When a solitary pulmonary lesion (SPL) is found in

GENERAL THORACIC Resection of Solitary Pulmonary Lesion Is Beneficial to Patients With a History of Malignancy Miki Sakamoto, MD, Tomohiro Murakawa, MD, Kentaro Kitano, MD, Tomonori Murayama, MD, Takehiro Tsuchiya, MD, and Jun Nakajima, MD Department of Cardiothoracic Surgery, University of Tokyo, Bunkyo-ku, Tokyo, Japan Background. Solitary pulmonary lesion poses a diagnostic challenge, especially in patients with a history of malignancy. The purpose of this study was to evaluate the characteristics of solitary pulmonary lesions and the outcome of surgical resection. Methods. We retrospectively analyzed 243 patients with a history of cancer who underwent surgery for new-found solitary pulmonary lesion between January 1998 and December 2007. Results. The diagnosis was primary lung cancer in 92 patients, metastasis in 133, and benign lesions in 18. The 5-year survival rate was 67.9% in all patients, 74.6% in those with primary lung cancer, 62.8% in those with metastasis, and 79.9% in those with benign lesions (p 0.56). In metastasis patients, history of extrapulmonary recurrence and larger diameter lesion were risk factors for recurrence by multivariate analysis. History of cancers other than colorectal and bone and soft tissue sarcoma and shorter disease-free interval were indicators of poor prognosis. Pathologic stage was the only indicator of prognosis for primary lung cancer, and none of the factors concerning antecedent cancer influenced prognosis. Conclusions. Surgical resection of solitary pulmonary lesion is essential in patients with a history of cancer because substantial numbers of benign lesions are included. In the case of malignancy, metastasectomy had a life-prolonging effect for selected patients, and prognosis of primary lung cancer was no worse than for the general population if treated appropriately. It is important not to hesitate to take a surgical approach for a diagnosis and to treat with standard therapy for primary lung cancer. (Ann Thorac Surg 2010;90:1766 72) 2010 by The Society of Thoracic Surgeons When a solitary pulmonary lesion (SPL) is found in patients with a history of malignant disease, it poses a diagnostic problem. Metastasis from antecedent cancer is most likely, but primary lung cancer and even benign lesions are possible. As treatment options differ according to the etiology of SPL, therapeutic strategies should not be determined only by clinical and radiologic findings. Pathologic diagnosis of SPL by surgical resection is important in making therapeutic decision and in some types of cancers, resection of pulmonary metastasis itself has life-prolonging effects [1, 2]. To assess the characteristics of SPLs and the therapeutic impact of surgical resection, we analyzed the experience of our institute over the past 10 years. Patients and Methods We retrospectively reviewed the records of 243 patients with a history of cancer who underwent pulmonary resection of SPL at our institute between January 1998 and December 2007. Patients with a history of previous pulmonary metastasectomy were not included. When we encountered SPLs, we routinely performed computed tomography (CT) with 5- to 10-mm slice thickness, and high-resolution CT scan was added around the lesion Accepted for publication July 14, 2010. Address correspondence to Dr Murakawa, 7-3-1 Hongo, Bunkyo-ku, Tokyo, 113-8655, Japan; e-mail: murakawa-tky@umin.ac.jp. with 2- to 2.5-mm slice thickness. In most cases, follow-up CT was performed in 3 months, and patients whose lesion diameter became smaller were excluded from the indication for surgery. When infectious disease was suspected, bronchoscopic examination was performed to exclude mycobacteriosis. The evaluation of lymph nodes was made by enhanced CT and positron emission tomography (PET), and patients with lymphadenopathy were excluded from the indication. As for operative procedure, we performed lobectomy with node dissection for primary lung cancer except for pure