Lung Cancer xxx (2011) xxx xxx. Contents lists available at ScienceDirect. Lung Cancer. journal homepage:

Transcription

1 Lung Cancer xxx (2011) xxx xxx Contents lists available at ScienceDirect Lung Cancer journal homepage: Coexisting pulmonary nodules in operable lung cancer: Prevalence and probability of malignancy A.M. Ruppert a,, U. Lerolle a, M.F. Carette b, A. Lavole a, A. Khalil b, B. Bazelly c, M. Antoine d, J. Cadranel a, B. Milleron a a Service de Pneumologie, Hôpital Tenon APHP and Université Paris IV, France b Service de Radiologie, Hôpital Tenon APHP and Université Paris IV, France c Service de Chirurgie thoracique, Hôpital Tenon APHP and Université Paris IV, France d Service d Anatomopathologie, Hôpital Tenon APHP and Université Paris IV, France article info abstract Article history: Received 17 December 2010 Received in revised form 7 March 2011 Accepted 21 March 2011 Keywords: Lung cancer Pulmonary nodule Diagnosis Surgery Computed tomography scan Introduction: Coexistence of pulmonary nodules in operable non small cell lung cancer (NSCLC) may influence the therapeutic indication. The aim of this study was to evaluate prospectively the prevalence and the probability of malignancy of pulmonary nodules in operable lung cancer. Methods: From a prospective database, all surgically treated patients diagnosed with NSCLC from 1998 to 2003 were retrospectively reviewed. Patients presenting pulmonary nodule(s) were identified. Results: Two hundred thirty nine patients had a complete resection for a NSCLC and 56 patients (24%) presented altogether 88 nodules on thoracic CT. Twenty-four of these nodules (27%) were malignant, 28 (32%) benign and 36 (41%) of undetermined nature. Five factors associated with nodule s malignancy were identified: tumour histology (non-squamous (non-scc) 44% vs. SCC 7%, p = 0.001), localization of the nodules in an upper lobe (vs. other lobe, p = 0.004), co localization in the same lobe as the NSCLC (vs. another lobe, p = 0.03), nodule size (p = 0.05) and shape (speculated vs. non spiculated, p = 0.02). From these factors, a probability score was assessed with a malignancy rate in SCC of 0% in nodules presenting 1 feature, 33% with 2 features and 100% with 3 features and in non-scc of 40% with 1 feature, 82% with 2 features and 100% with 3 features. Conclusion: Diagnosis of satellite nodules associated with early stage NSCLC is common. We developed a predictive score to estimate the probability of malignancy which may be a precious aid in the management of pulmonary nodules associated to a NSCLC Published by Elsevier Ireland Ltd. 1. Introduction Recent advances in thoracic imaging with the widespread use of computed tomography (CT) has improved the management of non-small cell lung cancer (NSCLC) [1]. However, the sensitivity of this technique has increased and the detection of small pulmonary nodules is common during the staging evaluation for patients with lung cancer. In patients with an operable lung cancer, the coexistence of satellite lesions may change TNM status and influence the therapeutic indication. If a nodule is considered as malignant, NSCLC may be classified as stage T4 (nodule in the same lobe) or M1 (nodule in another lobe), or as synchronous multiple lung cancer in Corresponding author at: UF d Oncologie Thoracique, Service de pneumologie, Hôpital de Tenon, Université Paris VI et Assistance publique des Hôpitaux de Paris 4 rue de Chine, Paris, France. Tel.: ; fax: addresses: anne-marie.ruppert@tnn.aphp.fr, ruppertannemarie@yahoo.com (A.M. Ruppert). the 1997 classification [2] and T3 (nodule in the same lobe), T4 (ipsilateral nodules in different lobes) or M1a (controlateral pulmonary nodules) in the 2009 classification [3]. As prognosis is significantly better in patients with stage I (T1-2N0M0) or synchronous primary lung cancer than those with stage IIIB (T4N0M0) or stage IV (T1-2N0M1), the existence of malignant pulmonary nodules could be considered as a contraindication to surgery. Thus, to obtain a differential diagnosis between metastases and