Ann Thorac Surg 2005;79:1174-1179
© 2005 The Society of Thoracic Surgeons
Original articles: General thoracic
The Surgical Management of Superior Sulcus Tumors: A Retrospective Review With Long-Term Follow-Up
Melvyn Goldberg, MDa,*,
Dipin Gupta, MDa,
Aaron R. Sasson, MDa,
Benjamin Movsas, MDb,
Corey J. Langer, MDc,
Alexandra L. Hanlon, PhD, MDb,d,
Hao Wang, MSd,
Walter J. Scott, MDa
a Department of Surgical Oncology, Fox Chase Cancer Center, Philadelphia, Pennsylvania
b Department of Radiation Oncology, Fox Chase Cancer Center, Philadelphia, Pennsylvania
c Department of Medical Oncology, Fox Chase Cancer Center, Philadelphia, Pennsylvania
d Department of Biostatistics, Fox Chase Cancer Center, Philadelphia, Pennsylvania
Accepted for publication September 7, 2004.
* Address reprint requests to Dr Goldberg, Department of Surgical Oncology, Fox Chase Cancer Center, 333 Cottman Ave, Philadelphia, PA 19111-2497 (E-mail: m_goldberg{at}fccc.edu).
 |
Abstract
|
|---|
BACKGROUND: We reviewed our experience with multimodality therapy for superior sulcus tumors to identify aspects of treatment that impact survival.
METHODS: We retrospectively analyzed the records of 39 consecutive patients who underwent surgical resection in a single institution between 1993 and 2000.
RESULTS: Median age at presentation was 59 years (range, 40 to 77). Twenty-five patients (64%) were men. At presentation, 36 patients (92%) had clinical T3 tumors and 3 (8%) had clinical T4 tumors. Mediastinoscopy was negative in all patients. Thirty-one patients (79%) received preoperative radiotherapy (median dose, 4500 cGy). Chemotherapy was administered concurrently with radiotherapy in 27 patients (69%). Complete surgical resection was performed in 34 patients (87%). There were 2 (5%) postoperative deaths. Of the 31 patients who received preoperative therapy, 14 (45%) had their tumors downstaged and 9 (29%) demonstrated a complete pathologic response. Median follow-up (100%) was 69 months. Overall 5-year survival was 47.9%. Five-year survival was 52.1% in patients with negative resection margins (p = 0.005), and 60.6% in patients who demonstrated a response to induction chemoradiation therapy (p = 0.008). Independently, margin status and response to induction therapy are predictors of overall survival (p = 0.01 and p = 0.02, respectively). Multivariable analysis identified margin status as the only factor significantly associated with overall survival. Negative margins strongly correlated with the response to preoperative therapy (p = 0.004). Disease-free survival correlated well with the response to induction therapy (p = 0.03). The chemotherapy regimen, T status, operative procedure, and complete pathologic response did not correlate with survival.
CONCLUSIONS: The use of chemoradiation induction therapy may downstage tumors, enhance the ability to obtain a complete surgical resection, and prolong survival.
|
Introduction
|
|---|
Tumors of the superior sulcus represent less than 5% of lung malignancies. The distinctive symptomatology was first described by Hare in 1838 [1], and Pancoast identified these tumors as a clinical entity in 1932 [2]. In 1956 Chardack and MacCallum reported a long-term survival after surgical resection and postoperative irradiation therapy [3]. Subsequently, Shaw and Paulson identified that preoperative radiation and a well-defined resection were associated with a 5-year survival of 34% [4, 5, 6]. Based upon these studies, preoperative radiation has been part of the preoperative standard of care over the last 5 decades.
The more recent addition of concurrent chemotherapy has been based upon the synergistic effect of combined chemoradiation therapy validated by Dillman and associates [7]. Interest in trimodality therapy led to the Southwest Oncology Group (SWOG) 8805 study of induction chemoradiotherapy (cisplatin, etoposide, 45 Gy) followed by surgery for local advanced stage IIIA and IIIB non-small cell lung cancer that resulted in a complete response rate of 22% and encouraging survival rates [8].
