General Thoracic Surgery (General Thoracic Surgery (Shields)) [2 VOLUME SET]

Editors: Shields, Thomas W.; LoCicero, Joseph; Ponn, Ronald B.; Rusch, Valerie W.

Title: General Thoracic Surgery, 6th Edition

Copyright 2005 Lippincott Williams & Wilkins

> Table of Contents > Volume I - The Lung, Pleura, Diaphragm, and Chest Wall > Section III - Thoracic Imaging > Chapter 10 - Computed Tomography of the Lungs, Pleura, and Chest Wall

Chapter 10

Computed Tomography of the Lungs, Pleura, and Chest Wall

Lacey Washington

Wallace T. Miller Jr.

Computed tomography (CT) is now firmly established as an indispensable radiologic modality for the evaluation of the chest. The advantages of CT over plain radiography include resolution of smaller differences in density and the provision of greater anatomic detail as well as visualization of structures without superimposition. CT may be used for the detection of abnormalities in the mediastinum, lungs, pleura, and chest wall and in many cases is helpful for further characterization of processes seen in the mediastinum, lungs, and pleura on plain chest radiographs.

INDICATIONS FOR CHEST COMPUTED TOMOGRAPHY

Thoracic CT is, with the possible, controversial exception of lung cancer screening, not a screening procedure. The high radiation dose of CT when compared with chest radiography should limit its use to patients in whom a specific clinical question must be addressed. Additionally, although an institution may have a standard CT protocol, multiple operator-dependent variables need to be selected on each CT examination in order to optimize the information acquired in light of the particular clinical question posed. Variables that need to be addressed include slice thickness, slice spacing, field of view, and reconstruction algorithm, as well as whether to give intravenous contrast and how to time its administration. Sometimes, a limited CT of the chest may be performed to address a specific, localized problem (most commonly, to evaluate for stability of a previously seen nodule). For all of these reasons, it is important that adequate clinical information be provided with requests for CT of the chest.

There are multiple long-accepted indications for chest CT and a variety of indications that are more experimental or are gaining acceptance. Many common indications for CT of the chest are reviewed in the American College of Radiology (ACR) Appropriateness Criteria. This chapter addresses uses of CT in evaluation of the lungs, pleura, and chest wall. Indications for CT to evaluate mediastinal abnormalities are discussed in Chapter 158.

Lungs

In the evaluation of primary neoplasms of the lung, CT is used in staging, or to attempt to identify a primary neoplasm in a patient with positive sputum cytology and negative chest radiograph and bronchoscopy. Occasionally, CT may be used in an attempt to detect occult neoplasms of the lung. As shown by Katagiri (1999) and Chorost (2001) and their associates, this is often helpful when metastatic disease is detected elsewhere and a primary tumor is not evident after history, physical examination, and chest radiography have been performed.

As noted by the Society for Computed Body Tomography (1979) and Fraser and associates (1994), CT may be used to look for occult pulmonary metastases in patients with primary neoplasms elsewhere, when surgical intervention is planned for a primary neoplasm with a high rate of metastasis to the lungs. As Davis and associates (2000) stated, CT is also indicated to look for additional nodules when a chest radiograph demonstrates a single pulmonary nodule or equivocal findings in this setting. Depending on the propensity of a tumor to metastasize to the lungs, CT may also be indicated in following patients with extrathoracic primary malignancies. This is particularly true of those tumors in which metastasectomy may be considered. Nevertheless, as Davis and colleagues (2000) noted, CT is surprisingly insensitive for the detection of pulmonary metastases when compared with palpation of the lung.

CT is also used to assess response to treatment of both primary and metastatic malignancies within the chest.

P.146

When a solitary pulmonary nodule is identified on radiographs, CT is the established technique to assess for fat or calcification. Contrast-enhanced thin-section CT with images at multiple time points may occasionally be used to attempt to distinguish benign from malignant nodules.

In the setting of pneumonia, CT can be used to assess for reasons for a poor response to treatment, particularly central obstruction and pleural collections. CT is routinely used to assess for occult infections in immunocompromised patients in whom chest radiography may be insensitive for clinically significant infections. A special technique, high-resolution computed tomography (HRCT), is used for the detection and characterization of pulmonary emphysema, bronchiectasis, and interstitial lung disease, including some specific infections.

CT has become an established modality to evaluate for pulmonary emboli despite lingering controversy. The frequency with which this technique is used instead of ventilation-perfusion imaging will vary by institution; however, as Lesser (1992) and Goldberg (1994) and their colleagues have shown, ventilation-perfusion ([V with dot above]/[Q with dot above]) imaging has a lower rate of diagnostic results in patients with other pulmonary abnormalities. CT is therefore frequently used to assess for pulmonary emboli in these patients, and many other abnormalities that explain a clinical deterioration may also be identified on such studies.

Although these are the most common reasons to use CT to evaluate the lungs, many other pulmonary abnormalities may be seen at CT. For example, complications of trauma, such as lung lacerations and contusions, are frequently identified in patients who are undergoing CT to evaluate for mediastinal abnormalities after trauma.

In recent years, several groups of investigators have assessed CT as a screening modality to detect occult lung cancer in asymptomatic (usually high-risk) patients, as reviewed by Patz and co-workers (2001) and by Swensen (2002). This remains a very controversial indication, and the Society of Thoracic Radiology, as reported by Aberle and associates (2001), has issued a position statement against this use of CT until further investigation has been performed. Nevertheless, self-pay screening will almost certainly remain available, and patients will continue to present for advice regarding screening-detected nodules.

