DOI:10.2214/AJR.05.0465
AJR 2006; 187:1274-1279
© American Roentgen Ray Society
Esophageal Stents: Findings on Esophagography in 46 Patients
Gregory S. Anderson1,
Marc S. Levine1,
Stephen E. Rubesin1,
Igor Laufer1,
Gregory G. Ginsberg2 and
Michael L. Kochman2
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.
Received March 16, 2005;
accepted after revision June 7, 2005.
Address correspondence to M. S. Levine
(marc.levine{at}uphs.upenn.edu).
Abstract
OBJECTIVE. The purpose of this report is to assess the findings on
esophagography in patients with esophageal stents for palliation of malignant
tumors involving the esophagus.
CONCLUSION. Radiologists should be familiar with findings of little
importance (stent narrowing, flow of contrast medium around stent, stent
kinks, and apparent esophageal narrowing below stent because of incompletely
distended hiatal hernias) versus more important findings (polypoid defects
above or below stent, narrowing within or below stent, delayed stent emptying,
esophageal-airway fistulas, stent migration, and abutting of distal stent
against greater curvature of stomach) on esophagography after stent placement
to guide endoscopists regarding the need for stent revision.
Keywords: dysphagia • esophageal cancer • esophageal disease • esophageal stent • esophagography • esophagram
Introduction
Self-expanding metallic stents have increasingly been used as the treatment
of choice for inoperable esophageal cancer because of low complication rates
and their effectiveness in palliating dysphagia in patients with malignant
strictures
[1-4].
Patients with esophageal stents may undergo esophagography immediately after
stent placement to evaluate positioning of the stent and rule out perforation,
or they may undergo esophagography weeks to months after stent placement to
evaluate dysphagia and rule out esophageal-airway fistulas. The purpose of our
investigation was to assess the various findings on esophagography in a series
of patients with esophageal stents placed for palliation of inoperable
esophageal cancer or other malignant tumors involving the esophagus.
Materials and Methods
A review of our computerized radiology database revealed 116 esophagrams
after stent placement for inoperable esophageal cancer or other malignant
tumors involving the esophagus from 1996 to 2003. Fifty-seven studies were
excluded because of lack of availability of radiographic images or inadequate
follow-up. The remaining 59 studies were performed in 46 patients, including
38 who had one esophagram, five who had two esophagrams, two who had three
esophagrams, and one who had five esophagrams. These 46 patients comprised our
study group. Forty-four patients had esophagrams within 3 days of stent
placement (mean interval, 1 day) to evaluate positioning of the stent and to
exclude perforation. Two patients had their initial esophagrams and eight
patients had a total of 13 follow-up esophagrams 1-13 months after stent
placement (mean interval, 2 months). These 15 esophagrams were obtained
because of dysphagia (n = 13) or suspected esophageal-airway fistulas
(n = 2).
When esophagrams were obtained immediately after stent placement, the
patients initially were given a water-soluble contrast agent (diatrizoate
meglumine and diatrizoate sodium [Gastroview, Mallinckrodt]). If spot images
showed a leak into the mediastinum, the study was terminated. If spot images
did not show a leak, however, the patient was given a 250% weight/volume
barium suspension (E-Z-HD, E-Z-EM, Inc.), and additional images were obtained.
The studies were performed by residents, fellows, or attending
gastrointestinal radiologists, and all were interpreted by the attending
radiologists.
All stents were covered, self-expanding metallic stents, including
Ultraflex stents in 24 patients, Wallstent II stents in 11, Wallstent I stents
in seven, Z-stents in three, and a Flamingo stent in one (Z-stent, Wilson-Cook
Medical; all other stents, Boston Scientific). All of these stents have a
short uncovered segment at their ends to allow the stent struts to anchor to
the esophageal wall. The stents were located in the upper and mid esophagus in
10 patients; the mid and distal esophagus in nine; the upper, mid, and distal
esophagus in 10; and extended into the gastric fundus in 17. The stents had a
mean length of 11 cm (range, 4-15 cm).
