sign in sign up
<<
Home , Lung nodule, Lung tumor

Note: This copy is for your personal non-commercial use only. To order presentation-ready copies for distribution to your colleagues or clients, contact us at www.rsna.org/rsnarights.

The Solitary Pulmonary Nodule1 REVIEW FOR RESIDENTS Ⅲ

Helen T. Winer-Muram, MD The imaging evaluation of a solitary pulmonary is complex. Management decisions are based on clinical his- tory, size and appearance of the nodule, and feasibility of obtaining a tissue diagnosis. The most reliable imaging features are those that are indicative of benignancy, such as a benign pattern of calcification and periodic follow-up

REVIEWS AND COMMENTARY with computed tomography for 2 years showing no growth. Fine-needle aspiration and core biopsy are important procedures that may obviate if there is a specific benign diagnosis from the procedure. In using the various imaging and diagnostic modalities described in this review, one should strive to not only identify small malig- nant tumors—where resection results in high survival rates—but also spare patients with benign disease from undergoing unnecessary surgery.

௠ RSNA, 2006

1 From the Department of Radiology, Indiana University, Indianapolis, Ind. Received February 28, 2005; revision requested April 22; revision received April 28; accepted June 13; final version accepted August 11; final review and update by the author October 31. Address corre- spondence to the author, 11224 Clarkston Rd, Zionsville, IN 46077 (e-mail: [email protected]).

஽ RSNA, 2006

34 Radiology: Volume 239: Number 1—April 2006 REVIEW FOR RESIDENTS: The Solitary Pulmonary Nodule Winer-Muram

solitary pulmonary nodule (SPN) aging features, the radiologist often and 80 years of age (18). Patients with is a round or oval opacity smaller plays a major role in the care of patients the human immunodeficiency virus have Athan 3 cm in diameter that is com- with SPNs. In this article, some of the an increased risk for and pletely surrounded by pulmonary pa- clinical and radiographic features that may develop cancer at a younger age renchyma and is not associated with are important to consider when deter- (19). was once far more , , or pneu- mining the likelihood of of common in men than women, but in- monia (1) (Fig 1a). Larger are an SPN will be reviewed, and an algo- creased smoking rates among women not included in this definition because rithm will be proposed for the care of during the 1960s and 1970s have led to many of these lesions are malignant (2– patients with indeterminate nodules. an increased incidence of lung cancer in 4). An SPN is noted on up to 0.2% of women (20). The American Cancer So- chest radiographs (5,6) (Fig 2a). While ciety estimated that there would be the differential diagnosis for SPN is ex- Risk of SPN Malignancy about 172 570 new cases of lung cancer tensive (Figs 3, 4), most lesions are To understand the rationale underlying in 2005 (93 010 in men and 79 560 in found to be , lung , clinical and imaging work-up when an women). The chance of developing lung or (7,8) (Fig 5). Detection SPN is discovered, one must first recog- cancer is one in 13 in men and one in 18 and work-up of SPNs are critical be- nize the clinical factors that make lung in women. This incidence includes all cause SPNs may be malignant and lung cancer a more likely cause of SPN (Ta- people, and it does not take into ac- cancer has an overall mortality rate of ble). The likelihood of lung cancer in- count whether they smoke (21). up to 85% (3,9). Early detection of creases if a patient has a smoking his- small nodules may potentially reduce tory, and it is directly proportional to lung cancer–specific mortality; in time, the number of pack-years as a smoker SPN Size data from the National Lung Screening (12). While many physicians have be- At chest radiography, an SPN is seldom Trial may be used to prove this hypoth- lieved that smoking cessation produces evident until it is at least 9 mm in diam- esis. a progressive reduction in lung cancer eter (22). Moreover, even larger nod- While one may not be able to estab- incidence, this concept has been chal- ules may be missed with radiography, lish a diagnosis based solely on the im- lenged (13). The incidence of lung can- unless prior chest radiographs are avail- cer does not increase after smoking ces- able for comparison. SPNs are fre- sation, but it never equals that for indi- quently detected because they are ei- Essentials viduals who have never smoked. ther absent on previously obtained Ⅲ While the differential diagnosis Consequently, one commonly sees pa- radiographs or present and not recog- for SPN is extensive, most SPNs tients with newly diagnosed lung cancer nized until the current radiographs are found to be granulomas, lung who stopped smoking years or even de- show enlargement. Nearly 90% of cancers, or hamartomas. cades earlier (14). newly discovered SPNs on chest radio- Ⅲ Benign nodules can be confidently Lung cancer risk also increases if graphs may be visible in retrospect on diagnosed if the is smaller the patient has a history of primary pul- prior radiographs (23). Failure to detect than 3 cm in diameter and exhib- monary or extrapulmonary cancer or an SPN is directly related to obscuration its one of the following patterns of pulmonary fibrosis (eg, idiopathic fibro- of the nodule by overlying structures, calcification: central nidus, lami- sis or fibrosis due to exposure, failure to compare the current radio- nated, popcorn, or diffuse. collagen vascular disease, adult respira- graph with prior radiographs, or use of Ⅲ The probability of malignancy is tory distress syndrome, or radiation) a faulty search pattern (24). Prior chest high (90% if the patient is older (10,15) (Fig 6). An SPN is unlikely to be radiographs are also needed because a than 60 years) with positive FDG a in the absence of a known nodule that is unchanged on chest radio- PET findings and low (Ͻ5%) with prior malignancy, and a routine search graphs for 2 years is almost certainly negative FDG PET findings. for an extrathoracic primary tumor is benign and requires no further imaging. Ⅲ Nodules with low likelihood for not cost-effective (16) (Fig 7). In pa- malignancy that are at least 5 mm tients with , , or tes- and smaller than 10 mm can be ticular , a malignant SPN is observed with CT for a 2-year 2.5 times more likely to be a metastasis period, while nodules with inter- than a primary lung cancer; however, in mediate or high likelihood for ma- patients with head and squamous Published online 10.1148/radiol.2391050343 lignancy can be sampled with cell carcinoma, a malignant SPN is eight FNAB or resected. times more likely to be a primary lung Radiology 2006; 239:34–49 Ⅲ A new persistent nodule that de- cancer (17). Abbreviations: velops during observation is wor- Onset of lung cancer before the age FDG ϭ fluorine 18 fluorodeoxyglucose risome for malignancy and war- of 40 years is rare; however, its inci- FNAB ϭ fine-needle aspiration biopsy rants intervention. dence increases steadily between 40 SPN ϭ solitary pulmonary nodule

Radiology: Volume 239: Number 1—April 2006 35 REVIEW FOR RESIDENTS: The Solitary Pulmonary Nodule Winer-Muram

Key Point larger the nodule (approaching 3 cm in The prevalence of cancer in SPNs Always compare current radiographs diameter), the more likely it is to be smaller than 1 cm in diameter is un- with previous radiographs (if available). malignant. More than 90% of nodules known. Of noncalcified nodules smaller The size of the SPN is not a reliable that are smaller than 2 cm in diameter than 1 cm, 42%–92% have been found to predictor of benignity (4); however, the are benign (10,25). be benign (3,4,26). The large variability reflects selection bias, and reports from Figure 1 surgical series tend to show higher preva- lence of malignant lesions than do reports

Figure 2

Figure 1: Chest CT scans (5-mm section width) in a female 48-year-old former smoker (9 pack-years) with a history of remote purified protein derivative conversion. (a) Transverse cardiac screening scan shows a 10-mm solid nodule (arrow) in the right lower lobe. (b) Transverse thin-section (1.25-mm section width) scan shows irregu- lar margins and central lucency. (c) Thin-section scan shows central lucency (Ϫ208 HU), which indicates air bron- Figure 2: (a) shows an inci- chiolograms or early cavitation. (d) Three-dimensional CT scan obtained with volume rendering shows the nodule dental small nodule (arrow) at the left costophrenic volume to be 531 mm3. FNAB was performed, and atypical cells were seen. Because of the nonspecific diagnosis, angle. (b) Thin-section CT scan shows central fat repeat FNAB was performed, and no malignant cells were seen. The nodule will be observed with serial CT during attenuation (Ϫ43 HU) in the nodule. the next 24 months. If the nodule remains stable for 24 months, no further intervention will be performed. was diagnosed.

