opinions/hypotheses

Managing the Small Pulmonary Nodule Discovered by CT*
Daniel M. Libby, MD, FCCP; James P. Smith, MD, FCCP; Nasser K. Altorki, MD, FCCP; Mark W. Pasmantier, MD; David Yankelevitz, MD; and Claudia I. Henschke, PhD, MD

Objectives: To review the Early Lung Cancer Action Project experience and the medical literature from 1993 to 2003 on detection of the small, noncalcified pulmonary nodule by CT in order to formulate a management algorithm for these nodules. Design: Prospective noncomparative study of smokers without prior malignancy and a review of the medical literature of CT screening of lung cancer. Interventions: Chest CT and, where appropriate, CT observation for nodule growth, antibiotics, CT-guided fine-needle aspiration (FNA) biopsy, fiberoptic bronchoscopy, and video-assisted thoracoscopic surgery (VATS). Results: The following factors influence the probability of malignancy in a CT-detected, small, noncalcified pulmonary nodule: size, change in size, age, smoking history, density, number of nodules, gender, circumstance of the CT, spirometry, occupational history, and endemic granulomatous disease. The two diagnostic techniques most useful in evaluating the CT-detected, small, noncalcified nodule are short-term observation of nodule growth by CT and CT-guided FNA. Due to small nodule size and the frequent finding of nonsolid or part-solid nodules, positron emission tomography, fiberoptic bronchoscopy, and VATS were less useful. Conclusions: Pulmonologists are frequently asked to evaluate the CT-detected, small, noncalcified nodule invisible on standard chest radiography. Immediate biopsy is justified if the likelihood of cancer is high, but if that likelihood is low or intermediate, a period of observation by CT is appropriate. VATS or thoracotomy are rarely necessary for a diagnosis of lung cancer in the CT-detected small pulmonary nodule. (CHEST 2004; 125:1522–1529)
Key words: CT-detected pulmonary nodule; early lung cancer detection; management Abbreviations: CXR chest radiography; ELCAP Early Lung Cancer Action Project; FNA fine-needle aspiration; PET positron emission tomography; 3D three dimensional; VATS video-assisted thoracoscopic surgery

lung is not the most common A lthough cancer,cancerthe leading cause of cancer type of it is death in the United States.1 This may result from
*From the Department of Medicine (Drs. Libby and Smith), Division of Pulmonary and Critical Care Medicine; Department of Surgery (Dr. Altorki), Division of Cardiothoracic Surgery; Department of Medicine (Dr. Pasmantier), Division of Hematology-Oncology; and Department of Radiology (Drs. Yankelevitz and Henschke), Division of Thoracic Radiology, Weill Medical College of Cornell University, New York, NY. Manuscript received May 9, 2003; revision accepted October 16, 2003. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (e-mail: permissions@chestnet.org). Correspondence to: Daniel M. Libby, MD, FCCP, 407 East Seventieth St, New York, NY 10021
1522

intrinsic aggressive biological behavior, late diagnosis and, until recently, little attention to screening for lung cancer. As with other epithelial cancers, there is likely a wide range of biological behaviors intrinsic to lung cancer. In patients whose lung cancer is detected in stage I, the cure rate is 70%.2 However, trials to determine if screening reduced lung cancer mortality in the 1970s and 1980s, utilizing the best technology of the day, chest radiography (CXR), and sputum cytology, did not demonstrate a decrease in lung cancer mortality, perhaps because only 30% of lung cancers discovered with CXR screening were in stage I.3– 6 Fewer than 10% of patients whose lung cancer was discovered by the development of symptoms (no screening) were stage I.4 Over the past 10
Opinions/Hypotheses

