| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2001;71:1797-1802
© 2001 The Society of Thoracic Surgeons
a Section of Thoracic Surgery, University of Pittsburgh, Pittsburgh, Pennsylvania, USA
Address reprint requests to Dr Luketich, Presbyterian Hospital, University of Pittsburgh Medical Center, 200 Lothrop St, Pittsburgh, PA 15213
e-mail: luketichjd{at}msx.upmc.edu
Presented at the Thirty-sixth Annual Meeting of The Society of Thoracic Surgeons, Fort Lauderdale, FL, Jan 31–Feb 2, 2000.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
Methods. Over a 48-month period, 127 stents were placed in 100 patients with dysphagia from esophageal cancer (93%) or lung cancer. Most had undergone prior treatment. Dysphagia scores, duration of palliation, complications, and reintervention were evaluated.
Results. Immediate improvement in dysphagia was observed in 85% of patients with no procedure-related deaths. Dysphagia score decreased from 3.3 before stent to 2.3 (p < 0.005). Average interval to reintervention was 80 days. In 40 patients surviving more than 120 days, 31 (78%) required reintervention. Major complications occurred in 3 patients receiving poststent chemoradiation (tracheoesophageal fistula, T1 vertebral body abscess, mediastinal abscess). Other complications included unsatisfactory deployment requiring immediate removal (3 patients), migration (11 patients), pain requiring removal (2 patients), food impaction (10 patients), and tumor ingrowth (37 patients).
Conclusions. Expandable metal stents offer excellent short-term palliation of malignant dysphagia. In long-term survivors, recurrent dysphagia requiring reintervention is common. In a small subset of patients receiving chemoradiation after stent placement, major complications were observed.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
Options for palliation of malignant esophageal obstruction include operation, dilation, radiation therapy, laser ablation, and esophageal stents. Effective endoscopic interventions such as laser therapy or expandable metal stents have all but eliminated esophagectomy for palliation alone, although esophageal exclusion and bypass may be an option in highly selected cases. Laser ablation of tumor by photodynamic therapy or other means offers good palliation for some patients but is less effective in the setting of malignant strictures or extrinsic compression [6].
Expandable metal stents offer good short-term relief of malignant dysphagia and have been shown to be easier to deploy and associated with less complications than plastic endoprostheses [7–10], but complications in larger series and long-term outcomes have not been well described. The aim of this report was to summarize our experience with expandable metal stents for palliation of malignant dysphagia in our first 100 patients.
|
Material and methods |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
Expandable metal stents (Microvasive Ultraflex, Watertown, MA; Schneider Wallstent, Minneapolis, MN) were inserted under fluoroscopic and endoscopic guidance with the patient under conscious sedation. Bronchoscopy was performed on all upper and middle third tumors before and after stent placement to evaluate for airway encroachment. If there was any concern over airway compression, a bougie of diameter similar to an expanded metal stent was placed across the esophageal obstruction, and a simultaneous bronchoscopy was performed to assess airway patency before actual stent deployment. External markers were placed under fluoroscopic guidance while performing endoscopy to mark the proximal and distal extent of the malignant stricture. Stents were selected that were approximately 4 cm longer than the stricture. If the endoscope could not be passed through the stricture, balloon dilation (Microvasive) was performed to allow passage of the endoscope distal to the stricture. Next, a guidewire was passed through the narrowed esophagus, and the compressed stent on the delivery device was passed through the stricture and deployed under fluoroscopic guidance.
Palliation of dysphagia was evaluated using a standard dysphagia score (Table 1), which has been previously reported [11]. Dysphagia score was measured before and at 48 hours after stent insertion. Patients were followed until either death or surgical intervention.
|
|
Results |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
|
|
|
Immediate improvement in dysphagia was seen after initial stent deployment in 85% of the entire group of 100 patients. This was defined as an improvement in dysphagia score of at least one grade after stent placement (Table 1). The mean prestent dysphagia score was 3.3, which improved to 2.3 after stent placement (p < 0.005, Wilcoxon rank sum test). The reasons identified for initial stent failure included inability to take oral alimentation after stenting because of extremely poor performance status (n = 5), intractable reflux or pain requiring stent removal (n = 2), poor stent expansion because of extremely bulky tumors or inadequate dilation (n = 3), and stent malposition (n = 5). In patients with poor performance status no further treatment was instituted. In patients with poor stent expansion or stent malposition, stents were removed and photodynamic therapy was performed.
There were 94 stent placements, which were immediately successful in patients who were not planning on undergoing esophagectomy. More than half (51 patients) required reintervention at a mean interval of 82 days. The most frequent reintervention was photodynamic therapy. In the other half, the stent provided palliation until death for a mean interval of 125 days.
