The Role of Endobronchial Ultrasound-guided Transbronchial Needle Aspiration in the Diagnosis of Recurrent Non-small Cell Lung Cancer after Surgery

ORIGINAL ARTICLE The Role of Endobronchial Ultrasound-guided Transbronchial Needle Aspiration in the Diagnosis of Recurrent Non-small Cell Lung Cancer after Surgery Seo Goo Han 1, Hongseok Yoo 1, Byung Woo Jhun 1,HyeYunPark 1, Gee Young Suh 1, Man Pyo Chung 1, Hojoong Kim 1, O Jung Kwon 1, Joungho Han 2 andsang-wonum 1 Abstract Objective Obtaining an accurate histopathological diagnosis is mandatory for the optimal treatment of patients who are suspected of having recurrent lung cancer. The purpose of this retrospective study was to investigate the usefulness of endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) for the diagnosis of recurrent non-small cell lung cancer (NSCLC) among patients who undergo curative surgical resection. Methods Consecutive patients who underwent convex probe EBUS-TBNA for mediastinal or hilar lymph node and peribronchial lung parenchymal lesions between May 2009 and May 2011 were included. The diagnostic sensitivity, specificity, accuracy, positive predictive value (PPV) and negative predictive value (NPV) were calculated on a per-lesion and per-patient basis. Results Forty-two patients who were suspected of having recurrent NSCLC underwent EBUS-TBNA to assess 53 mediastinal and hilar lymph nodes and seven peribronchial lung parenchymal lesions. Among the 60 lesions, recurrence of malignancy was confirmed in 41 lesions on EBUS-TBNA (36 lymph nodes and five peribronchial lung lesions). On a per-lesion basis, the diagnostic sensitivity, specificity, accuracy, PPV and NPV of EBUS-TBNA for confirming recurrence were 95.3%, 100%, 96.6%, 100% and 88.9%, respectively. On a per-person basis, the diagnostic sensitivity, specificity, accuracy, PPV and NPV were 94.3%, 100%, 95.2%, 100% and 77.8%, respectively. No serious complications related to the procedures were observed. Conclusion Convex probe EBUS-TBNA is a sensitive method for diagnosing recurrent NSCLC in patients with lymph node and peribronchial lung parenchymal lesions. Therefore, EBUS-TBNA should be considered first for the cytopathological diagnosis of recurrent NSCLC. Key words: endobronchial ultrasound-guided transbronchial needle aspiration, mediastinum, non-small cell lung cancer, recurrence, surgery (Intern Med 52: 1875-1881, 2013) () Introduction Obtaining an accurate histopathological diagnosis is mandatory for the optimal treatment of patients who are suspected of having recurrent non-small cell lung cancer (NSCLC). The American College of Chest Physicians guidelines recommend the use of invasive staging with tissue confirmation of suspected metastatic mediastinal lymph nodes (1). Mediastinoscopy is the diagnostic standard for mediastinal staging (2); however, it is associated with some limitations. Most patients receive mediastinoscopy at the time of the initial mediastinal staging of NSCLC (3). Therefore, the use of mediastinoscopy to assess suspected mediastinal recurrence in these patients is usually performed as a second procedure or remediastinoscopy. Performing remedi- Division of Pulmonary and Critical Care Medicine, Department of Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, Korea and Department of Pathology, Samsung Medical Center, Sungkyunkwan University School of Medicine, Korea Received for publication November 26, 2012; Accepted for publication March 4, 2013 Correspondence to Dr. Sang-Won Um, sangwonum@skku.edu 1875