Pleural Space

CT can be used to evaluate for the presence of a pleural effusion in cases in which its detection is made difficult by overlying pulmonary opacities, and particularly when ultrasound is confusing, as discussed by the Society for Computed Body Tomography (1979) and Fraser and co-workers (1994). This is particularly likely to occur when a pleural effusion contains multiple locules or is loculated in a medial or subpulmonic location. CT is also helpful in the detection of pleura-based neoplasm and can be used to identify asbestos-related pleural plaques.

CT is more sensitive than radiography for the detection of pneumothoraces, particularly when only supine radiographs can be performed. The clinical importance of occult pneumothoraces, defined as those seen only on CT, will depend on the clinical setting, particularly the presence of positive-pressure ventilation.

Chest Wall

CT can be used to identify osseous destruction as well as soft tissue masses in the evaluation of neoplastic involvement of the chest wall, whether by chest wall primary tumors or by metastases. In the absence of bony destruction, however, CT can be inaccurate in assessing direct chest wall extension of primary lung tumors. CT can be used to detect the involvement of the chest wall by infection, again, particularly where osseous destruction occurs. In both of these cases, bony abnormalities may be evident on CT before they are detectable on chest radiographs. CT may also be used to assess for chest wall fluid collections that require drainage, particularly in a postoperative setting and when infection is suspected.

As discussed by Westcott and associates (2000), CT usually provides a thorough evaluation of rib fractures when performed to evaluate for other injuries in the setting of trauma.

Clinical Settings

Mirvis and colleagues (1987) advocate the use of CT in critically ill and ventilator-dependent patients who can only be otherwise imaged by portable radiography, whenever there is an unexplained clinical deterioration that may be caused by an abnormality within the chest. As Naidich and co-workers (1990) indicate, another common clinical question that may be addressed by CT is the evaluation of hemoptysis. This most commonly results from either a malignancy or bronchiectasis, both of which may be seen on CT.

Interventions

A wide variety of interventions in the chest may be performed under CT guidance, as reviewed by Klein (1995) and Giron (1996) and their co-workers. Most commonly, CT can be used to guide percutaneous biopsy of lung nodules and masses for the diagnosis of malignancy and infection. Drainage of pleural fluid collections and, less commonly, pneumothoraces and lung abscesses can also be performed with CT guidance.

P.147

TECHNIQUES

Standard and Helical Computed Tomography

A CT image is acquired using the same physical principles as plain radiography, the passage of x-rays through tissue, allowing the detection of differences in density or attenuation of x-ray beams caused by different atomic numbers. CT passes multiple, highly collimated x-ray beams at various angles through an anatomic plane, and the attenuation is then measured by an array of detectors opposite the x-ray source. The geometry of the arrangement of source and detectors has evolved since the original CT scanners were developed, allowing for decreased scan times. This has been particularly helpful in improving image quality in the chest, by allowing imaging of the whole chest during a single breath-hold. As scanners become more and more rapid, the ability to limit artifacts from cardiac pulsation will allow for wider applications of CT. The information acquired by the array of detectors is then reconstructed with complex computer algorithms to create an image of the plane through which the beams have passed. This image is essentially a map of tissue densities. As Curry and associates (1990) noted, density differences of only 0.5% can be recognized with CT, in contrast to the 10% density differences required for visualization on plain radiographs.

Although early CT scanners acquired one slice of data at a time, more recent technology has allowed scanners to acquire spiral or helical data sets. A more recent improvement in CT technology is the development of multidetector CT (MDCT) scanners, a newer kind of helical scanner. With single or multidetector helical or spiral scanners, the table moves continuously as the data are acquired. The computer then reconstructs axial images, which appear similar to images that are acquired axially. Vock (1990) and Kalender (1990) and their co-workers reviewed the mechanism of spiral CT scanning, and Berland and Smith (1998) discuss the technologic changes that are introduced by MDCT. With the first single-detector spiral CT scanners, the entire chest could usually be imaged in one or two breath-holds, reducing or eliminating the problem of respiratory misregistration. This means that, unlike with conventional scanners, the same volume of anatomy is much less likely to be imaged twice, or not at all. With multidetector scanners, the entire chest can be imaged with a single breath-hold, with both a decrease in the thickness of slices and a decrease in the total scan time. One distinct advantage of these techniques, as Remy-Jardin and associates (1993) have shown, is that small pulmonary nodules are less likely to be missed.

Helical and multidetector scan acquisition also allows for three-dimensional reconstruction of complex anatomy, which sometimes may help elucidate the answers to specific clinical questions. If finely detailed multiplanar reformatted images are to be obtained, thinner sections should be acquired through the area of interest, and the specific indication for the study should therefore be known in advance. With multidetector scanners, thinner sections may sometimes be created retrospectively; however, this depends on the way in which images are initially acquired, so that it remains best to know the indication in advance. In assessment of the lungs, multiplanar and three-dimensional reformatted images are most commonly utilized in a dedicated examination of the airways, which may be performed to evaluate for a central airway mass (e.g., in a patient presenting with hemoptysis or evidence of airway obstruction) or airway stenoses.

After the data have been acquired, the calculated tissue densities are converted into a scale of CT numbers or Hounsfield units (named after G. N. Hounsfield, the inventor of the original CT scanner). The scale is designed such that the CT number for air is -1,000 and the CT number for dense bone is +1,000, with water having a CT number of 0. On this scale, fat has a CT number of approximately -100. The wide range of density values detected by CT cannot be displayed on a single set of images, and therefore multiple windows are used to display a single scan. In the chest, a set of mediastinal windows, with relatively high contrast, is used to demonstrate the narrow density differences in the mediastinum (between fat and soft tissue). Similarly, a set of lung windows is used to display the lung parenchyma, and a third set of bone windows should be examined to assess for osseous abnormalities. A window is specified by a level, or the density that is approximately average for the tissue being displayed, which will be displayed at a medium shade of gray, and a width, which determines the range of CT numbers that will be displayed in gray, between the values assigned to black and white. Different viewers and therefore different institutions will have different preferences for the exact levels and window widths at which images are displayed.

