| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Clinical Observations |
1 Department of Radiology, Hospital of the University of Pennsylvania, 3400
Spruce St., Philadelphia, PA 19104.
2 Department of Surgery, Hospital of the University of Pennsylvania,
Philadelphia, PA.
Received March 26, 2007;
accepted after revision May 19, 2007.
Address correspondence to M. S. Levine
(marc.levine{at}uphs.upenn.edu).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CONCLUSION. Our findings indicate that intraluminal migration of a surgical drain after transhiatal esophagogastrectomy is an infrequent but serious phenomenon that hinders or prevents healing of an anastomotic leak. Radiologists should be aware of this phenomenon and should be able to recognize the findings of an intraluminal drain on contrast examinations. When such drains are identified, we believe that they should be promptly withdrawn or removed to facilitate healing of anastomotic leaks.
Keywords: anastomotic leaks • barium studies • esophageal cancer • esophagogastrectomy • gastric cancer • radiography • surgical complications
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Although percutaneous surgical drains have a beneficial role in the management of patients after transhiatal esophagogastrectomy, we have encountered patients who were adversely affected by intraluminal migration of the drain. The purposes of this investigation were to review our experience with a group of patients with contrast studies after esophagogastrectomy that revealed intraluminal migration of the drain and to describe the radiographic appearance and clinical relevance of this phenomenon.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Examination Technique
All four patients had postoperative upper gastrointestinal studies with a
water-soluble contrast agent (diatrizoate meglumine and diatrizoate sodium
[Gastroview, Mallinckrodt]). If the water-soluble contrast study showed no
evidence of perforation, the patient was then given a 250% weight/volume
barium suspension (E-Z-HD, E-Z-EM) to increase radiographic sensitivity for
the detection of anastomotic leaks
[5]. Spot images of the
anastomosis were routinely obtained with the patient in semiupright frontal,
left posterior oblique (LPO), and right posterior oblique (RPO) positions. All
of the studies were performed by radiology residents or one of three
gastrointestinal attending radiologists, and all were interpreted by one of
the attending radiologists.
The index study was defined as the first postoperative contrast study that showed intraluminal migration of the surgical drain. The four patients had a total of 14 contrast studies. Four studies were performed before the index studies (mean number of studies, 1; range, 0–2 studies), and six studies were performed after the index studies (mean number of studies, 1.5; range, 0–4 studies). The mean interval from surgery to the index study was 19 days (range, 11–27 days).
Review of Images
In the four patients with an anastomotic leak and intraluminal migration of
the drain, contrast studies were reviewed retrospectively by a consensus of
two experienced gastrointestinal radiologists who had no knowledge of the
clinical course. The images from the index studies were reviewed to determine
the size and location of the leak and to classify the leak as a collection,
track, fistula, or free perforation. The location of drain erosion into the
lumen was also noted. The images from follow-up studies were then reviewed to
determine whether there were any changes in the nature, size, and location of
the leak; whether the leak worsened, improved, or resolved; and whether the
drain remained intraluminal or had been withdrawn, removed, or replaced in
comparison with previous studies. Any other abnormalities were also noted.
Study Design
Two criteria were used to identify intraluminal migration of the surgical
drain. If the distal tip of the drain was seen overlying the lumen on 90°
perpendicular LPO and RPO images from the same examination, the drain was
considered to be intraluminal. The drain was also considered to be
intraluminal if the distal end appeared as an intraluminal filling defect in
the surrounding pool of contrast material.
The computerized medical records of the four patients were reviewed by one author to determine the clinical presentation, treatment, and patient course. The radiographic findings were then correlated with the clinical findings in these four patients.
Institutional Review Board Approval
Our institutional review board approved all aspects of this retrospective
study and did not require informed consent from any of the patients included
in our study.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Radiographic Findings
In all four patients, the index contrast study revealed that the distal end
of the surgical drain had migrated from the mediastinum into the lumen at the
site of the esophagogastric anastomosis (Figs.
1A,
1B,
1C–3A,
2A,
2B,
2C3B,
3C,
3D). In one patient, the
distal end of the drain was located at the anastomosis, and in the remaining
three, the distal end of the drain had migrated into the proximal
intrathoracic stomach below the anastomosis (Figs.
1A,
1B,
1C–3A,
3B,
3C,
3D). In all four patients, the
distal end of the drain was seen overlying the lumen on 90° perpendicular
LPO and RPO views and was also seen as an intraluminal filling defect in the
surrounding pool of contrast material (Figs.
1A,
1B,
2B,
2C,
3B, and
3C).
|
|
|
|
|
|
|
|
|
|
Correlation of Radiographic Findings with Treatment and Course
When index contrast studies revealed intraluminal migration of a surgical
drain, the drain was promptly removed in two of the four patients with an
anastomotic leak. Both patients who had the drain removed were discharged the
same day as drain removal and both were tolerating a regular diet without
complications on their 2-week outpatient follow-up visits. Resolution of the
leak was confirmed on a follow-up contrast study in one patient 15 days after
removal of the drain (Fig. 1C).
However, that study also revealed the development of a 5-mm-wide stricture at
the esophagogastric anastomosis (Fig.
1C).