if it doubles in volume in 1 month. Number of Nodules: In the initial ELCAP report.org 1523 . The resulting analysis was organized into two sections: factors influencing the probability of lung cancer in a pulmonary nodule. immediate biopsy is no longer recommended for nodules 15 mm found on baseline screening. the use of CT led to increasing recognition of subsolid (nonsolid and part-solid) nodules (Fig 2). For example.5 to 2. It is important to remember that CT offers a two-dimensional view. size. unless three-dimensional (3D) software is available.15. offering promise in what has been a disease with a dismal outlook.8 The importance of size in determining the likelihood of lung cancer in a pulmonary nodule is influenced by the age of the subject and the circumstance of the study. depending on the clinical circumstances) is appropriate (Fig 1. This is a review of the principles of diagnosis and management of small pulmonary nodules gained through our experience in patients with the Early Lung Cancer Action Project (ELCAP) and the subsequent projects (New York ELCAP and International ELCAP).chestjournal. and treat ever-smaller pulmonary nodules discovered on CT performed either for lung cancer detection.7.17 but until CT there were little data on nodules 1. as we lowered the minimum age for lung cancer screening from 60 to 40 years. A partsolid nodule contains a solid component that obliterates the aerated lung and also contains a nonsolid CHEST / 125 / 4 / APRIL. and it is safe to repeat the CT at 1 year in these patients (Fig 1. its growth rate is uncharacteristic of lung cancer.20 although slowergrowing lung cancers have been described. Thus. and this has been confirmed after years of followup. A nodule that grows at a rate consistent with cancer (doubling time of 30 to 360 days) should be sampled for biopsy or resected.16 Materials and Methods We reviewed our experience with lung cancer screening by CT in patients without prior malignancy at the Weill Medical College of Cornell University ELCAP from 1993 to the present. coronary artery disease screening. lung cancer screening studies using CT have detected up to 85% of lung cancers in stage I.8 Change in Size: A nodule that increases in size is assumed to be active. top). it has been recognized that nodule size correlates with the risk of cancer. more than six nodules were thought to indicate inflammatory lung disease. a positive finding was defined as one to six noncalcified pulmonary nodules. We follow up any new nodule on repeat screening and all nodules 5 mm on baseline.7 Synchronous lung cancers are increasingly recognized as CT screening has become more prevalent. If such software is not available. particularly in women who never smoked or quit cigarette smoking many years earlier. we found lung cancer incidence in nodules 15 mm in diameter on baseline screening to be as low as 5%.years. or for other reasons. and multifocal bronchoalveolar cell or adenocarcinoma is often the histology.19 New nodules discovered on a 1-year repeat CT more frequently contain cancer and at smaller size than on the baseline CT. and performed a review of the literature using MEDLINE to evaluate all published studies of CT screening for lung cancer to determine the best approach to the diagnosis and management of the small pulmonary nodule.18 It is difficult to recognize lung cancer by CT in nodules 5 mm in diameter among the many nodules of this Table 1—Factors Influencing the Probability of Cancer in a Pulmonary Nodule Size Change in size (growth) Number Density Circumstance of CT Patient age Gender Cigarette smoking history Spirometry Occupational history Endemic granulomatous disease www. in ELCAP. 2004 Results Factors Influencing the Probability of Lung Cancer in a Pulmonary Nodule Size: Since CXR studies have shown the likelihood of cancer in solitary pulmonary nodules. Density: While CXR only permits identification of solid nodules. which is less accurate. estimates of nodule growth must rely on diameter rather than direct measurement of volume in assessment of doubling time. A small pulmonary nodule was defined as an opacity in the pulmonary parenchyma 3 cm in greatest dimension with a density greater than the surrounding lung tissue.21 A nonsolid nodule (previously termed ground-glass opacity) is a density through which aerated lung parenchyma is visible.0 cm in diameter (Table 1). top). and techniques utilized in the diagnosis of the small pulmonary nodule detected by CT. manage. Lung cancer appears to be rare in nodules 5 mm in size. A period of radiographic observation for growth (with or without antibiotics.7.7–14 Physicians are therefore increasingly asked to diagnose.