36 Radiology: Volume 239: Number 1—April 2006 REVIEW FOR RESIDENTS: The Solitary Pulmonary Nodule Winer-Muram

from screening studies. In 64 patients smaller than 1 cm in diameter were ma- viewing chest images, as most missed with SPNs 1 cm or smaller in diameter lignant (26). lung cancers are located in the right up- who were referred for video-assisted tho- per lobe (24). As benign nodules are racoscopic surgery, 58% of SPNs, includ- Key Point equally distributed throughout the upper ing six that were smaller than 5 mm in Nodules approaching 3 cm in diameter and lower lobes, location alone cannot be diameter, were malignant (27). Other se- are more likely to be malignant, while used as an independent predictor of ma- ries have found a similar frequency of ma- nodules smaller than 1 cm in diameter lignancy (16). In patients with idiopathic lignancy (26–29). In comparison, the are more likely to be benign. pulmonary fibrosis, lung cancers more Early Lung Cancer Action Project screen- commonly involve the periphery of the ing study showed that only 8% of lesions lower lobe, a site in which fibrosis is most SPN Location likely to occur (15). Approximately half of Lung cancer is 1.5 times more likely to primary pulmonary Figure 3 occur in the right lung than in the left lung manifest as isolated peripheral SPNs, (30). Studies have shown that 70% of while squamous cell that lung cancers are located in the upper manifest as SPNs are more likely to be lobes and occur most frequently in the centrally located (33). right lung (31,32). Thus, one should care- Increasing use of low-tar and filtered fully scrutinize the upper lobes when re- cigarettes in the United States has been

Figure 5

Figure 3: CT scan shows a markedly enhanc- ing right lower lobe nodule with a feeding artery and draining vein (not shown). Pulmonary arterio- venous malformation was diagnosed.

Figure 4

Figure 4: CT scan in a 90-year-old woman with chronic congestive heart failure shows a tiny nod- ule adjacent to the right major fissure that is likely to represent a congested intrapulmonary lymph node (arrow). Follow-up CT was not performed because of the patient’s advanced age. Figure 5: Chart shows differential diagnosis of SPN.

Radiology: Volume 239: Number 1—April 2006 37 REVIEW FOR RESIDENTS: The Solitary Pulmonary Nodule Winer-Muram

Figure 6 Figure 7

Hierarchy of Likelihood Ratios for Malignancy Likelihood Characteristic Ratio

Cavity wall thickness (mm) Ͼ16 37.97 Ͼ4–16 0.72 Յ4 0.07 Size (cm) Ͼ3.0 5.23 Figure 7: Transverse CT scan of the chest 2.1–3.0 3.67 shows a solitary right lower lobe nodule. FNAB 1.1–2.0 0.74 of the nodule revealed renal cell carcinoma Յ1.0 0.52 metastasis. PET standardized uptake value Ͼ2.5 4.30 Յ2.5 0.04 Age (y) Figure 8 Ͼ70 4.16 50–70 1.90 30–39 0.24 20–29 0.05 Figure 6: Transverse CT scan in a 75-year-old Growth rate (d) man with idiopathic pulmonary fibrosis shows a Ͼ465 0.01 solid left lower lobe nodule (arrow). FNAB of the 7–465 3.40 nodule revealed squamous cell carcinoma. Ͻ70 Enhancement (HU) Ͼ15 2.32 associated with changes in the histologic Յ15 0.04 type and anatomic distribution of lung Irregular spiculated edge 5.54 cancer, such as decreasing incidence of History of malignancy 4.95 small-cell carcinoma (the cell type most Current smoker 2.27 closely associated with cigarette smoking) Never smoked 0.19 Indeterminate calcification at CT 2.20 and increasing incidence of adenocarci- Upper and/or middle lobe location 1.22 noma (34). Adenocarcinomas are fre- Smooth nodule at CT 0.30 quently peripheral lung tumors rather Benign calcification at CT 0.01 than central lung tumors because people Figure 8: Transverse CT scan shows a 1-cm- who smoke low-tar filtered cigarettes in- Source.—References 10 and 11. diameter left lower lobe nodule with central nidus hale deeply and distribute smoke to the calcification. This finding is indicative of benign lung periphery. People who smoke high- disease. tar unfiltered cigarettes do not inhale as SPNs from malignant SPNs is calcifica- deeply and are more likely to develop tu- tion. Benign nodules can be diagnosed mors associated with central lesions, such confidently if the lesion is smaller than 3 that is clearly seen is likely to be dif- as squamous cell carcinomas (34). cm in diameter and exhibits one of the fusely calcified and benign. For larger following patterns of calcification: cen- SPNs, however, detection of calcifica- Key Point tral nidus, laminated, popcorn, or dif- tion with radiography is less certain. In The right upper lobe is the most com- fuse. When one of these patterns is a study where the mean SPN diameter mon location of lung cancer. seen, the likelihood of benignity ap- was 13 mm, sensitivity and specificity of proaches 100% (3,4). Popcorn calcifica- radiography in the detection of calcifica- tions are observed in one-third of tion were 50% and 87%, respectively SPN Internal Attenuation hamartomas (7), and the other patterns (35) (Fig 9). When computed tomogra- Characteristics are seen with histoplasmosis and tuber- phy (CT) was used, 7% of these defi- culosis (Fig 8). nitely calcified nodules were not calci- Calcification SPNs smaller than 9 mm in diame- fied and were, therefore, potentially The most important imaging feature ter are rarely visible on chest radio- malignant. that can be used to distinguish benign graphs; therefore, a nodule of that size While low-voltage radiography and

38 Radiology: Volume 239: Number 1—April 2006 REVIEW FOR RESIDENTS: The Solitary Pulmonary Nodule Winer-Muram

dual-energy computed radiography the nodule, with standard soft-tissue re- nign calcified lesion (4) (Fig 10). Fur- (and perhaps in the future, tomosynthe- construction. If a benign pattern of cal- thermore, central calcification in a spic- sis) can be used to show calcification cification involving more than 10% of ulated SPN should prompt concern for within SPNs (36,37), CT has largely re- the cross-sectional area of the nodule is malignancy, as most benign SPNs have placed radiography for this purpose. present, malignancy is unlikely and ob- smooth or minimally lobulated margins. Assessing calcification with CT requires servation is appropriate (38). Calcification in lung cancers may ap- the use of sequential thin “cuts” through While CT studies have shown that pear amorphous, stippled, or diffuse up to 13% of lung cancers have some (40). Some lung cancers can have dense Figure 9 calcification (4,39,40), this is true of foci of calcification or be entirely calcified, only 2% of lung cancers smaller than 3 with a pattern resembling that of benign cm in diameter (39). Eccentric calcifica- disease. Both of these patterns can be tion should not be considered a benign seen in , metastatic osteosarco- finding. It may represent a benign lesion mas, and chondrosarcomas (Figs 11, 12). that has calcified in an eccentric fashion A stippled appearance or psammomatous or a malignant lesion that has dystro- calcification can be seen in SPNs that are phic calcification or has engulfed a be- metastases from mucin-secreting tumors, such as colon or ovarian cancers. In pa- tients with a history of these tumors and a Figure 10 benign-appearing SPN, CT cannot be used to reliably determine benignity and biopsy may be necessary. Unfortunately, calcification is often not useful, as about 45% of benign nod- ules are not calcified (4); thus, other imaging features associated with benig- nity must be sought.