bronchioloalveolar carcinoma, if the patient s general condition permitted. Metastasis was resected by wedge resection when it was located in the periphery of the lung, and by segmentectomy or lobectomy when it was deep in the pulmonary parenchyma. We did not perform hilar or mediastinal lymphadenectomy at metastasectomy [3]. We recorded patient age, sex, primary sites, smoking history, physiologic state (history of coronary artery disease and diabetes mellitus, preoperative forced expiratory volume in 1 second), SPL diameter, preoperative CT diagnosis by radiologists, maximum standard uptake value of PET, history of extrapulmonary recurrence, and pathologic diagnosis of SPL. Disease-free interval (DFI) before resection was defined as the time between the day of SPL resection and the day of previous treatment, surgery of primary cancer, or local treatment of extrapulmonary recurrence. Disease-free survival (DFS) was the 2010 by The Society of Thoracic Surgeons 0003-4975/$36.00 Published by Elsevier Inc doi:10.1016/j.athoracsur.2010.07.054

Ann Thorac Surg SAKAMOTO ET AL 2010;90:1766 72 SOLITARY PULMONARY LESION time between the date of SPL resection and the date when recurrence was detected or the patient died. Before the study, the Research Review Board at our institution examined and approved our research protocol in accordance with the Declaration of Helsinki. All patients provided written informed consent before surgery. Differences between groups were tested by analysis of variance and Pearson s 2 test. Survival rates were calculated using the Kaplan-Meier method. Differences in survival curves were assessed by the log rank test. Multivariate analysis was performed according to the Cox proportional hazards model to estimate relative risk with 95% confidence interval (CI). A p value less than 0.05 was considered significant. Statistical analysis was performed using JMP 6 software (SAS Institute, Cary, NC). Results 1767 In the 243 patients, the diagnosis was primary lung cancer (n 92, 37.9%), metastasis (n 133, 54.7%), or benign lesions (n 18, 7.4%). In all, 162 patients were men and 81 were women. The median age of the cohort was 66 years (range, 9 to 86), but was significantly older for patients diagnosed with primary lung cancer than for those with metastasis and benign lesions. Respiratory function was worse in patients with primary lung cancer than in patients with metastasis and benign lesions (Table 1). The primary tumor sites of previous cancers included a wide range of organs (Table 1). A CT scan was performed preoperatively in all patients, and median diameter of SPL on CT was 1.5 cm (range, 0.2 to 9.5). The diameter was larger in primary lung cancer lesions than in metastasis and benign le- GENERAL THORACIC Table 1. Backgrounds of Patients and Characteristics of Solitary Pulmonary Lesion (SPL) Primary Lung Cancer Metastasis Benign Lesions p Value Age, median (range) 71 (40 85) 63 (9 82) 64 (33 75) 0.0001 Sex 0.11 Male 68 81 13 Female 24 52 5 Smoking history 0.0001 Smoker, current or former 73 61 11 Never smoker 19 72 7 Physiologic status Diabetes mellitus 19 20 2 0.43 Coronary artery disease 6 5 0 0.39 FEV 1, L, median (range) 2.26 (0.9 3.77) 2.33 (0.73 4.69) 2.36 (1.35 3.81) 0.03 Colorectal 14 53 6 Gastric 17 2 2 Hepatocellular carcinoma 7 10 2 Bone and soft tissue 1 17 1 Head and neck SCC 5 8 Primary lung cancer 7 3 Breast 2 5 1 Kidney 3 4 1 Uterine 4 2 Pancreas 5 Esophagus 3 1 Thyroid 1 3 Others 6 7 2 Multiple origins 22 13 3 SPL diameter, cm, median (range) 2.4 (0.4 12) 1.5 (0.2 7.5) 0.9 (0.4 3.2) 0.0001 MaxSUV PET, median (range) 7.8 (0 36.7) 3.9 (0 24.1) 0 (0 9.9) 0.26 History of extrapulmonary recurrence 8 56 7 0.0001 Timing of SPL detection 0.001 Synchronous 17 5 3 Metachronous 75 128 15 5 years from previous cancer 36 29 4 Follow-up, months, median (range) 36 (0 129) 37 (0 132) 39 (8 117) 0.68 FEV 1 forced expiratory volume in 1 second; MaxSUV maximum standard uptake value; PET positron emission tomography; SCC squamous cell carcinoma.

