| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Heart, Lung, and Esophageal Surgery Institute, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania
b Department of Surgery, University of Michigan, Ann Arbor, Michigan
c Department of Medicine, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania
d The University of Pittsburgh Cancer Institute Biostatistics Facility, Pittsburgh, Pennsylvania
Accepted for publication January 23, 2008.
* Address correspondence to Dr Luketich, The Heart, Lung, and Esophageal Surgery Institute, University of Pittsburgh, Suite C-800, 200 Lothrop Street, Pittsburgh PA 15213 (Email: luketichjd{at}upmc.edu).
Presented at the Forty-third Annual Meeting of The Society of Thoracic Surgeons, San Diego, CA, Jan 29–31, 2007.
| Dr Pennathur discloses that he has a financial relationship with Boston Scientific.
|
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
Methods: A total of 58 Polyflex stents were placed in 38 patients over a two-year period. There were 24 men and 14 women, with a median age of 63 years (range, 25 to 83). The most common indication for placement was an esophageal stricture in 25 patients (66%); other causes included leak in 8 (21%) and tracheoesophageal fistula (TEF) in 5 (13%). We evaluated the hospital course, complications, and outcomes.
Results: The median postoperative stay was one day. Complications included migration in 38 stents (63%) (28 patients; 73%), retrosternal chest discomfort in nine, reflux in four, airway obstruction in one, and food impaction in three. Continued leak or a persistent TEF occurred in five patients (38%). Reintervention was required predominantly due to migration of the stent at a mean interval of 46 days (range, 1 to 353). Patients with dysphagia improved significantly with dysphagia scores (1 = no dysphagia; 5 = unable to swallow saliva) improving from 3.44 to 2.15 (p < 0.0001).
Conclusions: Polyflex stents were effective in the relief of dysphagia due to strictures. They were less effective in esophageal perforations or leaks. Their primary disadvantage is a high migration rate and further improvements in design are required to decrease this high incidence of migration.
There has been a dramatic increase in the incidence of esophageal neoplasm in the United States over the last two to three decades, with an increase in the incidence of adenocarcinoma while the incidence of squamous cell cancer has decreased. The reported overall five-year survival of esophageal neoplasm is 14% [1]. The majority of patients are diagnosed at an advanced stage, and those with advanced disease typically present with dysphagia. The ideal palliation in these patients should provide relief of dysphagia and ensure quality of life. The various options for palliation include dilation, radiotherapy, ablative therapy, surgery, and stents [2].
Esophageal stents have been used with successful relief of dysphagia in many of these patients [2]. In the past, conventional plastic stents were commonly used. These semirigid plastic stents required dilation and were associated with an increase in complication rate. Subsequently, self-expanding metal stents (SEMS) were developed and the advantages of this design included a smaller, more flexible delivery system and increased ease of deployment. In 1993, Knyrim and colleagues [3] performed a randomized study comparing conventional plastic stents and uncovered SEMS. Although relief of dysphagia was similar, complications after plastic stent placement were significantly higher and associated with a longer hospital stay. These authors concluded that SEMS, despite their higher initial cost, were more cost effective in the management of symptomatic, unresectable esophageal cancer, because of lesser complications and a shorter hospital stay. The SEMS have become very popular and have been used widely in the last decade.
The SEMS are not without morbidity, and major complications including tumor ingrowth and overgrowth with resultant obstruction, hemorrhage, and perforation have been reported [3]. Covered SEMS have a decreased the tumor ingrowth, but have a higher migration rate. In addition, SEMS are difficult to remove.
Recently, a new generation of the self-expanding plastic esophageal stent (Polyflex; Boston Scientific Corp, Natick, MA) has been introduced in Europe and the United States. This removable stent was designed to combine the features of both the plastic and self-expanding metallic stent, with potential use in patients not only with malignant but also with benign diseases. The main objective of this study was to evaluate our initial experience with the use of Polyflex self expandable plastic stents in the treatment of esophageal disease from two institutions.
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
Characteristics of Stent
The Polyflex stent (Boston Scientific Corp) is a self-expanding plastic stent made of a polyester netting embedded in silicone. Its wall thickness is 0.45 mm and it has radio-opaque markers at both the ends and in the middle. It is available in three diameters with a proximal flared end (16, 18, and 21 mm diameter with corresponding 20, 23, and 25 mm proximal flared end); available lengths are 9 cm to 15 cm. The delivery system for the stent varies from 12 to 14 mm and strictures are typically dilated 1 mm larger than the planned diameter of the delivery device.