The largest series of 225 superior sulcus tumors undergoing resection indicated that actuarial survival at 5 years was 46% for stage IIB, 0% for stage IIIA, and 13% for stage IIIB (T4) [9]. Recently, Rusch and colleagues reported the experience of the SWOG 9416 phase II trial of 111 patients with Pancoast tumors undergoing preoperative concurrent chemoradiotherapy (cisplatin, etoposide, 45 Gy). The 2-year survival was 55% for all eligible patients and 70% for patients who had a complete resection [10].
These previous reviews led us to assess our single institution results in patients undergoing surgical extirpation of superior sulcus tumors with and without induction therapies. Our goal was to relate recurrence and survival to the variety of induction regimens and surgical procedures used in our patient care that might not have been addressed adequately in previous publications.
|
Material and Methods
|
|---|
This study retrospectively reviewed the records of 39 patients who underwent surgical resection for superior sulcus tumors at Fox Chase Cancer Center from April 1993 to December 2000. We have defined a superior sulcus tumor as an apical pulmonary malignancy with either signs or symptoms of pain in the arm or shoulder, or both, upper extremity weakness, or Horner syndrome; or chest wall, skeletal axis, or vascular involvement substantiated by radiographic assessment. Tumors with neurologic and rib involvement entirely caudad to T2 were excluded from study.
Before induction therapy, staging was performed by bone scan, brain scan, and a high-resolution computed tomographic (CT) scan of the chest. A CT scan was repeated after induction therapy was completed and before surgical resection. Magnetic resonance imaging (MRI) of the chest was used in selected patients. Bronchoscopy and mediastinoscopy were performed when mediastinal lymphadenopathy was suspected on CT scan of the chest and before the initiation of induction chemoradiotherapy. This experience predated our use of the positron emission tomography scanner.
With institutional review board (IRB) approval and IRB waiving patient consent, a detailed retrospective chart review was performed and the following items recorded: patient age and sex; histologic tumor type; preoperative and postoperative tumor-node-metastasis (TNM) status; type of preoperative and postoperative treatment; type of operation, including extent of rib resections, nerve resections, and neurolysis; completeness of resection; postoperative complications and mortality; disease status at the last follow-up contact; and when appropriate, dates and sites of relapse. Overall survival was recorded from the time of diagnosis until death or last follow-up. Disease-free interval was defined as the length of time from the initiation of treatment until recurrence (local or distant). Local failure was defined as tumor recurrence arising from the surgical resection site. A resection was considered complete if the surgeon recorded that all gross tumor was removed and the surgical margins were histologically free of disease (negative). A complete response to induction therapy was defined as the absence of viable tumor cells in the resected specimen. Pathologic staging of all specimens was performed using the American Joint Commission on Cancer (AJCC) system [11]. When not available by chart review, follow-up information was obtained by communication with the patient, family, or referring physician.
Patients had either no induction therapy or a varied combination of preoperative chemotherapy and radiation therapy. The chemotherapy regimens consisted of
- 1 cisplatin (50 mg/m2) on days 1, 8, 28, and 36 and etoposide (50 mg/m2) on days 1 to 5 and 29 to 33, or
- 2 carboplatin (AUC2) and paclitaxel (50 mg/m2) given weekly, or
- 3 carboplatin (AUC5) and paclitaxel (175 mg/m2) every 3 weeks.
Chemotherapy, if given, was administered concurrently with external beam radiotherapy delivered in daily fractions of 180 to 200 cGy for a total dose that ranged between 44 and 60 Gy.
Overall survival and disease-free interval were calculated using the Kaplan-Meier method [12]. Log-rank tests and Cox proportional hazards methods were used to conduct between-group comparisons in overall survival and disease-free distributions. All tests were two-sided with a type I error of 5%. Stepwise selection technique in Cox proportional hazards model was applied for multivariable analysis. Factors considered in the model for selection included induction therapies, resection margin status, response to induction therapy, clinical stage, and pathologic stage.
|
Results
|
|---|
Thirty-nine patients were in the study cohort (14 women, 25 men). The median and mean age was 59 years (range, 40 to 77 years). The predominant presenting symptom was ipsilateral pain in 77% of patients (anterior chest, shoulder, back, arm). Horner syndrome and no symptoms (incidental findings) were equally present in 10% of patients. Local involvement was identified in rib(s) in 33 (85%), subclavian artery in 1 (2.5%), stellate ganglion in 4 (10%), vertebral body in 2 (8%), and resected metastasis to brain in 2 (5%). Bone scans were locally positive in only 7 patients and in no patients with proven preoperative vertebral body involvement.