astinoscopy is technically more difficult than performing the first procedure (4) and cannot be used to approach the interlobar and hilar lymph nodes or peribronchial lung lesions (5). Recently, EBUS-TBNA was introduced as a minimally invasive technique for nodal staging. Previous studies have shown that EBUS-TBNA exhibits excellent diagnostic performance (6). Additionally, the use of EBUS-TBNA allows for access to the hilar and interlobar lymph nodes, which are inaccessible with mediastinoscopy (7). However, the role of EBUS-TBNA in diagnosing recurrent lung cancer is not well known. The purpose of this retrospective study was to investigate the usefulness of EBUS-TBNA in the diagnosis of recurrent NSCLC among patients who undergo curative surgical resection. Patients Materials and Methods A retrospective review was performed to evaluate the EBUS-TBNA yield for a diagnosis of locoregional recurrence of NSCLC. From May 2009 to May 2011, 42 consecutive patients who had previously undergone surgery for primary NSCLC and who underwent EBUS-TBNA to confirm the presence of recurrence in the mediastinal or hilar lymph nodes and peribronchial lung parenchymal lesions were included in this study. This retrospective study was conducted with the approval of the Institutional Review Board of Samsung Medical Center, which waived the requirement for informed consent by individual patients given the retrospective nature of the study. Patients with lymph node or peribronchial lung parenchymal lesions suspected of having recurrent NSCLC were enrolled when (1) a new or growing lymph node or peribronchial lesion with a short axis diameter of 10 mm was identified on chest CT or (2) lesions with increased fluorodeoxyglucose (FDG) uptake compared with the surrounding tissue were identified on positron emission tomography (PET)/CT scans, regardless of size. All patients were previously diagnosed with NSCLC based on histology and initially treated with surgical treatment or combined modalities of surgery, chemotherapy and radiation therapy. The patients underwent potentially curative resection of stage I-IIIA NSCLC between 2002 and 2010. The pathologic staging of the patients followed the revised tumor-node-metastasis system (8). EBUS-TBNA procedures EBUS-TBNA was performed using a convex probe EBUS-TBNA (CP-EBUS; BF-UC260F-OL8, Olympus Medical Systems, Tokyo, Japan) with the patient receiving local anesthesia and conscious sedation. During the procedure, all accessible mediastinal and hilar lymph nodes and peribronchial lesions were imaged systematically, and target nodal stations considered to be necessary were selected. Each target nodal station was punctured at least twice, and one or more tissue core specimens were obtained with a dedicated 22-gauge needle (NA-201SX-4022, Olympus Medical Systems, Tokyo, Japan). If no tissue core specimens could be obtained from the first two aspirations, then three or more lymph node aspirations were performed until enough tissue was obtained (9). The aspirate was smeared onto glass slides, air-dried, fixed immediately with 95% alcohol and stained with hematoxylin and eosin (H&E). The histological cores were fixed with 10% neutral buffered formalin and stained with H&E. An epidermal growth factor receptor gene analysis of the EBUS-TBNA specimen was performed if adenocarcinoma was revealed in the obtained specimen. All EBUS-TBNA procedures were performed by two bronchoscopists (SW Um and HY Park). Rapid on-site cytological examinations were not available. Definitions The cytopathological specimens were categorized as malignant (exhibiting the presence of malignant cells), benign (exhibiting lymphocyte or lymphoid tissue without malignant cells) or inadequate (containing no or scanty cellular components, blood only, mucus or benign bronchial cells only). The malignant and benign samples diagnosed on EBUS-TBNA were defined as adequate samples. Tumorpositive findings in the EBUS-TBNA samples were not surgically validated. The diagnostic standard for a tumornegative result was pathological or microbiological confirmation of a specific benign disease, surgical confirmation of lesions showing no malignant cells or a lack of evidence of lymph node progression for >6 months after EBUS-TBNA. The lesions were considered to be malignant when malignancy