Survival After Extended Resection for Mediastinal Advanced Lung Cancer: Lessons Learned on 167 Consecutive Cases

Transcription

1 Survival After Extended Resection for Mediastinal Advanced Lung Cancer: Lessons Learned on 167 Consecutive Cases Lorenzo Spaggiari, MD, PhD, Adele Tessitore, MD, Monica Casiraghi, MD, Juliana Guarize, MD, Piergiorgio Solli, MD, PhD, Alessandro Borri, MD, Roberto Gasparri, MD, Francesco Petrella, MD, Patrick Maisonneuve, Eng, and Domenico Galetta, MD, PhD Division of Thoracic Surgery, European Institute of Oncology, University School of Milan, and the Division of Epidemiology and Biostatistics, European Institute of Oncology, Milan, Italy Background. Extended resections (ER) for lung cancer may improve survival in selected patients. However, analysis on large series is still lacking. We reviewed our experience to identify prognostic factors useful for patient selection. Methods. Between 1998 and 2010, 167 patients with involvement of one or more mediastinal organs underwent operations with the intent to perform ER. At thoracotomy, 42 patients (25%) were considered unresectable (explorative thoracotomy [ET]), and 125 (75%) underwent ER. The types of ER were superior vena cava in 43 patients (34.4%), carina in 33 (26.4%), combined with superior vena cava in 18 (14.4%), with the left atrium in 35 (28%), and with the aorta in 14 (11.2%). We excluded Pancoast tumors and vertebral resections. The minimum follow-up was 6 months. Kaplan-Meier method and logrank test were used for statistical analysis of survival. Results. There were 136 men (81.4%), with mean age of 63 years (range, 36 to 81 years). Of the 167 patients, induction chemotherapy was administered in 119 (71.3%), including 34 ET patients (81%) and 85 ER patients (68%). Complete resection was achieved in 106 patients (84.8%). The overall 5-year survival was 23% (27% in ER and 13% in ET, p [ 0.41). Overall 30-day mortality was 4.8% and morbidity was 34.1%. Factors affecting survival were complete resection (p < 0.01), pstage 0-I-II disease (p < ), and age younger than 60 years (p < 0.01). Conclusions. ER for lung cancer invading mediastinal organs could improve long-term survival (46% at 5-years in pn0). The best surgical candidates are young patients without lymph nodes involvement who undergo radical resection. Multimodality treatment is suggested in case of mediastinal lymph node involvement. (Ann Thorac Surg 2013;95: ) Ó 2013 by The Society of Thoracic Surgeons Today, management of locally mediastinal advanced non-small cell lung cancer (NSCLC) is a matter of intense debate, and factors affecting survival are still unclear. These lung cancers are a heterogeneous group of tumors invading the left atrium, great vessels, trachea, and carina, and they identify a particular group of tumors with specific aspects of the surgical technique and different oncologic behaviors [1]. The oncologic results of such patients who are medically treated are disappointing, with 5-year survival of 3% to 17% for unresected disease [2 4]. During the past 2 decades, advances in surgical technique have made feasible complete resection of patients Accepted for publication Jan 29, Presented at the Forty-eighth Annual Meeting of The Society of Thoracic Surgeons, Fort Lauderdale, FL, Jan 28 Feb 1, Address correspondence to Dr Spaggiari, European Institute of Oncology, Division of Thoracic Surgery, Via Ripamonti 435, Milan, Italy; lorenzo.spaggiari@ieo.it. requiring en bloc removal of adjacent structures, such as the left atrium, great vessels, and trachea, and a survival benefit using these techniques has been demonstrated in highly selected patients [5, 6]. However, considerable uncertainty still exists concerning the role of surgical resection in the different types of locally advanced disease and the prognostic significance of lymph node status and completeness of resection in these patients as well as the role of complementary therapies (radiotherapy or chemotherapy). Moreover, the increased postoperative morbidity and mortality raise concerns about the real benefit of extended resection (ER). Up to now, except for one report [7], no consecutive series with more than 100 patients has been available. In addition, no informations concerning the outcome of candidates for ER, but who are not resected, are available. This may be the group of reference of major interest with respect to ER that should be compared. The aim of this work was to analyze our experience with particular attention to oncologic results of such a population. Ó 2013 by The Society of Thoracic Surgeons /$36.00 Published by Elsevier Inc