In 16 patients, a stent was placed as temporary palliation for malignant dysphagia while undergoing neoadjuvant therapy for possible later esophagectomy. The stent was successful in relieving dysphagia and avoiding enteral or intravenous nutrition up to the time of esophagectomy decision in 14 patients (88%). In 2 patients persistent dysphagia was present, requiring a feeding tube for nutritional supplementation. In 6 of the 16, chemoradiation was delivered after stent placement and 2 patients experienced major complications at the site of the stent deployment. One patient had the stent placed in the proximal esophagus and developed stent erosion into the T1 vertebral body, which was discovered as an occult contained perforation at the time of resection. This required subsequent neurosurgical drainage, but the patient ultimately recovered. The second complication in the chemoradiation group included a mediastinal abscess that was again related to an occult stent perforation and erosion noted at the time of thoracotomy and resection. A recurrent mediastinal abscess occurred after esophagectomy, and the patient ultimately succumbed to sepsis. No local complications were noted in those receiving chemotherapy alone.
Complications in the entire group are listed in Table 5. One patient died as a delayed complication of stent placement because of the mediastinal abscess described above. Four stents were not adequately positioned on initial deployment: three were removed and immediately replaced, and a second stent was placed in a fourth patient. The most common complication seen was tumor ingrowth or overgrowth (33% of patients). All patients with stents across the gastroesophageal junction were counseled to keep the head of the bed elevated and were treated with proton pump inhibitors. Significant symptomatic reflux persisted in 11% of patients in spite of these interventions. Stent migration requiring removal or replacement was seen in 11 patients (8.7%). In our experience, a greater percent of covered stents migrated (7 of 30, 23%) compared with only 4 of 97 (4.1%) uncovered stents. Location of the stent also influenced migration rates: 0 of 22 proximal stents, 2 of 32 (6.3%) mid esophageal stents, and 9 of 73 (12.3%) distal esophageal stents underwent migration. There were no stent removals because of airway compression. There were 3 patients with proximal esophageal tumors that were evaluated and not stented because of the concern over possible airway compression, and these patients were referred for photodynamic therapy. One patient experienced a contained perforation seen on poststent barium swallow, which did not lead to any clinical problems and oral diet was advanced without sequelae. One patient exhibited a tracheoesophageal fistula requiring exclusion and surgical bypass after definitive chemoradiation.
|
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
More than half of the patients stented had already received previous treatment for their esophageal carcinoma. In 15% of patients, a stent was placed as salvage therapy in those who had received photodynamic therapy and subsequently developed recurrent dysphagia with good results. Sixteen patients with potentially resectable disease had a stent placed to improve oral intake and nutritional status while receiving induction chemotherapy before planned esophagectomy. Palliation of dysphagia was observed in 14 of 16 patients for a mean interval 76 days before esophagectomy. It was noted by the surgeon that the dissection of the periesophageal planes was more difficult, but no significant complications were observed.
A total of 15 patients had a stent placed in association with chemoradiation. Localized esophageal perforations were noted in 3 of these patients at the time of esophagectomy. Free perforation was not noted preoperatively, but during dissection of the esophagus the stent was noted to have eroded through the wall of the esophagus with contained localized infection present. One patient had an associated erosion into the posterior membranous trachea and was successfully treated with an esophageal bypass. One patient with a proximal esophageal cancer experienced a T1 vertebral body abscess that was successfully treated with esophagectomy and drainage of the abscess. A third patient with a mid esophageal tumor was noted to have a contained perforation at the time of esophagectomy and had a delayed mediastinal abscess and ultimately died of sepsis. All 3 of these patients had received chemoradiation, 2 before stenting and 1 after stenting. In the other 12 patients who had received chemoradiation and subsequent stenting, no complications were observed. In addition to our findings, others have reported complications in the setting of combined chemoradiation and expandable metal stents [14]. Although the actual incidence of complications in a large controlled series is not known, other options for palliation in this setting should be considered. Perforation as a direct complication of stent placement was seen in only 1 patient. This was a contained leak, which responded to subsequent placement of a covered stent. This incidence of perforation compares favorably with other studies [8, 14–19].
