| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1
Department of Radiology, Hospital of the University of Pennsylvania, 3400
Spruce St., Philadelphia, PA 19104.
2
Department of Medicine, Hospital of the University of Pennsylvania,
Philadelphia, PA 19104.
3
Department of Pathology and Laboratory Medicine, Hospital of the University of
Pennsylvania, Philadelphia, PA 19104.
Received September 14, 2000;
accepted after revision October 31, 2000.
Address correspondence to M. S. Levine.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. A review of pathology and endoscopy data revealed 142 patients with short-segment Barrett's esophagus, which was defined as columnar epithelium in the distal esophagus extending 3 cm or less above the gastroesophageal junction at endoscopy with histopathologic confirmation of intestinal metaplasia. Twenty of these patients underwent double-contrast esophagography. These 20 patients comprised our study group. The original radiology reports and images were reviewed to determine the findings on double-contrast esophagography. Medical records were also reviewed to determine the clinical findings and treatment.
RESULTS. Double-contrast esophagrams revealed hiatal hernias in 18 patients (90%), gastroesophageal reflux in 16 (80%), reflux esophagitis in seven (35%), peptic scarring or strictures in 11 (55%), and a reticular mucosal pattern in none. A total of 14 patients (70%) had morphologic findings of reflux disease with esophagitis alone (three patients), peptic scarring or strictures alone (seven patients), or both (four patients), but the remaining six (30%) had hiatal hernias or gastroesophageal reflux as the only radiographic finding.
CONCLUSION. Double-contrast esophagography revealed morphologic findings of reflux disease with esophagitis, peptic scarring or strictures, or both in 70% of patients with short-segment Barrett's esophagus. Thus, the absence of esophagitis or peptic scarring or strictures on double-contrast esophagography does not exclude the possibility of short-segment Barrett's esophagus.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Recent data suggest that short-segment Barrett's esophagus is a premalignant condition associated with an increased risk of developing esophageal adenocarcinoma via a dysplasia-carcinoma sequence identical to that described for long-segment Barrett's esophagus [7,8,9,10,11]. As a result, some investigators advocate endoscopic surveillance of both patients with short-segment and long-segment types of Barrett's esophagus to detect dysplastic changes before overt adenocarcinoma develops [2, 4, 9].
The classic findings of long-segment Barrett's esophagus on double-contrast esophagography include a mid esophageal stricture or ulcer associated with a hiatal hernia or gastroesophageal reflux [12, 13]. A distinctive reticular pattern of the mucosa has also been described as a relatively specific sign of long-segment Barrett's esophagus, particularly if the pattern is seen adjacent to the distal aspect of a mid esophageal stricture [14]. Other more common but less specific radiographic findings include hiatal hernia, gastroesophageal reflux, reflux esophagitis, and peptic scarring or strictures in the distal esophagus [15, 16]. To our knowledge, however, the radiographic features of short-segment Barrett's esophagus have not been reported previously. The purpose of this study therefore was to describe the findings of short-segment Barrett's esophagus on double-contrast esophagography.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Subsequent review of the radiology database at our hospital revealed that 20 of the 142 patients with short-segment Barrett's esophagus had double-contrast studies (including 17 double-contrast upper gastrointestinal tract examinations and three double-contrast esophagrams) that had been performed within 1 year of endoscopy. These 20 patients comprised our study group. Sixteen had barium studies performed before endoscopy, and four had such studies performed after endoscopy. The average interval between the barium studies and endoscopy was 6 months (range, 1-11 months).
The original radiology reports were reviewed retrospectively to determine the radiographic findings in these 20 patients. Gastroesophageal reflux in the patients was classified as mild, moderate, or marked, depending on the terminology used to describe this reflux in the reports. The radiographic images were also reviewed to further characterize the various morphologic abnormalities, including reflux esophagitis, peptic scarring or strictures, a reticular mucosal pattern, and mid esophageal ulcers or strictures. In patients in whom esophagitis was present, we evaluated each radiographic study for various findings of esophagitis, including mucosal granularity or nodularity, thickened folds, ulceration, and inflammatory esophagogastric polyps. Finally, we reviewed the radiographs to determine the technical quality of the double-contrast study for each patient.