2 1718 SPAGGIARI ET AL Ann Thorac Surg EXTENDED RESECTION FOR LUNG CANCER 2013;95: Abbreviations and Acronyms CI = confidence intervals CT = computed tomography ER = extended resections ET = explorative thoracotomy HR = hazard ratio NSCLC = non small cell lung cancer PET = positron emission tomography PTFE = polytetrafluoroethylene SVC = superior vena cava Patients and Methods This study was approved by the Institutional Review Board. Individual consent for this study was obtained. Between 1998 and 2010, 167 patients with locally advanced mediastinal NSCLC underwent resection. All of these patients were candidates for ER after a careful preoperative evaluation. However, after surgical exploration, 42 patients (25%) were considered unresectable and underwent only exploratory thoracotomy (ET). Thus, only 125 patients (75%) underwent ER and are the subjects of this study. Preoperative Workup All patients underwent a preoperative evaluation that included a chest computed tomography (CT) scan, brain CT, or magnetic resonance imaging and positronemission tomography (PET) scan. Routine biochemical profile, fibroscopy, pulmonary function tests with spirometry, and arterial blood gas analysis at rest were required in all patients. Mediastinoscopy was performed in patients with a mediastinal lymph node larger than 1 cm or PET-positive. We excluded in this study patients who underwent vertebral or esophageal resection and those with Pancoast tumor. Intrapericardial resections of the pulmonary artery were not considered ER. Surgical Technique A lateral thoracotomy was used in 108 patients (86.4%), a hemi-clamshell incision in 11 (8.8%), a posterolateral thoracotomy in 3 (2.4%), and an anterolateral thoracotomy in 3 (2.4%). Type of lung resection is summarized in Table 1. No extracorporeal circulation was used. Superior Vena Cava Invasion Briefly, when superior vena cava (SVC) involvement was less than 50% of the vessel circumference, resection and direct repair were the preferred technique using mechanical suture or hand-suture with nonabsorbable continuous suture on a vascular clamp. An autologous or heterologous pericardial patch was used to repair a vascular resection that was too large. When the SVC was not extensively infiltrated close to one of the brachiocephalic trunks (left or right), the entire venous trunk was removed without subsequent reconstruction. When SVC involvement was more than 50% of the circumference, prosthetic vascular replacement by cross-clamping technique was preferred. During the first period of the study (1998 to 2002), the prosthetic SVC replacement was performed with a polytetrafluoroethylene (PTFE) graft. Since 2003, all patients have received SVC replacement using a biological, custom-made bovine pericardial tube. The details of our surgical techniques have been largely reported in other studies [8, 9]. Left Atrium Resection When a left atrium resection is planned, the pericardium is fully opened to identify the extent of atrial involvement before the lung resection begins. The origin of the contralateral veins is also identified, avoiding their occlusion during clamping. Right-side tumors usually invade the left atrium widely owing to the shortness of the right upper pulmonary vein. To increase the length of the right atrial cuff, the Sondergaard technique is used, which involves dissection of the epicardium between the right and left atrium (the interatrial groove) to lengthen the left atrial cuff to about 2 cm. This increases the margin of resection, avoiding the risk associated with clamping the right atrium. Afterward, a large Satinsky clamp is put in place to evaluate the surgical feasibility and the cardiocirculatory consequences of atrial volume reduction. Atrial resection is performed after lung resection is completed. After detachment of the lung, 2 running stitches of monofilament nonabsorbable suture are placed on the left atrium. The operative steps are virtually the same for left-side tumors, except for