|
|
||||||||
Department of Thoracic and Cardiovascular Surgery, St. Vincent Hospital, Indianapolis, Indiana
Accepted for publication February 7, 2011.
* Address correspondence to Dr Freeman, 8433 Harcourt Rd, Indianapolis, IN 46260 (Email: rfreeman{at}corvascmds.com).
Presented at the Fifty-seventh Annual Meeting of the Southern Thoracic Surgical Association, Orlando, FL, Nov 3–6, 2010.
| Abstract |
|---|
|
|
|---|
Methods: Over a 4-year period, patients found to have an acute, significant intrathoracic anastomotic leak after esophagectomy for benign or malignant disease undergoing surgery at or transferred to a single institution were offered endoluminal esophageal stent placement as initial therapy. Stents were placed endoscopically utilizing general anesthesia and fluoroscopy. Leak occlusion was confirmed by esophagram. Patients were followed until their stent was removed and their anastomotic leak had resolved.
Results: Seventeen patients had an esophageal stent placed for an anastomotic leak during the study period. Leak occlusion occurred in all 17 patients. One patient was found to also have a perforation of the gastric conduit and underwent operative repair. Fourteen patients (82%) were able to initiate oral nutrition within 72 hours of stent placement. Stent migration occurred in 3 patients (18%), requiring repositioning in 2 and replacement in 1. All stents were removed at a mean of 17 ± 9 days after placement.
Conclusions: Endoluminal esophageal stent placement is a safe and effective method for the treatment of an intrathoracic anastomotic leak after esophagectomy. This treatment resulted in rapid leak occlusion, provided the opportunity for earlier oral nutrition, and avoided the potential morbidity of reoperative repair or esophageal diversion.
| Introduction |
|---|
|
|
|---|
| GENERAL THORACIC SURGERY:
The Annals of Thoracic Surgery CME Program is located online at http://cme.ctsnetjournals.org. To take the CME activity related to this article, you must have either an STS member or an individual non-member subscription to the journal.
|
Intrathoracic leak after esophagogastrostomy is associated with a significant increase in patient morbidity and mortality. Traditional therapy has been continued drainage of the pleural space and antimicrobial therapy for small leaks and operative repair or, if unsuccessful, esophageal diversion for larger leaks. Such treatment prolongs hospitalization and significantly delays oral hydration and nutrition. In the case of an unsuccessful anastomotic repair, esophageal diversion commits the patient to further surgery to reestablish foregut continuity, which, itself, is associated with a significant rate of morbidity.
Experience with endoluminal esophageal stents for perforation and chronic fistulae by our group, and others, has resulted in the recognition of their value in the treatment of complex esophageal disorders [1–4]. A natural progression of esophageal stent use has been to question their effectiveness in treating patients with an anastomotic leak after esophagectomy. Reports early in the learning curve of modern esophageal stents use produced encouraging but mixed results in these patients [5, 6]. However, an assessment of the efficacy of treating an anastomotic leak after esophagogastrostomy using a hybrid endoluminal and operative approach, which has gained favor for other esophageal disorders, is lacking. This investigation reviews our experience treating patients with an intrathoracic anastomotic leak after esophagectomy using such a hybrid approach.
| Material and Methods |
|---|
|
|
|---|
The protocol utilized is as follows. The presence of an anastomotic leak was documented and localized by gastrograffin or barium esophagram, or both, before any treatment (Fig 1 ). To be considered a significant leak eligible for treatment other than observation, contrast had to be seen leaving the lumen of the anastomosis into the mediastinum or pleural space. Additionally, all patients being considered for stent placement underwent computer-aided tomographic imaging of the neck, chest, and abdomen. Excluded from participation in this investigation were patients who also had a leak from their gastric conduit distal to and remote from the anastomosis recognized in the preoperative evaluation, complete dehiscence of their anastomosis, or had something besides the stomach used as a conduit to reestablish gastrointestinal continuity.
|
|
|
Continuous data for this investigation are expressed as the mean plus or minus the standard deviation of the mean except where otherwise indicated.
| Results |
|---|
|
|
|---|
The patients' mean age was 62 ± 13 years (range, 48 to 68 years). Patients underwent esophagectomy for esophageal cancer in 16 patients (94%) and for chronic, recalcitrant strictures after lye ingestion in 1 patient (Table 1). Fifteen patients (88%) received neoadjuvant chemotherapy and radiation therapy in preparation for surgery. The regimen varied secondary to the different referring institutions but all involved cisplatinum-based chemotherapy. Total radiation dose was generally 45 Gy for patients treated at other institutions and 60 Gy for patients treated at the study institution. Two patients (12%) underwent operative repair of their anastomotic leak before transfer to our facility, without success.