The majority of stents used in our series of patients were uncovered. In our initial experience we observed a higher rate of stent migration using covered stents, especially at the gastroesophageal junction. This has been reported by other groups [3, 4]. It was our clinical suspicion that this was because of failure of significant fibrosis or ingrowth through the covered stent and lack of fixation that led to the migration. The incidence of stent migration overall was 8% and compares favorably with series in which covered stents were used [4, 13]. The disadvantage of uncovered stents is tumor ingrowth and recurrent obstruction. We observed a combination of ingrowth or overgrowth above or below the stent in 33% of patients in our series. This was the most common cause of recurrent dysphagia requiring reintervention. Because of the retrospective nature of this series, it was not possible to determine how many patients had actual isolated ingrowth through the stent as a cause of recurrent dysphagia or whether it was a combination of ingrowth and tumor progression above and below the stent. Our group and others have used photodynamic therapy with good results to treat these patients [10, 11]. Use of covered stents may help decrease tumor ingrowth, although it will not affect tumor progression above or below the stent. Covered stents were preferred in the setting of tracheoesophageal fistula or if there was concern that the patient was at risk for this because of the location of the tumor. In this study, 3 patients had a tracheoesophageal fistula and all were palliated with a covered stent.
Most patients in this series had obstruction at the level of the gastroesophageal junction. Placement of an expandable metal stent in this location renders the lower esophageal sphincter incompetent and virtually all patients will have a degree of gastroesophageal reflux and regurgitation [3, 15]. Most can be reasonably well managed, but in spite of lifestyle changes such as small meals, avoidance of late night meals, elevating the head of the bed, and proton pump inhibitors, the incidence of severe recalcitrant reflux was 11% in our series. The development of new generation stents with an antireflux valve, such as the Dua stent (Wilson-Cook, Winston-Salem, NC) may reduce this complication. Proximal stents were placed in 17% of patients and mid esophageal stents in 25%. Placement of an expandable metal stent in proximity to the airway can be associated with a risk of secondary airway compromise [1, 2]. This can be dramatic and life-threatening in some cases. We performed bronchoscopy in all patients with a tumor in proximity to the airway before stent placement. If there was concern over airway compression, a bougie of equal size to the planned stent was passed with the bronchoscope in place to evaluate for airway compromise. We excluded several patients from stent placement on the basis of these findings. Using this approach we had no episodes of acute airway compromise after stent placement. The majority of patients in this study were admitted to the hospital after stent placement. This facilitates intravenous hydration, intensive dietary counseling, close evaluation of oral intake, and management of gastroesophageal reflux. With experience and expansion of our outpatient support, a higher rate of outpatient procedures has been achieved, making esophageal stenting more cost-effective.
Of the 94 stent placements that were immediately successful in patients who were not planning on undergoing esophagectomy, more than half (51 patients) required reintervention at a mean interval of 82 days. In the 40 patients who survived more than 120 days, 78% required reintervention, which emphasizes the need for close clinical follow-up of this group of patients. Secondary interventions such as photodynamic therapy for tumor stent ingrowth or endoscopic treatment of food impaction can result in a significant extension of palliation of dysphagia and maintenance of oral intake. Using this approach of aggressive secondary intervention, only 7 patients ultimately required a feeding tube; all other patients were able to maintain oral intake until time of death, planned esophagectomy, or time of last follow-up.
In summary, several factors should be considered before applying palliative therapy for malignant esophageal obstruction. Factors such as medical comorbidity and overall expected duration of survival are important. Tumor characteristics such as size, shape, and length also need to be considered. Local expertise and the availability of different modalities may also be important. The limitations of expandable metal stents in our series have been summarized in this report, but major advantages were also noted, which include ease of insertion, the avoidance of general anesthesia, availability as an outpatient procedure, and immediate palliation of dysphagia in 85% of patients. For patients with a limited lifespan, this may fulfill the major objectives of palliation, which include satisfactory relief of dysphagia with minimal morbidity in a single procedure. Intensive dietary counseling and close follow-up should be exercised because recurrent dysphagia is common in long-term survivors and reintervention is frequently successful. Caution should be exercised when using expandable metal stents in the setting of chemoradiation, and other palliative options should be considered.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
I have two questions. Fortunately, we have not encountered your problem with tumor ingrowth. I think that is because we uniformly use coated stents. Would you agree that would obviate your problem?
DR CHRISTIE: A lot of these patients have tumors at the gastroesophageal junction, and the incidence of stent migration with coated stents at the gastroesophageal junction is high. It was that experience that led to the majority of our stents, at least for the gastroesophageal junction, being uncovered stents. In our experience, treatment with photodynamic therapy for tumor ingrowth has been effective. But I acknowledge that uncovered stent use may lead to high tumor ingrowth rate. Covered stents will not help with tumor overgrowth, and a significant number of these patients had tumor overgrowth over the ends of the stent.
DR WETSTEIN: My second question is a technical question. You stated that in approximately 8 patients you stented patients subsequent to resection. That appears to be very difficult because you are sort of stenting a residual stomach, a stomach tube, or colon, versus an esophagus. Technically, how do you do that?