Conventionally, CT images were printed on film and stored as hard copy. However, CT data are inherently digital and may be stored on a variety of electronic media. It is becoming increasingly common for radiology studies, including CT, to be stored on a picture archiving and communication system (PACS). Images that are stored electronically and recovered may be displayed in any window. Although lung, mediastinal, and bone windows should all be examined, it is common practice at this time not to film bone windows unless there is a specific bony abnormality; the images can be recalled and displayed at a future time if this is necessary.

Intravenous Contrast Media

In planning a CT examination of the chest, one important consideration is whether intravenous contrast is to be administered. Iodinated contrast media increase the density

P.148

Evaluation of hilar structures, unlike most mediastinal structures, depends more heavily on the use of intravenous contrast. Hilar lymph nodes can be very difficult to distinguish from blood vessels in the absence of intravenous contrast enhancement, and it is therefore important to use intravenous contrast if possible in any patient in whom assessment for hilar lymphadenopathy is important (Fig. 10-1).

Special uses of intravenous contrast include CT for pulmonary emboli and CT of pulmonary nodules. Pulmonary emboli are seen as nonenhancing areas within otherwise enhanced vessels. Images are obtained with thin-section imaging during administration of intravenous contrast at high injection rates (3.5 to 5 mL/s), with various techniques used to optimize the timing of acquisition so that the contrast densely enhances the vessels.

When intravenous contrast is used to evaluate a pulmonary nodule, serial images through the nodule are obtained over a time interval of about 5 minutes. The change in attenuation is measured over time.

High-Resolution Computed Tomography

HRCT is most commonly described as a form of imaging of the lung developed to allow improved diagnostic accuracy, sensitivity, and specificity in the evaluation of diffuse lung disease, when compared with plain radiography. (The term may also, more broadly, be used for any thin-section imaging, for example, images obtained through a pulmonary nodule to evaluate for calcification or fat.) As Webb and colleagues (2001) state, when compared with conventional CT, HRCT requires thinner collimation, resulting in marked improvement in spatial resolution. The algorithm by which the images are reconstructed is modified; specifically, a high-spatial frequency reconstruction algorithm is used. This improves evaluation of lung detail at the expense of increased image noise, which limits soft tissue evaluation. HRCT images are similar in appearance to macroscopic pathologic specimens. Other modifications that may be employed include an increased peak kilovoltage or milliamperage technique, a large matrix size, and targeted image reconstruction.

Continuous scanning throughout the lungs with such thin sections was once considered prohibitive in terms of dose to the patient, time of scanning, and tube current; therefore, a protocol that acquires discontinuous slices has been adopted as a common practice. With newer multidetector scanners, thin sections could be achieved throughout the chest, but in attempts to detect diffuse lung disease, contiguous images are not considered necessary and are therefore avoided in order to reduce patient dose. Most commonly, 1.0- or 1.5-mm thick sections are obtained at 10-mm intervals through the lungs. Studies by Leung (1991) and Kazerooni (1997) and their co-workers suggest that very few images are actually necessary in the detection of most diffuse lung diseases. No intravenous contrast is administered with a dedicated HRCT; however, some HRCT images are sometimes obtained in conjunction with another CT protocol, depending on the clinical setting.

Electron-Beam Computed Tomography

Electron-beam CT is a fairly uncommonly used form of CT, which uses an electron gun to produce an x-ray beam from a tungsten target ring, eliminating mechanical motion within the CT gantry. This technique allows the acquisition of images in as little as 100 milliseconds. Its primary uses are in the mediastinum, specifically in cardiac imaging; however, centers that have these scanners may use them for other applications in which a short scan time is helpful. For example, electron-beam CT may be used in dynamic imaging of the airways, to see the change in airway caliber over time, or in the evaluation of pulmonary emboli.

P.149

FINDINGS ON COMPUTED TOMOGRAPHY: LUNG PARENCHYMA

Diseases involving the lung parenchyma may be divided into those that primarily involve the air spaces, such as bacterial pneumonias, and those that involve predominantly the interstitium. Diseases may also be divided by their morphology into those that are nodular or masslike and those that are more infiltrative. The margins of any focal abnormality may be evaluated. Diseases may also be defined as focal, multifocal, or diffuse.

In the setting of air space disease, CT provides little characterizing information but may be helpful in determining the extent and location of disease. CT may be helpful for evaluation of the relationship of a given parenchymal abnormality to structures such as the bronchi, vessels, mediastinum, pleura, or chest wall. Localization of an abnormality may be useful in selecting an optimal approach for diagnostic bronchoscopy. When characterization of diffuse lung disease is required, HRCT is most helpful and is particularly useful in limiting the differential diagnosis of interstitial lung diseases.

Pulmonary Nodules

A wide variety of pathologic conditions can present as a solitary pulmonary nodule or mass. According to the Fleischner Society glossary of terms for CT of the lungs as discussed by Austin and co-workers (1996), a nodule is less than 3 cm in diameter, with rounded opacities greater than this size usually being referred to as masses. The differential diagnosis for any given focal lung nodule will depend in part on the clinical setting in which it is discovered, as well as on its appearance.

If a nodule is discovered on radiography as part of a workup for metastatic disease, the first step in further evaluation should be a CT to identify additional nodules. If there are multiple nodules seen throughout both lungs at CT, the probability that the nodule represents metastatic disease increases, although biopsy is required for definitive diagnosis. CT is very sensitive for the detection of pulmonary nodules. However, in geographic areas endemic for histoplasmosis, granulomas or subpleural lymph nodes may constitute a large proportion of lung nodules, and these are indistinguishable from metastases unless they are calcified. In the series reported by Munden and associates (1997), almost 40% of solitary pulmonary nodules smaller than 1 cm represented primary lung malignancy; however, in screening CT studies, this number has been consistently much smaller. For example, in the Mayo Clinic lung cancer screening trial as discussed by Swensen (2002), nearly 99% of nodules detected in smokers without an extrapulmonary primary malignancy were estimated to be benign.

