| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Clinical Observations |
1 All authors: Department of Radiology, Hospital of the University of Pennsylvania, 3400 Spruce St., Philadelphia, PA 19104.
Received September 13, 2006;
accepted after revision December 6, 2006.
M. S. Levine and S. E. Rubesin are consultants for E-Z-EM.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CONCLUSION. Nonanastomotic strictures usually appear on upper gastrointestinal tract radiography as relatively long segments of smooth, tapered narrowing involving the interposed colon, most likely resulting from chronic ischemia. Unlike strictures at the esophagocolic or cologastric anastomosis, these long nonanastomotic strictures generally have a poor response to endoscopic dilatation procedures and are more likely to necessitate surgical revision of the colonic interposition.
Keywords: colonic interposition • esophageal atresia • esophageal cancer • esophageal replacement surgery • gastrointestinal radiology • ischemia • swallowing disorders
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Although anastomotic strictures are a well-recognized complication of colonic interposition, we have encountered a number of patients who developed nonanastomotic strictures of the interposed colon. To our knowledge, this finding has been described only anecdotally in the radiology literature [11, 15]. We therefore report a series of patients with nonanastomotic strictures after colonic interposition and present the clinical and radiographic findings in these patients.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The seven patients in our study group underwent a total of 14 upper gastrointestinal radiographic examinations during the study period; four patients had multiple studies (mean, 2.8 studies; range, 2–4 studies). When multiple studies were performed, the first study in which a nonanastomotic stricture was detected was designated as the index study. Three of the four patients with multiple radiographic studies had one or more follow-up examinations, and one patient had undergone a previous examination before the index study. The mean duration from surgery to the time of the index study was 8.5 years (range, 7 days–28 years). Four of the 14 radiographic examinations were performed with a 250% weight/volume (w/v) barium suspension (E-Z-HD, E-Z-EM), two with a 100% w/v barium suspension (Solopaque, Lafayette Pharmaceutical), and eight with a water-soluble contrast agent (diatrizoate meglumine and diatrizoate sodium [Gastroview, Mallinckrodt]) followed by barium if there was no evidence of a leak.
All the studies were performed with digital fluoroscopy equipment (Diagnost 76, Philips Medical Systems; or Sireskop SD, Siemens Medical Solutions) using a combination of spot images, rapid sequence imaging, and video and DVD recordings. Swallowing function was evaluated initially with the patient in the upright frontal and lateral positions; for subsequent evaluations of the interposed colon, proximal and distal anastomoses, and stomach, the patient underwent imaging in the upright and in various recumbent positions.
The images from the index examinations were reviewed retrospectively by consensus of two authors, both of whom are experienced gastrointestinal radiologists, to determine the morphologic features of these nonanastomotic strictures, including their length and width (accounting for magnification), contour (smooth vs irregular), and proximal and distal margins (tapered vs abrupt). The strictures were also assessed for the presence of obstructive features, including proximal dilatation and delayed emptying. The anastomoses and interposed colon were evaluated for the presence of anastomotic leaks or fistulas, ulceration, spasm, thickened or effaced haustral folds, and other abnormalities. Finally, the studies and study reports were evaluated for the presence of gastrocolic reflux, aspiration, or other types of swallowing dysfunction.
Medical, endoscopic, surgical, and pathologic records were reviewed to determine the indications for colonic interposition; type of colonic interposition performed (including the portion of colon used for the conduit, the orientation of the interposed colon, and whether the interposed colon was clamped at surgery); nature and duration of symptoms after colonic interposition; subsequent treatment (including endoscopic dilatation procedures or additional surgery); histopathologic findings (if a portion of the interposed colon was resected); and patient course. The mean duration of clinical follow-up from the time of the index examination was 1.2 years (range, 12 days–5.2 years). When two or more radiographic examinations were performed, serial examinations were reviewed to determine the course of the strictures over time.
Our institutional review board approved all aspects of this retrospective study and did not require informed consent from the patients included in our study.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The left colon was used as the neoesophageal conduit in three patients and the right colon in three. We were unable to determine which portion of the colon was used as the conduit in the remaining patient. The interposed colon had an isoperistaltic orientation in three patients, an antiperistaltic orientation in two, and an unknown orientation in two. The interposed colon was not clamped at surgery in any of the seven patients.
Clinical Findings
Four (57%) of the seven patients with colonic interposition had dysphagia
as the major presenting symptom at the time of the index radiographic studies.
Two patients had dysphagia for solids only, and two had dysphagia for solids
and liquids. The mean duration of dysphagia was 6.6 years (range, 3
days–14 years). The remaining three patients (43%) with colonic
interposition had clinical signs and symptoms (fever, leukocytosis, and
increased wound drainage) of anastomotic leaks at the time of the index
radiographic studies, which were performed much earlier in the postoperative
course; the mean duration from colonic interposition to the index studies was
only 3 months (range, 7 days–8 months) for this group versus 16.7 years
(range, 2 months–28 years) for the group with dysphagia. Subsequently,
all three patients in the group with clinically suspected leaks also developed
dysphagia.
Two patients (29%) had clinical signs of aspiration, and one (14%) had a wound infection.
Radiographic Findings
Index examination—The mean length of the nonanastomotic
strictures in the seven patients with colonic interpositions was 8 cm (range,
3.5–14 cm), and the mean width was 0.9 cm (range, 0.3–1.8 cm).
These strictures therefore were relatively long segments of narrowing
involving the interposed colon (Figs.