was revealed on subsequently performed mediastinoscopy or an increase in the size of the lesion was noted on a follow-up imaging study after EBUS-TBNA. If no follow-up imaging study was available over six months, then the targeted lesion was regarded as an undetermined lesion and excluded from the analysis. Statistical analysis The data are presented as numbers (%) or medians (range). The diagnostic sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and accuracy of EBUS-TBNA were calculated according to standard definitions. The data were analyzed using the PASW statistics 17 software program (SPSS Inc., Chicago, IL). Results Baseline characteristics of the patients EBUS-TBNA was performed in 42 patients (32 men and 10 women) with suspected local recurrence of NSCLC between May 2009 and May 2011. The baseline clinical characteristics of the patients are summarized in Table 1 and Figure. The median age was 62.5 years (range, 44-79 years). The histology of the primary tumor was adenocarcinoma in 1876

Table 1. Clinical Characteristics of the 42 Study Patients Characteristics Median (range) or number (%) Gender Male 32 (76.2) Female 10 (23.8) Age at EBUS-TBNA, years 62.5 (44-79) Histology of primary tumor Adenocarcinoma 24 (57.1) Squamous cell carcinoma 11 (26.2) Large cell carcinoma 4 (9.5) Adenosquamous carcinoma 3 (7.1) Pathological stage at the initial diagnosis IA 10 (23.8) IB 7 (16.7) IIA 7 (16.7) IIB 1 (2.4) IIIA 17 (40.5) Initial treatment Surgery 42 (100) Neoadjuvant concurrent chemoradiation followed by surgery 10 (23.8) Surgery followed by adjuvant therapy 23 (54.8) Surgical modalities Wedge resection 3 (7.1) Lobectomy 35 (83.3) Pneumonectomy 4 (9.5) Mediastinal lymph node dissection 35 (83.3) Mediastinoscopy at initial diagnosis 23 (54.8) EBUS-TBNA at initial diagnosis 3 (7.1) Mediastinoscopy or mediastinal lymph node dissection 35 (83.3) EBUS-TBNA: endobronchial ultrasound-guided transbronchial needle aspiration Figure. Diagnostic algorithm of endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) in the 42 study patients who were suspected of having recurrent lung cancer after surgery. 24 patients, squamous cell carcinoma in 11 patients, large cell carcinoma in four patients and adenosquamous carcinoma in three patients. Mediastinoscopy was performed in 23 patients at the time of the initial lung cancer diagnosis, and mediastinal lymph node dissection was performed in 35 patients. 1877

Table 2. Characteristics of the 53 Lymph Nodes and Seven Peribronchial Lung Lesions Examined by EBUS-TBNA Table 3. Histological and Clinical Data on 53 Lymph Nodes and Seven Peribronchial Lung Lesions Characteristics Median (range) or number (%) Examined lesions per each patient 1 (1-4) Needle passes per lesion 2 (1-5) Lymph nodes (n=53) Highest mediastinal (#1) 4 (7.5) Upper paratracheal (#2) 7 (13.2) Retrotracheal (#3P) 1 (1.9) Lower paratracheal (#4) 23 (43.4) Subcarinal (#7) 11 (20.8) Hilar (#10) 2 (3.8) Interlobar (#11) 5 (9.4) Peribronchial lesions (n=7) Right upper lobe 1 (14.3) Right middle lobe 2 (28.6) Right lower lobe 2 (28.6) Left upper lobe 1 (14.3) Left main bronchial stump 1 (14.3) Size of lymph node, mm (n=53) Short axis 13 (5-22) Long axis 18 (8-44) Size of peribronchial lesion, mm (n=7) Short axis 15 (8-22) Long axis 22 (8-31) EBUS-TBNA: endobronchial ultrasound-guided transbronchial needle aspiration Characteristics of the EBUS-TBNA procedures The characteristics of the EBUS-TBNA procedures are summarized in Table 2. EBUS-TBNA was performed to assess 53 lymph nodes and seven peribronchial lung lesions. The median number of examined lymph nodes was one node per patient, and the median number of needle passes was two passes per lesion. The median size of the lymph nodes was 13 mm and that of the peribronchial lesions was 15 mm. A pneumothorax occurred in one patient after EBUS-TBNA of the right highest mediastinal lymph node, and tube thoracostomy was