Frequently, once a nodule is discovered, comparison with older films will show that the nodule was present in retrospect and is stable. It has been widely accepted, based on knowledge of tumor doubling times, that a nodule that remains stable for 2 years is benign. As Yankelevitz and Henschke (1997) indicate, most malignant pulmonary nodules have doubling times between 20 and 400 days. Nodules that double more quickly than this may be metastases but are usually infectious in nature, whereas those that have longer doubling times are usually benign. However, as these authors and Winer-Muram and colleagues (2002) point out, occasional lung carcinomas have much longer doubling times, and longer follow-up intervals may be necessary to detect the most indolent carcinomas.

If a nodule is truly solitary, it is important to evaluate its margins. As Huston and Muhm (1987) showed, a nodule with spiculated margins suggests malignancy, with a high, about 90%, probability (Fig. 10-2).

In a solitary nodule less than 3 cm in diameter, with smooth margins, it is helpful to evaluate for calcification. Contiguous thin-section (HRCT) images should be obtained through the nodule. The use of this technique prevents partial-volume averaging of surrounding lung. A nodule with central, popcornlike, concentric rings or diffuse calcification may be confidently diagnosed as benign (Fig. 10-3). Eccentric calcification may be seen in some carcinomas, and therefore the presence of calcification alone does not specifically indicate a benign process. The presence of fat within a nodule is considered virtually diagnostic of benignity and specific for the diagnosis of a hamartoma.

Historically, one technique that was used to help with the diagnosis of solitary pulmonary nodules was the CT reference phantom. This was developed to evaluate for calcification that was not visually obvious; the phantom was scanned and CT numbers higher than those of the phantom were considered diagnostic of calcification. This technique

P.150

A nodule without calcification, but otherwise meeting the above criteria, is frequently followed with serial CT examinations to evaluate for growth. Fine-needle aspiration (FNA) biopsy may also be performed using CT or fluoroscopic guidance; however, as Klein and co-workers (1996) reported, the yield of a confident diagnosis is lower for benign etiologies than for malignancy.

Occasionally, specific0 morphologic features allow for confident characterization of a solitary pulmonary nodule or mass based on a single CT examination alone. Specifically, a nodule may be diagnosed as representing a pulmonary arteriovenous malformation or rounded atelectasis. Remy (1992) has suggested that CT may be superior to the more traditional study, angiography, for the diagnosis of arteriovenous malformations (Fig. 10-4). These can frequently be diagnosed even without contrast enhancement of vessels, based on the demonstration of a feeding arterial vessel and a large draining vein. Administration of intravenous contrast, if performed, should demonstrate arterial-phase dense enhancement unless the malformation is thrombosed. Schneider and colleagues (1980) described the imaging features of rounded atelectasis. This is an abnormality that occurs in lung adjacent to an area of pleural thickening, most commonly in patients with asbestos-related pleural disease, but also in patients with pleural thickening of other etiologies. The mass is rounded and abuts the pleura; bronchovascular markings in the adjacent lung appear to curl into it, creating a comet tail sign. Finally, two new techniques have shown some promise in the evaluation of solitary pulmonary nodules. Swensen and coworkers (2000), in a large, multicenter trial, have shown that dynamic contrast administration may be used to exclude malignancy. Thin-section images are initially obtained through a nodule to assess for fat or calcium. If neither is present, intravenous contrast is administered, and images are obtained every minute for 4 minutes. Nodules that enhance by 15 Hounsfield units or less can be considered benign, with a sensitivity of 98%. Active inflammatory lesions will yield a false-positive result, and the specificity of the technique with this threshold is therefore only 58%. This technique has not gained widespread popularity, in part, perhaps, because of the advent of positron emission tomography (PET) scanning. Fluorodeoxyglucose (FDG) PET scanning also assesses metabolic activity of pulmonary nodules and may similarly be used to try to distinguish benign, inactive, from malignant and other active nodules, as shown by Dewan (1993), Gupta (1992), Hagberg (1997), and Patz (1993) and their co-workers.

P.151

Lung Cancer Screening

An understanding of the potential uses of CT in nodule evaluation is particularly helpful in light of the recent controversy over lung cancer screening. It is no surprise that CT is more sensitive than chest radiography for the detection of pulmonary nodules.

CT might, therefore, in theory, be used to identify lung cancer at an earlier stage than chest radiographs, and its proponents hope that this earlier diagnosis will lead to improved survival (see Chapter 99).

A discussion of the difficulties with validating a screening examination is beyond the scope of this chapter. However, as has been seen with mammography, proving the value of a screening examination is more difficult than might be expected. There are multiple forms of bias that are introduced by the screening setting, which can lead to false apparent survival advantage. One important source of bias is called overdiagnosis. Overdiagnosis represents the identification of histologic disease that would never actually harm the patient. Sone and co-workers (1998), in a series from Japan, have identified the same rate of histologic lung cancer in their smoking and nonsmoking screened patients. Because many fewer nonsmoking patients die from lung cancer, it seems possible that this shows overdiagnosis bias, that is, that some of the identified histologic lung cancers are not truly lethal lesions.