the dissection of the interatrial groove [10]. Carina Resection The surgical approach varied according to the type of carinal invasion. Options included tracheal sleeve right or left pneumonectomy with tracheobronchial anastomosis, tracheal sleeve lobectomy/bilobectomy with the remnant lobar bronchus reimplanted on the lateral part of the trachea or of the already reimplanted contralateral main bronchus, and wedge resection with direct suture [11 13]. Aortic Resection To define an extended aortic resection, it is necessary that the specimen at least include the adventitia. Aortic invasion was approached by intraadventitial dissection and excision of superficially invading tumors, by limited vascular excision requiring direct suture or patch reconstruction, or by en bloc resection with prosthetic replacements of the portion of the aorta with PTFE conduit [14]. Preoperative Therapy In the 125 ER patients, 78 (62.4%) were treated with preoperative cisplatin-based chemotherapy, 1 (0.8%) was treated with preoperative radiotherapy, and 7 (5.6%) with preoperative chemoradiotherapy. In ET group, 29 (69%) received chemotherapy and 5 (12%) received chemoradiotherapy. Postoperative Therapy Among the ER patients, adjuvant chemotherapy was administered to 11 (8.7%), radiotherapy to 33 (26.1%) and

3 Ann Thorac Surg SPAGGIARI ET AL 2013;95: EXTENDED RESECTION FOR LUNG CANCER 1719 Table 1. Clinicopathologic Characteristics of the 167 Patients Who Underwent Resection for T4 Non-Small Cell Lung Cancer With Mediastinal Invasion Structures Resected Characteristics a ER ET Carinal SVC Aorta Atrium Patient (26.4) 43 (34.4) 14 (11.2) 35 (28) Sex Male 101 (80.2) Female 24 (19.8) Age, y 63 (36 81) 63 (36 78) Pulmonary resection 125 Lobectomy 25 (20) Bilobectomy 4 (3.1) Pneumonectomy 52 (41.6) Segmentectomy 1 (0.8). 1.. Sleeve Lobectomy 17 (13.6) Pneumonectomy 22 (17.6) Bilobectomy 4 (3.2) Pathology Squamous cell 62 (49.6) Adenocarcinoma 50 (40) Adenosquamous Large cell Pleomorphic NSCLC N status N0 18 (14.4) N1 54 (43.2) N2 49 (39.2) N3 4 (3.2) Resection R0 106 (84.8) R1 18 (14.4) R2 1 (0.8)... 1 Neoadjuvant therapy Yes 86 (68.8) No 39 (31.2) Morbidity 57 (34.1) 17 (51.5) 19 (44.1) 7 (50) 14 (40) Mortality 8 (4.8) 6 (18.1) 2 (4.6) 0 0 a Categoric data are presented as number (%) and continuous data as median (range). ER ¼ extended resection; ET ¼ explorative thoracotomy; NSCLC ¼ non-small cell lung cancer; SVC ¼ superior vena cava. chemoradiotherapy to 3 (2.3%). In the ET group, 5 (11.9%) received chemoradiotherapy, 20 (47.6%) received radiotherapy, 3 (7.1%) received chemotherapy, and 14 (31.9%) received supportive care due to their poor performance status. Surgical Resection Patients were divided into five subgroups according to the type of extended lung resection: SVC resection in 43, left atrium resection in 35, SVC and carina resections in 18, carina resection in 15, and aorta resection in 14. In the 61 patients who underwent SVC resection, vascular resection was partial in 45 (73.8%), and 5 patients needed a patch of autologous pericardium for reconstruction, a synthetic vascular prosthesis was used in 6 patients (9.8%), and a bovine pericardial tube was used in 10 (16.4%). Tangential resection of the left atrium was performed in all 35 patients with atrial invasion, and among these, 2 patients needed a concurrent aortic resection. Carina resection was done in 33 patients: a tracheal sleeve pneumonectomy was performed in 23 (69.7%), a tracheal sleeve lobectomy in 6 (18.2%), a tracheal sleeve bilobectomy in 1 (3.0%), and a wedge carina resection in 3 (9.1%). In 14 patients with aortic invasion, 10 (71.4%) underwent adventitial dissection of the aorta because of superficial invading tumor, and 2 (14.3%) required a partial resection of the aorta with direct running suture. A circumferential