particularly tuberculosis (typical and atypical) and mycoses. and the risk of cancer rises as the size of the nodule. most lung cancers are found in solid nodules.21 Although solid nodules are the most common. rises.184 repeat screenings). it may contain premalignant lesions such as atypical adenomatous hyperplasia or bronchoalveolar hyperplasia. The partsolid lesion often contains invasive adenocarcinoma in the solid component. Despite the lower proportion of solid nodules that are cancer. the new repeat screen CT-diagnosed nodules resolved 4 to 6 weeks later.7 However. it was found that in 12 of the 30 patients. large-cell anaplastic. a course of antibiotics has been suggested (with a repeat diagnostic CT 4 to 6 weeks later).5 cm in diameter to contain lung cancer (approximately 40 to 50%). carcinoid (typical or atypical) or. neuroendocrine. The nonsolid nodule may represent. Part-solid nodules are the most likely of nodules 1. a lower proportion are cancer.component. the number of new nodules discovered on the repeat screening study a year later was only 30 (2. particularly the solid component. There appears to be a progression from nonsolid to part-solid to solid nodule. in some instances.8 For a new nodule discovered on the 1-year interval screening study. Circumstance of the CT Study: If chest CT is performed to screen for lung cancer.000 subjects undergoing the initial baseline screen CT. increases. inflammatory disease. Top: Prebiopsy algorithm for management. 1524 . Noguchi et al23 devised a histopathologic classification (types A to D) describing progressively greater malignant potential and propensity to locoregional metastases. particularly adenocarcinoma with bronchoalveolar cell features. permitting their designation as benign. the proportion that is cancer increases. small-cell carcinoma. Inflammatory diseases of the lung. 233 subjects (23.5 cm.3%) had one to six noncalcified pulmonary nodules. the initial baseline study may contain nodules resulting from prior inflammatory disorders or lung cancer. Nonsolid nodules possess a relatively low risk of cancer (approximately 15%). indeed. but as solid nodule size increases. Bottom: Postbiopsy algorithm for management. In the ELCAP series of 1. because there are many more solid than subsolid nodules. 7 Opinions/Hypotheses Figure 1. as their size rises 1. Only approximately 15% of solid nodules 1 in diameter contain cancer. in others.8 Of the 18 patients with new persistent nodules on the 1-year interval screening study. the risk of malignancy. Solid nodules containing cancer may contain adenocarcinoma or other cell types such as epidermoid. least likely.5% of 1. generally produce solid nodules that can be expected to eventually calcify.22 The nonsolid nodule may also represent bronchoalveolar carcinoma or invasive adenocarcinoma with bronchoalveolar cell features.

Studies of CT screening for lung cancer from Japan9. www. Bottom: Solid nodule.1 This depends. which contains a solid component and a nonsolid component. Gender: As cigarette smoking has increased in women. however. given the same risk profile of age and smoking history. aerated lung parenchyma is visible within the nodule.chestjournal. filter cigarettes inhale more deeply and retain smoke in more peCHEST / 125 / 4 / APRIL. Chest CT performed incidentally in an individual without respiratory signs or symptoms probably possesses the same chance of containing a pulmonary nodule as a baseline screening study (23%).subjects (39%) had lung cancer. Top: Nonsolid nodule. It is thought that smokers of low-tar. it is common to encounter patients with newly diagnosed lung cancer who stopped smoking many years or decades earlier.10. the cell type most closely associated with cigarette smoking. The location of the primary tumor has mirrored these changes in histology and smoking habits. a baseline screening CT is much more likely to have a false-positive result than the 1-year interval screening study.13 that included patients aged 40 years had a much lower incidence of cancer in pulmonary nodules than did screening studies7. 