ATS
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to Personal Folders
Right arrow Download to citation manager
Right arrow Author home page(s):
Gregory D. Trachiotis
Right arrow Permission Requests
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Alexander, E. P.
Right arrow Articles by Wadleigh, R. G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Alexander, E. P.
Right arrow Articles by Wadleigh, R. G.
Related Collections
Right arrow Esophagus - cancer

Ann Thorac Surg 2001;71:1640-1644
© 2001 The Society of Thoracic Surgeons

Original article: general thoracic

Evolving management and outcome of esophageal cancer with airway involvement

E. Pendleton Alexander, MDa, Gregory D. Trachiotis, MDa, Timothy O. Lipman, MDa, Robert G. Wadleigh, MDa

a Divisions of Cardiothoracic Surgery, Gastroenterology, and Medical Oncology, Washington DC Veterans Administration Medical Center, and George Washington University Medical Center, Washington, DC, USA

Address reprint requests to Dr Alexander, Department of Cardiothoracic Surgery, George Washington University, 50 Irving St, NW, Washington, DC 20422
e-mail: epalexander{at}med.va.gov

Presented at the Forty-sixth Annual Meeting of the Southern Thoracic Surgical Association, San Juan, Puerto Rico, Nov 4–6, 1999.


   Abstract
Top
 Abstract
Introduction
 Patients and methods
 Results
 Comment
 References
 
Background. Esophageal cancer with airway involvement, including patients with esophagorespiratory fistula (ERF), has been associated with a poor prognosis. Multimodality treatment, self-expanding metal stents, and improved supportive therapy may be impacting outcome in these patients. There is concern for the development of ERF during therapy.

Methods. We retrospectively studied 74 consecutive male patients at a single institution presenting between 1/85 to 12/98 with bronchoscopic, endoscopic or radiographic confirmation of airway involvement with esophageal cancer, including 35 patients with ERF. Comparison was made between the first 35 patients (group I) and the last 39 patients (group II) with regard to antineoplastic therapy, stent placement, and survival.

Results. Treatment in group I included supportive care in 17 of 35 patients, plastic stent in 7 of 35 patients, and radiation or chemotherapy in 9 of 35 patients. In group II, radiation or chemotherapy was offered to 33 patients, and self-expanding metal stents were placed in 10 of 39 patients. Surgical resection was possible after neoadjuvant therapy in 13 of 39 patients in group II, including 2 initially presenting with ERF. Median survival in group I was 16 weeks and in group II was 37 weeks. Comparison of Kaplan-Meier survival estimates using log rank testing demonstrated improved survival in group II (p = 0.0026). Long-term survival in 4 group II patients initially presenting with ERF and receiving multimodality treatment was observed. Development of ERF during treatment occurred in 3 group II patients. Treatment failure was predominantly local in group I and local and distant in group II.

Conclusions. More aggressive treatment may favorably influence outcome in esophageal cancer with airway invasion. Long-term survival and the development of ERF during therapy occurred at similar rates.


   Introduction
Top
 Abstract
 Introduction
Patients and methods
 Results
 Comment
 References
 
Esophageal cancer with airway involvement, particularly with progression to an esophagorespiratory fistula (ERF), has been associated with a poor prognosis. Traditionally, management has been palliative or supportive, with little expectation for successful surgical therapy or long-term survival. Therapeutic modalities used in this setting have included comfort measures, esophageal bypass, diversion, radiation therapy, chemotherapy, intubation, and attempted resection [1–4]. Therapy of ERF has been primarily directed at attenuating respiratory infection with an increasing dependence on endoesophageal and endobronchial prostheses [5–7]. Choice of prosthetic stent has changed in most centers as self-expanding metallic stents have been used in place of conventional plastic prostheses [5, 6].

Although increasingly accepted in less advanced cases of esophageal carcinoma, multimodality therapy has not been widely used in the setting of airway invasion. Several reports of long-term survival exist [8–12], although pessimism over the potential impact of therapy is generally coupled with anxiety over the development or extension of ERF with chemoradiotherapy, and the potential for irreparable airway injury with attempted resection [2–4, 11–19].

We have employed an increasingly aggressive approach to esophageal cancer with airway invasion, in addition to using newer stents and other supportive therapies. This study is a retrospective analysis of esophageal cancer with airway invasion from 1985 to 1998. To delineate changes in outcome and management, comparison is made between patients presenting before and after 1992.