4 1720 SPAGGIARI ET AL Ann Thorac Surg EXTENDED RESECTION FOR LUNG CANCER 2013;95: Table 2. Postoperative Complication in Resected Patients Resection, No. (%) Variable Total (%) Carinal SVC Aorta Atrium Death Cardiac complications 125 (100) 33 (26.4) 43 (34.4) 14 (11.2) 35 (28) No. (%) Arrhythmia 26 (20.63) Cardiogenic shock 1 (0.79) Cardiac dislocation 2 (1.59) Other 4 (3.17) Pulmonary complications Atelectasia 9 (7.14) Air leak >7 days 4 (3.17) Pyothorax/septic shock 2 (1.59) Bronchial fistula 5 (3.97) Postpneumo pulmonary edema 2 (1.59) Prolonged intubation 1 (0.79) 1... Tracheostomy 3 (2.38) Other 6 (4.76) Bleeding 11 (8.73) Overall patients 57 (45.6) 17 (51.5) 19 (44.1) 7 (50) 14 (40) 8 (6.4) SVC ¼ superior vena cava. aortic resection was performed in 2 patients (14.3%). For reconstruction in 1 patient, an end-to-end anastomosis was feasible; in the other patient, a synthetic prosthesis was used. The clinicopathologic characteristics of the 125 patients are summarized in Table 2. Follow-Up Patients were assessed at 1 month and every 4 months afterward. All patients received a contrast chest CT scan before discharge and before each follow-up visit, coupled with a CT scan of the upper abdomen. Information for the present study was obtained by directly contacting the patient or the referring physician. Statistical Analysis Postoperative morbidity and mortality included all adverse events occurring within the first 30 days after operation or during the same hospitalization. Patient survival rates were determined using the Kaplan-Meier method, and the prognostic significance of potential variables was determined by means of univariate analysis (log-rank and Wilcoxon test). A multivariate analysis was conducted to adjust for possible confounders, using a Cox proportional hazard regression. The hazard ratio (HR) and 95% confidence intervals (CI) for each variable included in the model were calculated. Results Morbidity and Mortality The overall morbidity was 34.1% (57 of 167) and the mortality rate was 4.8% (8 of 167). In the ER group of 125 patients, morbidity was 45.6% and mortality was 6.4%, and 90-day mortality after ER was 11.2% (14 patients). The most common complications were pulmonary and Fig 1. (A) Overall survival of 167 patients with mediastinal advanced lung cancer. (B) Survival according to extended resection (ER) and exploratory thoracotomy (ET).

5 Ann Thorac Surg SPAGGIARI ET AL 2013;95: EXTENDED RESECTION FOR LUNG CANCER 1721 cardiovascular disorders. The most frequent surgical complication was postoperative bleeding, resulting in 8 patients needing repeat thoracotomy for treatment, persistent air leak, and bronchopleural fistula. Overall, 16 patients (28% of complications) required an operation to treat their complication (Table 2). The median intensive care unit stay was 1 day (range, 0 to 60 days). The median postoperative stay was 7 days (range, 4 to 75 days). Survival The final date for follow-up was June 30, The median follow-up from the date of the operation was 20 months (range, 1 to 146 months) for the 167 patients. The median follow-up was 27 months (range, 2 to 146 months) for the ER group and 19 months (range, 1 to 124 months) for the ET group. The overall 5-year survival rate for the entire sample was 23% (95% CI, 16% to 30%) with a median survival of 23 months (Fig 1A). The overall 5-year survival for the 125 patients who underwent resection was 27% (95% CI, 17% to 34%) with a median survival of 23 months (Fig 1B). The 5-year survival rate in the ET group was 13% (95% CI, 1% to 24%), with a median survival of 22 months (Fig 1B). There was no statistically significant difference in survival between the two patient groups (p ¼ 0.41). In the group of resected patients, overall 5-year survival rates, according to the nearby organ invasion, were aorta invasion, 37%; atrium invasion, 25%; carina involvement, 22%; and SVC invasion, 26%. The aorta resection group showed a better survival results than the other groups, mostly when compared with the carina resection group (p ¼ 0.01; Fig 2A). Univariate analysis found the following factors were significantly associated with survival: residual tumor R0 vs Rþ (p ¼ 0.041), pathologic stage 0 to II vs III to IV (p ¼ ), and lymph node status pn0 vs pnþ (p ¼ 0.034; Fig 2B D). Considering only the 106 patients with R0 resection, the median and 5-year survivals were, respectively, 50 months and 46% for N0, and 17 months and 26% for pnþ (p ¼ 0.059). The multivariate Cox regression analysis for the 167 patients showed that advanced age (60 years), advanced Fig 2. Overall survival of 125 patients who underwent extended resection according to (A) type of T4 disease, (B) type of resection, (C) tumor stage, and (D) pathologic nodal involvement. (SVC superior vena cava.)