required to treat the pneumothorax. Except for the one patient with a pneumothorax, EBUS- TBNA carried no particular risks, resulted in no complications and caused little discomfort. Histological diagnoses The EBUS-TBNA cytology and histology examinations revealed 41 of the 60 lesions as being malignant (Figure and Table 3). The EBUS-TBNA cytology and histology of the specimens confirmed recurrent lung cancer in 36 lymph nodes and five peribronchial lesions (Table 3). Of the 19 EBUS-negative lesions, mediastinoscopy was sequentially performed to histologically confirm five mediastinal lymph nodes. The histological examinations revealed one additional malignant lymph node and four benign lymph nodes. Mycobacterium tuberculosis was identified in the culture of one aspiration cytology sample (one lymph node), and benignity was confirmed microbiologically. We followed the remaining 12 lesions for up to six months using CT scans and detected a size increase in one lymph node. Eleven lesions Median (range) or number (%) EBUS-TBNA cytology Adenocarcinoma 18 (30) Squamous cell carcinoma 4 (6.7) Large cell carcinoma 0 (0) NSCLC, unspecified 12 (20) Atypical cells 4 (6.7) No evidence of malignancy 19 (31.7) Inadequate sample 3 (5) EBUS-TBNA histology Adenocarcinoma 26 (43.3) Squamous cell carcinoma 9 (15) Large cell carcinoma 2 (3.3) Adenosquamous carcinoma 0 (0) NSCLC, unspecified 2 (3.3) No evidence of malignancy 21 (35) Adequate samples from EBUS-TBNA 57 (95) Malignancy confirmed by the cytology and histology results from the samples obtained by 41 (68.3) EBUS- TBNA Final result of benign EBUS-TBNA findings 19 (31.7) Malignancy confirmed by mediastinoscopy 1 (1.7) Benignity confirmed by mediastinoscopy 4 (6.7) Culture proven tuberculosis 1 (1.7) Stationary lesions during clinical follow-up 11 (18.3) Enlarged lesion during clinical follow-up 1 (1.7) No available follow-up data > 6 months 1 (1.7) Duration of clinical follow-up (N=11), days 476 (196-583) EBUS-TBNA: endobronchial ultrasound-guided transbronchial needle aspiration showed no size increases at the follow-up imaging studies. One patient had two enlarged lymph nodes; one lesion was confirmed to be malignant on EBUS-TBNA, while the other was not confirmed to be malignant on EBUS-TBNA. The patient received radiation therapy based on these results, and the two lesions were included in the radiation field. There were no increases in the size of the EBUS-negative lymph node during follow-up. However, we classified the EBUSnegative lymph node as an undetermined lesion since we were unable to assess the false-negative result for the lymph node on the follow-up CT scan obtained after radiation therapy. Thirty-three of 42 patients were diagnosed with recurrent NSCLC based on the EBUS aspiration cytology and histology examinations. Among the remaining nine patients with EBUS-negative results, two were diagnosed with recurrent NSCLC based on mediastinoscopy and follow-up imaging studies and seven showed no evidence of tumor recurrence. Therefore, recurrent NSCLC was confirmed in 35 patients. No patients were diagnosed with second primary lung cancer. Diagnostic yield of EBUS-TBNA The diagnostic sensitivity, specificity, accuracy, PPV and NPV of EBUS-TBNA for the confirmation of recurrence were 95.3%, 100%, 96.6%, 100% and 88.9%, respectively, 1878