Placement of Stent
Stents were placed either under monitored anesthesia care with sedation (n = 3) or in the operating room under general anesthesia in the reminder of patients. After evaluation with flexible esophagogastroduodenoscopy (EGD), the proximal and distal extent of the lesion was evaluated. The applicator system varies in diameter from 12 to 14 mm and strictures were dilated to accommodate the stent delivery system. If dilation was required endoscopic dilation was performed over a guide wire, typically with fluoroscopy guidance prior to placement of the stent. Bronchoscopy was performed in all upper and middle third tumors before and after stent placement to assess for airway compression. If there was concern of airway compression, a bougie of a diameter similar to the stent was placed to assess airway compression prior to placement of the stent. The stent was loaded in the applicator system and placed under fluoroscopic guidance. We then performed an EGD to verify the position. A contrast esophagogram was performed postoperatively in most patients.
Data Collection and Statistical Analysis
Data, including patient demographics, tumor characteristics, and concomitant treatment, were collected. Dysphagia scores, ranging from a scale of 1 to 5, were recorded preoperatively and after stent placement (Table 1). At the University of Pittsburgh, the dysphagia scores are recorded as part of the routine data collection in the clinic. When they were not specifically available, the scores were determined retrospectively by review of clinic notes and hospital charts. At the University of Michigan, the dysphagia scores were determined retrospectively. These scores were determined both prior to stent placement and after stent placement. Specific endpoints studied were relief of dysphagia, complications, requirement for reintervention, and the time to reintervention. The probabilities of survival and time to reintervention were estimated by the Kaplan-Meier method with Greenwood confidence limits. Association between categoric variables was tested with the Fisher exact test or the exact 
2 test. Changes in dysphagia score were evaluated with the signed rank test. All tests were two tailed.
|
|
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
|
Survival and Relief From Dysphagia
During the follow-up period, there were 14 deaths with a median survival of 18 months (95% confidence interval: 10- not reached). Patients who are alive have been followed-up for a median of 11 months (range, 0.3 to 36 months). The median survival in patients treated for malignant disease was 5.4 months (95% confidence interval: 3.3 months- not reached) (Fig 1). Mean dysphagia score improved from 3.5 to 2.2 (p < 0.0001).
|
|
|
|
A total of 12 patients underwent stent placement for benign stricture (esophagus n = 9, gastric conduit n = 3). Of the nine patients undergoing esophageal stent placement, three patients with laryngopharyngeal neoplasms had stents placed for postradiation upper esophageal stricture and six stents were placed for other indications, primarily reflux-related stricture. Dysphagia improved in eight of nine patients (89%). Migration was again the most common complication, occurring in seven patients. The stricture resolved in two patients and the others required reintervention for stricture. Three patients had stents placed in the gastric conduit after esophagectomy; all these stents migrated and required reintervention for stricture. In some patients, migration was associated with resolution of the stricture.
Specific Outcomes After Stent Placement for Leaks or Fistulae
A total of 13 patients underwent stent placement for leaks or malignant tracheoesophageal fistula. There were eight patients who underwent stent placement for leaks, including five patients with an anastomotic leak and three for iatrogenic injury. Leaks persisted in two of eight patients (25%); the stents ultimately succeeded in three of eight patients (38%). A total of five patients underwent stents for tracheoesophageal fistula. There was a progression of the fistula requiring removal in one patient and erosion in one patient. Recurrent or persistent fistula was noted in two patients and one patient was palliated until death. Overall the leaks and fistulae persisted in a total of 5 of 13 patients (38%).
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
In the combined early experiences from two institutions reported in this article, we evaluated the use of a SEPS. These stents were placed primarily for esophageal stenosis. Dysphagia score improved significantly in patients with strictures. In our institution, tumors with a predominant endoluminal component are primarily treated with photodynamic therapy and expandable stents are used in patients with significant extrinsic compression or patients who are not willing to avoid sun-exposure for six weeks. This approach may have resulted in patients with more advanced disease who were treated with stents. Dysphagia score, however, improved significantly in patients with both malignant and benign strictures. This relief of dysphagia is comparable with our experience with SEMS [4].
The main complication we encountered with these stents was migration of the stent, which occurred in 63% of the stents and 73% of patients. Ott and colleagues [5] reported a 37% migration rate, and a 45% migration when placed at the gastroesophageal junction. Similarly, Evrard and colleagues [6] reported migration in 11 patients (57%) out of 21 patients. Five of these patients had the SEPS stent placed within a metal stent for hyperplastic tissue ingrowth. Excluding these patients, the incidence of migration was 69% (11 of 16 patients). Other authors have reported a lesser incidence of migration. Dormann and colleagues [7] reported their results in 33 patients with malignant strictures with two migrations (6.0%).