GENERAL THORACIC 1768 SAKAMOTO ET AL Ann Thorac Surg SOLITARY PULMONARY LESION 2010;90:1766 72 Table 2. Diameter of Solitary Pulmonary Lesion Diameter (cm) Primary Lung Cancer Metastasis Benign Lesions 0.9 or less, n 3 39 10 1.0 to 1.9, n 36 60 4 2.0 to 2.9, n 27 18 2 3.0 or more, n 26 16 2 sions, and about 60% of the lesions larger than 3 cm were primary lung cancer (Table 2). That would relate to the percentage of patients who had an interval of 5 years or more from previous cancer: 39.1%, 21.8%, and 22.2% for primary lung cancer, metastasis, and benign lesions, respectively (Table 1). The CT diagnosis was made by radiologists based on morphology of the lesions on CT, and the sensitivity and specificity were, respectively, 93.5% and 88.2% for primary lung cancer and 90.2% and 82.0% for metastasis. Fifty-six patients (36 with primary lung cancer, 17 with metastasis, and 3 with benign lesions) underwent PET scan, and the sensitivity and specificity for malignancy were 73.6% and 66.7%, respectively, by the threshold of maximum standard uptake value of 2. Bronchoscopic examination was performed in 61 patients, and led to the diagnosis of primary lung cancer in 8, metastasis in 3, and malignant disease of unknown origin in 23. Percutaneous needle biopsy was performed in 7 patients, 5 of whom received a diagnosis of primary lung cancer. The rest also had a diagnosis of malignancy although the origin was unclear. Sputum cytology revealed malignant cells in a patient with primary lung cancer. There were no deaths within 30 days of surgery. Complications occurred in 29 (arrhythmia in 5 patients, pneumonia in 3, persistent air leak in 3, and chylothorax, wound infection, perineal nerve palsy, pulmonary embolism, and bronchopulmonary fistula in 1 patient each with primary lung cancer; and persistent air leak in 3 patients, pneumonia in 2, and cerebral infarction, postoperative bleeding, atelectasis, duodenal ulcer, chylothorax, ileus, urinary tract infection, and empyema in 1 patient each with metastasis). The morbidity was 17.4%, 9.8%, and 0% in patients with primary lung cancer, metastasis, and benign lesions, respectively. The death of 1 patient who had bronchopulmonary fistula who underwent lobectomy for primary lung cancer was related to surgery. Benign lesions included epithelioid granuloma (n 4), fibrosis (n 4), hamartoma (n 3), organizing pneumonia (n 2), and atypical adenomatous hyperplasia, cryptococcal granuloma, lymphoid hyperplasia, amyloidoma, and ossification nodule (n 1 each). In patients with metastasis, the origin was most commonly colorectal cancer (n 59), bone and soft tissue sarcoma (n 17), hepatocellular carcinoma (n 11), and head and neck squamous cell carcinoma (n 10). Other less frequent sites of origin were breast (n 6), pancreas (n 6), kidney (n 5), lung (n 4), thyroid gland (n 3), stomach, uterine, parotid gland, urinary tract, biliary tract (n 2 each), and anal canal, testis, and esophagus (n 1 each). Of 133 patients, 56 (42.1%) had a history of extrapulmonary recurrence. After resection of SPL, 46 patients (34.6%) underwent adjuvant therapy. Eightynine patients (66.9%) experienced recurrence after SPL resection, 44 of whom achieved local control by repeated surgery. Primary lung cancer was classified as stage I (n 65), stage II (n 17), and stage III or IV (n 10). Histology confirmed the presence of adenocarcinoma (n 64), squamous cell carcinoma (n 22), large cell carcinoma (n 2), small cell carcinoma (n 2), adenosquamous cell carcinoma (n 1), and carcinoid (n 1). Adjuvant chemotherapy was performed only in 7 patients, with oral tegafur/uracil or a platinum-based combination regimen. The median follow-up period for all patients was 37 months (range, 0 to 132). The 5-year survival rate was 67.9% for all patients; and 74.6%, 62.8%, and 79.9% for patients with primary lung cancer, metastasis, and benign lesions, respectively (Fig 1). According to univariate analysis, the only variable that influenced overall survival was diameter of SPL (Table 3). Multivariate analysis, including histology of SPL, primary tumor site, SPL diameter, and timing of SPL detection, revealed that diameter and timing of SPL detection were risk factors, and histology of SPL had no significant influence. The relative risk for diameter (by 1-cm increment) was 1.38 (95% CI: 1.18 to 1.59, p 0.0001) and that for DFI less than 1 year (versus DFI 1year) was 1.99 (95% CI: 1.10 to 3.51, p 0.02). In subgroup analysis of metastasis patients, the 5-year survival rates were 75.5%, 69.9%, and 50.1% for patients with previous bone and soft tissue sarcoma, colorectal cancer, and other cancers, respectively (p 0.04). History of cancers other than colorectal and bone and soft tissue sarcoma, and DFI less than 1 year were poor prognostic factors by univariate analysis. With regard to DFS, history of extrapulmonary recurrence and DFI less than 1 Fig 1. Survival curves by patient group using the Kaplan-Meier method. There is no significant difference in survival curves (p 0.56). SPL solitary pulmonary lesion.