Table 4. Diagnostic Yield of EBUS-TBNA for Recurrent Lung Cancer On a per-lesion basis On a per-person basis Sensitivity 95.3% 94.3% Specificity 100% 100% Positive predictive value 100% 100% Negative predictive value 88.9% 77.8% Diagnostic accuracy 96.6% 95.2% EBUS-TBNA: endobronchial ultrasound-guided transbronchial needle aspiration on a per-lesion basis (Table 4). The diagnostic sensitivity, specificity, accuracy, PPV and NPV of EBUS-TBNA for the confirmation of recurrence were 94.3%, 100%, 95.2%, 100% and 77.8%, respectively, on a per-person basis (Table 4). Role of EBUS-TBNA in patient management Treatment plans were determined in 40 patients (95.2%) based on the EBUS-TBNA results (Table 5). Among 33 patients whose recurrences were confirmed on EBUS-TBNA, 31 patients began to receive new treatments: concurrent chemoradiotherapy in 15 patients, radiation therapy in seven patients, palliative cytotoxic chemotherapy in seven patients, palliative epidermal growth factor receptor - tyrosine kinase inhibitor therapy in one patient and surgical treatment in one patient. Two patients received conservative treatment due to their age and performance status. One patient who was diagnosed with tuberculous lymphadenopathy received antituberculous medication. We performed regular follow-up for six patients with no recurrence. Discussion Recurrence remains a major obstacle to achieving a cure in patients with NSCLC. The reported recurrence rate after complete surgical resection is 30-75% (10-12), depending on the final pathological stage. The incidence of locoregional recurrence following curative resection of NSCLC is 4.9-22.7% at the time of NSCLC recurrence diagnosis (13-15). Mediastinal involvement in patients with recurrent NSCLC is common, and accurately detecting mediastinal involvement is crucial for the curative treatment of recurrent NSCLC. In this study, 42 patients were suspected of having recurrent lung cancer based on chest CT or PET/CT scan findings. Subsequent studies, including EBUS-TBNA and mediastinoscopy, revealed that 35 patients (83.3%) had recurrent lung cancer and seven (16.7%) did not. PET/CT imaging plays a role in distinguishing persistent or recurrent tumors from posttreatment scarring or fibrosis. It is more sensitive than chest CT and conventional radiography for detecting recurrent tumors (sensitivity, 97-100%) (16, 17). However, it has a specificity of 62-100%, sometimes yielding falsepositive results due to active inflammation, particularly in the acute postoperative or postradiation stage. Therefore, it is important to confirm recurrence histopathologically in order to select the most appropriate treatment. In this study, the diagnostic sensitivity of EBUS-TBNA for diagnosing recurrent NSCLC was 95.3% and 94.3% on a per-lesion and per-person basis, respectively, among the patients who were suspected of having recurrent lung cancer following curative resection. These results are comparable to those of EBUS-TBNA for the initial mediastinal staging of NSCLC. In 14 studies with a pooled cohort of 1,658 patients from eight different countries, the pooled sensitivity of EBUS-TBNA was 92% (18). The gold standard for mediastinal evaluation is mediastinoscopy (2); however, this procedure is associated with some limitations for diagnosing locoregional recurrence of NSCLC. First, it requires general anesthesia, which carries a risk of potentially significant complications, a morbidity of 0.6-3%, a risk of hemorrhage of 0.1-0.6% and a mortality of 0-0.3% (19-23). Second, performing remediastinoscopy is technically difficult in patients who have previously undergone mediastinoscopy, as mediastinal staging and the sensitivity and accuracy of remediastinoscopy are lower after the first procedure (24). Third, obtaining tissue specimens from the hilar and interlobar lymph nodes is not appropriate during mediastinoscopy (25). There were no significant side effects of EBUS-TBNA, except in one patient who developed a pneumothorax in the right hemithorax following EBUS-TBNA of the righthighest mediastinal lymph node. This lymph node was very close to the right visceral pleura, and we speculate that the EBUS- TBNA needle may have penetrated the lymph node and injured the right visceral pleura near the lymph node. In this study, 23 patients (54.8%) underwent first mediastinoscopy at the time of the initial diagnosis of NSCLC, and 10 patients (23.8%) underwent neoadjuvant concurrent chemoradiotherapy followed by surgery as initial treatment for