DR CHRISTIE: Well, those generally were patients who had a very bulky tumor recurrence, or in the situation where there is mediastinal mass and extrinsic compression, and we did not have any problems in terms of stent positioning.
DR CAROLYN REED (Charleston, SC): I enjoyed your talk, and some of your findings mirror our findings at the Medical University. Despite the enthusiasm for expandable metal stents, you have shown clearly, as we have found, that there is actually a very high complication rate and a high reintervention rate, and these patients continue to be difficult to treat.
I have several questions for you. I echo what Dr Wetstein says. We would certainly favor using coated stents for the very reason that you alluded to in that you have a high ingrowth rate.
I also have several other questions. One, I would like a little more information about your use either before or after photodynamic therapy. I hasten to say that when you use an expensive expandable metal stent and then you are having to use photodynamic therapy also, you have upped the cost considerably, and I wonder why you did not use something like YAG laser, which might have been less expensive.
Second, I would like to know what dysphagia score you used. Did you study any other quality-of-life variables?
I enjoyed your paper. Thank you very much.
DR CHRISTIE: Well, with respect to the dysphagia score, we used a scale of 1 to 5, 1 being normal swallowing and 5 being the inability to swallow your own secretions, and this is a modification of a standard published score. With respect to other end points, we have not looked at nutritional measures. We have looked at the dysphagia score and survival. We are in the process of prospectively collecting information in terms of quality-of-life variables.
Now, in terms of photodynamic therapy and stenting, again, I acknowledge what you and the other discussants have said, that using covered stents may result in a lower rate of tumor ingrowth. Again I will say that what drove us to do this was that there was quite a significant rate of stent migration in patients who had bulky tumors at the gastroesophageal junction, and I think that to some extent you have to individualize your treatment in these patients. If you have a patient who, based on your assessment, you think has a very poor survival and you do not anticipate that they are going to live very long, then the issues of tumor ingrowth and overgrowth become less important. If you believe that you have a patient who does have a potentially long length of survival, I think other issues are important. Because of the long-term complications of stents that we have seen (such as esophageal erosion and mediastinal sepsis) and our good experience with photodynamic therapy, we might be tempted to use that treatment as a first line. Our bias in this study is that most of these stented patients had bulky tumors and some component of extrinsic compression, because most patients who have primarily endoluminal tumors in our practice we would treat with laser therapy, such as photodynamic therapy.
|
References |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
Related Article
This article has been cited by other articles:
|
|
 |
|
  A. Pennathur, A. C. Chang, K. M. McGrath, G. Steiner, M. Alvelo-Rivera, O. Awais, W. E. Gooding, N. A. Christie, S. Gilbert, R. J. Landreneau, et al. Polyflex expandable stents in the treatment of esophageal disease: initial experience. Ann. Thorac. Surg., June 1, 2008; 85(6): 1968 - 1972. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J.-H. Guo, G.-J. Teng, G.-Y. Zhu, S.-C. He, W. Fang, G. Deng, and G.-Z. Li Self-expandable Esophageal Stent Loaded with 125I Seeds: Initial Experience in Patients with Advanced Esophageal Cancer Radiology, May 1, 2008; 247(2): 574 - 581. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. D. Luketich and A. Pennathur How to Keep the Treatment of Esophageal Disease in the Surgeon's Hands Ann. Thorac. Surg., February 1, 2008; 85(2): S760 - S763. [Full Text] [PDF]
|
|
|
|
 |
|
 W Harms, R Krempien, C Grehn, C Berns, F W Hensley, and J Debus Daytime pulsed dose rate brachytherapy as a new treatment option for previously irradiated patients with recurrent oesophageal cancer Br. J. Radiol., March 1, 2005; 78(927): 236 - 241. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. C. Fields, C. Choong, S. Edmundowicz, and B. F. Meyers Combined Antegrade and Retrograde Endoscopy with Radiofrequency Ablation and Dilatation for Complex Esophageal Obstruction Chest Meeting Abstracts, October 1, 2004; 126(4): 951S - 951S. [Abstract] [PDF]
|
|
|
|
|
|
M J Metcalfe, A C Steger, and A Leslie Benign complications of expandable metal stents used in the palliation of oesophageal carcinoma: two case reports Br. J. Radiol., March 1, 2004; 77(915): 245 - 247. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. R. Litle, J. D. Luketich, N. A. Christie, P. O. Buenaventura, M. Alvelo-Rivera, J. S. McCaughan, N. T. Nguyen, and H. C. Fernando Photodynamic therapy as palliation for esophageal cancer: experience in 215 patients Ann. Thorac. Surg., November 1, 2003; 76(5): 1687 - 1693. [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 |