Endoscopy and pathology reports for all 20 patients were also reviewed to determine the endoscopic and histopathologic findings. When esophagitis was found at endoscopy, we classified it as being grade 1-4, using the Hetzel grading system [17]. Fourteen patients underwent follow-up endoscopy an average of 10 months (range, 1-23 months) after the initial endoscopic examination; the results of these studies were also reviewed. Finally, medical records were reviewed to determine the clinical presentation and treatment.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Fourteen patients were treated with proton pump inhibitors or histamine receptor antagonists before undergoing esophagography or endoscopy. Of these 14 patients, two were treated with proton pump inhibitors for an average of 6 months (range, 3-9 months) before esophagography, and seven were treated with proton pump inhibitors for an average of 8 months (range, 2-12 months) before endoscopy. The remaining five patients were treated with histamine receptor antagonists for an average of 5 months (range, 1-10 months) before endoscopy.
Radiographic Findings
The quality of the double-contrast esophagrams was rated as good to
excellent in 18 patients (90%) and as suboptimal in two patients
(10%)—one with reflux esophagitis and one without evidence of
esophagitis or peptic scarring or strictures—because the pooling of
barium in the distal esophagus prevented adequate visualization of this region
on double-contrast images.
The double-contrast esophagrams revealed (Table 1) hiatal hernias in 18 patients (90%), gastroesophageal reflux in 16 (80%), reflux esophagitis in seven (35%) (Figs. 1,2,3), and peptic scarring or strictures in 11 (55%) (Figs. 3,4,5). A total of 14 patients (70%) had morphologic findings of reflux disease with esophagitis alone (three patients), peptic scarring or strictures alone (seven patients), or both (four patients), but the remaining six (30%) had hiatal hernias or gastroesophageal reflux as the only radiographic findings (Fig. 6). Inability to visualize esophagitis or peptic scarring or strictures in these six patients was not primarily related to technical factors; five of the six had technically satisfactory double-contrast examinations.
|
|
|
|
|
|
|
The radiographic findings in seven patients with reflux esophagitis included mucosal granularity or nodularity in four (Fig. 1), thickened folds in one (Fig. 2), superficial ulceration in one (Fig. 3), and an inflammatory esophagogastric polyp in one. These conditions were confined to the distal esophagus within 3 cm of the gastroesophageal junction in three patients and extended into the distal third of the thoracic esophagus in four others.
The radiographic findings in 11 patients with peptic scarring or strictures included circumferential strictures in eight patients (Figs. 3,4,5) and asymmetric flattening of one wall of the distal esophagus in three. The strictures had an average length of 2.2 cm (range, 0.5-5 cm) and an average diameter of 1.4 cm (range, 0.8-1.8 cm), not accounting for magnification. The strictures all had tapered borders, which had smooth contours in seven patients and had an irregular contour in the remaining one. All were located above hiatal hernias. Two of the strictures had associated esophageal intramural pseudodiverticula (Fig. 4). This peptic scarring and formation of strictures were confined to the distal esophagus within 3 cm of the gastroesophageal junction in seven patients and the distal third of the thoracic esophagus in four. Thus, both the esophagitis and peptic scarring often extended proximally above the level of involvement by short-segment Barrett's esophagus.
One patient with a peptic stricture in the distal esophagus also had a mid esophageal stricture associated with esophageal intramural pseudodiverticulosis. Another patient with a normal distal esophagus had a smooth, tapered radiation-induced stricture in the upper thoracic esophagus unrelated to short-segment Barrett's esophagus. None of the remaining 18 patients had strictures in the upper or mid esophagus, and no patients had a reticular mucosal pattern or mid esophageal ulcer visible on double-contrast esophagrams.
Endoscopic and Histopathologic Findings
The endoscopically visualized columnar epithelium in the distal esophagus
was located within 0.5 to 1 cm of the gastroesophageal junction in 12 patients
with short-segment Barrett's esophagus, 1 to 2 cm in seven, and 2 to 3 cm in
one. The columnar epithelium was characterized at endoscopy by fingerlike
projections of velvety red Barrett's mucosa in the distal esophagus in 18
patients and by circumferential extension of Barrett's mucosa in the distal
esophagus in the remaining two.
Hiatal hernias were observed at endoscopy in 10 (50%) of the 20 patients with short-segment Barrett's esophagus. Reflux esophagitis was observed in the distal esophagus in three patients (15%); the esophagitis was classified as grade 1 with erythema, friability, or nodularity of the mucosa in two of these patients and as grade 2 with mucosal erosions in one. Peptic strictures were detected at endoscopy in four patients (20%); two of the strictures were located in the distal esophagus within 3 cm of the gastroesophageal junction, and two were located in the distal third of the thoracic esophagus. Endoscopy also revealed a radiation-induced stricture in the upper thoracic esophagus of one patient.