NSCLC. Remediastinoscopy is crucial for tissue confirmation to diagnose mediastinal recurrence of NSCLC, while EBUS-TBNA enables the cytopathological diagnosis of mediastinal recurrence without the need for remediastinoscopy. No patients underwent remediastinoscopy in this study. Two patients underwent mediastinoscopy following EBUS-TBNA as a first mediastinoscopy. The NPV of this study was 88.9% on a per-lesion basis. Although all the negative EBUS-TBNA results were not verified surgically, we confirmed the absence of lymph node progression for >6 months after EBUS-TBNA in these patients. The NPV can be influenced by the prevalence of disease in the population tested. Because we performed diagnostic tests in a high-prevalence setting of recurrent NSCLC, the increased prevalence resulted in a decreased NPV. However, the NPV was still comparable with previously reported NPVs for remediastinoscopy of 73.3-84.5% (26-28). There were two false-negative lesions (4L and 1R) in this study. The short axis diameters of these lymph nodes were greater than 10 mm. The location of the lymph node was found to be a risk factor for false-negative results in previous studies (29, 30). In this study, the 4 L lymph node was 1879

Table 5. The Role of EBUS-TBNA in Making Treatment Decisions Number (%) Treatment decisions influenced by EBUS-TBNA 40 (95.2) Diagnosis of recurrent NSCLC 31 (73.8) Salvage CCRT 15 (35.7) Chemotherapy 7 (16.7) Radiation therapy 7 (16.7) EGFR-TKI 1 (2.4) Surgical treatment 1 (2.4) Conservative management 2 (4.8) Antituberculosis chemotherapy 1 (2.4) No recurrence and observation 6 (14.3) Treatment decisions uninfluenced by EBUS-TBNA 2 (4.8) Diagnosed by subsequent mediastinoscopy* 1 (2.4) Diagnosed by the other modality 1 (2.4) EBUS-TBNA: endobronchial ultrasound-guided transbronchial needle aspiration, CCRT: concurrent chemoradiotherapy, EGFR-TKI: epidermal growth factor receptor-tyrosine kinase inhibitor. *Mediastinoscopy revealed malignancy: and EBUS-TBNA result was a false negative. The patient received radiation therapy. Bronchoscopic mucosal biopsy was performed for the endobronchial mucosal lesion at the time of EBUS-TBNA. The bronchoscopic biopsy revealed a malignancy: and the patient received conservative management. located very close to the subaortic area, deep within the trachea. Therefore, visualizing and sampling the 4 L lymph node using EBUS-TBNA was relatively more difficult. Furthermore, the 1R lymph node was located at the posterior lateral aspect of the trachea. In order to visualize the 1R lymph node, the scope needed to be bent at an acute angle; therefore, it was technically difficult to target this lesion. The role of EBUS-TBNA in patients with previously treated lung cancer was recently reported. In two studies, the authors included a heterogeneous population of patients who developed recurrent disease after undergoing surgery or progressive disease after receiving chemotherapy (31, 32). Anraku et al. reported that the diagnostic sensitivity and accuracy of EBUS-TBNA were 93.1% and 95.1%, respectively, among 44 patients with previously treated lung cancer (31). Their results were similar to those of our study. The strength of our study is that we evaluated the role of EBUS-TBNA in a homogeneous population of patients who were suspected of having recurrent NSCLC following curative resection. In this study, EBUS-TBNA was performed for seven peribronchial lung lesions. EBUS-TBNA was performed on these lesions without any complications, and no falsenegative results were observed for these lesions. The role of convex probe EBUS-TBNA in diagnosing peribronchial lung parenchymal lesions has also been reported in previous studies, with a diagnostic sensitivity of more than 80% (33, 34). In this study, we obtained both cytological specimens and histological cores in all patients. There were some discrepancies in diagnosis between the cytological