophageal cancer [5, 12,13,14,15], CT findings in esophagitis are not well documented in the radiology literature. In the past, it has been suggested that 3 mm or 5 mm be considered the maximum normal values for esophageal wall thickness on CT and that a greater wall thickness may represent a sign of esophagitis [1, 2]. In our study, 5 mm proved to be a more useful threshold for esophageal wall thickening in patients with esophagitis; 55% of all patients with esophagitis had an esophageal wall thickness of 5 mm or greater on CT (Figs. 1A,1B,2,3), whereas only 4% of controls had a wall thickness exceeding this value (Fig. 4). Subjective evaluation of the CT findings by an experienced thoracic radiologist identified esophageal wall thickening in 45% of patients with esophagitis. These data suggest that careful observation alone will identify many patients with esophagitis.
Esophageal wall thickening is a nonspecific response to a veriety of
conditions involving the esophagus. However, benign or malignant tumors of the
esophagus usually are manifested on CT by focal, asymmetric thickening of the
esophageal wall [1,
2], whereas wall thickening in
esophagitis was concentric and circumferential in all patients in our series
and involved a relatively long segment of the esophagus in 75% of patients.
Esophageal wall thickening was found in all types of esophagitis, so a
specific cause of esophagitis cannot be suggested on the basis of CT findings.
In any case, we believe that the CT finding of relatively diffuse,
circumferential esophageal wall thickening (5 mm) should suggest the
diagnosis of esophagitis in the proper clinical setting. Occasionally,
however, malignant esophageal tumors may be manifested on CT by concentric
wall thickening over relatively long segments
[16] so that further
investigation with double-contrast esophagography or endoscopy may be required
in patients who do not have typical clinical features of esophagitis.
Some investigators previously noted that the apparent thickness of the bowel wall anywhere in the gastrointestinal tract is related to the degree of luminal distention; a collapsed viscus often appears to have greater wall thickening than a dilated viscus [2]. Because our thoracic CT was performed without prior administration of an effervescent agent, the esophagus often was collapsed during these examinations. In fact, a collapsed lumen represents the usual appearance of the esophagus encountered by radiologists who interpret thoracic CT. This circumstance may have influenced the maximum normal values chosen for esophageal wall thickness in earlier studies [1, 2, 10]. In our study, we found that only 4% of controls had a thickened esophageal wall, using 5 mm as the maximum normal value for wall thickness, whereas we found that 42% of controls had a thickened wall using 3 mm for this threshold. Thus, 5 mm rather than 3 mm seems to be more appropriate as the maximum normal value for esophageal wall thickness.
Focal soft-tissue thickening at the gastroesophageal junction is a well-known phenomenon on CT that sometimes results in the development of a pseudomass lesion in this region [8, 9]. Similarly, we found a statistically significant difference between the degree of wall thickening at the gastroesophageal junction and elsewhere in the esophagus both in patients with esophagitis and in controls.
In our study, a target sign on CT was found in the esophagus in 17% of patients with esophagitis (all of whom also had a thickened esophageal wall) (Fig. 3) and in none of our controls. The target sign is a well-recognized CT finding caused by the combination of mucosal enhancement and a hypodense submucosa [2, 10]. In the small bowel and colon, a target sign almost always indicates benign disease associated with submucosal edema resulting from inflammation or ischemia [2, 11]. Although a target sign in the esophagus on CT has not been described previously, we believe that it represents a relatively specific CT finding of esophagitis.
In conclusion, CT detected abnormalities in the esophagus in the majority of patients with esophagitis. The most common CT findings were a thickened esophageal wall and a target sign, which were present in 55% and 17% of patients, respectively. Although barium studies and endoscopy are more sensitive modalities for detecting this condition, the CT finding of a relatively long segment of circumferential esophageal wall thickening, with or without a target sign, should suggest the diagnosis of esophagitis in the proper clinical setting.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  S. Umeoka, T. Koyama, K. Togashi, T. Saga, G. Watanabe, Y. Shimada, and M. Imamura Esophageal Cancer: Evaluation with Triple-Phase Dynamic CT--Initial Experience Radiology, June 1, 2006; 239(3): 777 - 783. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