   Patients and methods
Top
 Abstract
 Introduction
 Patients and methods
Results
 Comment
 References
 
Patient population
From 1985 to 1998, 372 patients with biopsy-proven carcinoma of the esophagus were seen at a single center, the Washington, DC VAMC. During this period late presentation of advanced locoregional disease was common, with a plethora of comorbidities and a predominance of squamous histology (327 patients, 88%).

A retrospective review of all patients with radiographic, endoscopic, bronchoscopic or postmortem evidence of airway involvement was performed. Sources included medical records, imaging studies, tumor registry, and pathologic data. Recorded data included details of presentation, treatment, clinical course, and eventual survival.

Airway involvement was defined as invasion, infiltration, or the presence of an ERF. Juxtaposition, abutment, or the presence of a bulge or deviation alone were not considered adequate evidence of airway involvement. Using these criteria, of 372 patients seen over the 14-year study period, 74 (20%) were identified with airway invasion. Thirty-five (47%) of these patients had an ERF for a total ERF incidence of 35 of 372 patients (9.4%). In the absence of ERF, airway involvement was proven with biopsy in 29 of 39 patients (74%). In the remaining 10 patients incontrovertible endoscopic, bronchoscopic radiographic, or postmortem evidence was required to document airway involvement. In 6 of these 10 patients bronchoscopic biopsy of obvious treacheal "cobble stoning" was not done to avoid the potential for creating an ERF. To delineate potential changes in presentation, management, and outcome, the study group was divided into two chronological halves: group I, 1985 to 1991 and group II, 1992 to 1998. The increasingly aggressive approach to advanced esophageal cancer adopted in 1992 was concurrent with the initiation of a previously reported study of multimodality therapy for stage III esophageal carcinoma, including the stage groupings of T4N0M0 and T4N1M0 [10].

Statistical methods
{chi}2 and Fisher’s exact test were used to test whether relationships existed between nominal variables. Survival was measured from the date of diagnosis to the date of death or October 1, 1999. Actuarial survival was calculated using the Kaplan-Meier method and comparisons made using the log rank test. Significant predictors from a univariable analysis of independent variables were subjected to multivariable analysis.


   Results
Top
 Abstract
 Introduction
 Patients and methods
 Results
Comment
 References
 
Patient characteristics
Table 1 summarizes patient characteristics for the group as a whole, group I and group II. The incidence of airway involvement and ERF are not significantly different from group to group. Age and Eastern Cooperative Oncology Group performance status were similar for both groups. Histology was predominately squamous in both groups.


View this table:
[in this window]
[in a new window]
  		Table 1. Patient Characteristics

 
The respiratory location of airway involvement was constant throughout the study period, with predominately intrathoracic tumors. More than 50% of the involvement of intrathoracic airways involved the bronchi with slightly more involvement of the left bronchus than the right bronchus. Although different staging modalities, including thoracoscopy, laparoscopy, and endoscopic ultrasound, were used frequently in group II patients, there was only a small increase in the identification of nodal metastases. Remarkably, disease was limited to the chest in approximately 65% of patients, including those presenting with an ERF.

Management
Table 2 summarizes management strategies used in the group as a whole, group I, and group II. Supportive care measures, including intravenous hydration, feeding tubes, antibiotics, analgesics, and comfort measures, were employed much more commonly in group I.


View this table:
[in this window]
[in a new window]
  		Table 2. Management

 
Stent placement for palliation of malignant dysphagia or occlusion of an ERF evolved significantly over the study period (Table 2). Plastic stents, typically the flanged Celestin, were used predominately in group I. Self-expanding metal stents were used in the latter part of the group II study period. Patient selection for stent placement evolved as more aggressive therapy was selected over the course of group II.

Gastrostomy technique changed significantly over the study period (Table 2). Percutaneous endoscopic gastrostomy replaced open surgical gastrostomy in 1992, reflecting the initiation of a practice to place a percutaneous endoscopic gastrostomy in the majority of patients with esophageal cancer.

During the group II study period, laser treatment of esophageal cancer was used in multiple stages of disease for debulking and palliation of malignant dysphagia including patients with airway invasion. Four of 39 (10%) group II patients received laser treatment; however, more liberal use in "T4 airway" patients was curtailed after the development of an ERF in 1 patient shortly after a laser treatment.