2004 Figure 2. has become the least common histology. Middle: Part-solid nodule.24 Women may be at higher risk than equally smoking men of the same age. After age 80 years. the nodules discovered would contain the same proportion of lung cancers as a baseline screening study (12%). Cigarette Smoking History: The incidence of lung cancer directly correlates with the pack-years of cigarettes smoked.org 1525 . but its incidence increases steadily from 40 to 80 years.12. the incidence of lung cancer levels off and may decline.27 Small-cell carcinoma. The change in smoking habits in the United States toward low-tar and filter cigarettes has influenced the clinical presentation of lung cancer. Changing smoking habits are associated with the rise of adenocarcinoma to the most common histologic subtype. Age: Lung cancer is rare before the age of 40 years. but it does not drop to the levels of individuals who never smoked. Consequently.25 While it had been thought that smoking cessation produced a progressive drop in lung cancer incidence. Density of pulmonary nodules seen on CT. a much higher proportion than was found at baseline (27 of 233 subjects [12%]).8 from the United States that included patients 60 years old. Thus. on the age when smoking started. this concept has been questioned. perhaps related to genetically linked differences in enzymes involved in the detoxification of carcinogens in cigarette smoke. for peripheral adenocarcinomas have become much more common than large airway lesions.26 It appears that the incidence of lung cancer stops increasing after smoking cessation. the incidence of lung cancer has risen.

older women in particular. Part-solid or solid nodules 5 to 9 mm in size and some nodules 10 mm in size have an intermediate likelihood of malignancy.11 Although this provides for a higher risk group. has been observed in apparently healthy people. A nodule 5 mm in diameter or a nonsolid nodule 5 to 9 mm in diameter possess a very low risk of malignancy. This has led to the recommendation that the spirometric finding of an obstructive ventilatory impairment should be present with equal frequency (stratification) in studies comparing screening with no screening and in lung cancer prevention studies.30 Endemic Granulomatous Disease: In studies of CXR screening for lung cancer in the 1950s and 1526 1960s.7 as an entry criteria. It is fortuitous that CT has come into common use concurrently with this change in smoking habits in the United States.7–14 Spirometric Findings: There is a statistically significant association between a spirometrically demonstrated obstructive ventilatory impairment and lung cancer. a higher false-positive rate will be an expected finding due to the high incidence of tuberculosis. top). an FEV1/FVC of 0. When CT screening for lung cancer was performed in the Midwest United States. most lung nodules were due to Mycobacterium tuberculosis. thus explaining the different proportions of patients with cancer in pulmonary nodules reported from Japan. and is often associated with peripheral bronchiectasis and mucoid impaction.ripheral portions of the lung (and thus acquire peripheral adenocarcinomas) than do smokers of nonfilter. and Europe. usually due to Mycobacterium avium intracellulare or Mycobacterium kansasii. Occupational History: Asbestos exposure. that it may circumvent the need for other diagnostic techniques (see below). after a latency period of 20 to 40 years. Workers exposed to respirable radioactive gas in the production and disposal of fissionable materials have an increased risk of lung cancer. Uranium miners and individuals working with heavy metals such as cadmium and nickel are known to have an increased lung cancer risk. with or without the use of 3D software. In order to study the highest risk group possible (and thereby limit the false-positive rate of the screening test).29 International ELCAP is currently studying this group with CT. in studies currently underway in China. Individuals with idiopathic pulmonary fibrosis and pneumoconioses probably have an increased risk of acquiring adenocarcinoma or bronchoalveolar cell carcinoma. particularly lung cancer involving large airways. therefore. high-tar cigarettes (who acquire central airway lesions of other cell types). As the nationwide incidence of tuberculosis has declined. If doubling times (estimated by conventional CT or 3D software) are in the range of 30 to 360 days. M tuberculosis has become a less common etiology of pulmonary nodules.11 Similarly. some investigators have included. since it is more accurate in detecting peripheral lung cancers. a repeat CT in 6 weeks to assess for nodule