and histological specimens obtained with EBUS-TBNA. Among the cytological specimens, the lesions were classified as NSCLC, unspecified, exhibiting atypical cells or inadequate samples in 12, four and three cases, respectively. However, among the histological specimens, the lesions were classified as NSCLC, unspecified, exhibiting atypical cells or inadequate samples in two, zero and zero cases, respectively. Therefore, we believe that obtaining enough histological cores is very important for the accurate diagnosis of lung cancer subtypes. EBUS-TBNA also has several limitations in the diagnosis of recurrent NSCLC. When lesions are located in the aortopulmonary window or paraesophageal area, they are not accessible with EBUS-TBNA, although they might be accessible with EUS-FNA. In addition, EBUS-TBNA also has a limitation in accessing the lesion in cases in which the distance between the lesion and the bronchial wall is more than 1.5-2 cm. The limitations of the present study include its retrospective nature and potential selection bias. Therefore, the data in this study should be interpreted conservatively. In conclusion, EBUS-TBNA is a sensitive method for diagnosing recurrent NSCLC in lymph node and peribronchial lung parenchymal lesions. Therefore, EBUS-TBNA should be considered first for the cytopathological diagnosis of recurrent NSCLC. The authors state that they have no Conflict of Interest (COI). Acknowledgement This study was supported by the Samsung Biomedical Research Institute (CA92092). References 1. Silvestri GA, Gould MK, Margolis ML, et al. Noninvasive staging of non-small cell lung cancer: ACCP evidenced-based clinical practice guidelines (2nd edition). Chest 132: 178S-201S, 2007. 2. Detterbeck FC, Jantz MA, Wallace M, Vansteenkiste J, Silvestri GA. Invasive mediastinal staging of lung cancer: ACCP evidencebased clinical practice guidelines (2nd edition). Chest 132: 202S- 220S, 2007. 3. Pretreatment evaluation of non-small-cell lung cancer. Am J Respir Crit Care Med 156: 320-332, 1997. 4. Pauwels M, Van Schil P, De Backer W, Van den Brande F, 1880

Eyskens E. Repeat mediastinoscopy in the staging of lung cancer. Eur J Cardiothorac Surg 14: 271-273, 1998. 5. Gomez M, Silvestri GA. Endobronchial ultrasound for the diagnosis and staging of lung cancer. Proc Am Thorac Soc 6: 180-186, 2009. 6. Medford AR, Bennett JA, Free CM, Agrawal S. Mediastinal staging procedures in lung cancer: EBUS, TBNA and mediastinoscopy. Curr Opin Pulm Med 15: 334-342, 2009. 7. Ernst A, Eberhardt R, Krasnik M, Herth FJ. Efficacy of endobronchial ultrasound-guided transbronchial needle aspiration of hilar lymph nodes for diagnosing and staging cancer. J Thorac Oncol 4: 947-950, 2009. 8. Goldstraw P, Crowley J, Chansky K, et al. 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 2: 706-714, 2007. 9. Lee HS, Lee GK, Kim MS, et al. Real-time endobronchial ultrasound-guided transbronchial needle aspiration in mediastinal staging of non-small cell lung cancer: how many aspirations per target lymph node station? Chest 134: 368-374, 2008. 10. al-kattan K, Sepsas E, Fountain SW, Townsend ER. Disease recurrence after resection for stage I lung cancer. Eur J Cardiothorac Surg 12: 380-384, 1997. 11. Martin J, Ginsberg RJ, Venkatraman ES, et al. Long-term results of combined-modality therapy in resectable non-small-cell lung cancer. J Clin Oncol 20: 1989-1995, 2002. 12. Martini N, Bains MS, Burt ME, et al. Incidence of local recurrence and second primary tumors in resected stage I lung cancer. J Thorac Cardiovasc Surg 109: 120-129, 1995. 13. Ginsberg RJ, Rubinstein LV. Randomized trial of lobectomy versus limited resection for T1 N0 non-small cell lung cancer. Lung Cancer Study Group. Ann Thorac Surg 60: 615-622; discussion 622-623, 1995. 14. Harpole DH Jr, Herndon JE 2nd, Young WG Jr, Wolfe WG, Sabiston DC Jr. Stage I nonsmall cell lung cancer. A multivariate analysis of treatment methods and patterns of recurrence. Cancer 76: 787-796, 1995. 