6 1722 SPAGGIARI ET AL Ann Thorac Surg EXTENDED RESECTION FOR LUNG CANCER 2013;95: Table 3. Univariate and Multivariate Analysis of Factors Affecting Overall Survival by the Kaplan Meier Method Characteristics No. (%) Patients 5 year Survival Rate (%) Multivariate Cox Log Rank Test Regression Multivariate p Value HR (95% CI) p Value Patient 167 (42 ET) 23 (16 30) Gender 125 Male 101 (80.8) 27 (17 36) Female 24 (19.2) 21 (4 39) 1.21 ( ) Age 125 <60 years 41 (32.8) 37 (21 52) years 84 (62.7) 20 (10 29) 1.95 ( ) Pathology (125 patients) Squamous cell 62 (49.6) 21 (10 32) Adenocarcinoma 49 (39.2) 28 (14 42) 0.99 ( ) 0.96 Other types 14 (11.2) 36 (11 61) 1.07 ( ) 0.88 Surgery Explorative thoracotomy 42 (25.2) 13 (1 24) Extended resection 125 (74.8) 26 (17 34) ( ) 0.9 Carina resection 33 (26.4) SVC resection 43 (34.4) Aorta resection 14 (11.2) Left atrium resection 35 (28) Pneumonectomy 125 Yes 74 (59.2) 19 (9 29) ( ) 0.68 No 51 (40.8) 34 (20 47) p Stage 0 I II 27 (16.2) 46 (25 65) III IV 98 (58.7) 20 (11 29) 2.92 ( ) Explorative thoracotomy 42 (25.2) 13 (1 24) 3.14 ( ) 0.02 N status 125 N0 18 (14.4) 46 (22 70) Nþ 107 (85.6) 22 (14 31) ( ) 0.89 N1 54 (43.2) N2 49 (39.2) N3 4 (3.2) Resection 125 R0 106 (84.8) 29 (20. 39) ( ) 0.16 R (15.2) 10 (0 23) Neoadjuvant treatment 125 Yes 86 (68.8) 24 (14 33) ( ) 0.15 No 39 (31.2) 32 (14 49) Adjuvant treatment a 125 Yes 48 (38.4) 15 (3 26) ( ) 0.75 No 72 (57.6) 32 (21 44) a Missing data. CI ¼ confidence interval; ET ¼ exploratory thoracotomy; HR ¼ hazard ratio; SVC ¼ superior vena cava. stage (III to IV), and exploratory thoracotomy were independent predictors of survival (Table 3). Comment Locally mediastinally advanced NSCLC has been considered an unresectable disease for a long time, with 5-year survival of 3% to 17% after medical treatment [7, 15 19]; however, there may be a different biologic behavior in patients with local T4 disease with limited lymph node involvement (stage IIIA) and in those with N2 or N3 nodal disease (stage IIIB). T4 tumors with limited lymph node involvement are supposed to be biologically different from tumors with extensive lymph node involvement because they show fewer propensities to metastasize. Accordingly, patients with T4 locally advanced tumors without mediastinal lymph mode involvement could be considered as surgical candidates,

7 Ann Thorac Surg SPAGGIARI ET AL 2013;95: EXTENDED RESECTION FOR LUNG CANCER 1723 even if extended resection of tumor invading nearby mediastinal structure is technically demanding and few series have been published so far [7, 15 19]. Some authors reported a survival benefit after surgical ER, but the prognosis of surgically treated T4 NSCLC varies considerably, and an appropriate therapeutic strategy for each T4 patient has not yet been established. Moreover, arguments against a surgical approach are based on the observed increase in the morbidity and mortality rates with little objective benefit in survival. Previous published studies reported a morbidity rate varying from 20% to 50% and a postoperative mortality rate from 8% at 30 days to 18% at 90 days, even in experienced hands [7, 15 19]. Today some questions remain open: should chemotherapy be used in all cases of lung cancer infiltrating the mediastinal structures? Does chemotherapy increase the selection of patients, surgical complications, or postoperative mortality? What factors influence the survival of patients? In our experience, chemotherapy did not influence in any way the survival of patients, postoperative complications, or death. Certainly, however, it can determine an accurate selection of the patient and excludes from operations those patients with a rapid progression. In our study, 16.2% of patients who received ER had pathologic stage III disease, but this was the effect of response to induction chemotherapy. The patients in our study with resected mediastinal advanced NSCLC showed an overall 5-year survival of 27% (95% CI, 17% to 34%) compared