growth or resolution (with or without antibiotics) may be useful (Fig 1. Volumetric assessment of nodule size may permit estimation of doubling times within 6 to 12 weeks when change in size is not apparent on standard two-dimensional CT Table 2—Low-Risk Noncalcified Pulmonary Nodules Detected by CT Nodule with specific benign diagnosis after biopsy Patients with nodules that have resolved with or without antibiotics Nodules 5 mm in diameter Nonsolid nodules 5 to 9 mm in diameter Opinions/Hypotheses . so ELCAP recommends a repeat CT in 1 year (Table 2). the United States. smokers with reduced FEV1/FVC ratios have a higher risk of acquiring lung cancer. although the risk has not been as well quantified as the asbestos risk. Atypical mycobacterial infection. along with age and smoking history. it biases these studies toward including a higher percentages of patients with large airway cancers. than smokers without an obstructive ventilatory impairment. Lung cancer screening studies that include smokers and nonsmokers would be expected to show a lower incidence of cancer in pulmonary nodules than studies that require entrants to have smoked 10 to 20 pack-years. predisposes to lung cancer acting synergistically with the risk posed by cigarette smoking.28 After controlling for the number of cigarettes smoked. a high false-positive rate was observed. biopsy or resection may be reasonable options. If a nodule does not grow in volume in 6 months. except notably in immigrants from endemic areas and in HIV-positive individuals. Diagnostic Techniques for Managing the Small Pulmonary Nodule Detected by CT CT Assessment of Nodule Growth: Assessment of nodule growth by CT has become such an important predictive parameter. an area endemic with histoplasmosis. the risk of malignancy is very small ( 10%). Further CT assessment for nodule growth after 3 months is reasonable for those nodules that are “nonspecific benign” on fine-needle aspiration (FNA) and for those nodules that partially resolve or do not change after antibiotics on the 6-week CT.

nocardia. particularly if it is solid and 1 cm in diameter. Flexible Fiberoptic Bronchoscopy: Flexible fiberoptic bronchoscopy has limited usefulness in the diagnosis of small peripheral nodules. The treatment implications of a malignant diagnosis such as carcinoma or lymphoma are clear. nodules of 5 to 15 mm in diameter are found. The medical risks of biopsy by FNA. Fluoroscopically guided FNA was popularized in Sweden in the 1970s. repeat biopsy or resection are indicated. an infectious etiology may be present. The sensitivity of PET in adenocarcinoma with alveolar cell features has also been questioned.33 False-positive PET findings in active granulomatous disease and benign tumors have been described. and the financial expense of positron emission tomography (PET) may be avoided by rapid and accurate assessment of nodule growth. 2004 1527 . If the nodule is judged to have an intermediate likelihood of being lung cancer.images.34 Clinically occult extrathoracic metastases or synchronous extrathoracic primary malignancies may be discovered by PET. If lung cancer is highly suspect and the patient is considered fit for thoracic surgery. spirometric findings. Biplanar fluoroscopy and the introduction of biopsy needles procuring a core of tissue for histopathologic evaluation were further refinements. While this represents a great improvement over no screening. PET: PET using fluorodeoxyglucose is useful in assessing the likelihood of malignancy in a pulmonary nodule. etc. nonsolid lesions often produce a normal appearance on surgical inspection and may prove difficult to localize during VATS. biopsy should be undertaken. or surgery. but if the nodule is subsolid or 1 cm in diameter. less than one third of lung cancers were discovered in stage I.34 PET has limited usefulness in early lung cancer detection where many nodules are 1 cm in diameter and may be subsolid. or study performed for other reasons). Percutaneous FNA Biopsy: FNA has been used in the diagnosis of pulmonary nodules for 25 years (Fig 1). a fact documented in many thousands of patients in the United States.31 With CT. VATS: VATS may be useful in the diagnosis of peripheral pulmonary nodules eluding diagnosis by other techniques. A nonspecific benign diagnosis. the number of nodules. and the density of the nodule (solid. particularly