benign lesions is essential to choose the best therapeutic strategy. Few studies have focused on pulmonary nodules in lung cancer patients. However, the diagnosis of these nodules coexisting with a known primary lung cancer remains a diagnostic challenge. Fiberoptic bronchoscopy is not often helpful in the diagnosis of peripheral pulmonary nodules [4] and the diagnostic sensibility of CT guided transthoracic needle aspiration is conflicting. For certain authors, transthoracic needle aspiration is less accurate in small pulmonary nodules [5], whereas other authors report a good diagnostic value, but a high rate of pneumothorax [6]. Interest in non-invasive procedures as 18F-fluorodeoxyglucose positron /$ see front matter 2011 Published by Elsevier Ireland Ltd. doi: /j.lungcan

2 2 A.M. Ruppert et al. / Lung Cancer xxx (2011) xxx xxx emission tomography (PET-FDG) is high, because of a good sensitivity (96%) and specificity (73%) [7]. However pulmonary nodules which are less than 1 cm in size or show ground glass opacities on CT cannot be evaluated accurately by PET [8]. The aim of this study was to evaluate prospectively the prevalence of pulmonary nodules coexisting in operable lung cancer, the rate and the probability of malignancy of these nodules. 2. Patients and methods 2.1. Patients All patients diagnosed with lung cancer are registered in a prospective database. Data from surgically treated patients diagnosed with NSCLC between 15/08/1998 and 15/08/2003 were retrospectively reviewed to identify patients presenting at least one pulmonary nodule coexisting with lung cancer on initial thoracic CT Preoperative evaluation and follow up All patients presented histological proven NSCLC according to the World Health Organization guidelines and tumour stage was defined by the 7th TNM classification [2,9]. Preoperative assessment included fiberoptic bronchoscopy, thoracic CT, abdominal CT, cerebral CT or MRI, and bone scan if bone metastases were suspected. Realization of 18F-fluorodeoxyglucose positron emission tomography (PET-FDG) was recommended but not mandatory, as access to this investigation was difficult at that time. Postoperative follow up included clinical examination and chest X-ray every three months and clinical examination and thoracic CT every six months during the first two years, and clinical examination and chest X-ray every six months and clinical examination and thoracic CT every year during the three following years Investigation of pulmonary nodules The diagnostic procedures recommended for nodules coexisting with NSCLC depend on their localization with regard to the tumour. For pulmonary nodules situated in the same lung as the tumour, resection was considered. Contralateral nodules with a size 10 mm were evaluated by thoracic CT during follow up. Contralateral nodules with a size >10 mm were assessed by PET-FDG. If PET-FDG shows metabolic hyperactivity, resection was recommended. PET-FDG negative controlateral nodules were monitored during follow up Analyzed parameters For each patient, the following characteristics were analyzed: age, sex, smoking status, performance status, personal history of cancer, tuberculosis (active or sequela) or pneumoconiosis, tumour histology, clinical and post operative TNM tumour stage (established without considering the pulmonary nodule(s)), type of surgery, localization of the tumour, PET-FDG results concerning the tumour, number of pulmonary nodules identified on thoracic CT and patient s outcome. For each pulmonary nodule, its anatomical localization (upper lobe, lower lobe and middle lobe), its localization with regard to the tumour (same lobe, different lobes in ipsilateral or controlateral lung), its size, rate of calcifications, shape, and PET-FDG results were noted. During postoperative follow up, any modification in size of the nodule on thoracic CT was mentioned. Fig. 1. Study algorithm Classification of pulmonary nodules and definition of patient groups Three groups of pulmonary nodules according to their histological nature: benign, malignant or undetermined were considered. The diagnosis of malignancy of a nodule was established on histological analyses obtained by surgery, fiberoptic bronchoscopy or CT-guided transthoracic needle aspiration. A nodule was considered as benign when its benignity was histologicaly proven or when a nodule spontaneously disappeared or was stable in size over a period of at least 24 months. A nodule was considered of undetermined nature if the nodule cannot be clearly identified by the pathologist in the resected lobe when the nodule was in the same lobe or in the ipsilateral nodules in different lobes biopsied during the thoracotomy. Furthermore, three groups of patients were identified according to the histological nature of the nodule(s): group of patients with malignant nodules, group of patients with benign nodules and a group of patients with nodules of undetermined nature. Patients presenting several nodules were put into the group of patients with malignant nodules if at least one nodule was malignant Statistical analyses Statistical analyses for comparisons between groups were realized using Mann Whitney s U test for continuous variables and Fisher s exact test for proportions. A value of p < 0.05 was considered statistically significant. The date of point was December 30, Results 3.1. Patient s characteristics Between 15/08/1998 and 15/08/2003, 604 patients were diagnosed with NSCLC and 239 patients had complete resection. Seven of these patients were not included (five presented active pulmonary tuberculosis with multiple pulmonary nodular lesions, and data of two patients were incomplete). Fifty-six patients (24%) presented at least one pulmonary nodule coexisting with NSCLC on initial thoracic CT (Fig. 1). No signifi-

3 A.M. Ruppert et al. / Lung Cancer xxx (2011) xxx xxx 3 Table 1 Clinical and pathological characteristics of the patients with NSCLC. Characteristics Patients without nodule n = 176 Patients with nodule(s) n = 56 Comparison patients with vs. without nodules Median (IQ) or n (%) p Age Years 62 (19) 62 (13) 0.34 Gender Male 139 (79%) 44 (79%) Female 37 (21%) 12 (21%) 0.95 Smoking status Never smokers 17 (10%) 2 (4%) Current or former smokers 159 (90%) 54 (96%) 0.15 Personal history of cancer Yes 15 (9%) 4 (7%) No 161 (91%) 52 (93%) 0.74 Performance status (WHO) (84%) 49 (88%) 1 23 (13%) 7 (13%) (3%) 0 ctnm I 67 (38%) 23 (41%) II 47 (27%) 16 (29%) III 61 (35%) 17 (30%) 0.87 IV 1 (1%) 0 Tumour histology Adenocarcinoma 67 (38%) 19 (34%) Bronchiolo-alveolar carcinoma 4 (2%) 4 (7%) Squamous cell carcinoma 75 (43%) 23 (41%) 0.35 Large cell carcinoma 12 (7%) 6 (11%) Other 18 (10%) 4 (7%) cant difference in age, gender, smoking status, histology or tumour stage was noted between the patients with or without a pulmonary nodule (Table 1). Thirty-six patients presented one nodule and 20 patients presented several nodules (14 patients had two nodules, 2 patients had three nodules and 4 patients had four to six nodules). Altogether, 88 nodules were identified. Twenty-four nodules (27%) were malignant, 28 nodules (32%) were benign, and 36 (41%) were of undetermined nature. Among the malignant nodules, 18 (75%) presented the same histology as the NSCLC, and six (25%) were of a different histological type. For benign lesion, histological proof was obtained in seven cases (two hamartomas, one fibrotic nodule, one normal pulmonary lymph node, one granulation tissue, one silicotic nodule and one nodule with hyalin sclerosis), thoracic spiral CT diagnosed one arteriovenous fistula and follow up during 24 months showed stability or regression of the other 20 nodules. Among the 36 nodules of undetermined nature, 22 were resected with the NSCLC but could not be identified by the surgeon or the pathologist (10 patients) and follow up in 12 nodules did not attain 24 months (death or long-lost in 4 patients) Factors associated to nodules malignancy Patient s characteristics According to the histological nature of the nodule(s), three groups of patients were created: the group of patients with malignant, with benign or indeterminate nodule(s) (Table 2). The group of patients with malignant nodules included 18 patients (32%) and 24 malignant nodules. Seven patients in this group presented also nodules of undetermined nature (16 nodules). The group of patients with benign nodule(s) included 24 patients (43%, 28 nodules), and 13 patients (20 nodules) formed the group with nodules of undetermined nature. Patients in the group with malignant nodules were significantly younger than patients in the group with benign nodules (61 (11) vs. 69 (13) years, p = 0.03). No significant difference was noted between the two groups in terms of gender, smoking status, personal history of cancer and performance status. Four patients (22%) in the group with malignant nodules had more than two nodules vs. two patients (8%) in the group patients with benign nodules. However, this difference was not statistically significant. Tumour histology differed significantly in the two groups of patients: 44% of patients presented adenocarcinoma (ADC) in the group with malignant nodules vs. 17% in the group with benign nodules, p = However, size and localization of the NSCLC as well as ctnm stage were not significantly different between the two groups of patients Nodules characteristics Comparing benign to malignant nodules showed significant differences in the anatomical localization, localization with regard to the NSCLC, size and shape (Table 3). Malignant nodules were more often detected in an upper lobe than benign nodules (21 (88%) vs. 12 (43%), p = 0.004) and in the same lobe as the NSCLC (8 (33%) vs. another lobe 2 (7%), p = 0.03). Only 2 malignant nodules (8%) were situated in a lower lobe compared to 12 benign nodules (43%). Malignant nodules were significantly larger than benign nodules (median size 9 (9) vs. 6 (8) mm, p = 0.05). Malignant nodules were frequently spiculated (8 (33%) vs. 2 (7%), p = 0.02). Only 9 malignant nodules (38%) were well limited compared to 22 benign nodules (79%). Results of PET-FDG were not significantly different between benign and malignant nodules Predictive score of malignancy Five factors associated with malignancy seemed particularly interesting in order to predict malignancy of pulmonary nodules coexisting with NSCLC: tumour histology (ADC, BAC or LCC), localization of the nodule in an upper lobe, localization of the nodule in the same lobe as the NSCLC, nodule s size >10 mm and its spiculated shape. The percentage of malignancy of the nodules was evaluated according to the number of the features (0 5) presented by each nodule (Table 4). In our study, the observed percentage of malignancy was 0% in nodules without any of these 5 characteristics. The risk of malignancy was different in SCC than in non-scc (ADC, BAC or LCC). In SCC, the evaluated risk was 0% if only one feature was detected, 33%

4 4 A.M. Ruppert et al. / Lung Cancer xxx (2011) xxx xxx Table 2 Factors associated with nodules malignancy. Characteristics Patients with nodules n = 56 a n (%) Patients with benign nodules n = 24 Patients with malignant nodules n = 18 Comparison patients with malignant vs. benign nodules p Number of nodules per patient 1 2 nodules 50 (89%) 22 (92%) 14 (78%) >2 nodules 6 (11%) 2 (8%) 4 (22%) 0.2 Age Year 62 (13) 69 (13) 61 (11) 0.03 Gender Men 44 (79%) 18 (75%) 13 (72%) Women 12 (21%) 6 (25%) 5 (28%) 0.84 Tobacco Smoker or former-smoker 54 (96%) 24 (100%) 17 (94%) Non smoker 2 (4%) 0 1 (6%) 0.24 Performance status (WHO) 0 49 (88%) 23 (96%) 15 (83%) 1 7 (13%) 1 (4%) 3 (17%) 0.17 Histology Adenocarcinoma 19 (34%) 4 (17%) 8 (44%) Squamous cell carcinoma 23 (41%) 16 (67%) 3 (17%) Large cell carcinoma 6 (11%) 0 4 (22%) Bronchioloalveolar carcinoma 4 (7%) 1 (4%) 3 (17%) Other 4 (7%) 3 (13%) 0 Tumour size Size (mm) 40 (22) 39 (24) 40 (19) 0.69 ctnm I 23 (41%) 12 (50%) 8 (44%) II 16 (29%) 7 (29%) 4 (22%) III 17 (30%) 5 (21%) 6 (33%) IV ptnm I 18 (32%) 9 (38%) 7 (39%) II 14 (25%) 6 (25%) 4 (22%) III 18 (32%) 5 (21%) 6 (33%) IV 2 (4%) 0 1 (6%) Complet response 4 (7%) 4 (17%) 0 Follow-up Days 677 (629) 863 (928) 614 (435) a Patients (14) with nodules of undetermined nature were not used for comparison. if two features were detected and 100% if three or more features were detected. In non-scc, the risk of malignancy was estimated at 40% with 1 feature, 82% with 2 features and 100% with 3 or more features. The low number of nodules in some groups of patients may not allow predicting the risk of malignancy. A prospective assessment in a large cohort is mandatory. 4. Discussion In patients with a potentially operable lung cancer, the diagnosis of coexisting small nodules of 1 cm or less is critical to avoid insufficient or excessive surgical procedures. Previous studies estimated the prevalence of pulmonary nodules in the global population between 8 and 51%, with a rate of malignancy of 1, 1 12% [10 14]. This disparity depends on the used CT methodology and the population study, particularly the percentage of smokers and the frequency of granulomatoses [12]. In the Early Lung Cancer Action Project (ELCAP), prevalence of pulmonary nodules at the initial screening CT was 23% and 12% of these nodules were malignant [15]. However this data cannot be surrogated to cancer patients. In this study, prevalence of pulmonary nodules associated to lung cancer was 24% and 27% of these nodules were malignant. Only four previous studies reported prevalence of pulmonary nodules in operable NSCLC [16 19]. Results are conflicting. Similar results are found by Yuan et al. with a prevalence of pulmonary nodules of 26% in operable NSCLC and a rate of malignancy of 25% [16]. Caretta et al. reported 32 nodules in 29 patients who underwent surgery for NSCLC. Thirty-two percent of these nodules were malignant [17]. However in a recent Korean study, a much higher (44%) prevalence of nodules was noted and 93% of theses nodules were benignant [18]. This high prevalence of benign nodules may be related to the high prevalence of tuberculosis in Korea. In a British study, prevalence of pulmonary nodules in a pre-operative setting was only of 16% with 11% of malignancy [19]. Few data on factors associated to nodule malignancy associated to NSCLC are available. Seven factors associated to nodule malignancy in NSCLC were identified in this study: patient s age, tobacco exposure, tumour histology, localization of the nodules in a superior lobe, co localization in the same lobe as the NSCLC, nodule s size and nodule s shape. No association between tumour stage or number of nodules and malignancy were noticed. As expected, tumour histology was a predominant characteristic. A high prevalence of malignant nodules was mainly observed in ADC. These tumours are often multifocal [20]. Data on BAC or LCC have to be taken with caution, since the number of patients with this histology is limited. Yuan et al. reported that colocalization of the nodule in the same lobe as the NSCLC was significantly associated to malignancy [16]. However, nodule s size and tumour stages were not associated to malignancy and tumour histology of NSCLC was not reported. In the study of Caretta et al., no significant correlation between nodule malignancy and nodule size, localization with regard to the tumour, tumour histology or ctnm was detected [17]. The recent recommendations on the management of pulmonary nodules by the Fleischner Society are established on the nodule s size and the risk factors of the patient (tobacco status, asbestos exposure) [21]. Thus pulmonary nodules of 0.5 cm or smaller are recommended for CT follow-up, those larger than 0.8 cm for further investigation. In lung cancer patients, a substantial number of

5 A.M. Ruppert et al. / Lung Cancer xxx (2011) xxx xxx 5 Table 3 Characteristics of the nodules. N (total) a Benign nodules n = 28 (%) b Malignant nodules n = 24 (%) b Comparison benign vs. malignant nodules Anatomical localization a Upper lobe (n total = 54) 12 (22%) 21 (39%) p = Median lobe (n = 8) 4 (50%) 1 (13%) Lower lobe (n = 26) 12 (46%) 2 (8%) Localization/NSCLC Same lobe (n = 24) 2 (8%) 8 (33%) p = 0.03 Other ipsilateral lobe (n = 14) 7 (50%) 2 (14%) Contralateral lung (n = 50) 19 (38%) 14 (28%) Size 10 mm (n = 71) 24 (34%) 14 (20%) p = mm (n = 14) 4 (29%) 7 (50%) mm (n = 3) 0 3 (100%) Shape Well limited (n = 55) 22 (40%) 9 (16%) p = 0.02 Spiculated (n = 13) 2 (15%) 8 (62%) Lobulated (n = 16) 4 (25%) 6 (38%) Ground glass opacities (n = 4) 0 1 (25%) a Nodules of undetermined nature (36) were not used for comparison. b % corresponds to: benign or malignant nodules/all nodules for each localization. indeterminate pulmonary lesions are of a benign nature. However 20% of nodules of less than 10 mm were considered malignant in this study. These results confirm previously published data that in NSCLC even small pulmonary nodules must be explored, as small size is not a sufficient argument for benignity [17]. We developed a predictive score to estimate the probability of malignancy of pulmonary nodules in NSCLC. Five features associated to nodule malignancy were selected: tumour histology, localization of the nodules in an upper lobe, co localization in the same lobe as the NSCLC, nodule s size larger than 1 cm and its spiculated shape. The malignancy rate in SCC was estimated of 0% in nodules presenting 1 feature, 33% with 2 features and 100% with 3 features and in non-scc of 40% with 1 feature, 82% with 2 features and 100% with 3 features. A prospective assessment of this score in a large cohort is mandatory. This score may be a precious aid in the management of pulmonary nodules associated to a NSCLC. However even if the risk of malignancy of a nodule with 0 1 feature is low, exploration of the nodule is still recommended if it is larger than 1 cm [21]. The main limitation in this study is the high rate of nodules of indeterminate origin (36 nodules or 41%). In the study of Yuan et al., only 15% of nodules of indeterminate origin were reported [16]. This high rate may be linked to several reasons. Clinicians usually consider only the largest nodule and pay no attention to secondary small nodules. However in this study all nodules were identified. A histological proof in small nodules is often not available. Without a histological proof, a close follow-up of these nodules is necessary. Nodules were considered benignant when they disappeared or were stable on thoracic CT during a 24 month follow-up. Furthermore, 22 nodules were resected with the NSCLC but could not be identified by the surgeon or the pathologist. They were considered as of indeterminate origin, as a clear histological diagnosis cannot be obtained. In conclusion, characterisation of indeterminate satellite lesions in patients with early stage NSCLC is critical. In this study, prevalence of pulmonary nodules associated to lung cancer was 24% and 27% of these nodules were malignant. Even small pulmonary nodules must be explored, as small size is not a sufficient Table 4 Predictive score of malignancy. Histology 0 1 feature 2 features 3 5 features SCC ADC BAC LCC SCC ADC BAC LCC All Malignancy in % (n malignant/n detected) 0% (0/19) 40% (2/5) 33% (2/6) 82% (9/11) 100% (11/11) Features associated to malignancy: - tumour histology: adenocarcinoma (ADC) or bronchiolo-alveolar carcinoma (BAC) or large cell carcinoma (LCC). - localization of the nodule in an upper lobe. - localization of the nodule in the same tumour lobe. - size of the nodule >10 mm. - shape speculated or lobulated.

6 6 A.M. Ruppert et al. / Lung Cancer xxx (2011) xxx xxx argument for benignity. We developed a score based on five features to estimate the probability of nodules malignancy. This score may help in the management of nodules associated with a NSCLC. Conflict of interest statement None declared. Acknowledgments This work was supported by the Hôpital Tenon, Assistance Publique des Hôpitaux de Paris, France and the Faculté de Médecine Université PARIS VI, France. We thank Eliane Bertrand for technical work. References [1] Costello P, Anderson W, Blume D. Pulmonary nodule: evaluation with spiral volumetric CT. Radiology 1991;179(3): [2] Mountain CF. Revisions in the international system for staging lung cancer. Chest 1997;111(6): [3] Groome PA, Bolejack V, Crowley JJ, Kennedy C, Krasnik M, Sobin LH, et al. The IASLC lung cancer staging project: validation of the proposals for revision of the T N, and M descriptors and consequent stage groupings in the forthcoming (seventh) edition of the TNM classification of malignant tumours. J Thorac Oncol 2007;2(8): [4] Torrington KG, Kern JD. The utility of fiberoptic bronchoscopy in the evaluation of the solitary pulmonary nodule. Chest 1993;104(4): [5] Li H, Boiselle PM, Shepard JO, Trotman-Dickenson B, McLoud TC. Diagnostic accuracy and safety of CT-guided percutaneous needle aspiration biopsy of the lung: comparison of small and large pulmonary nodules. Am J Roentgenol 1996;167(1): [6] Wallace MJ, Krishnamurthy S, Broemeling LD, Gupta S, Ahrar K, Morello FA, et al. CT-guided percutaneous fine-needle aspiration biopsy of small (< or =1-cm) pulmonary lesions. Radiology 2002;225(3): [7] Gould MK, Maclean CC, Kuschner WG, Rydzak CE, Owens DK. Accuracy of positron emission tomography for diagnosis of pulmonary nodules and mass lesions: a meta-analysis. JAMA 2001;285(7): [8] Nomori H, Watanabe K, Ohtsuka T, Naruke T, Suemasu K, Uno K, et al. Evaluation of F-18 fluorodeoxyglucose (FDG) PET scanning for pulmonary nodules less than 3 cm in diameter, with special reference to the CT images. Lung Cancer 2004;45(1): [9] Travis W, Cloby T, Corrin B, Shimosato Y, Brambilla E. Histological typing of lung pleura Dans: Sobin LH. World Health Organization: international histological classification of tumors. 3rd éd. New York: Springer Verlag; [10] Wahidi MM, Govert JA, Goudar RK, Gould MK, McCrory DC. Evidence for the treatment of patients with pulmonary nodules: when is it lung cancer?: ACCP evidence-based clinical practice guidelines (2nd edition). Chest 2007;132(Suppl. 3):94S 107S. [11] Swensen SJ, Silverstein MD, Ilstrup DM, Schleck CD, Edell ES. The probability of malignancy in solitary pulmonary nodules. Application to small radiologically indeterminate nodules. Arch Intern Med 1997;157(8): [12] Swensen SJ, Jett JR, Hartman TE, Midthun DE, Sloan JA, Sykes A, et al. Lung cancer screening with CT: Mayo Clinic experience. Radiology 2003;226(3): [13] Gohagan J, Marcus P, Fagerstrom R, Pinsky P, Kramer B, Prorok P, et al. Baseline findings of a randomized feasibility trial of lung cancer screening with spiral CT scan vs. chest radiograph: the Lung Screening Study of the National Cancer Institute. Chest 2004;126(1): [14] Gould MK, Fletcher J, Iannettoni MD, Lynch WR, Midthun DE, Naidich DP, et al. Evaluation of patients with pulmonary nodules: when is it lung cancer?: ACCP evidence-based clinical practice guidelines (2nd edition). Chest 2007;132(Suppl. 3):108S 30S. [15] Henschke CI, McCauley DI, Yankelevitz DF, Naidich DP, McGuinness G, Miettinen OS, et al. Early Lung Cancer Action Project: overall design and findings from baseline screening. Lancet 1999;354(9173): [16] Yuan Y, Matsumoto T, Hiyama A, Miura G, Tanaka N, Emoto T, et al. The probability of malignancy in small pulmonary nodules coexisting with potentially operable lung cancer detected by CT. Eur Radiol 2003;13(11): [17] Carretta A, Ciriaco P, Canneto B, Nicoletti R, Del Maschio A, Zannini P, et al. Therapeutic strategy in patients with non-small cell lung cancer associated to satellite pulmonary nodules. Eur J Cardiothorac Surg 2002;21(6): [18] Kim YH, Lee KS, Primack SL, Kim H, Kwon OJ, Kim TS, et al. Small pulmonary nodules on CT accompanying surgically resectable lung cancer: likelihood of malignancy. J Thorac Imaging 2002;17(1):40 6. [19] Keogan MT, Tung KT, Kaplan DK, Goldstraw PJ, Hansell DM. The significance of pulmonary nodules detected on CT staging for lung cancer. Clin Radiol 1993;48(2):94 6. [20] Zwirewich CV, Miller RR, Müller NL. Multicentric adenocarcinoma of the lung: CT-pathologic correlation. Radiology 1990;176(1): [21] MacMahon H, Austin JHM, Gamsu G, Herold CJ, Jett JR, Naidich DP, et al. Guidelines for management of small pulmonary nodules detected on CT scans: a statement from the Fleischner Society. Radiology 2005;237(2):