A major difficulty with the use of CT for lung cancer screening is more obvious than subtle biases such as overdiagnosis. This is the high incidence of false-positive screening examinations; that is, the high number of patients with benign nodules. As the techniques used in these screening studies have improved (with workstation review and the use of MDCT to obtain thinner sections), the number of false-positive examinations has increased, even beyond the high numbers seen in early trials. For example, in the Mayo Clinic study, as reported by Swensen and colleagues (2002), over the course of 3 years, 71% of patients were found to have pulmonary nodules, and a total of more than 2,800 noncalcified nodules were found in their 1,520 participants. Only 41 lung cancers were identified. Promoters of lung cancer screening initially touted the use of a low radiation dose, but full doses are used for follow-up scans. There is therefore clearly a significant radiation exposure to the majority of these patients, most of whom do not have lung cancer. In the Mayo Clinic trial, 20% of the operations that were performed were for benign disease; hence, the morbidity of unnecessary surgery must also be weighed in evaluating the utility of lung cancer screening.

There are large trials in progress that will attempt to address the utility of CT lung cancer screening; however, these will take 7 to 10 years to yield adequate results. In the short run, demand for lung cancer screening will undoubtedly continue, and such scans will probably remain available. It is therefore important to be aware of the very high rate of detection of benign lesions and the concomitant risks to the patient.

Bronchial Carcinoma

CT of the chest has become a routine part of preoperative staging for lung carcinoma. Contrast-enhanced CT is performed at many centers to facilitate the detection of hilar and mediastinal lymph nodes and sometimes to better delineate the relationship of a centrally invading tumor to the aorta or pulmonary vessels. Patz and associates (1999) have suggested that the additional information resulting from contrast administration is minimal. Therefore, practices may vary in their use of contrast in this setting, some considering that the risks for intravenous contrast administration outweigh its benefits for these patients.

CT provides information about the size of the primary lesion and may suggest the possibility of direct extension of tumor to the contiguous structures of the chest wall, mediastinum, and diaphragm, thus providing information about its T status. CT is also used to evaluate for enlarged hilar and mediastinal lymph nodes (N status) and for hematogenous dissemination of tumor elsewhere within the lungs or to the liver, adrenals, and bones (M status). Although CT definitely gives additional information over chest radiography in all of these areas, it is important to understand its limited sensitivity and specificity for all aspects of lung cancer staging.

Even defining the size of a tumor may be difficult, particularly in the setting of postobstructive consolidation and atelectasis. It may be difficult, in fact, to establish the presence of a tumor in this setting. Onitsuka and colleagues (1991) noted that rapid scanning following administration of intravenous contrast material may help with the distinction of tumor from surrounding consolidated lung because tumors tend to show less enhancement than the normal lung. More rapid scanning is becoming routine with today's faster scanners.

Unfortunately, it has been well demonstrated that CT is neither sensitive nor specific for invasion of the chest wall by lung carcinoma. Pennes and co-workers (1985) found that CT had an accuracy of only 39%. Glazer and co-workers (1985) found better results: In their study of chest wall invasion, CT was shown to have an 87% sensitivity, 59% specificity, and 68% accuracy. In this study, localized chest wall pain was shown to be both more specific (94%) and more accurate (85%) for the detection of chest wall invasion, although less sensitive. CT signs of chest wall invasion that are very specific include bony destruction and presence of a soft tissue mass in the chest wall. Ratto and associates (1991) also found that obliteration of the extrapleural fat plane was 85% sensitive and 87% specific. This same study also suggested that a ratio of tumor-pleura contact to tumor diameter of greater than 0.9 was moderately sensitive and specific for chest wall invasion.

P.152

As with chest wall invasion, contiguity of tumor with the mediastinum cannot be equated with mediastinal invasion, nor is CT sensitive for invasion. Herman and colleagues (1994) reported 40% sensitivity and 99% specificity of CT for mediastinal invasion if 90 degrees of contact between the mass and a mediastinal structure was used as the criterion. In this study, all structures with more than 180 degrees of contact were invaded, but distortion of the structure and intraluminal tumor were, surprisingly, less specific for invasion.

Overall, Webb and co-workers (1991) demonstrated a sensitivity of only 62% for CT in distinguishing T3 to T4 tumors from T0 to T2 tumors. These authors also found that the overall accuracy of CT and magnetic resonance (MR) imaging in the staging of lung carcinoma were similar. However, when contiguity of tumor to chest wall or mediastinum is seen on CT scans, MR imaging may be better than CT in assessment of local invasion. This has historically been particularly true in the case of superior sulcus (Pancoast's) tumors, in which the axial orientation of CT slices limits evaluation of chest wall extension at the apex. Multiplanar reformatted images with MDCT significantly improve visualization of this region, and it is uncertain whether the multiplanar capability of MR imaging (which acquires data directly in the coronal plane) still allows for better visualization when compared with CT.

Although the greatest contribution of CT to evaluation of lung carcinoma has been the detection of mediastinal lymphadenopathy, there are severe limitations to the sensitivity and specificity of CT for metastasis to mediastinal lymph nodes. The difficulty with the CT staging of bronchogenic carcinoma is that enlarged nodes may be negative for tumor, and normal size lymph nodes, those less than 1 cm, may contain microscopic metastases, as shown by Libshitz (1984) and Arita (1995) and their co-workers. Nevertheless, CT is considered useful by directing biopsy so that appropriate nodes are evaluated before definitive surgery is attempted. FDG PET imaging, which detects metabolic activity, may detect activity secondary to metastatic tumor in normal-sized lymph nodes. As von Haag and co-workers (2002) reported, PET has greater accuracy and sensitivity and specificity than CT size criteria alone for assessment of nodal metastases. However, CT can identify the precise mediastinal location of lymph nodes, giving anatomic information that is missing on PET studies. The location of suspicious lymph nodes can influence the staging procedure used. For this reason, CT is increasingly used in conjunction with FDG PET imaging. As D'Amico and co-workers (2002) reported, combination (fused) PET CT imaging allows the detection of metabolically active, normal-sized lymph nodes. Although a small percentage of these will represent inflammatory nodes, the use of this technique increases the accuracy of preoperative lung cancer staging.