Surgical bypass and diversion have been used extensively in patients with airway involvement [2, 3], but were performed in only 2 group I patients.

Antineoplastic therapy in general was used much more liberally in group II (Table 2). Chemotherapy use quadrupled and radiotherapy use tripled over the course of the study, with highly statistically significant differences between group I and group II. Radiation doses varied from 25 to 60 Gy and a wide range of chemotherapeutic agents were used. A number of group II patients received continuous infusion of 5-fluorouracil with radiation as described in a previously reported protocol [10]. Surgical exploration was undertaken in 20 of 74 patients (27%) of the entire study group. Resections were performed in 15 of 74 (20%), with the majority of these in group II.

Resection was attempted predominately in patients who were thought to be complete responders to induction chemoradiotherapy. In these patients, resection was facilitated by the ability to develop a surgical plane in fibrous tissue between the airway and the esophagus. Surgical complications included death [1], airway injury [2], chylothorax [1], and anastomotic leak [1].

Impact of therapy on ERF
Sixteen of 35 patients (45.7%) with an ERF received chemotherapy or radiotherapy or a combination of both. Three patients had closure of their fistula during therapy. Two of 3 of these patients subsequently had successful resections.

Three patients without ERF at presentation developed ERF during chemoradiotherapy. As noted, development of a fistula was temporally related to laser therapy in 1 patient. None of the fistulas developing during treatment subsequently reclosed.

Survival
Kaplan-Meier actuarial survival for the entire group, group I, and group II are displayed in Figure 1. Median survival for the entire group was 22 weeks, 16 weeks for group I, and 37 weeks for group II. The difference in survival from group I to group II was highly significant (p = 0.0026).



View larger version (20K):
[in this window]
[in a new window]
  		Fig 1. Actuarial survival for all patients in the entire study group: group I (1985 to 1991) and group II (1992 to 1998). The difference between group I and group II was significant (p = 0.0026).

 
Actuarial survival for patients with an EGF is depicted in Figure 2. Median survival for the group as a whole was 16 weeks; 11 weeks in group I and 46 weeks in group II. The difference in survival from group I to group II approached statistical significance (p = 0.058).



View larger version (16K):
[in this window]
[in a new window]
  		Fig 2. Actuarial survival for patients with an esophagorespiratory fistula in the entire study group: group I (1985 to 1991) and group II (1992 to 1998). The difference between group I and group II approached statistical significance (p = 0.058).

 
Univariable analysis demonstrated Eastern Cooperative Oncology Group performance status, presence of an ERF, radiation therapy, resection, and N0M0 stage grouping as predictive of survival at 52 weeks. However, multivariable analysis of these predictors demonstrated only Eastern Cooperative Oncology Group status (p = 0.001), N0M0 stage grouping (p = 0.05), and presence of an ERF (p = 0.005) as significantly related.

The impact of N0M0 stage grouping on actuarial survival is depicted in Figure 3. There is a highly significant difference among patients in group II and the group as a whole when N0M0 patients are compared to all other groups.



View larger version (19K):
[in this window]
[in a new window]
  		Fig 3. The impact of N0M0 stage grouping on survival for both the total group and group II. In each group the difference was highly significant (p = 0.0001).

 
Survival after stent placement is not significantly different with metallic and plastic stents, although long-term survival with stent placement in this series was observed only in patients with metallic stents (Fig 4).



View larger version (14K):
[in this window]
[in a new window]
  		Fig 4. Actuarial survival of patients receiving plastic and metallic stents. There was no significant difference though long-term survival was observed with a metallic stent.

 
Within group II, patients progressing to resection fared much better than the rest of the group (Fig 5). Median survival for resected patients in this group was 106 weeks, whereas median survival in all other patients was 17 weeks. Within group II the actuarial survival advantage for resection was significant (p = 0.01).



View larger version (16K):
[in this window]
[in a new window]
  		Fig 5. Actuarial survival of resection, nonresection, and all patients in group II. Survival in resected patients is significantly superior to nonresected (p = 0.01).

 
Pattern of failure
Pattern of failure, defined as the predominant location of disease causing death, evolved over the study period. In group I, local disease caused death in 25 of 35 patients (71.5%), but decreased to 17 of 37 patients (46%) in group II. Distant failure rates essentially doubled with the more common use of aggressive therapy (group I, 10 of 35 patients, 28%; group II, 20 of 57 patients, 54%). The change from predominately local failure to mixed local and distant was significant (p = 0.038).