After all staging procedures (including mediastinoscopy) were performed, Stage IIB (T3N0) was predominant in 33 patients (85%). Tumors were classified as T3 in 36 patients (92%), T4 in 3 patients (8%), N1 in 1 patient (2.5%), and M1 (brain) in 2 patients (5%). Twenty-eight patients underwent mediastinoscopy, and all were negative for metastatic disease. Four patients with a positive CT scan for N2 status were proven negative with mediastinoscopy.
The median size of tumor was 5.0 cm (range, 3 to 9 cm). Tumors were localized to the right side in 26 and the left side in 13. Histologic examination revealed adenocarcinoma, 12 (31%); squamous cell, 12 (31%); adenosquamous cell, 11 (28%); large cell, 2 (5%); bronchioloalveolar, 1 (2.5%); and spindle cell, 1 (2.5%). Patient characteristics are detailed in Table 1.
Induction Therapy
Superior sulcus tumor was not diagnosed preoperatively in 8 patients (21%), and no induction therapy was administered. Thirty-one patients (80%) received preoperative external beam radiotherapy with a median dose of 45 Gy (20, 44 to 45 Gy; 11, 50 to 60 Gy). Four of these 31 patients did not receive concurrent chemotherapy. Three patients received radiation therapy elsewhere, and 1 patient refused chemotherapy. Of the 27 patients (69%) who received combined chemoradiation therapy, 19 patients received cisplatin and etoposide (13 were also part of the SWOG 9416 intergroup study) and 8 received carboplatin and paclitaxel.
Operative Details
Contraindications to surgery included extensive involvement of the brachial plexus, invasion of the vertebral cancellous bone, spinal cord encroachment, and metastatic disease that included mediastinal lymph node involvement.
The operative technique resembles that described by Shaw and Paulson in their original operative description and includes the en bloc resection of lung, ribs, stellate ganglion, and C8 and T1 roots, as well as portions of transverse processes and vertebral bodies, if required. Intraoperative mediastinal lymph node sampling was performed in all instances. When positive, a complete mediastinal lymph node dissection was performed. Venous structures, if involved, were resected and reconstructed as necessary.
The primary operative resection was biopsy only in 1, wedge resection in 2, wedge resection en bloc with chest wall in 8, lobectomy in 1, lobectomy en bloc with chest wall in 25, pneumonectomy in 1, and pneumonectomy en bloc with chest wall in 1. Of these, three palliative procedures were performed because of encasement of the subclavian artery, encasement of the superior vena cava, and extensive multifocal disease. In these instances, biopsy only, wedge resection of lung, and palliative lobectomy were performed (Table 2).
A complete surgical extirpation required en bloc resection of ribs in 35 patients, with nearly all requiring multiple rib resections. The first rib was resected in 13 instances. Nerve resection of C8 was performed in one instance, T1 in 13, and T2 in 20. Neurolysis of the brachial plexus was performed in 13 patients and neurolysis of the phrenic nerve in 2 patients. We were unable to establish the true resectability rate of the stellate ganglia (Table 3).
Postoperative complications occurred in 12 patients (31%), with arrhythmia the most frequent (7). Acute respiratory distress syndrome developed in 3 patients, a left hemidiaphragm occurred in 1 patient because of a phrenic nerve injury, and another suffered a persistent bronchopleural fistula. There were 2 postoperative deaths (5%): 1 from sepsis after aspiration and 1 from lethal arrhythmia. The resectability rate was 92%.
Pathologic Response
Of the 34 patients undergoing potentially curative resections, 26 patients (76%) had microscopically negative margins (R0). Positive microscopic margins (R1) were detected in 8 patients (24%). Results of pathologic staging are depicted in Tables 4 and 5. Fourteen patients were downstaged to stage 0 [9], IA [3], and IB [2]. Four patients were upstaged to either N1 or N2 disease, and 3 patients were upstaged to T4 disease. In those patients with N2 disease at operation, a negative mediastinoscopy had been done before induction therapy. One can presume either inadequacy of mediastinoscopy or progression of disease during the induction therapy program.