Key Point With the exception of SPNs in patients Figure 10: CT scan in an 80-year-old man with a history of bone malignancy, SPNs shows a 2.2-cm-diameter nodule in the left upper lobe with eccentric calcification. FNAB of the nod- with a benign pattern of calcification are ule revealed . indeed benign. Fat Demonstration of fat may be difficult if the nodule is small, as partial volume Figure 11 inclusion of the lung may interfere with attenuation measurements. However, if one can determine that fat is present, hamartoma or lipoma (albeit less likely) become the most likely causes (Fig 2b). Some , such as a metasta- sis from liposarcoma or renal cell carci- noma, may occasionally contain fat (41). In patients without prior malig- nancy, focal fat attenuation (Ϫ40 to Ϫ120 HU) is a reliable indicator of a hamartoma and is seen in over 50% of Figure 9: (a) Chest radiograph shows a right hamartomas at thin-section CT (42). In upper lobe nodule with central calcification. The a series of 47 patients with hamarto- margins are irregular. (b) CT scan shows a right mas, both fat and calcium were seen in upper lobe nodule with irregular margins that 10 and fat alone was seen in 18 (42). represents pulmonary carcinoma (black arrow). Figure 11: CT scan shows eccentric dense The calcification seen on the radiograph is caused calcification in a right lower lobe tumor. Key Point by a calcified anterior to the tumor (Image courtesy of J. W. Gurney, MD, University of Demonstration of fat in an SPN in pa- (white arrow). Nebraska, Omaha, Neb.) tients without a history of liposarcoma

Radiology: Volume 239: Number 1—April 2006 39 REVIEW FOR RESIDENTS: The Solitary Pulmonary Nodule Winer-Muram

Figure 12 Figure 13 or renal cell carcinoma suggests that the SPN is benign.

Nodule Attenuation The advent of CT has led to improved recognition of the frequency with which nodules are nonsolid, partly solid, and solid. Aerated lung parenchyma is visi- ble through a nonsolid (ground-glass) nodule, while a partly solid nodule con- tains solid regions that mask an aerated lung. Approximately 34% of nonsolid nodules are due to malignancy (43). The risk of malignancy increases if the diam- eter of the SPN exceeds 1.5 cm or the nodule is round (43,44). Malignancies such as bronchioloalveolar carcinomas or invasive adenocarcinomas with bron- chioloalveolar cell features may appear to be nonsolid nodules (Fig 13). Non- Figure 12: CT scan shows calcified right lower solid nodules are often caused by benign lobe nodule that resembles a benign granuloma conditions, such as inflammatory dis- (arrow). The patient had a history of osteosarcoma. ease, and may contain premalignant le- Open revealed metastatic disease. Figure 13: CT scan in a 64-year-old man sions, such as atypical adenomatous hy- shows an oval 2.1-cm left lower lobe nonsolid perplasia or bronchoalveolar hyperpla- nodule (arrow). FNAB revealed adenocarcinoma. sia (45). Precursors of adenocarcinoma with increasing diameter (48). While are believed to begin in regions of bron- solid nodules are usually noncancerous choalveolar hyperplasia (46). Noguchi (granulomas), most lung cancers are et al (47) devised a histopathologic clas- found in solid nodules (Fig 6). Histologic Figure 14 sification system in which nodules are types of cancerous solid nodules include stratified according to their malignant adenocarcinomas and squamous cell, potential and propensity for local and large-cell anaplastic, neuroendocrine, regional metastases. carcinoid, and (rarely) small-cell carci- Partly solid nodules are more likely nomas (48). In addition, most meta- to be malignant than nonsolid nodules static nodules are solid in appearance, (Fig 14). Between 40% and 50% of with a partly solid appearance occurring partly solid nodules smaller than 1.5 cm less frequently. in diameter are cancerous, and the risk of cancer increases with increasing nod- Key Point ule size, particularly if the solid compo- While most cancerous nodules are nent is in the center of the nodule solid, partly solid nodules are most (43,44). This solid component often likely to be malignant. contains invasive adenocarcinoma. Although solid nodules are the most Air Bronchograms common type of nodule, they are less Air bronchograms and bronchiolograms likely to be malignant than are partly are seen more commonly in pulmonary solid or nonsolid nodules. Inflammatory carcinomas than in benign nodules (42) diseases of the lung, particularly tuber- (Figs 1b, 1c, 15b). In one review, air Figure 14: CT scan in an 81-year-old man culosis and mycoses, usually produce bronchograms were seen in approxi- shows a 2.8-cm irregular, partly solid left upper solid nodules that may eventually calcify mately 30% of malignant nodules but in lobe nodule with pleural tags. FNAB revealed bronchioloalveolar cell carcinoma. and permit the designation of benign only 6% of benign nodules (49). Air disease. Only 15% of solid nodules bronchiolograms, also referred to as smaller than 1 cm in diameter contain bubble-like lucencies or pseudocavita- malignant foci, but the proportion of tion, may simulate cavities and are seen nodules that contain such foci increases in up to 55% of bronchioloalveolar cell

40 Radiology: Volume 239: Number 1—April 2006 REVIEW FOR RESIDENTS: The Solitary Pulmonary Nodule Winer-Muram

carcinomas (42). This appearance is malignancy, it can occasionally be seen in Cavitation caused by a desmoplastic reaction to the granulomatous disease, lipoid pneumo- Both benign and malignant nodules can tumor that distorts the airways (50,51). nia, organizing , and progres- form a cavity. Up to 15% of lung cancers sive massive fibrosis (4,56). A smooth form a cavity, but most are larger than 3 Margin margin does not indicate benignity, as up cm in diameter (57). However, cavitation Edge characteristics indicative of malig- to one-third of malignant lesions have may be seen in SPNs as small as 7 mm in nancy include irregularity, spiculation, smooth margins and many of these tu- diameter. SPNs with irregular-walled cav- and lobulation (10) (Fig 1b). Edge irregu- mors are metastatic (7,9,54) (Figs 7, 16). ities thicker than 16 mm tend to be malig- larity and spiculation are associated with A lobulated margin indicates that the nant (84%–95% of SPNs), while benign the radial extension of malignant cells along nodule has uneven rates of growth (Fig cavitated lesions usually have thinner interlobular septa, lymphatics, small air- 16). In a series by Siegelman et al (3), smoother walls; approximately 95% of le- ways, or blood vessels and have been lik- approximately 40% of smooth-edged lob- sions with cavity walls thinner than 4 mm ened to the spokes of a wheel (52). ulated nodules were malignant. However, are benign (58–60). However, because In a study with thin-section CT, all close inspection of any nodule that ap- there is much overlap, cavity wall charac- nodules with a halo margin—97% with pears to be lobulated is necessary. Adja- teristics cannot be used to confidently dif- densely spiculated margins, 93% with cent tiny nodules, called satellite nodules, ferentiate benign and malignant SPNs ragged margins, and 82% with lobulated may mimic the appearance of a lobulated (Figs 18, 19). margins—were malignant (53). Nodule margin, and the presence of these nod- halos (peripheral nonsolid component) ules is strongly associated with benignity Key Point should not be confused with the corona (Fig 17). Even so, the presence of satellite Several imaging features (nodule atten- radiata, which is a radiolucent halo asso- nodules does not allow confident diagno- uation, presence of air bronchograms, ciated with paracicatricial emphysema sis of benignity, as 10% of dominant nod- edge characteristics, and cavity wall (52,54). The presence of spiculation has a ules with satellite nodules will be malig- thickness) must be considered when as- predictive value for malignancy of ap- nant (4,52). When cancerous, satellite sessing the likelihood of malignancy; proximately 90% and should prompt nodules are usually the result of periph- however, there is considerable overlap an aggressive work-up (3,4,10,42,55). eral foci of tumor or skip metastatic le- in the appearance of benign and malig- While an irregular margin is indicative of sions (52). nant lesions.