Ann Thorac Surg SAKAMOTO ET AL 2010;90:1766 72 SOLITARY PULMONARY LESION Table 3. Relative Risk of All Patients Variable RR 95% CI p Value Age, by 10-year increments 0.94 0.81 1.11 0.43 Sex, male 1.13 0.68 1.94 0.64 Smoking history 1.33 0.81 2.26 0.27 Histology Benign 1.00 Primary lung cancer 0.89 0.53 1.49 0.68 Metastasis 1.25 0.76 2.09 0.38 Physiologic status Diabetes mellitus 0.47 0.16 1.05 0.07 Coronary artery disease 0.77 0.16 1.05 0.07 FEV 1, by 100-ml increments 0.98 0.94 1.01 0.16 Colorectal 1.00 Gastric 0.89 0.21 2.66 0.85 Hepatocellular carcinoma 2.45 0.99 5.56 0.05 Bone and soft tissue 1.04 0.34 2.66 0.94 Others 1.76 0.94 3.40 0.08 Multiple origins 1.18 0.50 2.63 0.69 Extrapulmonary recurrence 1.15 0.67 1.91 0.61 SPL diameter, by 1-cm increments 1.32 1.15 1.48 0.0002 Timing of SPL detection DFI 1 year 1.00 DFI 1 year 1.70 0.96 2.93 0.07 Synchronous 1.77 0.80 3.52 0.15 CI confidence interval; DFI disease-free interval; FEV 1 forced expiratory volume in 1 second; RR relative risk; SPL solitary pulmonary lesion. year were risk factors for recurrence by univariate analysis (Table 4). Multivariate analysis was performed and included primary tumor site, history of extrapulmonary recurrence, SPL diameter, and timing of SPL detection. History of cancers other than colorectal and bone and soft tissue sarcoma (versus colorectal cancer) and DFI less than 1 year (versus DFI 1 year) were poor prognostic factors, with relative risks of 2.41 (95% CI: 1.19 to 5.12, p 0.01) and 2.41 (95% CI: 1.19 to 4.86, p 0.01), respectively. With regard to DFS, history of extrapulmonary recurrence and diameter (by 1-cm increment) were risk factors for recurrence, with relative risks of 2.14 (95% CI: 1.37 to 3.34, p 0.0008) and 1.27 (95% CI: 1.06 to 1.50, p 0.01), respectively. Among lung cancer patients, diameter and pathologic stage were poor prognostic factors by univariate analysis (Table 5). In multivariate analysis including pathologic stage, timing of SPL detection, history of extrapulmonary recurrence, and primary tumor site (gastric, others, or multiple origins versus colorectal), pathologic stage was the only risk factor, with a relative risk of 7.95 (95% CI: 2.72 to 23.6, p 0.0003) for stage II and 13.2 (95% CI: 3.90 to 45.1, p 0.0001) for stage III and IV. The 5-year survival rate was 89.8% for stage I and 28.9% for stage II. The 5-year survival rate for stage III and IV was not calculated because of the short follow-up period. Comment 1769 Multiple lung nodules found during follow-up of malignant disease usually suggest metastasis from the primary site. However, SPL poses a diagnostic challenge. It is essential to remove newly found SPL in patients with a history of malignant disease for the following reasons: for definite pathologic diagnosis, to obtain local control and perform adjuvant chemotherapy in the case of pulmonary metastasis, for curative surgery in the case of lung cancer, and to avoid unnecessary chemotherapy or radiotherapy in the case of benign pulmonary lesions. The prevalence of malignancy of SPL is reported to be approximately 60% in the general population [4, 5]. In our series, the percentage is much higher, that is, 92.6% of SPL were malignant. However, it is important to note that as many as 18 patients (7.4% of all cases) had benign GENERAL THORACIC Table 4. Relative Risk of Metastasis Patients Overall Survival DFI RR 95% CI p Value RR 95% CI p Value Age, by 10-year increments 0.98 0.82 1.19 0.82 1.09 0.96 1.24 0.19 Sex, male 0.87 0.46 1.66 0.67 0.8 0.53 1.21 0.28 Smoking history 1.30 0.69 2.45 0.41 0.91 0.60 1.36 0.65 Colorectal 1.00 1.00 Bone and soft tissue 0.92 0.26 2.60 0.88 0.80 0.39 1.49 0.49 Others 2.18 1.11 4.46 0.02 1.09 0.71 1.69 0.68 Extrapulmonary recurrence 1.23 0.65 2.31 0.52 2.13 1.42 3.21 0.0003 SPL diameter, by 1-cm increments 1.08 0.80 1.38 0.60 1.18 0.99 1.37 0.06 Timing of SPL detection DFI 1 year 1.00 1.00 DFI 1 year 2.18 1.13 4.17 0.02 1.60 1.02 2.47 0.04 Synchronous 2.19 0.35 7.57 0.34 2.13 0.65 5.20 0.19 CI confidence interval; DFI disease-free interval; RR relative risk; SPL solitary pulmonary lesion.