with 13% (95% CI, 1% to 24%) of patients who underwent ET (p ¼ 0.41). Given the short duration of follow-up of 27 months for ER group and 19 months for the ET group, the median duration of survival was the more reliable survival statistic, showing close similarity between the ET and ER cohorts. We failed to find a subgroup of T4 tumor with a clearcut prognostic advantage: only patients with aorta invasion showed a trend toward better long-term results, with a 5-year survival rate of 37% in the ER patients. The factors that were found to possibly affect survival were age, complete resection, and lymph node status (Table 3). Our data showed an overall morbidity rate of 45.2%, and postoperative mortality rates were 6.4% at 30 days and 11.2% at 90 days. These data are comparable with previous reports [14, 20, 21], and together with survival, are improved in the most recent period compared with the early experience. Patients who underwent complete resection with pn0 disease showed better survival than those with R1 resection and pnþ disease. Because patients with pathologic positive mediastinal lymph node metastases have a negative prognosis, pn2 disease should be considered a contraindication for ER. Considerable controversy exists regarding surgical treatment for patients with pn2 disease. N2 disease, as in the patients with incomplete resection, is the most powerful negative prognostic factor. Among the patients with T4 NSCLC, pn2 disease had an adverse impact on overall survival. Vansteenkiste and colleagues [22] reported that mediastinoscopy-negative N2 patients had a significantly better prognosis than mediastinoscopypositive N2 patients. Given the poor results of patients resected for T4 N2 NSCLC, mediastinoscopy is a critical preoperative assessment to exclude patients with occult N2 disease who would have no survival benefit from aggressive surgical therapy. In conclusion, surgical resection for lung cancer with involvement of mediastinal structures remains rare but can be accomplished in selected patients with acceptable morbidity and mortality rates and in highly specialized centers. Our series confirms that long-term results are closely linked to the nodal status and completeness of resection. Preoperative selection of surgical candidates remains difficult. Routine mediastinoscopy or endobronchial ultrasound or endoscopic ultrasound, or both, should be performed, and patients with pn2 disease should probably be treated by definitive combined modality chemoradiation therapy. Second, surgical resection might be proposed only in patients who have a good response to induction treatment. References 1. Thomas W. Rice, Eugene H. Blackstone. Radical resections for T4 lung cancer. Surg Clin N Am 2002;82: Fournel P, Robinet G, Thomas P, et al. Randomized phase III trial of sequential chemoradiotherapy compared with concurrent chemoradiotherapy in locally advanced non small cell lung cancer: Groupe Lyon Saint Etienne d Onco logie Thoracique Groupe Francais de Pneumo Cancerologie NPC Study. J Clin Oncol 2005;23: Jang RW, Le Maitre A, Ding K, et al. Quality adjusted time in non small cell lung cancer: an analysis of the National Cancer Institute of Canada Clinical Trials Group JBR.10 trial. J Clin Oncol 2009;27: Goldstraw P, Crowley J, Chansky K, et al; International Association for the Study of Lung Cancer International Staging Committee; Participating Institutions. The IASLC Lung Cancer Staging Project: proposals for the revision of the TNM stage groupings in the forthcoming (seventh) edition of the TNM Classification of malignant tumours. J Thorac Oncol 2007;2: Albain KS, Swann RS, Rusch VW, et al. Radiotherapy plus chemotherapy with or without surgical resection for stage III non small cell lung cancer: a phase III randomised controlled trial. Lancet 2009;374: Tomaszek SC, Wigle DA. Surgical management of lung cancer. Semin Respir Crit Car Med 2011;32: Yıldızeli B, Dartevelle PG, Fadel E, Mussot S, Chapelier A. 