Staging CT of the chest may also detect hematogenous metastases to the adrenal glands, liver, contralateral lung, and bones. Because of the high propensity of lung carcinoma to metastasize to the adrenals, CT in the setting of lung cancer is routinely obtained through the level of the adrenal glands to evaluate for masses. Unfortunately, benign adrenal masses (usually adrenal adenomas) are very common. On a routine contrast-enhanced CT of the chest, adrenal masses can be detected, but not characterized as benign or malignant.

Several imaging strategies for the characterization of adrenal masses have been developed. Mitchell and associates (1992) were the first researchers to recognize that adenomas could be distinguished from metastasis on the basis of lipid content. They performed MR chemical-shift imaging and showed that masses that lose signal on out-of-phase sequences compared with in-phase sequences are highly likely to represent adenomas. Masses that do not meet these criteria require biopsy. These findings have been confirmed by multiple other groups, including Bilbey (1995), Reinig (1994), and Tsushima (1993) and their co-workers.

Because lipid content also reduces CT attenuation, unenhanced CT can be used to distinguish between benign and malignant adrenal masses. On unenhanced images, a region of interest (ROI) is placed over the mass, and the Hounsfield unit measurement of the mass is obtained. According to Lee and colleagues (1991), a value of less than 10 Hounsfield units has a high positive predictive value for an adenoma, whereas a higher value indicates that the lesion may be a metastasis and should undergo biopsy.

Because both unenhanced CT and MR imaging require a patient to return for a second study after an initial contrast-enhanced CT has been performed, other authors have attempted to use delayed CT imaging after the initial contrast injection to characterize adrenal masses. Brodeur (1995) and Korobkin (1996) and their co-workers suggested that imaging after a delay of 1 hour could be used to characterize these masses, whereas Boland and co-workers (1997) suggested that imaging as soon as 12 to 18 minutes following the initial injection might be used. Caoili and associates (2002) have recently reported high accuracy for the diagnosis of adenomas when 60% or greater decrease in attenuation is measured on images obtained after a 15-minute delay.

Air Space Disease

Air space disease can have a variety of manifestations on CT, as described by Naidich and colleagues (1985). Findings include (a) air space nodules, which appear as poorly marginated opacities ranging up to 1 cm in size, caused by sublobular accumulations of fluid, hemorrhage, or cells; (b) coalescent densities that are usually the result of confluence of air space nodules (Fig. 10-5); (c) air bronchograms and air alveolograms; and

P.153

Air space patterns at CT are usually not specific for the etiology, and a wide variety of diseases can produce such patterns, including pneumonia, aspiration, hemorrhage, pulmonary edema, alveolar proteinosis, and bronchioloalveolar carcinoma. CT adds little to the characterization of these diverse disease processes. However, CT may be used to define the extent of disease, which it can do more accurately than plain radiography. The extent and distribution of disease may influence the differential diagnosis: diseases such as pulmonary edema and hemorrhage and pulmonary alveolar proteinosis are more likely to involve the pulmonary parenchyma diffusely. Multifocal involvement with areas of sparing is more likely to be caused by diseases such as bronchiolitis obliterans with organizing pneumonia (BOOP), vasculitis, alveolar sarcoidosis, and chronic eosinophilic pneumonia, bronchioloalveolar carcinoma, and lymphoma. Some forms of pneumonia are more likely to be diffuse, such as Pneumocystis carinii pneumonia, whereas most bacterial pneumonias are more likely to be unifocal or multifocal. As Epstein and co-workers (1982) have pointed out, careful demonstration of the extent of disease is particularly important for patients with bronchioloalveolar carcinoma.

CT may also limit or direct a differential diagnosis by demonstrating some features either not seen or not well delineated on plain film. A common example is occult cavitation in air space disease. Less common and more specific features that may be seen at CT would include a halo sign, described by Kuhlman and associates (1985), which is an area of ground-glass opacity surrounding an area of more dense consolidation. This suggests hemorrhage, frequently on the basis of aspergillosis. Another more specific feature within air space disease is high density within an area of focal consolidation suggesting the possibility of amiodarone toxicity.

Cavitary Lung Disease

CT may be helpful in identifying cavities within areas of known lung disease. Focal cavities are to be distinguished from diseases such as emphysema that cause diffuse low attenuation through the lungs. These, like most diffuse lung diseases, are best characterized on HRCT.

Focal areas of low attenuation within areas of abnormal lung usually represent areas of cavitation. The presence of

P.154

Localized thin-walled cavities may represent pneumatoceles, lacerations, or bullae. Bullae are usually seen in the setting of emphysema, discussed subsequently. Pneumatoceles represent dilated distal airways and are seen in patients with pneumonia and barotrauma, most commonly in pediatric patients. Pneumatoceles are distinguished from abscesses by their thin walls, but the distinction from abscess cavity may be difficult if there is extensive surrounding pneumonia. Lacerations represent disruption of lung parenchyma in the setting of trauma and are distinguished from pneumatoceles by the clinical setting in which they arise.

Pneumonia

CT usually adds very little diagnostic information in the setting of uncomplicated pneumonia. In this setting, the most common use of CT is to assess for complications that may be producing a poor response to treatment. These include empyema, lung abscess, and bronchopleural fistula (see under Pleural Processes). Additional reasons for a poor response to treatment include infection with an atypical organism or one that is resistant to the antibiotics being used, postobstructive pneumonia, and noninfectious pulmonary abnormalities that may mimic pneumonia on plain radiographs. Of these, CT is most likely to be helpful in identifying a central lesion that is associated with a postobstructive pneumonia. CT may also be useful in detecting occult pneumonias, undiagnosed by standard chest radiographs. This is most often encountered in intensive care unit patients with multiple other medical problems, as discussed by one of us (WTM) and colleagues (1998). Thus, in the setting of fever of unknown origin in debilitated patients, a chest CT may be warranted to search for occult pneumonia.