GENERAL THORACIC 1770 SAKAMOTO ET AL Ann Thorac Surg SOLITARY PULMONARY LESION 2010;90:1766 72 Table 5. Relative Risk of Primary Lung Cancer Patients RR 95% CI p Value Age, by 10-year increments 0.96 0.62 1.54 0.85 Sex, male 2.78 0.94 11.8 0.07 Smoking history 3.31 0.96 20.7 0.06 Colorectal 1 Gastric 1.11 0.21 6.01 0.90 Others 1.68 0.52 7.44 0.41 Multiple 1.09 0.27 5.34 0.90 Extrapulmonary recurrence 0.97 0.15 3.33 0.97 SPL diameter, by 1-cm increments 1.49 1.26 1.74 0.0001 Timing of SPL detection DFI 1year 1 DFI 1year 1.13 0.18 4.03 0.88 Synchronous 1.27 0.41 3.27 0.65 Histology Adenocarcinoma 1 Squamous cell carcinoma 1.36 0.49 3.37 0.53 Others 0.65 0.036 3.20 0.65 Pathologic stage I 1 II 7.13 2.57 19.8 0.0003 III and IV 11.5 3.63 35.0 0.0001 Operation Anatomical resection 1 Partial resection 0.84 0.30 2.04 0.71 CI confidence interval; DFI disease-free interval; RR relative risk; SPL solitary pulmonary lesion. lesions. Furthermore, 42% of SPL was metastasis even if more than 5 years had passed from previous cancer. It is difficult to predict the etiology of SPL preoperatively. Although CT diagnosis by radiologists was surprisingly accurate, it was not 100% accurate because of diagnostic limits. Recently, PET was employed to heighten the accuracy of diagnosis [5]; but PET was only performed in 56 patients because the method was not used in the early days of our series, and it was not of value for reference because of the low sensitivity and specificity. Generally, the negative predictive value was not sufficiently high, especially in the population with a high prevalence of malignancy. For this reason, PET could not replace pathologic diagnosis. Bronchoscopic examination is safe and in widespread use [6], but it provided a diagnosis in only 34 patients (less than 60%), in two thirds of whom it was difficult to distinguish between primary lung cancer and metastasis because of the small amount of available tissue. Percutaneous needle biopsy was performed in 7 patients, leading to a diagnosis of malignancy in all, but it is associated with a high complication rate of 24% [7]. Thoracoscopic surgery is minimally invasive [8], and is of equivalent therapeutic value to open thoracotomy for metastasectomy [3]. From the point of view of accurate diagnosis and safety, the surgical approach would be regarded as superior. According to subgroup analysis for metastasis patients, history of extrapulmonary recurrence and diameter were risk factors for recurrence, but they did not influence overall survival. On the contrary, history of cancers other than colorectal and bone and soft tissue sarcoma, which was a poor prognostic factor for survival, was not a risk factor for recurrence. This finding means that recurrence after SPL resection does not necessarily lead to death in some types of cancers. For example, repeated metastasectomy would contribute to longer survival for selected patients with colorectal cancer and soft tissue sarcoma [9, 10]. Shorter DFI was a poor prognostic factor, and if we seek a life-prolonging effect other than diagnosis in resection of SPL, patients with DFI of 1 year or more would be good candidates. In our series, the proportion of patients with primary lung cancer was very high, and two thirds of all cases were stage I. As CT screening of the general population yields many early-stage lung cancers [11], close follow-up of previous cancer by CT may have the same result. In cases of primary lung cancer, except for pure bronchioloalveolar carcinoma, we performed lobectomy with node dissection if the patient s general condition permitted. There is no need to take into account the status of the previous cancer for therapeutic decision making, because none of the variables concerning previous cancer had an influence on survival. It was reported that the 5-year survival rates for stage IA, IB, IIA, IIB, IIIA, IIIB, and IV primary lung cancer were 73%, 58%, 46%, 36%, 24%, 9%, and 13%, respectively [12]. The 5-year survival in the present study