|
|
|
There were no complications associated with stent placement or removal in the study population. None of these patients required further surgery for their anastomotic leak. One patient required a video-assisted thoracoscopic thoracic duct ligation for a chylothorax. Stent migration occurred in 3 patients (18%), requiring revision in 2 and replacement in 1. One patient (6%) had an esophageal stricture at the site of the anastomosis 2 months after stent removal that required a single endoscopic dilation. Other morbidities included respiratory failure (3), pneumonia (1), and deep venous thrombosis (1). There were no deaths in this series of patients.
| Comment |
|---|
|
|
|---|
Leak after intrathoracic esophagogastrostomy also remains a source of significant morbidity and mortality for patients undergoing esophagectomy. Junemann-Ramirez and colleagues [8] found hospital mortality to be 35.7% compared with 4.2% for patients with and without an anastomotic leak, respectively. Crestanello and coworkers [9] reported that an anastomotic leak was also a negative predictor of long term-survival.
The current treatment of an intrathoracic anastomotic leak after esophagectomy remains nonoperative therapy for small leaks and reoperative surgical intervention for all other leaks. Nonoperative therapy has been demonstrated to be successful in treating selected patients. However, many times the patients who would benefit the most from avoiding the physical stress and physiologic setback of further surgery are not appropriate for traditional nonoperative therapy. Furthermore, no attempt is made to seal the leak or prevent extension of the anastomotic dehiscence, resulting in the need for operative intervention in as many as 90% of patients initially treated in this manner [9].
Reoperative thoracotomy with revision of the intrathoracic esophagogastrostomy can also be unsatisfying. In addition to the obvious setback to the patient's recovery, primary repair of the anastomosis with or without the application of pedicled muscle has a failure rate reported conservatively as 25% [9]. Moreover, if the anastomosis cannot be repaired at the initial reoperation or if the initial repair fails, the surgeon is faced with the likelihood of esophageal diversion. Such a procedure is of concern for the surgeon as a significant percentage of patients never undergo the reestablishment of gastrointestinal continuity. Those that do face significant rates of morbidity and, not infrequently, marginal functional results [10].
Experience in treating patients with an esophageal perforation or fistula caused us to question whether esophageal stent placement could offer another treatment option for patients with an intrathoracic anastomotic leak after esophagectomy. Such a technique would provide the opportunity to immediately seal small, moderate, and large leaks, eliminate the need for reoperative thoracotomy, and allow earlier oral hydration and nutrition. Cost and hospital stay could also potentially be reduced when compared with traditional operative and nonoperative therapy.
Such a treatment strategy has been evaluated before in different contexts. Zisi and associates [5] reported a series of 8 patients treated for an anastomotic leak after esophagectomy using a self-expanding metal esophageal stent. That was successful in 7 patients (88%) without any stent migration, as expected with a metal stent. However, the metal stent proved difficult to remove and resulted in 1 patient having a tracheoesophageal fistula.
Langer and coworkers [6] reported a similar series of 13 patients who experienced an intrathoracic leak after esophagectomy. However, these patients had a silicone-coated, polyester esophageal stent placed. That resulted in leak occlusion in all 13 patients. However, stent migration occurred in 6 patients (46%) with the stent in place for a median of 57 days. Hospital mortality in this series was 25%.
This investigation summarizes a relatively large series of patients who had an intrathoracic anastomotic leak after esophagectomy treated with an esophageal stent. A high rate of leak occlusion was obtained while stent migration was minimized by oversizing the stent and minimizing the time the stent was used, respectively, as we have previously suggested. Furthermore, oral hydration and nutrition were able to commence quickly after stent placement. Hospital length of stay compared favorably to reports of patients treated by conventional nonoperative or operative therapy. Morbidity, and especially mortality, also compared favorably with other series of similar patients.
Although it would appear that endoluminal esophageal stent placement for anastomotic leak after esophagectomy could provide an additional treatment option beyond traditional operative or nonoperative therapy, this investigation has some weaknesses. While representing the largest number of such patients treated in this manner in the current literature, a study population of 17 remains a small number even for a relatively uncommon condition. Furthermore, no attempt was made to compare endoluminal therapy and traditional operative or nonoperative therapy in either a prospective or retrospective fashion.