Figure 15 CT Examination In any patient with a newly discovered SPN, a standard CT examination with- out contrast material enhancement may be performed in the chest. This exami- nation will ensure there are no other findings, such as additional nodules, lymphadenopathy, , chest wall involvement, or adrenal mass. Be- cause of concerns about radiation dose to the patient, subsequent follow-up CT may be limited to the nodule location. Thin-section CT scans obtained through the nodule provide information regard- ing nodule size (by using diameters from the largest cross-sectional area or vol- ume measurement), attenuation, edge characteristics, and the presence of calcification, cavitation, or fat (42) (Fig 1b, 1c). Sequential thin-section CT (1– 3-mm section width) performed through the entire nodule with a single Figure 15: (a) Transverse CT scan in a 75-year-old man shows a 2.0-cm-diameter nonsolid left upper lobe breath hold and without contrast mate- nodule. FNAB revealed no malignant cells; thus, no specific diagnosis was made. (b) The lesion was followed rial enhancement should allow these up with serial CT; 25 months later, the nodule was slightly increased in size and had converted to a partly solid features to be analyzed. Demonstration attenuation lesion with air bronchograms. Volumetric measurement showed the doubling time of the opacity of certain findings, such as fat or a be- to be 1375 days. Repeat FNAB showed bronchioloalveolar cell carcinoma. nign pattern of calcification, may be all

Radiology: Volume 239: Number 1—April 2006 41 REVIEW FOR RESIDENTS: The Solitary Pulmonary Nodule Winer-Muram

Figure 16 Figure 17 that is required to ascertain that the nodule is benign (42) (Figs 2b, 8). The cause of many SPNs, however, will remain undetermined after the ini- tial and thin-section CT examinations. If the nodule is at least 10 mm in diame- ter, a contrast material–enhanced ex- amination may be performed. The use of contrast material may be especially helpful in regions where granulomatous disease is endemic. Contrast enhance- ment is directly related to the vascular- ity of the nodule, and blood flow is usu- ally greater in malignant nodules than in benign nodules (61,62). To perform the study, the nodule is examined with 3-mm collimation before and after ad- ministration of a weight-related dose of intravenous contrast material. Con- Figure 17: CT scan shows a right lower lobe trast-enhanced examinations are per- subcentimeter nodule with adjacent tiny satellite formed at 1-minute intervals up to 4 nodules (arrow). The larger nodule was calcified minutes after injection of contrast mate- (not shown). The appearance is that of benign rial. Nodule enhancement is determined granulomatous disease. by subtracting baseline attenuation from peak mean attenuation during the Figure 16: CT scan in a 56-year-old man contrast-enhanced examinations (31). shows a 2.9-cm smooth lobulated central right The use of multi–detector row CT facil- middle lobe nodule (arrow). FNAB revealed squa- Figure 18 itates performance of this examination mous cell carcinoma. because the nodule is likely to be in the examined volume, despite varying breath holds. performed in nodules smaller than 10 Nodule enhancement of less than mm in diameter, cavitary lesions, or 15 HU after administration of contrast lesions with central necrosis. material is strongly indicative of be- nignity (positive predictive value, ap- Key Point proximately 99%). Rare false-negative When performing contrast-enhanced findings are associated with central CT, enhancement of less than 15 HU noncavitating necrosis and adenocar- indicates benignity. cinomas (especially bronchioloalveo- lar cell carcinoma), which may be re- lated to mucin production (31,63,64). Positron Emission Tomography Although enhancement of more than Positron emission tomography (PET) 15 HU is more likely to represent ma- is rapidly becoming a front-line modal- lignancy, only 58% of nodules are ma- ity in the evaluation of SPNs. Its diag- lignant; the remainder represent en- nostic ability is based on increased hancing lesions due to active inflam- glucose consumption of malignant matory disease that have increased cells. The radiopharmeutical fluorine blood flow, such as granulomas or or- 18 fluorodeoxyglucose (FDG) is a Figure 18: CT scan in an 83-year-old man ganizing (31,63,65). En- glucose analogue that is injected intra- shows a 2.3-cm left upper lobe cavitary nodule. hancing nodules should still be consid- venously, transported through the The wall is variable and the cavity wall is as ered indeterminate and require fur- cell membrane, and phosphorylated thick as 8 mm. FNAB revealed squamous cell ther work-up. In summary, nodule through normal glycolytic pathways, carcinoma. behavior after contrast material ad- remaining unmetabolized in the cell ministration is sensitive but not spe- (66). For solid pulmonary nodules 1–3 cific for malignancy. Contrast-en- cm in diameter, sensitivity and speci- hanced CT examinations should not be ficity are approximately 94% and

42 Radiology: Volume 239: Number 1—April 2006 REVIEW FOR RESIDENTS: The Solitary Pulmonary Nodule Winer-Muram

83%, respectively (11). Demonstra- the strategy of observing nodules that in the development of the model is con- tion of a hypermetabolic state, espe- are negative at PET. However, certain ditionally independent of all others. cially in smaller lesions, is of concern , such as carcinoid and bron- Investigators in one study analyzed for malignancy and necessitates either chioloalveolar cell carcinoma, have a many of the characteristics used in the intervention or close scrutiny. The low metabolic rate that may result in model and found that three clinical (age, probability of malignancy in associa- false-negative examinations. Further- smoking history, and history of previous tion with positive FDG PET findings is more, sensitivity and specificity are not malignancy) and three radiographic (di- high (90% if the patient is older than as high for nodules that are smaller than ameter, spiculation, and upper lobe lo- 60 years); likewise, the probability of 1 cm in diameter (69). cation) features are independent pre- malignancy in association with nega- dictors of malignancy (16). Investiga- tive FDG PET findings is low (Ͻ5%) Key Point tors in another study compared the (67,68) (Fig 20). Once lung cancer has FDG PET examination of an SPN larger assessment made with the clinical pre- been diagnosed in a solid nodule, par- than 1 cm in diameter is the best test to diction model with that made by the ticularly if FDG uptake is high, PET determine if a lesion is malignant or be- clinicians. Receiver operating charac- may be helpful in the detection of me- nign. teristic analysis showed no difference diastinal lymph node metastases, even between the assessments. However, when the nodes are not enlarged on physicians overestimated the probabil- CT scans (69). Occult extrathoracic Bayesian Analysis ity of a malignant lesion in patients with metastases and synchronous extratho- Bayesian analysis can be useful in the low risk of malignant disease. The au- racic primary malignancies also may evaluation of SPN. Bayesian analysis in- thors concluded that the logistic model be detected with PET. volves the use of likelihood ratios of nu- could be used to improve the care of False-positive PET findings are asso- merous imaging findings and clinical patients with SPNs that are likely to be ciated with focal infections, inflamma- features associated with SPNs to estab- benign (6). tion, and nonneoplastic diseases (eg, tu- lish the probability of malignancy Not all Bayesian models are effec- berculosis, , and rheumatoid (10,70) (Table). In addition to imaging tive: Comparison of a Bayesian ap- disease) and are more frequent in re- findings, age and smoking history are proach with FDG PET alone revealed gions with endemic fungal diseases such factors that are included in the Bayesian that FDG PET outperformed the model as Histoplasma and Coccidioides infec- analysis. A general principle in Bayesian (68). A standard uptake value of more tions. The high negative predictive value models is that each characteristic used than 2.5 with PET yields a likelihood of PET in low-risk populations supports ratio of 4.30 for malignancy, whereas a negative PET examination has a likeli- Figure 20 hood ratio of 0.04. Bayesian analysis is Figure 19 equivalent or slightly superior to evalua- tion by an experienced radiologist in the stratification of benign and malignant pulmonary nodules (69). An explana- tion of the mathematics of Bayesian modeling is beyond the scope of this article; however, for a particular pa- tient, one can easily determine the probability that an SPN is cancerous at the following Web site: http://www .chestx-ray.com.

Key Point A Bayesian analysis that includes PET findings is an effective method that can be used to calculate the probability of malignancy.