was poor for stage II disease, but it was comparable to that of the general population for stage I. Survival analysis showed that pathologic stage was the only risk factor, and none of the variables relating to previous cancer such as primary tumor site, timing of SPL detection, and history of extrapulmonary recurrence had any influence on survival. It is debatable whether primary lung cancer with antecedent cancer had a poor prognosis [13]. However a number of studies have suggested that cancer multiplicity is not associated with a worse prognosis [14, 15], and our results are in agreement with this. For this reason, standard therapy for primary lung cancer should also be indicated for patients with previous cancer, irrespective of the primary tumor site of the previous cancer, history of recurrence, and short DFI. Our study might be limited by selection bias. The patients who underwent surgical resection were in comparatively good general health and would not represent the whole population with SPL. Moreover, SPL apparently of benign nature was excluded from the indication of surgery, which would lead to high prevalence of malignancy. This selection bias was the other side of our policy that surgical indication should be examined strictly, and therefore would be accepted. In conclusion, surgical resection of SPL is an effective procedure to make a definite diagnosis and decide upon a therapeutic strategy. Radiologic findings are useful but cannot replace the pathologic diagnosis because 7.4% of patients had a benign lesion despite CT diagnosis of malignancy. In the case of metastasis, prognosis was influenced by DFI, and patients who had a history of

Ann Thorac Surg SAKAMOTO ET AL 2010;90:1766 72 SOLITARY PULMONARY LESION cancers other than colorectal and bone and soft tissue sarcoma had a worse prognosis compared with patients who had previous colorectal cancer. Large-diameter SPL and history of extrapulmonary recurrence led to early recurrence. With regard to primary lung cancer, two thirds were stage I, and patient survival was comparable to that of the general population. None of the variables concerning previous cancer had an influence on survival. It is essential to make a pathologic diagnosis of SPL by surgical resection and to treat for primary lung cancer with standard therapy regardless of a history of malignancy. References 1. Pfannschmidt J, Dienemann H, Hoffmann H. Surgical resection of pulmonary metastasis from colorectal cancer: a systematic review of published series. Ann Thorac Surg 2007; 84:324 38. 2. Billingsley KG, Burt ME, Jara E, et al. Pulmonary metastases from soft tissue sarcoma: analysis of patterns of disease and postmetastasis survival. Ann Surg 1999;229:602 12. 3. Nakajima J, Murakawa T, Fukami T, Takamoto S. Is thoracoscopic surgery justified to treat pulmonary metastasis from colorectal cancer. Interact Cardiovasc Thorac Surg 2008;7:212 6. 4. Mery CM, Pappas AN, Bueno R, et al. Relationship between a history of antecedent cancer and the probability of malignancy for a solitary pulmonary nodule. Chest 2004;125: 2175 81. 5. Grgic A, Yuksel Y, Groschel A, et al. Risk stratification of solitary pulmonary nodules by means of PET using 18Ffluorodeoxyglucose and SUV quantification. Eur J Nucl Med Mol Imaging 2010;37:1087 94. 1771 6. Trkanjec JT, Peros-Golubicic T, Grozdek D, Ivicevic A, Ailovic M. The role of transbronchial lung biopsy in the diagnosis of solitary pulmonary nodule. Coll Antropol 2002; 27:669 75. 7. Mitruka S, Landreneau RJ, Mack MJ, et al. Diagnosing the indeterminate pulmonary nodule: percutaneous biopsy versus thorocoscopy. Surgery 1995;118:676 84. 8. Salaini L, Prusciano F, Bagioni P, Di Francesco F, Salaini L, Poddie DB. Video-assisted thoracic surgery (VATS) of the lung: analysis of intraoperative and postoperative complications over 15 years and review of the literature. Surg Endosc 2008;22:298 310. 9. Welter S, Jacobs J, Krbek T, Krebs B, Stamatis G. Long-term survival after repeated resection of pulmonary metastases from colorectal cancer. Ann Thorac Surg 2007;84:203 10. 