A complete pathologic response was observed in 9 (29%) of the 31 patients who received preoperative therapy. Two of the 4 patients who received preoperative radiation therapy alone were upstaged and 1 was downstaged. Of the 27 patients who received a combination of preoperative chemoradiation therapy, 8 had no change, 6 were upstaged, and 13 were downstaged. Nine of the patients who received some form of induction therapy were upstaged (29%) and 14 (45%) were downstaged. This included the 9 patients who had a complete pathologic response.
Survival Analysis
Median follow-up of surviving patients was 69 months (5.75 years), ranging between 38 and 116 months. At the last follow-up, 17 of 39 patients were alive and well and free of disease. The 5-year overall survival for the entire cohort was 47.9%, with a median survival of 40 months. The 5-year survival for patients undergoing R0 and R1 resections was 52.1% and those that responded to induction therapy, 60.6%.
The completeness of resection determines survival (p = 0.0001), response to induction predicts survival (p = 0.02), completeness of resection influences time to recurrence (p = < 0.00001), and response to induction influences the time to recurrence (p = 0.01) (Figs 1–5). Neither age nor sex nor type of resection were significant prognostic predictors for overall or disease-free survival. Patients with T0, T1, and T2 pathologically staged tumors had a 5-year overall survival of 55% in contrast to 32% in patients with T3 and T4 lesions (p = 0.04).
View larger version (15K):
[in this window]
[in a new window]
|
Fig 1. Overall survival of all patients (solid line). Dashed lines represent the 95% limits of confidence.
|
|
After curative resections, recurrences occurred in 12 patients; 4 recurrences were local and 8 were distant. The median time to recurrence was 5 months (range, 0.5 to 22 months). Second primaries were identified in two patients, and in both instances the first primary tumor had not recurred. In 26 patients with R0 resections, 2 had local and 3 had distant recurrences. In 8 patients with R1 resections, 2 had local and 5 had distant recurrences. In no instance did any patient have both local and distant recurrence. Brain metastases were identified in 3 patients, 1 with an R0 resection and 2 with an R1 resection. Two patients presented with an initial brain metastasis. One was dead at 3 months, and the other was free of disease at 53 months. Of the 9 patients that had a complete response at operation, 7 remained free of disease at the last contact.
Individually, margin status and response to induction therapy each have a significant association with overall survival. As well, a negative margin correlates with response to induction therapy, and disease-free survival correlates with response to induction therapy. No correlation could be identified between survival and chemotherapy regimen, T status, operative procedure, or complete pathologic response. Multivariable analysis identified margin status as the only factor significantly associated with overall survival.
|
Comment
|
|---|
This series retrospectively reviewed our experience in the management of superior sulcus tumors during a 7-year period. During that time, two distinct regimens involving induction chemoradiotherapy (cisplatin/etoposide and carboplatin and paclitaxel) were available. Nine patients had a complete pathologic response, with no tumor detected in the resected specimen. The 29% complete response rate in this series is similar to the 34% reported by Rusch and colleagues using induction chemoradiation and cisplatin and etoposide [10]. Patients who had evidence of a response to therapy, either complete or partial, had an improved disease-free survival compared with patients who had no evidence of response.
Interestingly, Rusch and colleagues found that radiographic assessment of response to induction chemoradiation therapy did not correlate well with pathologic response. Over 50% of the patients thought to have radiographically stable disease had either a complete or a significant partial response [10]. In our series of patients, clinical staging before induction was compared with pathologic staging at operation. Comparisons were not made with radiographic staging before and after surgery.
At presentation, 2 of our patients had vertebral body involvement, 1 of whom also had subclavian vessel involvement. Both were treated with induction chemoradiation therapy, and the patient with vertebral body involvement only was resected with negative microscopic margins (R0). This patient had not recurred at the last contact. In their series Sartori and colleagues reported that none of the 8 patients with vertebral body involvement and only of 5 patients with subclavian vessel involvement survived more than 1 year [13]. Our experience is too small to suggest that complete resection of T4 disease with a durable survival can be achieved after preoperative chemoradiotherapy, but this was suggested in the early series reported from the Sloan Kettering Memorial Hospital and again in the more recent SWOG 9516 study [9, 10].
Previous reports suggested the potential benefits of both induction radiotherapy and resection over radiotherapy alone. Beyer and colleagues reported a 5-year survival of 45% for patients treated with induction radiotherapy and resection and 22% for those treated with radiotherapy alone [14]. Neal and Anderson and colleagues reported similar survival for radiotherapy alone [15, 16]. Hilaris and colleagues reported resectability of 23% in patients who did versus 9% in patients who did not receive preoperative radiotherapy [17].