if there is high fluorodeoxyglucose uptake. noncalcified nodule is not available. Europe. and CT-guided FNA has become a very important diagnostic technique. the physician must first assess the likelihood of lung cancer. Fig 1). requires careful clinical and radiographic follow-up. www. under present “routine care. it is not sufficient. A standard uptake value 3 is sensitive and specific for cancer. Discussion Lung cancer is common and.org PET may be helpful in detecting mediastinal lymph node metastases even when the nodes are not enlarged on CT.7–14 Faced with a small. For a nodule known to be new by CT. without organisms on smear or culture). endemic granulomatous disease. nonsolid. In CXR screening. and it can be helpful in diagnosing infectious diseases presenting with focal pulmonary nodules such as tuberculosis (typical or atypical) or mycoses. inflammation (granulomatous or other. and nonspecific benign. and empiric antibiotics may CHEST / 125 / 4 / APRIL. If further growth occurs after a nonspecific benign diagnosis is obtained from FNA. It is most accurate in peripheral (within 2 cm of the pleura) solid lesions. CT finds lung cancer in stage I 85% of the time. either PET or a period of radiographic observation with a repeat CT in 6 to 12 weeks should be suggested. mycoses. In institutions where CT-guided FNA is not performed. or benign intrapulmonary lymph node do not require surgery and may be medically treatable or safe to observe. nodule size. sensitivity and specificity decline. and Japan. an initial clinical impression is formulated (Table 1. atypical cells. smoking history. Once a diagnosis of lung cancer has been established in a solid nodule. specific benign. the circumstances of the CT (baseline screen.” commonly fatal. CT-detected pulmonary nodule. the presence of radiographic or clinical signs of inflammatory lung disease such as bronchiectasis or atypical tuberculosis. however. Specific benign diagnoses such as tuberculosis (typical or atypical). Three types of results may be obtained from FNA: malignant.36 Fiberoptic bronchoscopy has an important role in the uncommon circumstance of endobronchial lesions discovered on screening CT. interval 1-year screen.32 although its sensitivity and specificity in the very small CT-detected nodules 5 to 10 mm in diameter need verification. VATS and thoracotomy will probably be performed for benign nodules with somewhat greater frequency if CT-guided biopsy of the small. Nonspecific benign diagnoses might include atypical bronchoalveolar hyperplasia.chestjournal.35 although transbronchial needle aspiration biopsy increases the sensitivity. hamartoma. occupational history. there will likely be greater reliance on CT assessment of nodule growth. videoassisted thoracoscopic surgery (VATS). Utilizing the parameters of age. or part solid).

colorectal. Available at: http://www. Yankelevitz DF. Li F. Chest 1992. and the judicious use of empiric antibiotics and biopsy. Evaluation and management of solitary multiple pulmonary nodules.be justified. Mass screening for lung cancer with mobile spiral computed tomography scanner. while part-solid nodules are more commonly bronchoalveolar carcinoma and solid nodules are invasive adenocarcinoma. Zaman MB. and granuloma without organisms on special stains must be viewed with suspicion. and clinical and radiographic follow-up are essential. Mirtcheva R. management decisions are clear. Sloan JA. 351:1242–1245 11 Swensen SJ. Sanderson DR. et al. 89:3245–3265 6 Kubik A. Recent trends in lung cancer mortality in the United States. Survival and mortality from lung cancer in a screened population: the Johns Hopkins study. et al. either biopsy or follow-up CT may be reasonable. Am J Respir Crit Care Med 2002.org. cellular atypia. The effect of surgical treatment on survival from early lung cancer: implications for screening. et al. Clin Chest Med 1993. Br J Cancer 2001. International Early Lung Cancer Action Project. Some lung cancers appear to begin in areas of bronchoalveolar hyperplasia and subsequently develop features of bronchoalveolar carcinoma and. Woolner LB. particularly the widespread application of volumetric analysis capable of estimating doubling times in 6 to 8 weeks and the development of multislice scanners capable of producing extremely thin slices. finally. if it grows. et al. specific management recommendations are likely to change. 