Figure 20: FDG PET scan shows a lingular Growth Rate Measured from Serial Figure 19: CT scan in an 80-year-old man nodule (arrow) with a standardized uptake value of Studies shows a right upper lobe 2.9-cm cavitary nodule more than 2.5. FNAB revealed carcinoid tumor. The notion that an SPN exhibiting no with a smooth, uniform 2.5-mm-thick cavity wall. (Image courtesy of Dr E. Long, Goshen General growth for at least 2 years is benign is FNAB revealed non–small cell lung cancer. Hospital, Goshen, Ind.) widely accepted and is the standard of

Radiology: Volume 239: Number 1—April 2006 43 REVIEW FOR RESIDENTS: The Solitary Pulmonary Nodule Winer-Muram

Figure 21 care (71,72). Yankelevitz and Henschke is close to the CT section width (73) have challenged the basis of 2-year (32,74,76). Another limitation of the stability as a reliable indicator that automated volumetric assessment is the SPNs are benign. They used no growth potential to not include in the measure- as the predictor of benignity and calcu- ment the ground-glass component of a lated the predictive value as 65%, the partly solid nodule, which has a high sensitivity as 40%, and the specificity as frequency of malignancy. 72%. The 2-year rule is still being used; An early study showed that all nod- however, one should exercise caution ules doubling in size in less than 7 days and attempt to obtain radiographs and and almost all nodules doubling in size CT scans older than 2 years because in more than 465 days were benign even a substantial increase in volume (77). Some benign lesions, such as gran- may be missed in small nodules (73). ulomas and hamartomas, grow slowly; The time for a nodule to double in therefore, growth in and of itself cannot volume is referred to as the doubling be used to predict malignancy (78). time. For radiographic measurement, Many investigators have estimated the the nodule diameter is measured in at growth rates of various types of lung least two dimensions and averaged on cancer with chest radiography (79–84). two serial images. Doubling time (Td)is The volume-doubling time for most lung calculated with the following equation: cancers was reported to be between 1 ϭ ⅐ ⅐ ϭ Td Ti log 2/3 log(Di/Do), where Ti and 18 months. Few CT studies have ϭ interval time, Di initial diameter, and been reported, but they have shown ϭ Do final diameter. For example, a di- that lung cancer doubling times have a ameter increase of 26% corresponds to wider range than anticipated on the ba- a volume increase of 100%. Diameters sis of the previously reported chest radi- measured with electronic calipers are ography literature. CT-derived doubling preferable to diameters measured man- times have been reported to range from ually; however, it is difficult to reliably 32 days to virtually no detectable show changes in size that are smaller growth (32,85,86) (Figs 15, 21). It has than 2 mm (74). As new software be- been speculated that the wider range of comes more widely available, auto- growth rates may be related to the use mated volume measurement algorithms of a more sensitive modality in nodule will likely become more important in detection and measurement. The ma- the evaluation of indeterminate nodules jority of lung cancers have rapid or (75) (Fig 1d). Automated measurement moderately rapid growth rates; how- with multi–detector row CT (which ever, CT may preferentially depict slow- speeds up image acquisition, thus de- growing adenocarcinomas that are not creasing motion and partial volume ef- depicted with radiography (86). Tu- fects) may enable growth to be detected mors with doubling times of more than with serial CT examinations performed 730 days will appear stable during a as little as 4 weeks apart (48). 2-year observation period (87) (Fig 15). A limitation of nodule measurement Growth rate is an independent and im- is that adjacent inflammatory change, portant prognostic factor in patients atelectasis, or scars may inadvertently with lung cancer, so one can expect be included in the measurement, thus longer survival in patients with slow- making the lesion appear larger than it growing tumors (88). is. Tumors may undergo necrosis, hem- orrhage, or cavitation, any of which Key Point Figure 21: (a) CT scan in an 80-year-old man would alter their size. The partial vol- Cancerous tumors with doubling times of shows a 2.5-cm right upper lobe nodule at the ume effect may lead to substantial over- more than 730 days may appear stable posterior segment. (b) Repeat CT scan obtained estimation of lesion size, particularly if during the 2-year observation period. prior to treatment performed 2 months later shows thin CT sections are not used. Even with rapid interval enlargement. The volumetric dou- thin CT sections, volume measurements bling time was 26 days. FNAB revealed mixed small cell and non–small cell carcinoma. in very small nodules may show sub- SPN Biopsy stantial variability and be inaccurate, For nodules that have clinical and imag- especially if the diameter of the nodule ing features of malignancy, a tissue sam-

44 Radiology: Volume 239: Number 1—April 2006 REVIEW FOR RESIDENTS: The Solitary Pulmonary Nodule Winer-Muram

ple is required. There are many meth- formed because of their small size or coaxial technique. FNAB is optimally ods of obtaining tissue from an SPN, location. used in peripheral nodules, although bi- such as video-assisted thoracoscopic or FNAB has a sensitivity of 86.0% and opsies can be performed in more cen- open surgical biopsy; however, the a specificity of 98.8% in the diagnosis of tral lesions. Nodules that are in the method radiologists are frequently malignancy (92); however, in nodules lower lobes or adjacent to the heart may asked to perform is fine-needle aspira- 5–7 mm in diameter, sensitivity is only be difficult to access because of varying tion biopsy (FNAB). CT-guided FNAB 50% (93). Sensitivity of FNAB is also breath holds and diaphragmatic and has been a great advancement in the substantially lower (12%) in patients cardiac motion. Core-needle biopsy can diagnosis of small pulmonary nodules with , and core biopsy (sensi- be reserved for cases where FNAB fails larger than 5 mm in diameter. In pa- tivity, 62%) is recommended when pa- to provide a specific diagnosis, espe- tients who are not candidates for sur- tients are suspected of having lym- cially for cases of benign disease. gery because of comorbidities, FNAB phoma (94). The skills and experience Interpretation of the FNAB speci- can be used to diagnose malignancy and of the cytopathologist are critical in the mens falls into one of three categories: determine the histologic type of malig- interpretation of biopsy specimens. Im- malignant, specific benign, or nonspe- nancy. In patients who are candidates mediate cytologic interpretation in- cific benign. When the FNAB sample is for surgery, FNAB may be used to diag- creases the yield and accuracy of FNAB interpreted as malignant or if a specific nose benign disease, thus obviating sur- (95). benign condition is diagnosed, further gery (89–91) (Fig 1). Contraindications FNAB may be performed with fluo- decisions regarding care are dictated by to FNAB include inability of the patient roscopic, CT, or ultrasonographic (US) the diagnosis. On the other hand, when to cooperate (eg, inability to reproduce guidance. With fluoroscopy, FNAB can a nonspecific benign condition is diag- breath holds, lie immobile on the CT be performed quickly and with the pa- nosed, further evaluation is required. table for more than 30 minutes, or with- tient in a seated position, if necessary. The lesion may be malignant, but the hold coughing). Other relative contrain- CT allows localization of smaller nod- sample was obtained outside the nodule dications include bleeding diathesis, ules and enables planning of the needle or from a necrotic area, thus precluding previous pneumonectomy, severe em- path to avoid blebs and fissures. US may the pathologist from establishing a diag- physema, severe hypoxemia, pulmo- be useful when the lesion abuts the nosis. Core-needle biopsy is more likely nary artery hypertension, or nodules in pleura. Most malignant lesions can be than FNAB to provide a specific benign which successful biopsy cannot be per- diagnosed with FNAB and with use of a result. In a group of patients with be- nign nodules, a specific benign diagnosis Figure 22 was made in 69% of patients with core needle biopsy and in 31% of patients with FNAB (96). A nonspecific benign diagnosis re- quires careful clinical and radiographic follow-up. Diagnoses such as atypical bronchioloalveolar hyperplasia or in- flammation without organisms on a smear or a culture are considered non- specific. If further growth occurs after a nonspecific benign diagnosis is obtained with FNAB, repeat biopsy or resection is indicated. The most common complications of FNAB are and hemor- rhage (87,89). Pneumothorax occurs in approximately 25% of patients, but it is often not clinically important. Only about 7% of patients with a pneumotho- rax will eventually require a (92). The use of an autologous blood patch (injecting the patient’s own blood along the needle track while withdraw- ing the needle) has shown mixed results in reducing the rate of pneumothorax Figure 22: Flowchart for evaluation of the solitary indeterminate nodule for patients with intermediate and after FNAB (97). Presence of emphy- intermediate risk for malignancy. FNA ϭ FNAB. sema, deep lesion location, a lengthy