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 2008;86: Spaggiari L, Leo F, Veronesi G, et al. Superior vena cava resection for lung and mediastinal malignancies: a single center experience with 70 cases. Ann Thorac Surg 2007;83: Spaggiari L, Magdeleinat P, Kondo H, et al. Results of superior vena cava resection for lung cancer. Analysis of prognostic factors. Lung Cancer 2004;44: Spaggiari L, D Aiuto M, Veronesi G, et al. Extended pneu monectomy with partial resection of the left atrium, without cardiopulmonary bypass, for lung cancer. Ann Thorac Surg 2005;79: Lanuti M, Mathisen DJ. Carinal resection. Thorac Surg Clin 2004;14: Regnard JF, Perrotin C, Giovannetti R, et al. Resection for tumors with carinal involvement: technical aspects,

8 1724 SPAGGIARI ET AL Ann Thorac Surg EXTENDED RESECTION FOR LUNG CANCER 2013;95: results, and prognostic factors. Ann Thorac Surg 2005;80: Mitchell JD. Carinal resection and reconstruction. Chest Surg Clin N Am 2003;13: Rice TW, Blackstone EH. Radical resections for T4 lung cancer. Surg Clin N Am 2002;82: Osaki T, Sugio K, Hanagiri T, et al. Survival and prognostic factors of surgically resected T4 non small cell lung cancer. Ann Thorac Surg 2003;75: Izbicki JR, Knoefel WT, Passlick B, Habekost M, Karg O, Thetter O. Risk analysis and long term survival in patients undergoing extended resection of locally advanced lung cancer. J Thorac Cardiovasc Surg 1995;110: Martini N, Yellin A, Ginsberg RJ, et al. Management of non small cell lung cancer with direct mediastinal involvement. Ann Thorac Surg 1994;58: DISCUSSION DR NASSER ALTORKI (New York, NY): I enjoyed your paper. Can you comment on how you made the preoperative clinical diagnosis of invasion of the atrium or the aorta? DR GALETTA: As regards the aorta, in most cases, the diagnosis of the suspicion of infiltration is made on the computed tomog raphy (CT) scan images. If involvement of the diameter of the aorta more than 50%, we suspect that it is completely infiltrated; while, if this appearance on CT scan is less than 50%, we suppose that the tumor is attached to the aorta but not infiltrating it. As regards the atrium, usually the method we use to diagnose the atrial infiltration is the CT scan, and in some cases, magnetic resonance imaging. But in our experience, even if there is the suspicion of high probability of atrial infiltration on CT scan images (ie, the CT scan shows a tumor localized proximally to the atrial cuff), we have to verify if it really is infiltrated; because in many cases, above all on the right side, the CT image is more advanced than the real status, whereas in the left side, usually the infiltration on the CT scan is really like it is in the intra operative view. DR ALTORKI: And the second question, did you use bypass, partial or total, for some of the vascular reconstruction? DR GALETTA: In all our series of aortic and left atrial resection, only in few cases did we use bypass, and these are not included in this report. In general, for atrial resection, we perform a direct atrial resection and a double suture on a Satinsky clamp after having performed the Sondergaard technique. But, in every case, the decision to use bypass or not depends on the entity of the infiltration of the atrium. DR AKIF TURNA (Istanbul, Turkey): You stated that you refer the patients with bulky N2 disease to oncology first. How did you decide bulky N2 disease? Did you perform mediastinoscopy on these patients or did you only rely on the positron emission tomography (PET)/CT? I think mediastinoscopy in this type of patient is of utmost importance, because as you concluded, the N2 disease is the most dismal prognostic factor in these patients. And the second question is: how did you decide on resect ability? After the thoracotomy, you had 25% of exploratory thoracotomy. Did you look at only anatomic resectability during the thoracotomy, or did you sample lymph nodes also? DR GALETTA: Thank you for you questions. As regards to the first question, as shown by the previous slides, patients with bulky N2 disease are candidates for a definitive chemotherapy. 