Beyer and associates suggested a correlation between radiotherapy dose and survival [14]. Shahian and colleagues reported an 87% 5-year survival in patients with T3N0M0 tumors treated with induction radiotherapy (with or without chemotherapy), radical resection, and postoperative radiotherapy; their entire series had a 56% 5-year survival [18]. Devine and colleagues, however, were unable to show a difference in survival (or resectability) between patients given 30 to 35 Gy and patients given 45 to 50 Gy [19].
The median dose of preoperative radiotherapy given to our patients was 45 Gy (range 44 to 60 Gy). Of the 9 patients who received higher doses of induction radiotherapy, 1 received 60 Gy, 1 received 57.6 Gy, 6 received 54 Gy, and 1 received 50 Gy. Six of these 9 patients achieved R0 resection and 3 underwent R1 resection. Of the 14 patients who received 45 Gy and preoperative chemotherapy, 9 underwent R0 resections, 2 underwent R1 resections, one underwent a R2 resection, and 2 were unresectable at thoracotomy. Our experience does not suggest a dose-dependent advantage to induction radiotherapy.
Despite initial evidence showing that induction chemotherapy did not have beneficial effects on long-term survival [20], recent reports have revealed potential salutary effects on resectability. Yashar and associates reported that 31 of 36 patients treated with concurrent radiation and chemotherapy in the preoperative period were completely resected [21]. The SWOG reported that 55 of 68 patients with IIIA (N2) or IIIB disease who had received induction chemotherapy in addition to radiotherapy were completely resectable at the time of thoracotomy [9]. Rusch and associates reported that 76 of 83 patients enrolled in SWOG 9416 were completely resected after receiving preoperative chemoradiotherapy [10]. In our series, 27 patients received both induction chemotherapy and radiotherapy. Of these, 17 patients (63%) achieved R0 resection, 6 patients (22%) underwent R1 resection, 2 underwent R2 resection, and 2 patients were found to be unresectable at thoracotomy.
In our series, 5 of 26 patients with R0 resections recurred and 7 of 8 patients with R1 resections recurred. Three of the 5 patients who developed recurrences after R0 resection did not receive induction chemotherapy or radiotherapy. Two of the 5 recurrences were locoregional; other sites included the central nervous system (CNS), liver, and inguinal lymph nodes. Two of the 7 patients who developed local recurrence after R1 resection had not received induction chemotherapy or radiotherapy; CNS recurrence later developed in these patients. Local recurrence developed in two other patients who had only received induction therapy. These results suggest that the addition of induction chemotherapy to radiotherapy may help sterilize the tumor bed before resection.
Summary
Surgical resection of superior sulcus tumors after induction chemoradiotherapy can be accomplished with low morbidity and mortality. A complete response rate of 29% was observed, with an improvement in disease-free survival in patients who had either a partial or complete response to induction therapy. Furthermore, the effect of induction therapy on the tumor may increase resectability and improve the likelihood of achieving negative resection margins. Completeness of resection and response to induction therapy significantly affects patient overall survival and time to recurrence.
Conclusions
Complete surgical resection with negative margins for superior sulcus tumors has a significant impact on survival. Induction chemoradiation may downstage tumors, enhance the ability to obtain negative resection margins, and prolong survival. The use of induction therapy and completeness of resection have a significant impact on disease-free survival. Trimodality therapy for superior sulcus tumors has produced an overall 5-year survival of 60.6% in patients responding to induction chemoradiation.
|
Acknowledgments
|
|---|
We express our appreciation to Rose Vickers and Marguerite Gonzalez for their help in preparing this report, and the entire team of physicians, nurses, and other health care professionals who have contributed to the care of patients with superior sulcus tumors at the Fox Chase Cancer Center.
|
References
|
|---|
- Hare ES. Tumor involving certain nerves[letter] London Med Gaz 1838;1:16-18.
- Pancoast HK. Superior sulcus tumors: tumor characterized by pain, Horner's syndrome, destruction of bone, and atrophy of hand muscles JAMA 1932;99:1391-1396.
- Chardack WM, MacCallum JD. Pancoast tumor: five-year survival without recurrence or metastases following radical resection and post-operative radiation J Thorac Surg 1956;31:535-542.