View larger version (87K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1A —80-year-old woman with stent placed for palliation of
dysphagia caused by squamous cell carcinoma of esophagus. Left posterior
oblique scout image shows tapered narrowing (arrows) of midportion of
stent.
|
|
View larger version (84K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1B —80-year-old woman with stent placed for palliation of
dysphagia caused by squamous cell carcinoma of esophagus. Left posterior
oblique spot image from single-contrast esophagram shows tapered narrowing of
barium column (arrows) where lumen and stent are compressed by
surrounding esophageal tumor.
|
|
The radiographic images were interpreted by consensus retrospectively by
two gastrointestinal radiologists who were blinded to the clinical and
endoscopic findings. They reviewed the images to assess the flow of contrast
material through or around the stents, kinking or fracture of the stents,
filling defects or contour defects above or below the stents, the caliber and
contour of the lumen within and below the stents, emptying of contrast
material from the stents, stent migration, the presence or absence of
perforation or esophageal-airway fistulas, and any other findings. Medical
records were also reviewed to determine the indications for stent placement
and the clinical findings and course. Seventeen patients (37%) had follow-up
endoscopy after esophagography; the endoscopic and pathology reports were
reviewed and correlated with the radiographic findings in these patients.
Our institutional review board approved all aspects of this retrospective
study and did not require informed consent from any patients included in our
study.
Results
Clinical Aspects
Thirty patients (65%) were men, and 16 (35%) were women. The mean age was
67 years (range, 45-98 years). Twenty-six patients (57%) had esophageal
carcinoma, three (6%) had gastric carcinoma, four (9%) had lung cancer
invading the esophagus, one (2%) had metastatic endometrial sarcoma, and 12
(26%) had malignant strictures of uncertain origin. The stents were placed for
palliation of dysphagia in 35 patients (76%) and palliation of
esophagealairway fistulas in 11 (24%).
Twenty-nine (83%) of the 35 patients in whom stents were placed for
palliation of dysphagia had substantial relief or resolution of dysphagia.
Eight (28%) of these 29 patients developed recurrent dysphagia within 13
months (mean duration, 5 months) after stent placement. Seven (15%) of the 46
patients had additional stents placed because of intractable dysphagia
(n = 5) or continued esophageal-airway fistulas (n =2).
View larger version (85K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2 —70-year-old man with stent placed for palliation of dysphagia
caused by advanced malignant tumor of uncertain origin involving upper
thoracic esophagus. Left posterior oblique spot image from single-contrast
esophagram shows barium (arrows) flowing around left anterolateral
wall of proximal end of stent.
|
|
Radiographic Findings of Little Clinical Importance
Narrowing of stent caliber—Seventeen (29%) of the 59
esophagrams revealed tapered narrowing (less than 50% of the diameter of the
stent) in the midportion of the stent (Figs.
1A and
1B). All but two patients had
relief from dysphagia, so this "waist" characteristic was not
thought to be important.
Flow of contrast material around the sides of the
stent—Sixteen esophagrams (27%) revealed flow of contrast material
around one (n = 11) or both (n = 5) sides of the stent
(Fig. 2). In 15 of these
patients, this finding was not thought to be important because their dysphagia
resolved without further intervention. In the other patient, contrast material
passed around the stent into a tracheoesophageal fistula.
Stent kinks—Three esophagrams (5%) revealed kinking or
angulation of the stent (Fig.
3). This finding was not thought to be important because these
patients all had symptomatic improvement without further intervention.
View larger version (128K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3 —75-year-old man with stent placed for palliation of dysphagia
caused by adenocarcinoma of distal esophagus invading gastric cardia and
fundus. Steep right posterior oblique spot image from single-contrast
esophagram shows narrowing and kinking of stent (black arrow) by
surrounding tumor in distal esophagus. Note how stent traverses
gastroesophageal junction with distal end (white arrow) in gastric
fundus.
|
|
View larger version (77K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4A —60-year-old man with stent placed for palliation of dysphagia
caused by squamous cell carcinoma of esophagus. Steep right posterior oblique
spot image from single-contrast esophagram shows apparent narrowing of distal
esophagus (arrows) from distal end of stent to gastroesophageal
junction.
|
|
View larger version (123K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4B —60-year-old man with stent placed for palliation of dysphagia
caused by squamous cell carcinoma of esophagus. Frontal spot image from same
examination as A shows barium trapped between gastric folds of
incompletely filled hiatal hernia (arrows). Subsequent endoscopy
confirmed presence of hiatal hernia in this patient.
|
|
View larger version (81K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 5 —55-year-old man with stent placed for palliation of dysphagia
caused by squamous cell carcinoma of esophagus. Left posterior oblique spot
image from single-contrast esophagram shows asymmetric mass effect
(arrows) on right posterolateral wall of distal esophagus abutting
stent. This finding was caused by tumor overgrowth into distal end of
stent.
|
|
Incomplete distention of hiatal hernia mimicking narrowed distal
esophagus—Two esophagrams (3%) revealed apparent narrowing of the
distal esophagus just below the stent because of a partially collapsed hiatal
hernia with barium trapped in the folds of the hernia (Figs.