   Comment
Top
 Abstract
 Introduction
 Patients and methods
 Results
 Comment
References
 
Airway invasion in this series was identified in 20% of patients with biopsy-proven esophageal carcinoma. Esophagorespiratory fistula was noted in approximately half of these patients, occurring in 9.7% of esophageal cancer patients during the 14-year study period. Wide variations exist in the reported incidence of ERF, and the incidence is clearly dependent on the patient population being studied. Reported rates range from 0.9% to 18% [3]. The overall incidence of airway involvement is less well defined and is dependent on not only the patient population being studied, but also the diagnostic studies and criteria used. Reported rates of airway involvement range from 10.9% to 62% [1, 3]. In the time period of this study, there was no significant change in the incidence of airway invasion or of ERF.

The respiratory location of airway invasion in this series was consistent with previous reports [1–3]. In this series approximately 50% of esophageal cancers involving the thoracic airways involved the bronchi, with a slight predilection for the left bronchus. An ERF communicating peripherally into the lung parenchyma has been previously described [3], but was not noted in this series.

Stage grouping was relatively constant throughout the study period in spite of the application of newer staging modalities such as endoscopic ultrasound, thoracoscopy, and laparoscopy. There was a nonsignificant increase in the assignment of the N1 designator between group I and group II. Remarkably, disease was confined to the thorax in approximately 65% of patients in this series. This finding is concurrent with other reports [8, 11], and is consistent with the observation that left untreated, most patients will die from the local effects of the tumor rather than disseminated disease.

The appropriateness of cytotoxic treatment for airway invasion is controversial. Burt [2] and Fry [8] and their colleagues reported the resolution of an ERF after radiotherapy. Progression to ERF has also been reported with radiotherapy. Lewinsky and associates [18] noted the development of ERF in 9 of 85 patients. Eastridge and coworkers [19] also reported development of an ERF with radiotherapy. Three patients presenting with ERF in this series had documented closure with chemoradiotherapy and 2 patients were successfully resected. Three patients without ERF at presentation developed ERF during chemoradiotherapy, and none of these subsequently reclosed. It is likely that both closure and development or extension are possible with cytotoxic therapy. During radiotherapy, edema may transiently close or diminish the size of a fistula, with a consequent decrease in respiratory suppuration. Tumor necrosis and fibrous healing may be competing processes in the development, closure, or perpetuation of an ERF during cytotoxic therapy.

Stenting has been reported to be a helpful adjunct in patients with advanced locoregional esophageal cancer, for both the attenuation of respiratory soilage and the palliation of malignant dysphagia [4–6]. No survival advantage of metallic over plastic stents was demonstrated in this report, although long-term survival occurred in 2 patients with a metallic stent. A head-to-head comparison of stent types in this study is not possible because different patient selection criteria were used with poorer performance status patients receiving stents in the latter group.

Resection is feasible and associated with long-term survival in patients having a dramatic response to chemoradiotherapy. We do not believe primary resection in the absence of previous induction therapy should be undertaken, unless the airway involvement is very distal and esophagectomy can be accompanied by a pulmonary resection. Success with induction therapy followed by operation in this group of patients has been previously reported [8].

The prognosis of esophageal cancer with airway invasion remains poor, but long-term survival is possible, including patients with ERF. Aggressive treatment of local disease, applied liberally in group II, appears to favorably influence survival. Survival is highly dependent on Eastern Cooperative Oncology Group performance status, N0M0 stage grouping, and the presence of an ERF. Aggressive treatment significantly changed the pattern of failure from predominately local to mixed local and distant, as successful therapy of some primary tumors was achieved.