18. Pitz CC, Brutel de la Riviere A, van Swieten HA, Westermann CJ, Lammers JW, van den Bosch JM. Results of surgical treatment of T4 non small cell lung cancer. Eur J Cardiothorac Surg 2003;24: Watanabe Y, Shimizu J, Oda M, et al. Results of surgical treatment in patients with stage IIIB non small cell lung cancer. Thorac Cardiovasc Surg 1991;39: Dartevelle PG, Khalife J, Chapelier A, et al. Tracheal sleeve pneumonectomy for bronchogenic carcinoma: report of 55 cases. Ann Thorac Surg 1988;46: Mathisen DJ, Grillo HC. Carinal resection for bronchogenic carcinoma. J Thorac Cardiovasc Surg 1991;102: Vansteenkiste JF, De Leyn PR, Deneffe GJ, et al. Survival and prognostic factors in resected N2 non small cell lung cancer: a study of 140 cases. Ann Thorac Surg 1997;63: DR TURNA: On PET/CT? DR GALETTA: Certainly, the first diagnostic tool is the PET/CT. But, you have to prove the bulky nodal involvement by endo bronchial ultrasound or mediastinoscopy. Your second question was about the resectability. The explorative thoracotomies were due to the intraoperative evidence of bulky N disease and also to the local infiltration of mediastinal organs not clearly evident preoperatively on the CT scan. DR ANTHONY KIM (New Haven, CT): I really enjoyed your presentation. I have also enjoyed the work on this subject that your group has done over the years. My first question is a fol low up to Dr Altorki s question. In the patients on whom you used bypass, was that planned bypass or unplanned bypass? And then the second question is, based on the group s experience, did you come up with or identify an optimal strategy in dealing with the T4 N0 subgroup? And also, did that subgroup behave differently than, say, the T4 N2 group that was down staged after induction? Thank you, and nice work. DR GALETTA: As regards to your first question, in our experience, the bypass was used only in few cases, and in all of these, it was preoperatively planned and performed with cardiac surgeons. Then, as regards the evaluation of subgroups, we did not perform this kind of evaluation, so it is not possible for me to answer to your second question, I m sorry. DR CLARK FULLER (Los Angeles, CA): Bold strokes. I applaud your bravery in this difficult situation. A comment, however. In your high exploration rate, you might consider a thorascopic examination first prior to committing the patient to thoracotomy. The second is a question. In your pneumonectomy patients or even the other patients, did you use any flaps, muscle flaps, in coverage of the stumps? DR GALETTA: I m sorry, could you repeat the first question, please? DR FULLER: It was more of a comment. I was just suggesting thoracoscopy prior to thoracotomy. DR GALETTA: As regards your comment, I would like to underline that thoracoscopy was used only in those cases in which there was a suspicion of pleural invasion for the presence of pleural fluid, or in those cases in which we had a suspicion of

9 Ann Thorac Surg SPAGGIARI ET AL 2013;95: EXTENDED RESECTION FOR LUNG CANCER 1725 a bulky involvement of the tumor of the mediastinal organs. But we think that for this kind of surgery (extended surgery) the surgeon had to open the chest, had to put his hands inside, and verify the real involvement of the mediastinal organs and their resectability. We think that if we limit our decision of resectability only on the base of the CT scan images, we could judge the patient inoperable in the majority of the cases and deny to them the possibility of offering a possible curative surgical option. As regards to your second question, in every case when we perform a pneumonectomy or a sleeve lobectomy or a sleeve pneumonectomy, we usually prepare some vital tissue, like intercostal muscle or mediastinal fat pad flap, and we encircle the anastomosis or we cover the bronchus to protect them.