10. Weser MR, Downey RJ, Leung DHY, Brennan MF. Repeat resection of pulmonary metastases in patients with softtissue sarcoma. J Am Coll Surg 2000;191:184 90. 11. Veronesi G, Bellomi M, Scanagatta P, et al. Difficulties encountered managing nodules detected during a computed tomography lung cancer screening program. J Thorac Cardiovasc Surg 2008;136:611 7. 12. Rami-Porta R, Crowley JJ, Golodstraw P. The revised TNM staging system for lung cancer. Ann Thorac Cardiovasc Surg 2009;15:4 9. 13. Furak J, Trojan I, Szoke T, et al. Lung cancer as a second primary malignant tumor: prognostic values after surgical resection. Interact Cardiovasc Thorac Surg 2008;7:50 3. 14. Massard G, Ducrocq X, Beaufigeau M, et al. Lung cancer following previous extrapulmonary malignancy. Eur J Cardiothorac Surg 2000;18:524 8. 15. Aguilo R, Macia F, Porta M, Casamitjana M, Minguella J, Novoa AM. Multiple independent primary cancers do not adversely affect survival of the lung cancer patient. Eur J Cardiothorac Surg 2008;34:1075 80. GENERAL THORACIC INVITED COMMENTARY The article by Sakamoto and colleagues [1] is a retrospective review of 243 patients with a previous cancer diagnosis who were found to have a solitary pulmonary nodule. Only those patients who had a surgical resection are included in the analysis. The cohort is further refined by excluding patients who had prior lung metastases, any enlarged lymph nodes on computed tomographic scan, or a positive lymph node on positron emission tomographic scan, and nodules that were decreasing in size. Although the population is highly selected, there are some important lessons illustrated. The data regarding nodule size indicates that our follow-up of patients who have had a prior cancer needs improvement. The median diameter of a mass resected in this series was 1.5 cm. For those who had a primary lung cancer, the median size was 2.4 cm. Moreover, in patients with primary lung cancer, 60% of the patients had a tumor that was larger than 3.0 cm, and 30% of patients had a tumor that was more than stage I cancer, despite being in the system after their treatment for a cancer elsewhere. This clearly demonstrates the need for careful analysis of how we follow patients after a cancer in the lung or elsewhere in the body. To find that a patient has a 3-cm mass in the lung spread to mediastinal lymphatics is quite depressing and implies we could do better with our follow-up methods. In our own institution, follow-up is random and varies all the way from an aggressive plan of a computed tomographic scan every 3 months with intermittent positron emission tomographic scans to only on an as needed basis. Clearly, the correct follow-up is somewhere in between these two extremes. With a trial underway for screening of the general population for lung cancer, perhaps a more important population would be to screen those patients who have had a prior lung cancer. It would be an interesting randomized trial to perform to see whether or not there is a benefit. The other lesson to be learned in this series is that radiographic and clinical examinations are somewhat lacking. The specificity for a computed tomographic scan for those with primary lung cancer was only 88% and only 82% for those with a pulmonary nodule from metastatic disease. The positron emission tomographic scan was even worse, with only two-thirds being correctly predicted as a malignancy. Needle biopsy, bronchoscopy, and clinical impression were also equally misleading, although specific numbers are not given. In any event, the current technique of thoracoscopic wedge resection, which can be done with relatively low morbidity and mortality, can obtain near 100% specificity of a solitary pulmonary nodule. The practice of watching pulmonary nodules in patients who have a history of cancer because 2010 by The Society of Thoracic Surgeons 0003-4975/$36.00 Published by Elsevier Inc doi:10.1016/j.athoracsur.2010.08.019