In conclusion, this series demonstrates the effectiveness of a removable, occlusive esophageal stent in treating patients with an intrathoracic anastomotic leak after esophagectomy. Endoluminal stent placement provided rapid sealing of the leak, allowed patients to begin oral intake within 48 hours of stent placement, and eliminated the need for further surgery in the vast majority of patients while producing minimal morbidity and no mortality in this series. Stent placement also offered an appealing alternative to reoperative repair, and especially to esophageal diversion and subsequent reconstruction. Proper patient selection and stent sizing was also shown to minimize stent migration. Thoracic surgeons caring for patients with an intrathoracic leak after esophagectomy should be familiar with this technique and consider its use as an alternative to reoperation or extended observation of intrathoracic anastomotic leak after esophagectomy.
| Discussion |
|---|
|
|
|---|
DR FREEMAN: There are two of us who do exclusively noncardiac thoracic surgery in our practice. I generally wait 7 or 8 days and get an esophagram. My partner does not get an esophagram. He often sends the patients home and gets that as an outpatient. So unless they manifest any symptoms of a leak, he is not going to study them. A large percentage of these patients also came from outside our institution.
DR SEE: I don't know how you perform yours, but I do it as a side-to-side. Does the stent migrate in that setting any worse?
DR FREEMAN: Interestingly, because a lot of these people came from different hospitals, we saw just about every imaginable anastomosis, handsewn, stapled, side-to-side, the trapdoor anastomosis, and for the most part, it worked in all those patients, mainly because they had significant intrathoracic esophageal leak. Our technique is to leave two thirds of the stent in the remaining esophagus, with that other third bridging the anastomosis.
DR SEE: Thank you very much.
DR J. MICHAEL DIMAIO (Dallas, TX): Excellent job, as usual, and thanks for pushing this field along. This is a very difficult problem. This offers a very good alternative to the open repair. Two questions. Number one, what types of stents are you using? I imagine you are going to tell me "myriad stents." I know we have now progressed to the Boston Scientific Wallstent, because the part of the stent that is not covered ends up being able to grasp the tissue and not migrate. That's question number one. And number two, where do you position the stent in relation to the leak? And the third question, nasogastric tube placement after the stent.
DR FREEMAN: Well, I have to recognize Dr DiMaio, who helped me put in my first esophageal stent when I was in training.
We use mainly Polyflex stents. Again, neither my partner nor I are paid by any of these companies. I like the radial force of the Polyflex. That being said, we did use some Alveolus stents for a while. Right now the WallFlex, which is not the same thing as the self expanding metal Wallstent but a new, removable stent, has certainly gained some favor in our practice. We generally try to leave a nasogastric tube through the stent for a few days in contrast to our perforation patients, in whom we place a percutaneous endoscopic gastrostomy tube. I worry a lot about migration, so I tend to leave about two thirds of the stent in the intrathoracic esophagus—I like to see it just below the arytenoid when I come back with the scope, and have that one third bridge the anastomosis.
DR DIMAIO: Thank you, Richard. Good job.
DR ROBERT SHEN (Rochester, MN): I enjoyed your presentation as well. One of the late sequela of an anastomotic leak is that the patient is going to be at very high risk for having an anastomotic stricture. In many cases, the stricture becomes a much more difficult management issue than the leak itself. I think I heard you say that only 1 of the patients in this series actually required an esophageal dilation, and I was wondering if you think that the stent itself somehow changes the way an anastomotic leak heals, and that in doing so, this is a proposed benefit for using this strategy of managing the anastomotic leaks versus more traditional approaches.
DR FREEMAN: I think that is a very good question, and the short answer is I don't know. My personal experience is that before putting stents in these patients, many of them did have strictures that were very hard to treat, and I wonder if having a stent in there, so that the anastomosis heals open rather than granulates in a more narrow fashion, may be a benefit, but I don't think we have that answer.
DR SHANDA HALEY BLACKMON (Houston, TX): Great paper, Richard. Have you considered using muscle flap as an adjunct to stenting, and have you ever done that? And in big perforations or big holes, does that salvage the esophagus? Do you think it keeps you from having a stricture specifically and gives you healthier tissue in the area that heals, instead of just fibrosis? And my other question is, does that help it to heal faster?
DR FREEMAN: We have reoperated on some patients, obviously, and we put muscle on there if it is not already. In our practice, we put intercostal muscle on the anastomosis if the patient had chemoradiation before surgery. We have not gone back and put muscle on a stented patient. Surprising to me is the size of the anastomotic leak that the stent will seal, because, in theory, you are not really healing that area as much as you are just expanding it so that it contacts other tissue and seals itself.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
C. D. Lyons, M. P. Kim, and S. H. Blackmon A Novel Fixation Procedure to Eliminate Covered Self-Expanding Metal Stent Migration Ann. Thorac. Surg., November 1, 2012; 94(5): 1748 - 1750. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. K. Freeman, A. J. Ascioti, T. Giannini, and R. J. Mahidhara Analysis of Unsuccessful Esophageal Stent Placements for Esophageal Perforation, Fistula, or Anastomotic Leak Ann. Thorac. Surg., September 1, 2012; 94(3): 959 - 965. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |