Proportion and characteristics of transient nodules in a retrospective analysis of pulmonary nodules

Thoracic Cancer ISSN 1759-7706 ORIGINAL ARTICLE Proportion and characteristics of transient nodules in a retrospective analysis of pulmonary nodules Jin-Yeong Yu 1, Boram Lee 1, Sunmi Ju 1, Eun-Young Kim 1, Yoon-Hee Kim 1, Su-Young Chi 1, Hee-Jung Ban 1, Yong-Soo Kwon 1, In-Jae Oh 1, Kyu-Sik Kim 1, Yu-Il Kim 1, Sung-Chul Lim 1, Song Choi 2, Yun-Hyeon Kim 2 & Young-Chul Kim 1 1 Department of Pulmonology and Critical Care Medicine, Chonnam National University Medical School, Hwasun Hospital, Jeonnam, South Korea 2 Department of Radiology, Chonnam National University Medical School, Hwasun Hospital, Jeonnam, South Korea Keywords ground glass opacity; pulmonary nodule; solid; transient. Correspondence Young-Chul Kim, Lung Cancer Clinic, Pulmonary Medicine, Chonnam National University Medical School, Hwasun Hospital 160, Ilsim-ri, Hwasun, Jeonnam 519-809, South Korea. Tel: +82 61 379 7614 Fax: +82 61 379 7619 Email: kyc0923@chonnam.ac.kr Received: 14 November 2011; accepted 12 December 2011. doi: 10.1111/j.1759-7714.2011.00101.x Abstract Background: Pulmonary nodules manifest as pure or mixed ground glass opacities (GGOs), or solid nodules. Methods: We retrospectively surveyed 317 cases with pulmonary nodules to observe the proportion and predictive factors of transient lesions in patients with pulmonary nodules. Results: At the initial computed tomography scan, 63.7% showed solid nodules, while 20.2% had mixed GGOs and 16.1% of cases manifested as pure GGOs. Nodules from 114 cases (36%) disappeared or decreased in size during follow up, while in 203 cases (64%), they did not change or became enlarged. During follow up, more than half of the GGOs resolved (66.7% in pure GGOs, 54.7% in mixed GGOs), while only 22.3% of solid nodules resolved. Between transient and persistent pulmonary nodules, significant differences were observed in age, gender, smoking history, presence of eosinophilia, size, and radiologic attenuation of nodules (solid or GGO). In multivariate analysis, age ( 55 years), size of nodules (>15 mm), eosinophilia, and GGO were significant independent predictors of transient nodules. The main causes of transient nodules were pneumonia or eosinophilic pulmonary infiltrates. Conclusion: Thirty-six percent of pulmonary nodules resolved spontaneously or with medical treatment. Transient nodules showed different clinical and radiological characteristics from persistent nodules. Introduction In the National Lung Screening Trial, 1 the positive screening rate was 24.2%; 96.4% of the positive screening results were falsepositives.however,asa20%reductionintherateof death from lung cancer resulted from the use of low-dose computed tomography (CT) screening, chest CT scanning is likely to become a routine screening procedure. With the popularity of chest CT scanning, differential diagnosis of pulmonary nodules has become a great burden to chest physicians. CT scans can detect small nodules 2,3 and pure or mixed ground glass opacities (GGOs), 4 as well as large solid nodules. The major question that follows detection of a pulmonary nodule is whether the lesion may be malignant or not. According to the size, radiologic features, and clinical settings, nodules can be biopsied or followed with another CT scan. 5 7 However, nodules can regress spontaneously or disappear with a course of antibiotics. In a series of 186 cases with GGO lesions, 8 37.6% of pure GGOs and 48.7% of mixed GGOs had resolved at the time of a follow up CT scan, and peripheral blood eosinophilia was highly predictive of transient lesions. We performed this retrospective analysis to observe the proportion and predictive factors of transient lesions in a cohort of patients who visited our institution for the differential diagnosis of pulmonary nodules. Subject and methods Three hundred and seventy-nine patients with pulmonary nodules were screened by a search of electronic medical 224 Thoracic Cancer 3 (2012) 224 228 2011 Tianjin Lung Cancer Institute and Blackwell Publishing Asia Pty. Ltd

J-Y. Yu et al. Transient pulmonary nodules Assessed for eligibility (n=379) Excluded (n=31) (n=348) Excluded (n=31) gratory nodules* (n=7) Table 1 Clinical and radiological findings Total number 317 Age* (range) 57.1 11.3 (30 86) Sex (M/F) 197 (62.1%)/120 (37.9%) Smoking (Yes/No/Missing) 171 (53.9%)/137 (43.3%)/9 (2.8%) Follow up months* (range) 4.7 6.3 (0.2 46) Nodule size (mm)* (range) 14.8 11.6 (2 90) Solid/mixed GGO/pure GGO 202 (63.7%)/64 (20.2%)/51 (16.1%) Multiplicity (Solitary/Multiple) 169 (53.3%)/148 (46.7%) Transient Nodules (n=114) Pneumonia (n=69) Eosinophilic Infiltrates (n=38) Tuberculosis (n=4) Unknown (n=3) Analysed (n=317) Persistent nodules (n=203) Unknown (n=97) Malignancy (n=32) Fibrosis (n=15) Granuloma (n=48) Others (n=11) Figure 1 Consort diagram in this study. *The target lesion disappeared, but new lesions of a similar pattern occurred in other locations. coal worker s pneumoconiosis, aspergilloma with peribronchiolar lymphoid hyperplasia, scleosing hemangioma (2), pulmonary hamartoma, chondroid hamartoma, chronic inflammation (4), silicotic nodule. records from June 2004 to March 2011. Inclusion criteria for this analysis were pulmonary nodules surrounded by lung parenchyma or pleura, nodule diameters of less than 10 cm at the baseline CT, and a follow-up CT scan performed before biopsy or surgery. Typically, focal pulmonary lesions that are 3 cm in diameter are called nodules, and those that are >3 cm in diameter are called masses. 7,9 However, both were referred to as nodules in this study. Cases for evaluation of definitive lung cancer were not included for this survey, as they proceeded to biopsy and staging work up. A total of 317 cases were analyzed for this study, after exclusion of 31 cases which were biopsied without CT scan follow-up, and another 31 cases who were lost to follow-up or showed migratory nodules (Fig. 1). Pulmonary nodules were classified as solid, mixed GGO, or pure GGO nodules. Nodules were measured for the longest diameter, and for multiple nodules, the largest diameter was recorded. Eosinophilia was defined as a peripheral blood eosinophil count greater than 450/mm 3 and neutrophilia as a peripheral blood neutrophil count greater than 7500/mm 3. Peripheral blood cell count data and smoking history were not available in 46 and nine patients, respectively. At follow up CT scan, nodules which were decreased in size or disappeared were classified as transient nodules, and nodules which were enlarged or stable in size were grouped as persistent nodules. Statistical analyses were performed using SPSS 18.0. The Student s t-test, analysis of variance and Kruskal-Wallis test were used to compare age and diameter of nodules. The Chisquare test was used to compare sex, smoking history, eosinophilia, neutrophilia, and radiologic features of nodules. Univariate and multivariate logistic regression analyses were used to find predictive markers of transient nodules. Statistical significance was set at P < 0.05. The study protocol was approved by the Institutional Review Board of Chonnam National University Hwasun Hospital. Results Among 317 subjects, male patients comprised 62.1% (n = 197), and 53.9% (n = 171) of patients were smokers. The age distribution was 57.1 11.3 years (mean standard deviation, range 30 86), and mean follow-up duration was 4.7 months. Multiple nodules were found in 46.7% of cases (Table 1). At the initial CT scan, 63.7% showed solid nodules, while 20.2% had mixed GGOs, and 16.1% of cases manifested as pure GGOs. Cases with eosinophilia were more likely to have GGO nodules than solid nodules. Eosinophilic pulmonary infiltrates were the most common cause of pure GGOs (Table 2). Nodules from 114 cases (36%) disappeared or decreased in size during follow up, while in 203 cases (64%), they did not change or became enlarged. During follow up, more than half of the GGOs resolved (66.7% in pure GGOs, 54.7% in mixed GGOs), while only 22.3% of solid nodules resolved (Table 3). Final diagnoses for persistent nodules (n = 203) were made by percutaneous needle biopsy (n = 8), by bronchoscopy (n = 27), by surgical biopsy (n = 34), by cytology of pleural fluid (n = 1), or by radiologic findings only (n = 133). Thirty-two cases were confirmed to have primary (n = 25) or secondary (n = 7) lung malignancy (15.8% among persistent nodules). Other diagnoses were fibrosis (n = 15), granuloma (n = 49), unknown (n = 97), and others (Fig 1). The main causes of transient nodules were pneumonia (60.5%, n = 69) or eosinophilic pulmonary infiltrates (33.3%, n = 38). Other causes were tuberculosis (n = 4), and in three Thoracic Cancer 3 (2012) 224 228 2011 Tianjin Lung Cancer Institute and Blackwell Publishing Asia Pty. Ltd 225

Transient pulmonary nodules J-Y. Yu et al. Table 2 Comparison of clinical characteristics according to radiologic features of nodules Solid Mixed GGO Pure GGO P-value Number (%) 202 (63.7) 64 (20.2) 51 (16.1) Age (years)* 58.0 11.5 57.6 10.1 52.7 10.9 0.009 Sex (M/F) 120/82 40/24 37/14 0.224 Smoking (Yes/No) 108/91 31/30 32/16 0.214 Solitary/Multiple 118/84 17/47 34/17 <0.001 Size (mm)* 13.3 10.7 19.2 13.6 15.6 11.2 0.001 Eosinophilia (Yes/No) 25/141 13/44 20/28 <0.001 Neutrophilia (Yes/No) 10/156 9/48 1/47 0.021 Etiology Unknown (76) Pneumonia (20) Eosinophilic infiltrates (19) Granuloma (41) Unknown (15) Pneumonia (14) Pneumonia (35) Eosinophilic infiltrates (12) Unknown (9) Malignancy (24) Granuloma (6) Malignancy (7) Fibrosis (9) Fibrosis (5) Fibrosis (1) Eosinophilic infiltrates (7) Tuberculosis (2) Granuloma (1) Tuberculosis (2) Malignancy (1) Others (8) Others (3) cases the etiology was unknown. For transient nodules, empirical antibiotic treatment was prescribed in 45.6% (n = 52), praziquantel in 6.1% (n = 6), and anti-tuberculous treatment in 6.1% (n = 7). The rest of the transient cases (42.0%, n = 49) were followed up without any specific treatment. Among 38 cases with eosinophilic pulmonary infiltrates, enzyme-linked immunosorbent assay (ELISA) for parasitespecific antibodies (IgG) showed positive reaction for Clonorchis sinensis (CS, n = 4), Paragonimus westermani (PW, n = 2), both CS and PW (n = 1), and Sparganum (n = 1). Serologic tests for Toxocara canis antigen were performed in six patients, and all of them showed a positive reaction. Between transient and persistent pulmonary nodules, significant differences were observed in age, gender, smoking history, presence of eosinophilia, size, and radiologic attenuation of nodules (solid or GGO, Table 3). In results of multivariate analysis, age ( 55 years), eosinophilia, size (>15 mm), and GGO were significant independent predictors of transient nodules (P < 0.05, Table 4). Discussion In this retrospective analysis, 36% of pulmonary nodules showed transience. More than half of the GGO lesions and about 20% of solid nodules had resolved in follow up CT scans. The high rate of transience represents the prevalence of eosinophilic pulmonary infiltrates in Korea, as about 33% of transient nodules were eosinophilic pulmonary infiltrates in our series. This trend is not restricted to a region, as other reports 8,10 from hospitals in a metropolitan capital city also reported the high prevalence of transient nodules in association with eosinophilia. We believe it is related to a dietary habit of Korean people. However, even in the western population, a certain proportion of lung nodules show transient features. In an early lung Table 3 Clinical and radiological characteristics of transient and persistent nodules Transient nodules (n = 114) Persistent nodules (n = 203) P-value Sex (M/F) 82/32 115/88 0.007 Age (years)* 54.6 10.8 58.5 11.4 0.003 Smoking history (Yes/No) 72/41 99/96 0.028 Eosinophilia (Yes/No) 39/66 19/47 <0.001 Neutrophilia (Yes/No) 7/98 13/153 0.721 Multiplicity (Solitary/Multiple) 53/61 116/87 0.068 Size (mm)* 19.6 14.0 12.2 9.0 <0.001 Shape of nodules Pure GGO 34 17 <0.001 Mixed GGO 35 29 Solid nodule 45 157 226 Thoracic Cancer 3 (2012) 224 228 2011 Tianjin Lung Cancer Institute and Blackwell Publishing Asia Pty. Ltd

J-Y. Yu et al. Transient pulmonary nodules Table 4 Univariate and multivariate logistic regression analysis predicting transience of nodules Odds ratio 95% CI P-value Univariate analysis Sex (M/F) 1.961 1.197 3.213 0.008 Age ( 55/>55) 1.699 1.069 2.700 0.025 Smoking history (Yes/No) 1.703 1.059 2.739 0.028 Eosinophilia (Yes/No) 4.572 2.458 8.503 <0.001 Neutrophilia (Yes/No) 0.841 0.324 2.181 0.721 Multiplicity (Multiple/Single) 1.535 0.968 2.434 0.068 Size (>15/ 15 mm) 2.830 1.747 4.586 <0.001 GGO/Solid 5.586 3.382 9.227 <0.001 Multivariate analysis Age ( 55/>55) 1.884 1.030 3.447 0.040 Eosinophilia (Yes/No) 3.558 1.740 7.275 0.001 GGO/Solid 4.764 2.641 8.591 <0.001 Size (>15/ 15 mm) 2.894 1.566 5.347 0.001 CI, confidence interval; GGO, ground glass opacity. cancer action project series among 41 individuals who had follow-up CT scans within two months of initial CTs, 12 patients (29%) had complete or partial resolution. 11 Thus, the authors suggested that short-term follow-up CT within two months with or without antibiotics may circumvent the need for further evaluation in some individuals. In the National Lung Screening Trial, 1 as high as 24.2% of CT scans detected non-calcified nodules, and 96.4% of nodules were benign. The false positive rate should be higher in ethnic groups with high prevalence of transient lung nodules, and the high false positive rate should be considered before adopting CT screening as a screening procedure. In the present study, we also searched for predictors of transience of nodules. According to multivariate analysis, independent predictors of transient nodules were younger age ( 55 years), greater size (>15 mm), presence of eosinophilia, and GGO lesions. It is clearly understood that younger patients are more likely to have transient nodules compared to elderly people. In many previous studies, the probability of a pulmonary nodule being malignant rises with increasing patient age. 12 14 Generally, larger lesions are more likely to be malignant than smaller ones. 6 However, in this survey, as we excluded definitive cases with lung malignancy, nodules with larger sizes are more likely to be pneumonic or eosinophilic infiltrates. This explains why larger nodules are associated with transience. Thus size parameter could not be used in prospective clinical settings. Presence of eosinophilia was also predicted with the transience of nodules in another report from Korea, 8 in which 45% of GGO lesions were resolved after follow up CT scan, and eosinophilia highly predicted the transient features of GGO lesions. In our survey, cases with eosinophilia were more likely to have GGO lesions than solid nodules. This correlation explains the association of GGO lesions with transience of nodules. Recently, Lee et al. 10 reported a 69.8% of transience rate from 126 partly solid GGO nodules. They also observed that young patient age, blood eosinophilia, lesion multiplicity, large solid portion and ill-defined border were significant independent predictors of transience. This study has several limitations. First, as a retrospective survey, many cases should be omitted, and we should consider selection bias.in clinical settings,there should be a higher proportion of persistent nodules as this study excluded cases with lung cancer. Second, no consistency in CT scanning methods or intervals between first and follow-up CT scanning was maintained. Third, this is a study within the Korean population only, with a high incidence of eosinophilic pulmonary infiltrates. 15 Thus,interpretation of this study should consider characteristics of this specific ethnic population. In conclusion, 36% of pulmonary nodules resolved spontaneously or with medical treatment. Transient nodules showed different clinical and radiological characteristics from persistent nodules. We should consider the proportion of transient nodule in the diagnostic algorithm of CT screening practices. Disclosure No authors report any conflict of interest. References 1 CT Screening for lung cancer: diagnoses resulting from the New York Early Lung Cancer Action Project. Radiology 2007; 243: 239 49. 2 Kaneko M, Eguchi K, Ohmatsu H et al. Peripheral lung cancer: screening and detection with low-dose spiral CT versus radiography. Radiology 1996; 201: 798 802. 3 Henschke CI, McCauley DI, Yankelevitz DF et al. Early Lung Cancer Action Project: overall design and findings from baseline screening.lancet 1999; 354: 99 105. 4 Henschke CI, Yankelevitz DF, Mirtcheva R et al. CT screening for lung cancer: frequency and significance of part-solid and nonsolid nodules. AJR Am J Roentgenol 2002; 178: 1053 7. 5 Libby DM, Smith JP, Altorki NK, Pasmantier MW, Yankelevitz D, Henschke CI. Managing the small pulmonary nodule discovered by CT. Chest 2004; 125: 1522 9. 6 MacMahon H, Austin JH, Gamsu G et al. Guidelines for management of small pulmonary nodules detected on CT scans: a statement from the Fleischner Society. Radiology 2005; 237: 395 400. 7 Gould MK, Fletcher J, Iannettoni MD 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.): 108S 30S. Thoracic Cancer 3 (2012) 224 228 2011 Tianjin Lung Cancer Institute and Blackwell Publishing Asia Pty. Ltd 227

Transient pulmonary nodules J-Y. Yu et al. 8 Oh JY, Kwon SY, Yoon HI et al. Clinical significance of a solitary ground-glass opacity (GGO) lesion of the lung detected by chest CT.Lung Cancer 2007; 55: 67 73. 9 Austin JH, Müller NL, Friedman PJ et al. Glossary of terms for CT of the lungs: recommendations of the Nomenclature Committee of the Fleischner Society. Radiology 1996; 200: 327 31. 10 Lee SM,Park CM,Goo JM et al. Transient part-solid nodules detected at screening thin-section CT for lung cancer: comparison with persistent part-solid nodules. Radiology 2010; 255: 242 51. 11 Libby DM, Wu N, Lee IJ et al. CT screening for lung cancer: the value of short-term CT follow-up. Chest 2006; 129: 1039 42. 12 Toomes H, Delphendahl A, Manke HG, Vogt-Moykopf I. The coin lesion of the lung. A review of 955 resected coin lesions. Cancer 1983; 51: 534 7. 13 Trunk G,Gracey DR,Byrd RB.The management and evaluation of the solitary pulmonary nodule. Chest 1974; 66: 236 9. 14 Bourke W, Milstein D, Giura R et al. Lung cancer in young adults. Chest 1992; 102: 1723 9. 15 Lim JH, Lee KS. Eosinophilic infiltration in Korea: idiopathic? Korean J Radiol 2006; 7: 4 6. 228 Thoracic Cancer 3 (2012) 224 228 2011 Tianjin Lung Cancer Institute and Blackwell Publishing Asia Pty. Ltd