1A,
1B,
2, and
3). The strictures had a smooth
contour in six patients and an irregular contour in one. The proximal margins
of the strictures were tapered in all seven patients, and the distal margins
were tapered in six and abrupt in one. Haustral folds in the interposed colon
were effaced in six patients and were thickened in one. The strictures caused
partial obstruction in two patients, with proximal dilatation and delayed
emptying of contrast material into the stomach in both. One of the patients
with delayed emptying had a food impaction above the proximal end of the
stricture.
|
|
|
|
Serial examinations—In two of the four patients with nonanastomotic strictures in the interposed colon who had multiple studies, we observed progressive shortening and narrowing of the strictures, which decreased in both length and width on follow-up examinations. One of these patients also had a sealed-off leak from the esophagocolic anastomosis that subsequently progressed to a colocutaneous fistula on later examinations. In contrast, the third patient with multiple studies had a stricture that increased in length and width on follow-up examinations. The final patient with multiple studies had an earlier radiographic examination showing no evidence of a nonanastomotic stricture in the interposed colon 3 months before the index study revealed a stricture (Fig. 1B).
Treatment
Six (86%) of the seven patients with colonic interpositions underwent a
total of 27 endoscopic dilatation procedures (mean number of dilatations, 5;
range, 1–11). Four patients (57%) ultimately required surgical revision
of the colonic interposition (with resection of a portion of the interposed
colon in three) because of intractable dysphagia. Pathologic examination of
the resected specimens revealed extensive submucosal fibrosis in two patients
and erosion of the mucosa with hemosiderin-laden macrophages in one. When
surgery was repeated, the mean duration from the time of the original colonic
interposition to the time of the first surgical revision was 5.7 years (range,
4 days–15 years).
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
These nonanastomotic strictures after colonic interposition had characteristic findings on upper gastrointestinal radiographic studies: The nonanastomotic strictures appeared as relatively long segments of narrowing involving the interposed colon (Figs. 1A, 1B, 2, and 3). The strictures almost always had smooth contours, tapered margins, and effaced haustral folds. The location, length, and appearance of nonanastomotic strictures after colonic interposition therefore enable differentiation from anastomotic strictures on upper gastrointestinal radiography in almost all cases.
All seven patients with nonanastomotic strictures after colonic interposition presented with dysphagia at the time of the index examination (n = 4) or developed dysphagia after the index examination (n = 3). When dysphagia was present at the index examination, it usually occurred months to years after colonic interposition, and the strictures responded poorly to treatment, frequently necessitating multiple endoscopic dilatation procedures or even surgical revision of the interposed colon. In contrast, anastomotic strictures that develop after colonic interposition are often amenable to endoscopic dilatation without the need for multiple dilatation procedures or repeat surgery [11, 13, 14]. The development of nonanastomotic strictures in the interposed colon therefore has important implications for the long-term management of these patients.
Although the cause of nonanastomotic strictures after colonic interposition is uncertain, the frequent effacement or obliteration of haustral folds and the relatively long length of the narrowed colonic segment in our patients are characteristic of ischemic strictures involving the colon elsewhere [16]. In two cases, pathologic examination of the resected portion of the interposed colon revealed extensive submucosal fibrosis. Although none of the patients in our series had radiographic findings of acute colonic ischemia, other authors have described edema, spasm, ulceration, and thumbprinting as radiographic signs of acute ischemia in the interposed colon and actual perforation as a sign of ischemic necrosis of the interposed colon [11, 12]. The surgical literature also contains several reports of patients who developed dysphagia as a result of long nonanastomotic strictures after colonic interposition [17–19]. One patient had an acute hypotensive episode after surgery [17]; in another patient, a postoperative angiogram showed abnormal vascularization of the interposed colon [18]. We therefore believe these nonanastomotic strictures most likely develop as a result of chronic ischemia from inadequate perfusion of the surgically mobilized colon.
Nonanastomotic strictures after colonic interposition could also result from postoperative leaks with associated scarring and fibrosis, but most leaks involve the esophagocolic or cologastric anastomosis, so developing strictures are usually located at or near these anastomoses [2, 7, 11, 13]. Although two of our patients did have small sealed-off anastomotic leaks, we doubt that these leaks were an important contributing factor because of the long length of the strictures in our patients and because of the location of these strictures—that is, at a discrete distance from the proximal and distal anastomoses. Reflux of acid from the stomach via the cologastric anastomosis could also account for the development of nonanastomotic strictures in the interposed colon, but such reflux was not observed at fluoroscopy in any of our patients. Furthermore, one would expect reflux-induced strictures in the interposed colon to be located distally at or near the cologastric anastomosis, whereas all of the strictures in our patients were located well above the distal anastomosis.
Our investigation has the inherent limitations of a retrospective study. Selection bias may also have increased the prevalence of nonanastomotic strictures in our series because many patients were referred to our institution as a result of preexisting complications from failed colonic interpositions. The limited duration of follow-up for several patients also limits our ability to draw firm conclusions in all cases.
In summary, nonanastomotic strictures after colonic interposition have characteristic features on radiographic examinations of the upper gastrointestinal tract, appearing as relatively long segments of smooth, tapered narrowing involving the interposed colon that are separate from the proximal esophagocolic and distal cologastric anastomoses. These strictures most likely develop as a result of chronic ischemia of the surgically mobilized colon. Unlike strictures at the esophagocolic or cologastric anastomosis, nonanastomotic strictures after colonic interposition have a poor response to endoscopic dilatation procedures and are more likely to necessitate surgical revision of the colonic interposition.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  R. J. Stanley Commentary on Two Classic GI Studies Am. J. Roentgenol., July 1, 2007; 189(1): 24 - 24. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