Multiple factors may contribute to migration in these stents. The possible contributing factors for the increased incidence of stent migration include design features such as the complete silicone coverage of the stent, which is intended to limit ingrowth of tissue. The diameter of the delivery device ranges from 12 to 14 mm, which is larger compared with the SEMS. Therefore, one of the requirements for placement of the stent is to dilate the stricture more than is typically required prior to SEMS placement and this may contribute to the migration. In our series, almost all patients required dilation prior to placement of the stent. Other potential factors in migration may relate to the indication for the stent placement, presence or absence of a stricture, location of the stent, stricture characteristics, and the size of the stent. However, because this is a small series, in our analysis, several of these factors were evaluated, and none appear to have a significant association. In some instances, the resolution of the stricture may be associated with stent migration.
In our previously reported experience in 100 consecutive patients who had placement of a SEMS, the mean time to reintervention was 82 days [4]. The main reason for reintervention in these patients was tumor ingrowth or overgrowth in these stents. The most frequent reintervention was photodynamic therapy. In the current series with SEPS, the mean interval to reintervention was 46 days (range, 1 to 386) and the most common reason for reintervention was migration of the stent. Stent migration was addressed with either replacement, repositioning, or removal of the stent. Because these reports were in different time periods, and with the limitations of retrospective studies including selection bias, we cannot make direct comparisons between results of these studies.
These stents have also been placed for leaks and fistulae [8–11]. Hunerbein and colleagues [8] reported their experience in nine patients who had stents placed for anastomotic leaks after esophagectomy. Migration of the stents occurred in two patients; however, the stent was successful in eight of the nine patients. In our experience, the ultimate success of the use of the stent for leaks has been lower when compared with their efficacy in terms of relieving dysphagia. Caution should be exercised because the size of the disruption may worsen after stent placement. In patients with a tracheoesophageal fistulae, the use of the stent was associated with erosion in one patient and enlargement of the fistula in another requiring reoperation. Langer and colleagues [9] reported that in 2 of 24 patients who were treated with a SEPS for anastomotic leaks, the anastomotic disruption worsened, requiring reoperation. Several factors have to be taken into consideration in the treatment of leaks. Important considerations in the management of leaks include the nature of the presentation, the condition of the patient, etiology of the leak, the location of the leak, assessment as to whether the leak is contained, assessment of ischemia, adequate debridement and drainage, and surgical decision regarding surgical repair. The role of stents as an adjunct in selected patients with leaks has not been defined.
In terms of patient selection, the device may be more difficult to use for tight, fibrotic strictures (eg, postradiation), particularly because of its large delivery system. Another area where this stent should be used cautiously is for patients with upper esophageal stricture because of risk of airway compromise. These stents can be used for mid-distal esophageal stricture, without severe fibrosis, which can be predilated safely to accommodate the delivery system. Their primary disadvantage is a high migration rate and further improvements in design are required to decrease this high incidence of migration. One of the advantages of the expandable plastic stent is the ease of removal and hence may be potentially used in patients with benign disease, such as selected patients with benign strictures who are not operative candidates. In addition, because of the ease of removal, they may be potentially applicable in patients with malignant obstruction while undergoing neoadjuvant chemotherapy prior to surgical resection.
The current study has the limitations, such as selection bias, that are inherent to retrospective studies. The patients treated in this study comprise a very heterogeneous group, encompassing not only advanced esophageal cancer but also including patients with benign disease. In our experience although dysphagia was relieved, there was a high rate of migration. Further improvements in design are required to decrease the incidence of migration.
We have reported our initial experience with the use of SEPS from two institutions. This stent was effective in the relief of dysphagia due to strictures. However, it was less effective in the management of esophageal leak or tracheoesophageal fistula. Their primary disadvantage is a high migration rate and further improvements in design are required to decrease this high incidence of migration. There are several factors which merit further investigation, including optimal patient selection, evaluation of stent anchoring mechanisms, and use of this stent in combination with other therapies. Further prospective studies are required to address these issues and to define the role of self expanding plastic stents in the treatment of esophageal diseases.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
DR PENNATHUR: All 3 patients had problems with migration of the stent within a matter of 3 to 28 days, and essentially the stents were removed. It did help when the stent was in place. The reason for the stents was that one patient was about 7 years after esophagectomy but had somewhat of a redundant stomach, had a revisional surgery, but still had symptoms and it was felt that the stent may help. Two other patients, both were status post esophagectomy with some constriction right around the hiatus; these were stented, and it migrated in all of them.