45: 26 –33 7 Henschke CI. Accessed March 17. et al. Accessed March 17. 165:508 –513 12 Sobue T. Cytopathologic analysis of FNA specimens and histopathologic study of resected lesions have changed our concept of the natural history of some lung cancers. the risk of cancer in the nodule also increases. McCauley DI. 84:25–32 10 Sone S. small pulmonary nodules may not be a wise practice. Tarone RE.NYELCAP. CT Opinions/Hypotheses . no further evaluation is warranted. 201:798 – 802 14 Diederich S. Kaneko M. 92:153–159 9 Sone S. depending on the skill and experience of the radiologist and cytopathologist. 28:746 – 750 4 Melamed MR.IELCAP. physicians will be increasingly asked to manage the small pulmonary nodules discovered by CT in clinical practice. While a definitive answer to the question of whether CT screening reduces lung cancer mortality may not be forthcoming for several years. Naidich DP. Smith JP. Chu KC. blood. Yankelevitz DF. patients are spared unnecessary surgery. Peripheral lung cancer: screening and detection with low-dose spiral CT versus radiography. if it continues to grow despite benign biopsy findings. 14:111–119 18 Zerhouni EA. Chest 1986. Semik M. Results of three-year mass screening programme for lung cancer using mobile low-dose spiral computed tomography scanner. Flehinger BJ. et al. Lack of benefit from semi-annual screening for cancer of the lung: follow-up report of a randomized controlled trial on a population of high-risk males in Czechoslovakia. et al. Screening for lung cancer with low-dose spiral computed tomography. Early Lung Cancer Action Project: overall design and findings from baseline screening. 2004 17 Lillington GA. J Natl Cancer Inst 2001. Wormanns D. 1528 References 1 Jemal A. bronchoalveolar hyperplasia. When the result of the biopsy is clearly malignant or when a specific benign diagnosis is made from the FNA. J Occup Med 1986. depending on clinical factors. CT-guided FNA has been a great advance in the diagnosis of small pulmonary nodules. PET. Ohmatsu H. et al. et al. Khlat M. Yankelevitz DF. it must still be viewed with suspicion and either sampled again for biopsy or resected. Screening for early lung cancer with low-dose spiral CT: prevalence in 817 asymptomatic smokers. Widespread use of antibiotics for CTdetected. Lung Cancer 2002. Yang ZG. Nodules 5 mm in diameter may be amenable to biopsy by this technique. 20:911–920 13 Kaneko M. et al. Radiology 1986. Yankelevitz DF. 354:99 –105 8 Henschke CI. et al. 160:319 –327 19 Henschke CI. Eguchi K. 101:1013–1018 3 Fontana RS. By combining careful radiographic follow-up by CT. Early Lung Cancer Action Project: initial findings on repeat screenings. Jett JR.37 This may be analogous to the pathologic progression in breast. Lung cancer screening: the Mayo program. Moriyama N. Melamed MR. Determination of malignancy in small pulmonary nodules based on volumetrically determined growth rates: preliminary results. nonspecific benign findings such as inflammation. Li F. as it may exacerbate the problem of antibiotic resistance. Available at: http://www. Stilik FP. it is most likely benign. Radiology 2000. Reeves A. et al. 93:277–283 2 Flehinger BJ. Radiology 2002. Chest 1984.org. As technologic advances are applied to CT. bronchial cells. Screening for early lung cancer: results of the Memorial Sloan-Kettering study in New York. If resolution of the nodule occurs. Radiology 1996. 217:251–256 21 Henschke CI. CT of the pulmonary nodule: a cooperative study. Lancet 1998. et al. Siegelman SS. 86:44 –53 5 Tockman MS. Nonsolid nodules commonly contain premalignant bronchoalveolar hyperplasia. particularly if nonspecific benign. Int J Cancer 1990. invasive adenocarcinoma. and radiologic-pathologic correlations have emerged. Lancet 1999. Screening for lung cancer: the Early Lung Cancer Action Project approach. However. et al. 35:143–148 20 Yankelevitz DF. biopsy should then be done. As the size of any of these radiologic types of nodule increases. Kostis W. If a nodule fails to change over a 2-year period on CT. and cervical carcinomas. If no change occurs. Parkin DM. 222:773–781 15 New York Early Lung Cancer Action Project. Takashima S. Cancer 2001. Screening for lung cancer with low-dose helical computed tomography: anti-lung cancer association project. J Clin Oncol 2002. 2004 16 Henschke CI. Kimmel M. et al. Caskey CI.