15. Warren WH, Faber LP. Segmentectomy versus lobectomy in patients with stage I pulmonary carcinoma. Five-year survival and patterns of intrathoracic recurrence. J Thorac Cardiovasc Surg 107: 1087-1093; discussion 1093-1094, 1994. 16. Erasmus JJ, McAdams HP, Patz EF Jr. Non-small cell lung cancer: FDG-PET imaging. J Thorac Imaging 14: 247-256, 1999. 17. Hicks RJ, Kalff V, MacManus MP, et al. The utility of (18)F-FDG PET for suspected recurrent non-small cell lung cancer after potentially curative therapy: impact on management and prognostic stratification. J Nucl Med 42: 1605-1613, 2001. 18. Chandra S, Nehra M, Agarwal D, Mohan A. Diagnostic accuracy of endobronchial ultrasound guided transbronchial needle aspiration in mediastinal lymphadenopathy: A systematic review and meta-analysis. Respir Care 57 (3): 384-391, 2012. 19. Defranchi SA, Edell ES, Daniels CE, et al. Mediastinoscopy in patients with lung cancer and negative endobronchial ultrasound guided needle aspiration. Ann Thorac Surg 90: 1753-1757, 2010. 20. Hammoud ZT, Anderson RC, Meyers BF, et al. The current role of mediastinoscopy in the evaluation of thoracic disease. J Thorac Cardiovasc Surg 118: 894-899, 1999. 21. Lemaire A, Nikolic I, Petersen T, et al. Nine-year single center experience with cervical mediastinoscopy: Complications and false negative rate. Ann Thorac Surg 82: 1185-1190, 2006. 22. Park BJ, Flores R, Downey RJ, Bains MS, Rusch VW. Management of major hemorrhage during mediastinoscopy. J Thorac Cardiovasc Surg 126: 726-731, 2003. 23. Urschel JD. Conservative management (packing) of hemorrhage complicating mediastinoscopy. Ann Thorac Cardiovasc Surg 6: 9-12, 2000. 24. Van Schil P, De Waele M, Hendriks J, Lauwers P. Remediastinoscopy. J Thorac Oncol 2: 365-366, 2007. 25. Herth FJ, Eberhardt R, Krasnik M, Ernst A. Endobronchial ultrasound-guided transbronchial needle aspiration of lymph nodes in the radiologically and positron emission tomography-normal mediastinum in patients with lung cancer. Chest 133: 887-891, 2008. 26. De Waele M, Serra-Mitjans M, Hendriks J, et al. Accuracy and survival of repeat mediastinoscopy after induction therapy for nonsmall cell lung cancer in a combined series of 104 patients. Eur J Cardiothorac Surg 33: 824-828, 2008. 27. Marra A, Hillejan L, Fechner S, Stamatis G. Remediastinoscopy in restaging of lung cancer after induction therapy. J Thorac Cardiovasc Surg 135: 843-849, 2008. 28. Van Schil P, van Schoot J, Poniewierski J, et al. Remediastinoscopy after neoadjuvant therapy for non-small cell lung cancer. Lung Cancer 37: 281-285, 2002. 29. Jhun BW, Park HY, Jeon K, et al. Nodal stations and diagnostic performances of endobronchial ultrasound-guided transbronchial needle aspiration in patients with non-small cell lung cancer. J Korean Med Sci 27: 46-51, 2012. 30. Szlubowski A, Kuzdzal J, Kolodziej M, et al. Endobronchial ultrasound-guided needle aspiration in the non-small cell lung cancer staging. Eur J Cardiothorac Surg 35: 332-335; discussion 335-336, 2009. 31. Anraku M, Pierre AF, Nakajima T, et al. Endobronchial ultrasound-guided transbronchial needle aspiration in the management of previously treated lung cancer. Ann Thorac Surg 92: 251-255; discussion 255, 2011. 32. Chen F, Miyahara R, Sato T, et al. Usefulness of endobronchial ultrasound in patients with previously treated thoracic malignancy. Interact Cardiovasc Thorac Surg 14: 34-37, 2012. 33. Nakajima T, Yasufuku K, Fujiwara T, et al. Endobronchial ultrasound-guided transbronchial needle aspiration for the diagnosis of intrapulmonary lesions. J Thorac Oncol 3: 985-988, 2008. 34. Tournoy KG, Rintoul RC, van Meerbeeck JP, et al. EBUS-TBNA for the diagnosis of central parenchymal lung lesions not visible at routine bronchoscopy. Lung Cancer 63: 45-49, 2009. 2013 The Japanese Society of Internal Medicine http://www.naika.or.jp/imonline/index.html 1881