4A and
4B). In both of these patients,
endoscopy confirmed a hiatal hernia.
Radiographic Findings of Greater Clinical Importance
Defects above stent—Two (3%) of the 59 esophagrams revealed
polypoid defects (1 cm and 0.7 cm) in the esophagus abutting the proximal end
of the stents. Both patients had overgrowth of tumor into the proximal end of
the stent at endoscopy, so additional stents were placed.
View larger version (91K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 6 —53-year-old man with stent placed for palliation of dysphagia
caused by malignant tumor of uncertain origin encasing mid esophagus. Right
posterior oblique spot image from single-contrast esophagram shows focal
segment of marked luminal narrowing (black arrows) in distal end of
stent. Note irregular contour and abrupt, shelflike distal margins (white
arrows) of narrowed segment. At endoscopy, this finding was caused by
tumor ingrowth through uncovered distal end of stent.
|
|
Defects below stent—Two esophagrams (3%) revealed polypoid
defects (2 and 3 cm) in the esophagus abutting the distal end of the stents.
One patient had overgrowth of tumor into the distal end of the stent
(Fig. 5), and the other had
exuberant reactive fibrosis at endoscopy.
Narrowing of lumen within stent—Fourteen esophagrams (24%)
revealed esophageal narrowing within the distal end of the stent because of
tumor (n = 3), debris or blood clot (n = 3), epithelial
hyperplasia (n = 3), and unknown causes (n = 5). The
narrowed segment had an irregular contour and abrupt distal margins in
patients with tumor ingrowth (Fig.
6) and a smooth contour and more gradual margins in the patients
with epithelial hyperplasia (Fig.
7).
View larger version (133K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 7 —71-year-old woman with stent placed for palliation of
carcinoma of lung invading upper thoracic esophagus. Right posterior oblique
spot image from single-contrast esophagram shows narrowing of lumen
(arrows) in distal end of stent. Note relatively smooth contour and
tapered margins of narrowed segment. Endoscopic biopsy specimens from this
region revealed epithelial hyperplasia. (Note pneumomediastinum and
subcutaneous emphysema in soft tissues of neck from esophageal perforation
that occurred during endoscopic dilatation procedure before placement of
stent.)
|
|
Narrowing of lumen abutting distal end of stent—Fourteen
esophagrams (24%) revealed esophageal narrowing abutting the distal end of the
stent because of tumor overgrowth (n =3), tumor ingrowth (n
= 2), debris or blood clot (n = 3), epithelial hyperplasia
(n = 2), and unknown causes (n = 4). The narrowed segment
had an irregular contour with abrupt margins in both patients with tumor
ingrowth and a smooth contour with tapered margins in both patients with
epithelial hyperplasia.
Delayed emptying of stent—Three esophagrams (5%) revealed
delayed emptying of contrast material from the stent because of recurrent
tumor in two patients and blood clot in one.
Esophageal-airway fistulas—Four esophagrams (7%) revealed
contrast material entering fistulas (two tracheoesophageal and two
esophagobronchial fistulas). The stents had all been placed for palliation of
known fistulas. In two patients, the stents failed to protect the fistulas
because their distal ends were improperly positioned above the fistulas. In
the remaining two patients, the stents were properly positioned, but barium
passed around the stent into the fistula in one and through the uncovered
distal end of the stent into a new fistula in the other
(Fig. 8). In two patients,
additional stents were placed.
View larger version (85K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 8 —79-year-old man with stent placed for palliation of
tracheoesophageal fistula caused by squamous cell carcinoma of esophagus. Left
posterior oblique spot image from single-contrast esophagram shows irregular
luminal narrowing (white arrows) in distal end of stent. Also note
barium in left mainstem bronchus (black arrows) from
esophagobronchial fistula that presumably developed as a result of tumor
ingrowth through adjacent uncovered distal portion of stent.
|
|
Stent migration—One esophagram (2%) revealed that a stent
placed across the gastroesophageal junction for palliation of a carcinoma of
the cardia had migrated into the stomach
(Fig. 9A). A new stent was
therefore placed (Fig.