In the remaining two patients, the surgical team initially chose not to manipulate the drain after the index study showing intraluminal migration of the drain. In both patients, follow-up studies 10 days after the index examinations revealed a persistent anastomotic leak, with the drain continuing to have an intraluminal location. In one of the patients, subsequent removal of the drain led to resolution of symptoms on a follow-up outpatient visit 10 days after drain removal. In the other patient, increasing output from the drain indicated worsening of the leak. Two weeks after the follow-up study, the drain was removed and replaced twice by an interventional radiologist. Both of the new drains were also shown to have an intraluminal location on follow-up contrast studies, with a persistent anastomotic leak (Figs. 3B and 3C). The drain was finally removed 12 weeks after placement, resulting in resolution of the leak on a final follow-up contrast study (Fig. 3D). The final study also revealed a 3-mm-wide stricture at the esophagogastric anastomosis (Fig. 3D).
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
At our institution, esophagogastrectomies are routinely performed by a transhiatal approach in which the esophagogastric anastomosis is placed in the lower cervical region at or near the thoracic inlet. A stapled anastomosis is created because it is associated with a lower rate of anastomotic leaks than a sutured anastomosis [6]. A Jackson-Pratt drain is then placed prophylactically by one of our thoracic surgeons.
These patients typically undergo a radiographic examination with water-soluble contrast material or a trial of oral feeding 1 week after surgery. Both of these tests are used to determine the patency of the anastomosis and rule out perforation. The choice of a radiographic test or feeding trial depends on the surgeon's experience and preferences. If a contrast study shows no evidence of a leak, the drain is removed, and the patient is started on a clear liquid diet. If a contained anastomotic leak is detected, however, the drain is left in place and oral feeding is withheld until a follow-up contrast study shows healing of the leak or the patient's symptoms resolve. If, alternatively, a contrast study shows an anastomotic leak that is not properly drained by the Jackson-Pratt drain, the surgical team or an interventional radiologist can reposition or replace the drain to obtain adequate drainage.
Although surgical drains have a beneficial role in the healing of anastomotic leaks, four (7%) of our 57 patients with anastomotic leaks after esophagogastrectomy had intraluminal migration of the drain into the region of the esophagogastric anastomosis or proximal intrathoracic stomach on postoperative contrast examinations. In all four patients, the drain was seen overlying the lumen on 90° perpendicular LPO and RPO views, and the drain was seen as an intraluminal filling defect in the surrounding pool of contrast material (Figs. 1A, 1B, 1C, 2A, 2B, 2C, 3A, 3B, 3C, 3D). To our knowledge, this study is the first in the radiology literature showing intraluminal migration of surgical drains in patients with anastomotic leaks after esophagogastrectomy. However, cases have also been reported in the surgery literature in which anastomotic leaks were associated with intraluminal migration of drains after Ivor-Lewis esophagogastrectomy [7, 8]. Radiologists therefore should be aware of the radiographic features of intraluminal drain migration so this phenomenon can be recognized on contrast studies in patients with anastomotic leaks.
The pathophysiology for intraluminal migration of a surgical drain after esophagogastrectomy is uncertain. Because the drain is placed under direct visualization by a surgeon or under fluoroscopic guidance by an interventional radiologist, it is unlikely that the drain has an intraluminal location at the time of placement. Instead, we suspect that it migrates into the lumen in the region of the esophagogastric anastomosis at the site of a preexisting anastomotic leak. In fact, two of our four patients with an intraluminal drain on index contrast studies had prior studies showing an anastomotic leak with a properly positioned drain (Fig. 2A). In both of these patients, the drain therefore most likely migrated into the lumen at the site of the leak shown on the prior contrast studies (Figs. 2B and 2C).
When a surgical drain migrated into the lumen at the site of the anastomosis after transhiatal esophagogastrectomy, removal of the drain led to healing of anastomotic leaks in two patients, with resolution of symptoms in both and resolution of the leak on a follow-up contrast study in one (Fig. 1C). Conversely, both patients in whom the drain initially was not manipulated had a persistent leak on follow-up studies; in both cases, the leak healed only after removal of the drain (Fig. 3D). In the two previously reported cases, the leaks also resolved only after withdrawal or removal of the drains [7, 8]. These findings indicate that intraluminal migration of a drain after esophagogastrectomy may hinder or prevent healing of an anastomotic leak. We therefore believe that surgical drains that have migrated into the lumen should be promptly withdrawn or removed to facilitate healing of anastomotic leaks.
Two (50%) of our four patients with an anastomotic leak and intraluminal migration of the drain developed an anastomotic stricture on follow-up contrast studies (Figs. 1C and 3D). It is well known that anastomotic leaks are a major cause of anastomotic strictures, which were reported in 47% of patients with leaks in one study [9]. Because intraluminal surgical drains prevent anastomotic leaks from healing, we are concerned that stricture formation may be even more likely in this subset of patients.
Our investigation has the inherent limitations of a retrospective study, including selection bias and interpretation bias. The small size of our study group also limits the generalizability of our findings. Finally, many patients at our institution had bedside feeding trials to rule out perforation rather than radiographic contrast studies after esophagogastrectomy. As a result, we may have underestimated the true frequency of intraluminal drain migration in those patients.
In conclusion, our findings indicate that intraluminal migration of the surgical drain after esophagogastrectomy is an infrequent but serious complication of surgical drain placement that may hinder or prevent healing of an anastomotic leak. When an intraluminal drain is identified on contrast studies, we believe that prompt withdrawal of removal of the drain is needed to facilitate healing of anastomotic leaks. It therefore is important for radiologists to be aware of this phenomenon and to recognize the radiographic findings of intraluminal surgical drains, particularly in patients with persistent anastomotic leaks, so appropriate management can be undertaken for their expeditious resolution.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