Radiology: Volume 239: Number 1—April 2006 45 REVIEW FOR RESIDENTS: The Solitary Pulmonary Nodule Winer-Muram

procedure, many pleural punctures, their growth is difficult to detect. Thus, References and traverse of the fissure all increase thin-section CT examinations per- 1. Midthun DE, Swensen SJ, Jett JR. Ap- the risk for pneumothorax (98,99). As formed to show stability at 12 and 24 proach to the solitary pulmonary nodule. urgent intervention may be required, months are acceptable (102,103). Nod- Mayo Clin Proc 1993;68:378–385. radiologists who perform lung biopsy ules shown to be stable at 2-year fol- 2. Swensen SJ, Morin RL, Schueler BA, et al. procedures must be competent at aspi- low-up are considered benign and are Solitary pulmonary nodule: CT evaluation ration of pleural air and insertion of a no longer followed with CT. It is possi- of enhancement with iodinated contrast chest tube. Hemorrhage is almost al- ble that as volumetric measurements material—a preliminary report. Radiology ways self-limiting, although it can occa- become routine, the current follow-up 1992;182(2):343–347. sionally be life-threatening (100). Bleed- CT protocol (3-, 6-, 12-, and 24-month 3. Siegelman SS, Khouri NF, Leo FP, Fishman ing complications are more likely to oc- follow-up) will become compressed EK, Braverman RM, Zerhouni EA. Solitary cur in patients with bleeding diatheses (73). pulmonary nodules: CT assessment. Radi- ology 1986;160(2):307–312. or pulmonary artery hypertension. Nodules with low likelihood for ma- Air embolism is a rare complication; lignancy that are at least 5 mm in diam- 4. Zerhouni EA, Stitik FP, Siegelman SS, et al. patients can present with symptoms re- eter but smaller than 10 mm in diame- CT of the pulmonary nodule: a cooperative sembling those of stroke, transient ische- ter can be observed with CT for a 2-year study. Radiology 1986;160(2):319–327. mic attack, seizure, or cardiopulmonary period, while nodules with intermediate 5. Holin SM, Dwork RE, Glaser S, Rikli AE, collapse. Treatment includes placing the or high likelihood for malignancy may Stocklen JB. Solitary pulmonary nodules patient in the left lateral decubitus posi- be sampled with FNAB or resected found in a community-wide chest roentgen- ographic survey. Am Rev Tuberc 1959;79: tion or Trendelenburg position and ad- (104). Demonstration of nodule growth 427–439. ministering 100% oxygen. Hyperbaric should prompt intervention with either oxygen therapy is recommended. A di- FNAB or surgery. A new persistent nod- 6. Swensen SJ, Silverstein MD, Edell ES, et al. agnosis may be established by perform- ule that develops during observation is Solitary pulmonary nodules: clinical predic- tion model versus physicians. Mayo Clin ing immediate brain or cardiac CT to worrisome for malignancy and warrants Proc 1999;74:319–329. search for intravascular air bubbles intervention (104). (101). For a nodule at least 10 mm in diam- 7. Bateson EM. An analysis of 155 solitary lung lesions illustrating the differential diag- eter, PET and contrast-enhanced CT Key Point nosis of mixed tumors of the lung. Clin Ra- may be helpful in deciding between ob- diol 1965;16:51–65. FNAB results other than a specific ma- servation or intervention. Patients with 8. Murthy SC, Rice TW. The solitary pulmo- lignant or benign diagnosis should be negative PET findings and enhancement nary nodule: a primer on differential diag- viewed with caution. of less than 15 HU on contrast-en- nosis. Semin Thorac Cardiovasc Surg 2002; hanced CT scans may be observed, 14:239–249. while intervention and tissue diagnosis 9. Siegelman SS, Zerhouni EA, Leo FP, Imaging Management of Indeterminate are indicated in patients with positive Nodules Khouri NF, Stitik FP. CT of the solitary PET findings. pulmonary nodule. AJR Am J Roentgenol There is no uniform approach for the 1980;135:1–13. management of an indeterminate nod- Summary 10. Gurney JW. Determining the likelihood of ule. In patients with a high or intermedi- malignancy in solitary pulmonary nodules ate likelihood for malignancy, early in- In general, SPNs can be considered be- with Bayesian analysis. I. Theory. Radiol- tervention with FNAB or surgery is a nign if they exhibit a pattern of benign ogy 1993;186:405–413. good approach. In patients with a low calcification and/or show no growth for 11. Gould MK, Maclean CC, Kuschner WG, likelihood for malignancy, close obser- 2 years. When the imaging features in- Rydzak CE, Owens DK. Accuracy of vation with serial CT might be pre- dicate that the probability of malignancy positron emission tomography for diagno- ferred. In patients with a very low likeli- is high, tissue samples should be ob- sis of pulmonary nodules and mass lesions: hood for malignancy (eg, a healthy non- tained for diagnosis. a meta-analysis. JAMA 2001;285:914–924. smoker younger than 30 years with a In using the various imaging and diag- 12. Wynder EL, Graham EA. well-defined noncalcified nodule), one nostic modalities described in this review, as a possible etiologic factor in bronchio- might consider observation with radiog- one should strive to not only identify small genic carcinoma: a study of 684 proved raphy if the nodule is visible on the malignant tumors—where resection re- cases. J Am Med Assoc 1950;143:329–336. chest radiograph. sults in high survival rates—but also 13. Hrubec Z, McLaughlin JK. Former ciga- The flowchart presented in Figure spare patients with benign disease from rette smoking and mortality among US 22 is a guide proposed by the author for undergoing unnecessary surgery. veterans: a 26-year followup, 1954–1980. the care of patients with an SPN of in- In: Burns D, Garfinkel L, Samet J, eds. Changes in cigarette-related disease risks termediate or high likelihood for malig- Acknowledgment: Many thanks to S. Gregory and their implication for prevention and nancy. Note that nodules smaller than 5 Jennings, MD, for his assistance in editing this control. Bethesda, Md: U.S. Government mm in diameter are often benign, and manuscript. Printing Office, 1997; 501–530.