All seven patients with endoscopically diagnosed esophagitis or strictures also had esophagitis or strictures found on double-contrast esophagography. Conversely, only three of seven patients with radiographically diagnosed esophagitis had findings of esophagitis at endoscopy. The discrepancy in the remaining four patients may be related to the treatment they received before undergoing endoscopy. After being treated with proton pump inhibitors or histamine receptor antagonists for an average of 4 months (range, 2-8 months), their esophagitis may have healed. An alternative explanation is that the findings for these patients could represent false-positive barium studies for esophagitis.
All 20 patients with short-segment Barrett's esophagus at endoscopy had histopathologic findings of intestinal metaplasia on endoscopic biopsy specimens from the distal esophagus. One of these 20 patients also had low-grade dysplasia, but none had high-grade dysplasia or adenocarcinoma on biopsy specimens from the initial endoscopic examination.
Follow-up endoscopy in 14 patients revealed low-grade dysplasia in one; high-grade dysplasia was not found in any of the patients. An esophageal adenocarcinoma arising in short-segment Barrett's esophagus was found in one patient who had evidence of esophagitis on the initial barium study. The finding of esophagitis was confirmed when endoscopy was performed 5 months later. However, a second endoscopic examination 2 months after that revealed a 1-cm nodule in the distal esophagus near the gastroesophageal junction, and findings of endoscopic biopsy specimens indicated adenocarcinoma. This patient subsequently underwent esophagogastrectomy, which revealed a moderately well-differentiated adenocarcinoma (with invasion of the muscularis propria) within a 2-cm-segment of Barrett's mucosa adjacent to the gastroesophageal junction.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The pathophysiology of short-segment Barrett's esophagus is uncertain. Recent data suggest that it occurs as a complication of gastroesophageal reflux disease but that patients with short-segment Barrett's esophagus have a less severe form of reflux disease than patients with long-segment Barrett's esophagus [6]. The natural history of short-segment Barrett's esophagus also is not well understood. Nandurkar and Talley [6] have postulated that short-segment Barrett's esophagus represents part of a continuum of disease and that, over time, short-segment Barrett's esophagus eventually may progress to become long-segment Barrett's esophagus. In contrast, Cameron and Lomboy [20] believe that, in most patients, Barrett's esophagus rapidly progresses to full length with little subsequent variation in the extent of disease. Further investigation therefore is needed to better elucidate the relationship between the short-segment and long-segment forms of Barrett's esophagus.
In our study population, the ratio of patients with short-segment to long-segment Barrett's esophagus was 1.7:1. In other studies, the prevalence of long-segment Barrett's esophagus is approximately 1% at endoscopy for all indications [21] and 10% at endoscopy for symptomatic gastroesophageal reflux [22]. In contrast, the prevalence of short-segment Barrett's esophagus at routine endoscopy is as high as 2-12% [3, 5, 23]. Thus, recent literature suggests that short-segment Barrett's esophagus occurs with considerably greater frequency than the long-segment form of the disease.
Long-segment Barrett's esophagus has been proven to be a premalignant condition associated with an increased incidence of esophageal adenocarcinoma via a dysplasia—carcinoma sequence [1]. The risk of patients with long-segment Barrett's esophagus developing adenocarcinoma is thought to be 30-40 times greater than that of the general population [24]. Early data from prospective studies in patients with short-segment Barrett's esophagus have shown that these individuals are more likely to develop dysplasia than the general population but less likely to develop dysplasia than patients with long-segment Barrett's esophagus [9, 10, 25]. Nevertheless, cases of esophageal adenocarcinoma have been reported in patients with short-segment Barrett's esophagus [7,8,9, 11]. Although additional data are needed to determine the cancer risk in these individuals, some investigators believe that endoscopic surveillance is warranted not only for patients with long-segment Barrett's esophagus but also for patients with the short-segment form of the disease [2, 4, 9]. Because of uncertainty about the incidence of adenocarcinoma in patients with short-segment Barrett's esophagus, recommended intervals for endoscopic surveillance examinations range from every 12 months to every 5 years [25].