HIGH-RESOLUTION COMPUTED TOMOGRAPHY OF LUNG PARENCHYMA

The very thin sections obtained with HRCT allow for an anatomically accurate depiction of the fine structure of the lung parenchyma; this is of value in the characterization of many lung diseases, particularly diffuse lung diseases. Interstitial lung disease, emphysema, and bronchiectasis are very well evaluated with HRCT. However, the acquisition of noncontiguous sections means that large portions of the lung parenchyma are not imaged, so that small nodules can easily be missed. It also may be more difficult to distinguish small pulmonary vessels from nodules on HRCT because thin sections through vessels usually appear rounded, whereas thick sections may demonstrate more of the linear shape of the vessel.

Interstitial Lung Disease

HRCT has become an important modality for the evaluation of interstitial lung diseases, both in patients with confusing clinical findings and in patients with questionable abnormalities at chest radiography. HRCT was developed to help address the weaknesses of the chest radiograph, which, as Hansell and Kerr (1991) point out, may be normal in up to 10% of patients with biopsy-proven lung disease and may give false-positive impressions of diffuse lung disease, particularly in obese patients.

Diffuse lung diseases are best characterized on CT by their distribution within the anatomic unit of lung known as the secondary pulmonary lobule. This anatomic unit is the smallest unit of lung divided by connective tissue septa and varies in size from about 1 to 2.5 cm in diameter. The surrounding septa measure about 0.1 mm in thickness and are usually below the limits of HRCT resolution. When the septa are seen, they are usually abnormally thickened, and certain disease processes are suggested. Each secondary pulmonary lobule contains a centrilobular artery and bronchiole; other diseases tend to affect this centrilobular region.

As Webb and co-workers (2001) stated, the basic patterns of interstitial lung disease at HRCT include peribronchovascular thickening, interlobular septal thickening, intralobular interstitial thickening, nodular opacities, honeycombing, and ground-glass opacities. Small nodules can be distributed randomly throughout the lung or can be found in a predominantly centrilobular distribution.

Webb and associates (2001) discuss the most common HRCT patterns of diffuse lung disease as follows. Peribronchovascular thickening is most commonly seen in lymphangitic carcinomatosis, pulmonary edema, and sarcoidosis; interlobular septal thickening is also frequently seen in these diseases. Intralobular interstitial thickening is manifested by a fine, peripheral, netlike or reticular pattern and is most commonly seen in usual interstitial pneumonitis (UIP), nonspecific interstitial pneumonitis (NSIP), and other idiopathic interstitial pneumonias as well as in chronic hypersensitivity pneumonitis and asbestosis (Fig. 10-7). Small nodules distributed randomly are likely to be seen in miliary tuberculosis or fungal infections or in hematogenous metastases. Small nodules in a peribronchovascular distribution are seen in sarcoidosis, silicosis, lymphangitic carcinomatosis, and lymphocytic interstitial pneumonitis in patients with acquired immunodeficiency syndrome. Centrilobular nodules are primarily seen in endobronchial spread of tuberculosis or nontuberculous mycobacterial infections, in bronchopneumonia, in Langerhans' histiocytosis, and in hypersensitivity pneumonitis.

P.155

Honeycombing, a pattern of small cysts, has become an area of great interest in recent years. Although some honeycombing may be seen in a variety of diseases leading to fibrosis and end-stage lung, the pattern of predominantly subpleural cysts, most prevalent at the lung bases, is now considered to be characteristic of UIP. As Hunninghake and colleagues (2001) have shown, when CT scans of patients with possible UIP are interpreted by radiologists experienced with interstitial lung disease as highly suggestive of this diagnosis, lung biopsy is unnecessary. They suggest that this is a valuable finding, given the high rate of complications with surgical biopsy in this patient population.

Ground-glass opacities have been stated previously to be a manifestation of air space disease as well as interstitial lung disease. The list of diseases that cause ground-glass opacity is extremely long, and the differential diagnosis is often limited by the chronicity of disease. Acute diseases that are likely to cause ground-glass opacity include acute interstitial pneumonitis, adult respiratory distress syndrome, pulmonary edema and hemorrhage, acute eosinophilic pneumonia, and acute infectious pneumonias such as P. carinii, viral, and Mycoplasma pneumonias. Subacute and chronic causes of ground-glass opacity include NSIP, hypersensitivity pneumonitis, bronchiolitis obliterans organizing pneumonia, and pulmonary alveolar proteinosis (see Fig. 10-6).

The pattern of abnormality on CT can also be used to assess the optimal way to obtain tissue for diagnosis of a lung disease when a tissue diagnosis is required. For example, diseases that produce abnormalities in a predominantly peribronchovascular pattern affect the lung tissue adjacent to the bronchi, and transbronchial biopsy therefore has a high diagnostic yield for these diseases. Leung and co-workers (1993) historically suggested that ground-glass opacity may indicate the presence of an ongoing active disease and that biopsy directed toward areas of ground-glass opacity might therefore have a higher yield. More recently, attention has focused on the distinction of UIP, fibrotic, and cellular NSIP. Distinguishing among these diseases has a pronounced impact on treatment and prognosis. As Flaherty and colleagues (2001) stated, HRCT may direct biopsy toward areas of lung with the best diagnostic yield. Biopsy specimens should ideally be obtained from areas that show the entire range of manifestations of the lung disease. Areas that contain ground-glass opacity are therefore still a good target, as are areas of relatively preserved lung. Areas of severe honeycombing should be avoided because only end-stage lung will be identified. The lung specimen must have fibroticlung adjacent to normal lung for the pathologist to identify a UIP pattern. As these authors noted, multiple histologic patterns may coexist in the same lung, and it is important to identify the UIP pattern, which carries a poor prognosis if seen anywhere in the lung.