46 Radiology: Volume 239: Number 1—April 2006 REVIEW FOR RESIDENTS: The Solitary Pulmonary Nodule Winer-Muram

14. Tong L, Spitz MR, Fueger JJ, Amos CA. detected at CT: clinical importance. Radiol- 42. Zwirewich CV, Vedal S, Miller RR, Muller Lung carcinoma in former smokers. Cancer ogy 1997;202:105–110. NL. Solitary pulmonary nodule: high-reso- 1996;78:1004–1010. lution CT and radiologic-pathologic correla- 28. Zerhouni EA, Spivey JF, Morgan RH, Leo tion. Radiology 1991;179:469–476. 15. Lee HJ, Im JG, Ahn JM, Yeon KM. Lung FP, Stitik FP, Siegelman SS. Factors influ- cancer in patients with idiopathic pulmo- encing quantitative CT measurements of 43. Henschke CI, Yankelevitz DF, Mirtcheva R, nary fibrosis: CT findings. J Comput Assist solitary pulmonary nodules. J Comput As- et al. CT screening for lung cancer: fre- Tomogr 1996;20:979–982. sist Tomogr 1982;6:1075–1087. quency and significance of part solid and non-solid nodules. AJR Am J Roentgenol 16. Swensen SJ, Silverstein MD, Ilstrup DM, 29. Proto AV, Thomas SP. Pulmonary nodules 2002;178:1053–1057. Schleck CD, Edell ES. The probability of studied by computed tomography. Radiol- malignancy in solitary pulmonary nodules: ogy 1985;156:149–153. 44. Li F, Sone S, Abe H, MacMahon H, Doi K. application to small radiologically indeter- Malignant versus benign nodules at CT 30. Garland LH. Bronchial carcinomas: lobar minate nodules. Arch Intern Med 1997; screening for lung cancer: comparison of distribution of lesions in 250 cases. Calif 157:849–855. thin section CT findings. Radiology 2004; Med 1961;94:7–8. 233:793–798. 17. Quint LE, Park CH, Iannettoni MD. Solitary 31. Swensen SJ, Viggiano RW, Midthun DE, et pulmonary nodules in patients with ex- 45. Vazquez M, Flieder D, Travis W, et al. al. enhancement at CT: multi- trapulmonary neoplasms. Radiology 2000; Early lung cancer action project pathology center study. Radiology 2000;214:73–80. 217:257–261. protocol [letter]. Lung Cancer 2003;39: 32. Winer-Muram HT, Jennings SG, Tarver 231–232. 18. Jemal A, Chu KC, Tarone RE. Recent RD, et al. Volumetric growth rate of stage I trends in lung cancer mortality in the lung cancer prior to treatment: serial CT 46. Kerr KM. Pulmonary preinvasive neopla- United States. J Natl Cancer Inst 2001;93: scanning. Radiology 2002;223:798–805. sia. J Clin Pathol 2001;54:257–271. 277–283. 33. Quinn D. Gianlupi A, Broste S. The chang- 47. Noguchi M, Morikawa A, Kawasaki M, et 19. Parker MS, Leveno DM, Campbell TJ, ing radiographic presentation of broncho- al. Small adenocarcinoma of the lung. Can- Worrell JA, Carozza SE. AIDS-related genic carcinoma with reference to cell cer 1995;75:2844–2852. bronchogenic carcinoma: fact or fiction? types. Chest 1996;110:1474–1479. 48. Libby DM, Smith JP, Altorki NK, Pasman- Chest 1998;113:154–161. 34. Thun MJ, Lally CA, Flannery JT, Calle EE, tier MW, Yankelevitz D, Henschke CI. 20. Travis WD, Lubin J, Ries L, Devesa S. Flanders WD, Heath CW Jr. Cigarette Managing the small pulmonary nodule dis- United States lung carcinoma incidence smoking and changes in the covered by CT. Chest 2004;125:1522– trends: declining for most histologic types of lung cancer. J Natl Cancer Inst 1997;89: 1529. among males, increasing among females. 1580–1586. 49. Kui M, Templeton PA, White CS, Cai ZL, Cancer 1996;77:2464–2470. 35. Berger WG, Erly WK, Krupinski EA, Bai YX, Cai YQ. Evaluation of the air bron- 21. American Cancer Society. What are the Standen JR, Stern RG. The solitary pulmo- chogram sign on CT in solitary pulmonary key statistics for lung cancer? American nary nodule on chest radiography: can we lesions. J Comput Assist Tomogr 1996;20: Cancer Society detailed guide: lung can- really tell if the nodule is calcified? AJR 983–986. cer—2005. www.cancer.org. Accessed Am J Roentgenol 2001;176:201–204. 50. Weisbrod GL, Towers MJ, Chamberlain April 2005. 36. Kelcz F, Zink FE, Peppler WW, Kruger DW, Herman SJ, Matzinger FR. Thin- 22. Kundel HL. Predictive value and threshold DG, Ergun DL, Mistretta CA. Conventional walled cystic lesions in bronchioalveolar detectability of lung tumors. Radiology chest radiography vs dual-energy computed carcinoma. Radiology 1992;185:401–405. 1981;139(1):25–29. radiography in the detection and character- 51. Lee KS, Kim Y, Han J, Ko EJ, Park CK, 23. Muhm JR, Miller WE, Fontana RS, Sander- ization of pulmonary nodules. AJR Am J Primack SL. Bronchioloalveolar carcinoma: son DR, Uhlenhopp MA. Lung cancer de- Roentgenol 1994;162:271–278. clinical, histopathologic, and radiologic tected during a screening program using 37. Sone S, Kasuga T, Sakai F, et al. Digital findings. RadioGraphics 1997;17:1345– 4-month chest radiographs. Radiology tomosynthesis imaging of the lung. Radiat 1357. 1983;148:609–615. Med 1996;14:53–63. 52. Heitzman ER, Markarian B, Raasch BN, 24. Austin JH, Romney BM, Godsmith LS. 38. Colice GL. Chest CT for known or sus- Carsky EW, Lane EJ, Berlow ME. Path- Missed bronchogenic carcinoma: radiologic pected lung cancer. Chest 1994;106:1538– ways of tumor spread through the lung: findings in 27 patients with a potentially 1550. radiologic correlations with anatomy and respectable lesion evident in retrospect. pathology. Radiology 1982;144:3–14. Radiology 1992;182:115–122. 39. Grewal RG, Austin JH. CT demonstration of calcification in carcinoma of the lung. 53. Furuya K, Murayama S, Soeda H, et al. 25. Swensen SJ, Jett JR, Hartman TE, et al. CT J Comput Assist Tomogr 1994;18:867– New classification of small pulmonary nod- screening for lung cancer: 5-year prospec- 871. ules by margin characteristics on high-reso- tive experience. Radiology 2005;235(1): lution CT. Acta Radiol 1999;40:496–504. 259–265. 40. Mahoney MC, Shipley RT, Corcoran HL, Dickson BA. CT demonstration of calcifica- 54. Seemann MD, Seeman O, Luboldt W, et al. 26. Henschke CI, McCauley DI, Yankelevitz tion in carcinoma of the lung. AJR Am J Differentiation of malignant from benign DF, et al. Early lung cancer action project: Roentgenol 1990;154:255–258. solitary pulmonary lesions using chest radi- overall design and findings from baseline ography, spiral CT and HRCT. Lung Cancer 41. Muram TM, Aisen A. Fatty metastatic le- screening. Lancet 1999;354:99–105. 2000;29:105–124. sions in 2 patients with renal clear-cell car- 27. Munden RF, Pugatch RD, Liptay MJ, Sugar- cinoma. J Comput Assist Tomogr 2003;27: 55. Khan A, Herman PG, Vorwerk P, Stevens baker DJ, Linh UL. Small pulmonary lesions 869–870. P, Rojas KA, Graver M. Solitary pulmonary

Radiology: Volume 239: Number 1—April 2006 47 REVIEW FOR RESIDENTS: The Solitary Pulmonary Nodule Winer-Muram