Further complicating the debate over long-segment and short-segment Barrett's esophagus are recent data showing that intestinal metaplasia of the gastric cardia may occur as a distinct pathologic entity; endoscopic biopsy specimens taken from a patient with a normal-appearing gastroesophageal junction may reveal focal intestinal metaplasia in the cardiac region even with no endoscopic or pathologic evidence of Barrett's mucosa found elsewhere in the esophagus [26]. Several studies have suggested that gastroesophageal reflux disease and Helicobacter pylori infection of the cardia may be etiologic factors in the development of this condition [26, 27]. Intestinal metaplasia of the cardia may be even more common than short-segment Barrett's esophagus, with a reported prevalence of 6-23% on routine endoscopic biopsy specimens from the cardiac region [28,29,30]. Unlike patients with Barrett's esophagus, however, the risk of these individuals developing esophageal adenocarcinoma is unknown. As a result, endoscopic surveillance is not currently advocated for patients with intestinal metaplasia of the cardia without other endoscopic or pathologic findings of Barrett's esophagus [3, 4].
The radiographic features of long-segment Barrett's esophagus have been well documented [12,13,14,15,16]. To our knowledge, however, the findings of short-segment Barrett's esophagus on double-contrast esophagography have not been previously reported. In our study, 70% of patients with short-segment Barrett's esophagus had morphologic findings of reflux disease with esophagitis, peptic scarring or strictures, or both in the distal esophagus (Figs. 1,2,3,4,5). With the exception of a mid esophageal stricture associated with esophageal intramural pseudodiverticulosis and a radiation-induced stricture in the upper thoracic esophagus, all our patients with short-segment Barrett's esophagus had disease confined to the distal third of the thoracic esophagus. In contrast, long-segment Barrett's esophagus is sometimes manifested by the development of strictures, ulcers, or a reticular mucosal pattern in the mid esophagus [12,13,14]. Thus, the findings on double-contrast esophagography may be more specific for Barrett's esophagus in patients with long-segment disease than in those with short-segment disease. However, in patients with short-segment Barrett's esophagus, the length of involvement of the distal esophagus by esophagitis or peptic scarring may extend more than 3 cm above the gastroesophageal junction, and so in these patients, the diseased segment on esophagography does not necessarily correspond to the extent of Barrett's mucosa present.
Previous studies have shown that on double-contrast esophagography 97-99% of patients with long-segment Barrett's esophagus have morphologic findings of reflux disease, including reflux esophagitis, peptic scarring or strictures, mid esophageal strictures or ulcers, and a reticular mucosal pattern [15, 16]. In our study, 70% of patients with short-segment Barrett's esophagus had reflux esophagitis, peptic scarring or strictures, or both on double-contrast esophagography (Figs. 1,2,3,4,5), but 30% had hiatal hernias or gastroesophageal reflux as the only radiographic findings (Fig. 6). Therefore, the absence of either esophagitis or peptic scarring or strictures on double-contrast barium studies does not exclude the possibility of short-segment Barrett's esophagus. Nevertheless, the clinical significance of this observation remains uncertain because of the lower cancer risk of short-segment Barrett's esophagus compared with that associated with long-segment disease.
Ours was a retrospective study with a small sample size and no control group. Our study was also limited by the relatively long average interval of 6 months between double-contrast esophagography and endoscopy. Because esophagitis is a condition that fluctuates over time, this interval may explain the discrepancy between the number of patients with findings of esophagitis on double-contrast esophagography and endoscopy. Also, many patients received treatment with proton pump inhibitors or histamine receptor antagonists, and so possibly their esophagitis healed before endoscopic examinations were performed. Finally, our study had a selection bias for patients with symptomatic gastroesophageal reflux disease. As we discussed earlier, however, short-segment Barrett's esophagus has a prevalence of 2-12% at routine endoscopy [3, 5, 23]. Thus, prospective studies are needed to further elucidate the radiographic findings in both symptomatic and asymptomatic patients with short-segment Barrett's esophagus.
In conclusion, radiologists should be familiar with short-segment Barrett's esophagus, a recently described entity, which is more commonly found at endoscopy than long-segment Barrett's esophagus. Although the cancer risk in these patients remains uncertain, some investigators advocate endoscopic surveillance not only for patients with long-segment Barrett's esophagus but also for patients with short-segment disease. In our study, double-contrast esophagography revealed morphologic findings of reflux disease with esophagitis, peptic scarring or strictures, or both in 70% of patients with short-segment Barrett's esophagus, but 30% had hiatal hernias or gastroesophageal reflux as the only radiographic findings. Thus, the absence of esophagitis or peptic scarring or strictures on double-contrast esophagography does not exclude the possibility of short-segment Barrett's esophagus.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