   References
Top
 Abstract
 Introduction
 Patients and methods
 Results
 Comment
 References
 

  1. Altorki N.K., Migliore M., Skinner D.B. Esophageal carcinoma with airway invasion. Chest 1994;106:742-745.[Abstract/Free Full Text]
  2. Burt M., Diehl W., Martini N., Bains M., et al. Malignant esophagorespiratory fistula: management options and survival. Ann Thorac Surg 1991;52:1222-1229.[Abstract]
  3. Duranceau A., Jamieson G.G. Malignant tracheoesophageal fistula. Ann Thorac Surg 1984;37:346-354.[Abstract]
  4. Lolley D.M., Ray J.F., Ransdell H.T., Razzuk M.A., Urschel H.C. Management of malignant esophagorespiratory fistula. Ann Thorac Surg 1978;25:514-520.
  5. Raijman I., Lynch P. Coated expandable esophageal stents in the treatment of digestive-respiratory fistulas. Am J Gastroenterol 1997;92:2188-2191.[Medline]
  6. Siersema P.D., Hop W.C.J., Dees J., Tilanus H.W., Blankenstein M.V. Coated self-expanding metal stents versus latex prostheses for esophagogastric cancer with special reference to prior radiation and chemotherapy: a controlled, prospective study. Gastrointest Endosc 1998;47:113-120.[Medline]
  7. Reed C.E. Endoscopic palliation of esophageal carcinoma. Chest Surg Clin N Amer 1994;4:155-172.
  8. Fry W.A., Griem M.L., Adams W.E. Malignant tracheo-esophageal fistula treated by combined radiotherapy and surgical excision. Dis Chest 1968;54:384-387.
  9. Sileni V.C., Tomio L., Koussis H., et al. Concomitant radiotherapy and chemotherapy followed by surgery in esophageal carcinoma with bronchotracheal involvement. Proc Am Soc Clin Oncol 1996;15:199.
  10. Alexander E.P., Lipman T., Harmon J., Wadleigh R. Aggressive multimodality therapy for stage III esophageal cancer: A phase I/II study. Ann Thorac Surg 2000;69:363-368.[Abstract/Free Full Text]
  11. Martini N., Goodner J.T., D’Angio G.J., Beattie E.J., Jr Tracheoesophageal fistula due to cancer. J Thorac Cardiovasc Surg 1970;59:319-324.[Medline]
  12. Fitzgerald R.H., Jr, Bartles D.M., Parker E.F. Tracheoesophageal fistulas secondary to carcinoma of the esophagus. J Thorac Cardiovasc Surg 1981;82:194-197.[Medline]
  13. Gschossmann J.M., Bonner J.A., Foote R.L., Shaw E.G., Martenson J.A., Su J. Malignant tracheoesophageal fistula in patients with esophageal cancer. Cancer 1993;72:1513-1521.[Medline]
  14. Judd D.R., Codd J. Palliation for malignant esophagotracheal fistula. J Thorac Cardiovasc Surg 1967;54:751-755.[Medline]
  15. Belsey R.H.R. Palliative management of esophageal carcinoma. Am J Surg 1980;139:789-794.[Medline]
  16. Lambert A. Surgical management of malignant tracheoesophageal fistula. NY State J Med 1975;75:2369-2372.
  17. Little A.G., Ferguson M.K., DeMeester T.R., Hoffman P.C., Skinner D.B. Esophageal carcinoma with respiratory tract fistula. Cancer 1984;53:1322-1328.[Medline]
  18. Lewinsky B.S., Annes G.P., Mann S.G., et al. Carcinoma of the esophagus. Radiol Clin Biol 1975;44:192-196.
  19. Eastridge C.E., Greenberg B.E., Hughes F.A., Aslam P.A. Cancer of the esophagus. South Med J 1970;63:1135-1141.[Medline]



This article has been cited by other articles:


Home page
 Ann. Thorac. Surg.Home page
J. Wlodarczyk, H. Olechnowicz, and P. Kocon
Esophago-Pericardial Fistula During the Course of Primary Esophageal Carcinoma
Ann. Thorac. Surg., December 1, 2008; 86(6): 1967 - 1969.
[Abstract] [Full Text] [PDF]

Home page
 Ann. Thorac. Surg.Home page
M. Margolis, P. Alexander, G. D. Trachiotis, F. Gharagozloo, and T. Lipman
Percutaneous endoscopic gastrostomy before multimodality therapy in patients with esophageal cancer
Ann. Thorac. Surg., November 1, 2003; 76(5): 1694 - 1698.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to Personal Folders
Right arrow Download to citation manager
Right arrow Author home page(s):
Gregory D. Trachiotis
Right arrow Permission Requests
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Alexander, E. P.
Right arrow Articles by Wadleigh, R. G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Alexander, E. P.
Right arrow Articles by Wadleigh, R. G.
Related Collections
Right arrow Esophagus - cancer

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