nodules: comparison of classification with 70. Black WC, Armstrong P. Communicating duration of primary bronchial cancer. Am J standard, thin-section, and reference phan- the significance of radiologic test results: Roentgenol Radium Ther Nucl Med 1966; tom CT. Radiology 1991;179:477–481. the likelihood ratio. AJR Am J Roentgenol 96:604–611. 1986;147:1313–1318. 56. Huston J 3rd, Muhm JR. Solitary pulmo- 84. Weiss W. Peripheral measurable broncho- nary opacities: plain tomography. Radiol- 71. Gurney JW, Lyddon DM, McKay JA. De- genic carcinoma: growth rate and period of ogy 1987;163:481–485. termining the likelihood of malignancy in risk after therapy. Am Rev Respir Dis 1971; solitary pulmonary nodules with Bayesian 103:198–208. 57. Chaudhuri MR. Primary pulmonary cavitat- analysis. II. Application. Radiology 1993; ing carcinomas. Thorax 1973;28:354–366. 85. Aoki T, Nakata H, Watanabe H, et al. Evo- 186:415–422. lution of peripheral lung adenocarcinomas: 58. Theros EG. Varying manifestations of pe- 72. Tan BB, Flaherty KR, Kazerooni EA, Ian- CT findings correlated with histology and ripheral pulmonary neoplasms: a radiology- nettoni MD; American College of Chest tumor doubling time. AJR Am J Roentgenol pathologic correlative survey. AJR Am J Physicians. The solitary pulmonary nodule. 2000;174:763–768. Roentgenol 1977;128:893–914. Chest 2003;123(1 suppl):89S–96S. 86. Hasegawa M, Sone S, Takashima S, et al. 59. Woodring JH, Fried AM, Chaung VP. Soli- 73. Yankelevitz DF, Henschke CI. Does 2-year Growth rate of small lung cancers detected tary cavities of the lung: diagnostic implica- stability imply that pulmonary nodules are on mass CT screening. Br J Radiol 2000;73: tions of cavity wall thickness. AJR Am J benign? AJR Am J Roentgenol 1997;168: 1252–1259. Roentgenol 1980;135:1269–1271. 325–328. 60. Woodring JH, Fried AM. Significance of 87. Lillington GA. Management of solitary pul- wall thickness in solitary cavities of the 74. Jennings SG, Winer-Muram HT, Tarver monary nodules. Dis Mon 1991;37:271– lung: a follow up study. AJR Am J Roentge- RD, Farber MO. growth: as- 318. sessment with CT—comparison of diame- nol 1983;140:473–474. 88. Usuda K, Saito Y, Sagawa M, et al. Tumor ter and cross-sectional area with volume doubling time and prognostic assessment of 61. Swensen SJ, Brown LR, Colby TV, Weaver measurements. Radiology 2004;231:866– patients with primary lung cancer. Cancer AL. Pulmonary nodules: CT evaluation of 871. enhancement with iodinated contrast mate- 1994;74:2239–2244. 75. Yankelevitz DF, Reeves AP, Kostis WJ, rial. Radiology 1995;194:393–398. 89. Westcott JL, Rao N, Colley DP. Transtho- Zhao B, Henschke CI. Small pulmonary racic needle biopsy of small pulmonary nod- 62. Swensen SJ, Brown LR, Colby TV, Weaver nodules: volumetrically determined growth ules. Radiology 1997;202:97–103. AL, Midthun DE. Lung nodule enhance- rates based on CT evaluation. Radiology ment at CT: prospective findings. Radiology 2000;217:251–256. 90. Li H, Boiselle PM, Shepard JO, Trotman- 1996;201:447–455. Dickenson B, McLoud TC. Diagnostic accu- 76. Winer-Muram HT, Jennings SG, Meyer racy and safety of CT-guided percutaneous 63. Zhang M, Kono M. Solitary pulmonary CA, et al. Effect of varying CT section width needle aspiration biopsy of the lung: com- nodules: evaluation of blood flow patterns on volumetric measurement of lung tumors parison of small and large pulmonary nod- with dynamic CT. Radiology 1997;205: and application of compensatory equations. ules. AJR Am J Roentgenol 1996;167:105– 471–478. Radiology 2003;229:184–194. 109. 64. Yamashita K, Matsunobe S, Tsuda T, et al. 77. Nathan MH, Collins VP, Adams RA. Differ- 91. Moore EH. Needle-aspiration lung biopsy: a Intratumoral necrosis of lung carcinoma: a entiation of benign and malignant pulmo- comprehensive approach to complication potential diagnostic pitfall in incremental nary nodules by growth rate. Radiology reduction. J Thorac Imaging 1997;12:259– dynamic computed tomography analysis of 1962;79:221–232. solitary pulmonary nodules? J Thorac Im- 271. 78. Jensen KG, Schiodt T. Growth conditions aging 1997;12(3):181–187. 92. Lacasse Y, Wong E, Guyatt GH, et al. of hamartoma of the lung: a study based on Transthoracic needle aspiration biopsy for 65. Yamashita K, Matsunobe S, Tsuda T, et al. 22 cases operated on after radiographic ob- the diagnosis of localized pulmonary Solitary pulmonary nodule: preliminary servation for from one to 18 years. Thorax lesions: a meta-analysis. Thorax 1999;54: study of evaluation with incremental dy- 1958;13:233–237. namic CT. Radiology 1995;194:399–405. 884–893. 79. Steele JD, Buell P. Asymptomatic solitary 93. Wallace MJ, Krishnamurthy S, Broemeling 66. Erasmus JJ, McAdams HP, Patz EF Jr. pulmonary nodules: host survival, tumor LD, et al. CT-guided percutaneous fine-nee- Non-small cell lung cancer: FDG-PET imag- size, and growth rate. J Thorac Cardiovasc dle aspiration biopsy of small (Յ1-cm) pul- ing. J Thorac Imaging 1999;14:247–256. Surg 1973;65:140–151. monary lesions. Radiology 2002;225:823– 67. Gupta NC, Maloof J, Gunel E. Probability of 80. Weiss W, Boucot KR, Cooper DA. Growth 828. malignancy in solitary pulmonary nodules rate in the detection and prognosis of bron- 94. Goralnik CH, O’Connell DM, el Yousef SJ, using fluorine-18-FDG and PET. J Nucl Med chogenic carcinoma. JAMA 1966;198: Haaga JR. CT-guided cutting-needle biop- 1996;37:943–948. 1246–1252. sies of selected chest lesions. AJR Am J 68. Dewan NA, Shehan CJ, Reeb SD, Gobar 81. Weiss W. Tumor doubling time and sur- Roentgenol 1988;151:903–907. LS, Scott WJ, Ryschon K. Likelihood of vival of men with bronchogenic carcinoma. 95. Austin JH, Cohen MB. Value of having a malignancy in a solitary pulmonary nodule: Chest 1974;65:3–8. comparison of Bayesian analysis and results cytopathologist present during percutane- of FDG-PET scan. Chest 1997;112:416– 82. Garland LH, Coulson W, Wollin E. The rate ous fine-needle aspiration biopsy of lung: 422. of growth and apparent duration of un- report of 55 cancer patients and metaanaly- treated primary bronchial carcinoma. Can- sis of the literature. AJR Am J Roentgenol 69. Goldsmith SJ, Kostakoglu L. Nuclear medi- cer 1963;16:694–707. 1993;160:175–177. cine imaging of lung cancer. Radiol Clin North Am 2000;38:511–524. 83. Garland LH. The rate of growth and natural 96. Boiselle PM, Shepard JO, Mark EJ. Routine

48 Radiology: Volume 239: Number 1—April 2006 REVIEW FOR RESIDENTS: The Solitary Pulmonary Nodule Winer-Muram

addition of an automated biopsy device to 99. Kazerooni EA, Lim FT, Mikhail A, et al. ciousness of nodules according to size on fine needle aspiration of the lung: a pro- Risk of pneumothorax in CT-guided trans- baseline scans. Radiology 2004;231:164– spective assessment. AJR Am J Roentgenol thoracic needle aspiration biopsy of the 168. 1997;169:661–666. lung. Radiology 1996;198:371–375. 103. Macmahon H, Austin JH, Gamsu G, et al. 97. Lang EK, Ghavami R, Schreiner VC, et al. 100. Milner LB, Ryan K, Gullo J. Fatal intratho- Guidelines for management of small pulmo- Autologous blood clot seal to prevent pneu- racic hemorrhage after percutaneous aspi- nary nodules detected on CT scans: a state- mothorax at CT guided lung biopsy. Radiol- ration lung biopsy. AJR Am J Roentgenol ment from the Fleischner Society. Radiol- ogy 2000;216:93–96. 1979;132:280–281. ogy 2005;237(2):395–400. 101. Kodama F, Ogawa T, Hashimoto M, et al. 98. Laurent F, Michel P, Latrabe V, Tunon de 104. Henschke CI, Naidich DP, Yankelevitz DF, Fatal air embolism as a complication of CT Lara M, Marthan R. Pneumothoraces and et al. Early Lung Cancer Action Project: guided needle biopsy of the lung. J Comput chest tube placement after CT-guided initial findings on repeat screenings. Cancer Assist Tomogr 1999;23:949–951. transthoracic lung biopsy using a coaxial 2001;92:153–159. technique: incidence and risk factors. AJR 102. Henschke CI, Yankelevitz DF, Naidich DP, Am J Roentgenol 1999;172:1049–1053. et al. CT screening for lung cancer: suspi-

Radiology: Volume 239: Number 1—April 2006 49