22 23 24 25 26 27 28 screening for lung cancer: frequency and significance of part-solid and nonsolid nodules. JAMA 2001. 178:1053–1057 Vasquez M. Kuschner WG. 4:730 –744 Westcott JL. Jaurand MC. 1997. 1994. In: Samet JM. et al. Tobacco smoking as a possible etiologic factor in bronchogenic carcinoma: a study of 684 proved cases. Rao N. Flannery JT. 202:97–103 Yankelevitz DF. 397– 412 Lubin JH. Kawasaki M. MD: US Government Printing Office. Travis WD. Accuracy of positron emission tomography for diagnosis of pulmonary nodules and mass lesions: a meta-analysis. Morikawa A. Flieder D. 87:817– 827 Pairon JC. CT-guided transthoracic needle biopsy of small solitary pulmonary nodules. 2004 1529 . Cigarette smoking and changes in the histopathology of lung cancer. New York. Early Lung Cancer Action Project pathology protocol. et al. Small adenocarcinoma of the lung. J Clin Pathol 2001. The value of transbronchial needle aspiration in the diagnosis of peripheral pulmonary lesions. 116:704 –708 Kerr KM. Graham EA. 2004 Wynder EL. Brochard P. 89:1580 –1586 Tockman MS. National Cancer Institute Surveillance. Accessed March 17. 285:914 –924 Goldsmith S. et al. AJR Am J Roentgenol 2002. 38:511–524 Torrington KG. 501–530 Thun MJ. Former cigarette smoking and mortality among US veterans: a 26-year followup. Kostakoglu L. Garfinkel L.seer. et al. Weber J. et al. Silica and lung cancer: a controversial issue. 1954 –1980. Nuclear medicine imaging of lung cancer. J Natl Cancer Inst 1997. Lung cancer in radon-exposed miners and estimation of risk from indoor exposure.gov. Samet J. Tamm M. Bethesda.org CHEST / 125 / 4 / APRIL. McLaughlin JK. Epidemiology. Koizumi JH. Colley DP. et al. Henschke CI. and End Results (SEER) program 2001. Available at: http://www. Other host factors and lung cancer suscepti- 29 30 31 32 33 34 35 36 37 bility.chestjournal. Pulmonary preinvasive neoplasia. Transthoracic needle biopsy of small pulmonary nodules. eds. et al. Chest 1999. Epidemiology of lung cancer. Radiology 1997. Changes in cigaretterelated disease risks and their implication for prevention and control. Chest 1993. In: Burns D. et al. Lally CA. 104:1021–1024 Reichenberger F. Maclean CC. 75:2844 –2852 National Cancer Institute. The utility of fiberoptic bronchoscopy in the evaluation of the solitary pulmonary nodule.cancer. Edling C. Cancer 1995. NY: Marcel Dekker. JAMA 1950. ed. 143:329 –336 Hrubec Z. Radiol Clin North Am 2000. 54:257–271 www. 21:107–110 Gould MK. Clin Imaging 1997. Lung Cancer 2004 (in press) Noguchi M. Kern JD. Eur Respir J 1991. J Natl Cancer Inst 1995. Boice JD Jr.