- Shaw RR, Paulson DL, Kee Jr JL. Treatment of the superior sulcus tumor by irradiation followed by resection Ann Surg 1961;154:29-40.[Medline]
- Paulson DL. Carcinomas in the superior pulmonary sulcus J Thorac Cardiovasc Surg 1975;70:1096-1104.
- Shaw RR. Pancoast's tumor Ann Thorac Surg 1984;37:343-345.[Abstract]
- Dillman RO, Herndon J, Seagren SL, Eaton Jr WL, Green MR. Improved survival in stage III non-small cell lung cancer: seven-year follow-up of cancer and leukemia group B (CALBG) 8433 trial J Natl Cancer Instit 1996;88:1175-1177.[Free Full Text]
- Albain KS, Rusch VW, Crowley JJ, et al. Concurrent cisplatin/etoposide plus chest radiotherapy followed by surgery for stages IIIa (N2) and IIIb non-small cell lung cancer: mature results of Southwest Oncology Group Phase II Study 8805 J Clin Oncol 1995;13:1880-1892.[Abstract/Free Full Text]
- Rusch VW, Parekh KR, Leon L, et al. Factors determining outcome after surgical resection of T3 and T4 lung cancers of the superior sulcus J Thorac Cardiovasc Surg 2000;119:1147-1153.[Abstract/Free Full Text]
- Rusch VW, Giroux DJ, Kraut MJ, et al. Induction chemoradiation and surgical resection for non-small cell lung carcinomas of the superior sulcus: Initial results of Southwest Oncology Group Trial 9416 (Intergroup Trial 0160) J Thorac Cardiovasc Surg 2001;121:472-483.[Abstract/Free Full Text]
- Mountain C. Revisions in the International System for Staging Lung Cancer Chest 1997;111:1710-1717.[Abstract/Free Full Text]
- Kaplan EL, Meier P. Nonparametric estimation from incomplete observations J Am Stat Assoc 1958;53:457-481.
- Sartori F, Rea F, Calabro F, Mazzucco C, Bortolotti L, Tomio L. Carcinoma of the superior pulmonary sulcusResults of irradiation and radical resection. J Thorac Cardiovasc Surg 1992;104:679-683.[Abstract]
- Beyer DC, Weisenburger T. Superior sulcus tumors Am J Clin Oncol 1986;9:156-161.[Medline]
- Neal CR, Amdur RJ, Mendenhall WM, Knauf DG, Block AJ, Million RR. Pancoast tumor: radiation therapy alone versus preoperative radiation therapy and surgery Int J Radiat Oncol Biol Phys 1991;21:651-660.[Medline]
- Anderson TM, Moy PM, Holmes EC. Factors affecting survival in superior sulcus tumors J Clin Oncol 1986;4:1598-1603.[Abstract/Free Full Text]
- Hilaris BS, Martini N, Luomanen RJK, Batata M, Beattie Jr EJ. The value of pre-operative radiation therapy in atypical cancer of the lung Surg Clin North Am 1974;54:831-840.[Medline]
- Shahian DM, Neptune WB, Ellis Jr FH. Pancoast tumors: improved survival with preoperative and postoperative radiotherapy Ann Thorac Surg 1987;43:32-38.[Abstract]
- Devine JW, Mendenhall WM, Million RR, Carmichael MJ. Carcinoma of the superior pulmonary sulcus treated with surgery and/or radiation therapy Cancer 1986;57:941-943.[Medline]
- Ginsberg RJ, Martini N, Zaman M, et al. Influence of surgical resection and brachytherapy in the management of superior sulcus tumor Ann Thorac Surg 1994;57:1440-1445.[Abstract]
- Yashar J, Weitberg AB, Glicksman AS, Posner MR, Feng W, Wanebo HJ. Preoperative chemotherapy and radiation therapy for stage IIIa carcinoma of the lung Ann Thorac Surg 1992;53:445-448.[Abstract]
This article has been cited by other articles:
|
|
|
|
 
B. Yildizeli, P. G. Dartevelle, E. Fadel, S. Mussot, and A. Chapelier
Results of primary surgery with T4 non-small cell lung cancer during a 25-year period in a single center: the benefit is worth the risk.
Ann. Thorac. Surg.,
October 1, 2008;
86(4):
1065 - 1075.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