9B).
View larger version (132K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 9A —62-year-old man with stent placed for palliation of dysphagia
caused by carcinoma of gastric cardia invading distal esophagus. Left
posterior oblique spot image from single-contrast esophagram shows distal
migration of stent (white arrows) into gastric fundus. Note barium in
distal esophagus (black arrows).
|
|
View larger version (130K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 9B —62-year-old man with stent placed for palliation of dysphagia
caused by carcinoma of gastric cardia invading distal esophagus. Malpositioned
stent was removed, and a new stent was placed across gastroesophageal
junction. Repeat examination 1 day after first study shows proper positioning
of new stent (black arrows) with proximal half in distal esophagus
and distal half in proximal stomach. Note how distal end of stent (large
white arrow) directly abuts greater curvature of proximal stomach.
Despite this finding, patient's dysphagia was adequately palliated by stent.
Polypoid carcinoma (small white arrows) is seen at gastroesophageal
junction.
|
|
Distal end of stent abutting gastric wall— Five esophagrams
(8%) revealed stents traversing malignant strictures at the cardia, with the
distal end of the stent abutting the greater curvature of the proximal stomach
(Fig. 9B). This finding raised
concern about the possibility of impending obstruction, but none of these
patients had problems with stent function.
Stent Fractures
None of the 59 esophagrams revealed stent fractures after stent
placement.
Perforation
None of the 59 esophagrams revealed esophageal perforations after stent
placement.
Discussion
Expandable metallic stents have been used with increasing frequency for
palliation of dysphagia or esophageal-airway fistulas in patients with
inoperable esophageal carcinoma or other malignant tumors involving the
esophagus. These stents are often evaluated by esophagography, so it is
important for radiologists to be familiar with the findings after stent
placement.
In our study, we made a number of observations about esophageal stents on
esophagography that are of little clinical importance because these patients
rarely had dysphagia. Tapered narrowing in the midportion of the stent was
often seen in patients in whom dysphagia was palliated after stent placement
(Figs. 1A and
1B). This phenomenon is
probably secondary to impingement on the expanding stent by surrounding tumor.
Flow of contrast material around one or both sides of the stent was another
frequent finding of little importance (Fig.
2), presumably resulting from incomplete apposition of the stent
against the esophageal wall. In one patient, however, contrast material flowed
around the stent into an esophagobronchial fistula, necessitating placement of
a new stent. Finally, two esophagrams revealed apparent narrowing of the
distal esophagus below the stent (Fig.
4A), raising concern about tumor overgrowth. In both patients,
however, this finding was caused by trapping of barium in gastric folds within
an incompletely distended hiatal hernia
(Fig. 4B).
The most feared complication of stent placement is esophageal perforation.
Such perforations are usually caused by erosion of the stent through a friable
tumor or an esophageal wall already damaged by mediastinal irradiation or
laser therapy [2]. When
perforation is suspected, radiographic studies with water-soluble contrast
agents may show extravasation of contrast material into the mediastinum or
pleural space. Esophageal perforation rates have ranged from 0% to 14% after
stent placement [1,
5,
6]. However, we did not
encounter any patients with this complication, indicating that stent placement
is a safe procedure associated with a low perforation rate when the stents are
deployed by endoscopists experienced in performing this procedure.
An esophageal stent may fail to palliate dysphagia if the stent is not
properly positioned or if it migrates distally because of inadequate anchoring
to the esophageal wall (Fig.
9A). Recurrent dysphagia may also be caused by tumor overgrowth or
ingrowth, epithelial hyperplasia, or adherent debris or blood clot.
Therapeutic options for palliating the patient's dysphagia include balloon
dilatation, laser therapy, stent revision, and endoscopic removal of any
debris or blood clots.
Tumor overgrowth is defined as extension of a tumor into one end of the
stent with varying degrees of obstruction
[2]. In our series, tumor
overgrowth was characterized by a polypoid defect above or below the stent or
by asymmetric mass effect and narrowing below the stent
(Fig. 5). In contrast, tumor
ingrowth is defined as extension of a tumor directly into the lumen through
uncovered metallic stents or through the uncovered proximal or distal ends of
covered metallic stents [2].