DR DANIEL L. MILLER (Atlanta, GA): We all know that this is a very difficult group of patients to take care of and we're still looking for the ideal stent. You had a diverse group of patients with this stent. We're seeing more and more upper esophageal stenosis, and I think it came out in this study. Usually after patients have undergone chemoradiation for head and neck cancer, they've got a high cervical stenosis and you have to dilate them (inaudible). I saw in your series that 42% of them had upper stents. My question is, would you use a Polyflex stent for that area? It causes a tremendous amount of pain, the patients can't tolerate it, and so you've got to remove it almost 60%, 70% of the time. So we have not been very pleased with the Polyflex in that high position. I wanted you to comment on that. Now, after you've reviewed your data, what is your recommendation for the upper cervical stenosis?
DR PENNATHUR: Thank you, Dr Miller, for your questions. I agree that patients with a proximal stenosis can comprise of a difficult group of patients. The patients who were classified as proximal included some anastomotic strictures after esophagogastrectomy. The patients who were in this series form a heterogeneous group of patients, and there were some patients who had head and neck cancer, radiation, and these are a difficult group of patients to treat. The stent was placed primarily because these patients were getting dilated so frequently, and, thus, the stent was being placed as an alternative to see whether the stricture would remodel. Certainly the high proximal stricture is a difficult problem. There were a couple of patients in whom we had to remove it within a few days because of global sensation. We generally place a dilator of the proposed size of the stent in the esophagus and make sure there's not significant compression of the airway before placing the stent. Despite that, one patient did have some stridor, and so we removed the stent postoperatively. So I think that this group with a high stricture comprise a difficult group of patients. In general, when considering endoscopic therapies, for palliation of malignant disease: when the cause of the stricture is endoluminal disease, one can consider photodynamic therapy and in patients with extrinsic compression, stents can be considered. This study was not designed to assess all options, and comprised of a group of patients who were so heterogeneous. Therefore, I think I won't be able to give you a specific recommendation in terms of what to do with a difficult proximal stricture.
DR MILLER: Because the ideal stent is not ...
DR PENNATHUR: That's right, the ideal stent has not been discovered.
DR MILLER: And the last thing, in regards to your migration rate, which was 71%, how have you corrected that? You've got to do a better job in sizing. Most of the time, if we put them up high, we downsize, and if you put them down low, you increase in size, you know, so you don't get that migration into the stomach. How have you corrected that?
DR PENNATHUR: Well, there are a few things I think we need to do in terms of migration, first, design of the stent and second any technical issues we can address. One technical adjunct may be to anchor the stent with clips in the esophagus, which we have tried. We have also tried placing a stitch in the stent and have attempted to anchor the stent so that it doesn't slip. The other issue in terms of sizing is that to place a 21-mm stent, which has a 25-mm proximal flare, it requires a 14-mm delivery apparatus; and that requires us to dilate it about 45-French, prior to stent placement. It is sometimes hard to dilate immediately, particularly in cancer, so as to minimize the risk of perforation, so we might use a smaller diameter stent; and the other danger, I think, is particularly when using stents with perforations, which some investigators have used successfully; others have reported anastomotic disruption and so forth with the bigger stent. Finally we analyzed some of the factors, including size which may be associated with migration; we did not see a significant association with size. Thank you for your comments.
DR GEORGE B. HAASLER (Milwaukee, WI): When you have now finished looking at this series, do you think that the reason for the migration rate was more that they were oversized or they were undersized, or was it more design of the stent that you think is the main factor in the migration?
DR PENNATHUR: Thank you, Dr Haasler, for an excellent question. I think that it's a difficult thing to answer. That's why we did an analysis on all these patients looking at several factors including size, and the association with migration and we couldn't identify a specific factor associated with migration. I think the reason it migrates is multifactorial. I think we can make improvements in terms of the stent design; in terms of anchoring we can make some improvements; in terms of how we can size it; and, importantly, in terms of the delivery apparatus in by you have to deliver the Polyflex stent. In comparison, the self-expanding metal stents have a 5-mm to 6-mm delivery apparatus and you don't have to predilate as much. So I think that the reason the stents migrate is multifactorial. Thank you for your comments.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