Tumor ingrowth through the uncovered distal end of the stent was characterized
on esophagography by irregular luminal narrowing with abrupt distal margins
(Fig. 6). Finally, epithelial
hyperplasia is defined as exuberant tissue overgrowth as a reaction to
metallic esophageal stents [7,
8]. In previous studies, the
frequency of epithelial hyperplasia has ranged from 2% to 28% after esophageal
stent placement
[8-10].
Epithelial hyperplasia was characterized by esophageal narrowing with a smooth
contour and more tapered margins than tumor ingrowth
(Fig. 7).
Covered metallic stents are also placed for palliation of esophageal-airway
fistulas. Stent failure can result from an improperly positioned stent or from
stent migration below the fistula. In our study, however, one patient with a
properly positioned stent developed a new esophagobronchial fistula because of
tumor ingrowth through the uncovered distal end of the stent
(Fig. 8). Another had a
properly positioned stent that failed to palliate a fistula because of flow of
contrast material around the stent. Radiologists, therefore, should be aware
that a covered stent may not palliate all esophageal-airway fistulas despite
proper positioning of the stent.
Our investigation has the inherent limitations of a retrospective study,
such as selection bias. The frequency of various complications related to
stent placement may therefore be skewed by our study population, which did not
represent a random sample but rather a selected group of patients, most of
whom had esophagrams within 3 days of stent placement. A subset of patients
had follow-up esophagrams 1 month or longer after stent placement because of
dysphagia, and these individuals were more likely to have abnormalities.
Because of the retrospective nature of our investigation, these follow-up
esophagrams were not obtained at uniform time intervals after stent placement.
The presence of different types of stents in our study patients represented
another confounding variable. Finally, it was not possible to have a
pathologic diagnosis in approximately 25% of patients with luminal narrowing
or masses because biopsy or surgical specimens were not obtained in these
individuals.
References
- Schaer J, Katon RM, Ivancev K, Uchida B, Rosch J, Binmoeller K.
Treatment of malignant esophageal obstruction with silicone-coated metallic
self-expanding stents. Gastrointest Endosc1992; 38:7
-11[Medline]
- Gollub MJ, Gerdes H, Bains MS. Radiographic appearances of
esophageal stents. RadioGraphics 1997;17
: 1169-1182[Abstract]
- Cwikiel W, Tranberg KG, Cwikiel M, Lillo-Gil R. Malignant
dysphagia: palliation with esophageal stents—long-term results in 100
patients. Radiology 1998;207
: 513-518[Abstract/Free Full Text]
- Therasse E, Oliva VL, Lafontaine E, Perreault P, Giroux MF, Soulez
G. Balloon dilation and stent placement for esophageal lesions: indications,
methods, and results. RadioGraphics 2003;23
: 89-105[Abstract/Free Full Text]
- Song HY, Do YS, Han YM, et al. Covered, expandable esophageal
metallic stent tubes: experiences in 119 patients.
Radiology 1994;193
: 689-695[Abstract/Free Full Text]
- Knyrim K, Wagner HJ, Bethge N, Keymling M, Vakil N. A controlled
trial of an expansile metal stent for palliation of esophageal obstruction due
to inoperable cancer. N Engl J Med 1993;329
: 1302-1307[Abstract/Free Full Text]
- Mayoral W, Fleischer D, Salcedo J, Roy P, Al-Kawas F, Benjamin S.
Nonmalignant obstruction is a common problem with metal stents in the
treatment of esophageal cancer. Gastrointest Endosc2000; 51:556
-559[Medline]
- Vakil N, Gross U, Bethge N. Human tissue responses to metal stents.
Gastrointest Endosc 1999;9
: 359-367
- Wang MQ, Sze DY, Wang ZP, Wang ZQ, Gao YA, Dake MD. Delayed
complications after esophageal stent placement for treatment of malignant
esophageal obstructions and esophagorespiratory fistulas. J Vasc
Interv Radiol 2001; 12:465
-474[Medline]
- Siersema PD, Hop WC, Blankenstein M, et al. A comparison of 3 types
of covered metal stents for the palliation of patients with dysphagia caused
by esophagogastric carcinoma: a prospective, randomized study.
Gastrointest Endosc 2001;54
: 145-153[Medline]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Top
Abstract
Introduction