Emphysema

A variety of diseases can lead to diffusely distributed areas of decreased lung attenuation at CT. These include relatively uncommon diseases such as Langerhans' histiocytosis and lymphangiomyomatosis, large airways diseases such as bronchiectasis, small airways diseases such as bronchiolitis obliterans, and interstitial diseases with associated honeycombing. However, by far the most common disease to lead to decreased pulmonary parenchymal attenuation is emphysema.

There are three major subtypes of emphysema: centrilobular emphysema, panlobular emphysema, and paraseptal emphysema. According to Webb and co-workers (2001), the chief CT characteristic that distinguishes emphysema from diseases such as histiocytosis X or lymphangiomyomatosis is the absence of walls around the low-attenuation areas. The subtypes of emphysema are distinguished by their predominant location within the structure of the lung. In their milder forms, they are readily distinguished from one another, but as they become more severe, the findings become more similar. Centrilobular emphysema is the most common type of emphysema and is most strongly associated with cigarette smoking. It is characterized by small, round areas of low attenuation grouped near the centers of secondary pulmonary lobules. Although the margins of the secondary pulmonary lobule are not always seen discretely, the presence of small, scattered areas of emphysema is diagnostic of centrilobular

P.156

Bullae are defined as sharply demarcated areas of emphysema measuring 1 cm or greater in diameter with thin walls; these can become very large. Although most strongly associated with paraseptal emphysema, they can be seen in association with other types of emphysema as well. In patients in whom bullectomy is planned, HRCT can be used to define the extent of emphysema in the underlying lung.

HRCT can be useful in making the diagnosis of emphysema in patients who present with confusing clinical findings (e.g., an isolated low diffusing capacity on pulmonary function tests, which may have suggested interstitial lung disease). A finding of significant emphysema may obviate a lung biopsy. HRCT is also considered an essential component of planning for lung volume reduction surgery. As Hunsaker and associates (2002) indicate, heterogeneous, upper lobe predominant distribution of emphysema on HRCT is associated with better response to lung volume reduction surgery.

Bronchiectasis

Bronchiectasis is best evaluated with thin-section CT, although it can be seen on conventional CT examinations. CT has replaced the older technique of bronchography in establishing this diagnosis. Criteria for the diagnosis of bronchiectasis include airways with a diameter 1.5 or more times the diameter of the adjacent pulmonary artery, airways seen to extend into the outer one third of the lung, and airways imaged along their long axis that either fail to taper as they become more distal or actually become larger in diameter (Fig. 10-9). By definition, the airway dilatation should be irreversible. Because bronchi may become transiently dilated during an episode of acute pneumonia, the diagnosis should not be made at that time.

The most common cause of bronchiectasis is acute, chronic, or recurrent infection. Diffuse bronchiectasis is most likely to occur in patients with an underlying abnormality of host defenses against infection, such as cystic fibrosis, dyskinetic cilia syndrome (Kartagener's syndrome), or abnormalities of the immunoglobulin system. Focal bronchiectasis, in contrast, may be caused by disorders such as mycobacterial infection or a single episode of severe acute bacterial pneumonia. Allergic bronchopulmonary aspergillosis tends to present with central bronchiectasis. The finding of focal bronchiectasis in a patient who presents with hemoptysis may allow for treatment of the hemoptysis by resection of the area of involved lung. Similarly, chronic infection with associated bronchiectasis may be treated by

P.157

At CT, large, solid-appearing branching structures usually represent areas of bronchiectasis with mucoid impaction. Tiny branching peripheral structures may represent dilated bronchioles with inflammatory bronchiolar wall thickening (known as tree-in-bud opacities); this may be referred to as bronchiolectasis and is often associated with an acute infectious process.

PULMONARY VASCULAR LESIONS

CT can be extremely useful in the noninvasive diagnosis of vascular lesions of the lung, including pulmonary arteriovenous malformation, pulmonary varix, scimitar syndrome (partial anomalous pulmonary venous return), and sequestration. Contrast-enhanced studies can demonstrate the vascular nature of all these aforementioned lesions. Arteriovenous malformations present as a pulmonary nodule or mass. Pulmonary varix and partial anomalous pulmonary venous return have distinct morphologic appearances. In patients with a chronic lower lobe paravertebral mass or focal consolidation suspected of representing sequestration, the diagnosis may be established by CT if the abnormal systemic artery supplying the lesion is identified. This may obviate the need for more invasive conventional aortography.

In recent years, there has been increasing interest in the use of CT to evaluate for pulmonary emboli. There has been interest in the development of the technique largely because of dissatisfaction with ventilation perfusion scintigraphy. As Mayo and colleagues (1997) noted, at the typical tertiary care referral center, only 34% of cases imaged with ventilation-perfusion imaging have a combination of clinical suspicion and findings on [V with dot above]/[V with dot above] scans that give a high enough predictive value to obviate further imaging. Clinicians are often reluctant to order pulmonary angiograms because of the exaggerated perception of the morbidity and mortality of the procedure, even when it is theoretically indicated, as discussed by Sostman (1982) and by Henschke and associates (1995). There is therefore a widespread desire for a better noninvasive examination for pulmonary embolism.

CT examinations for pulmonary emboli are now most commonly performed with MDCT, although eln-beam CT has also been used for this indication. Studies must be tailored to detect pulmonary emboli, with thin sections and a rapid infusion of intravenous contrast, although large, central emboli may be incidentally appreciated on studies performed for other indications. CT for pulmonary emboli is sometimes called CT pulmonary arteriography, or CTPA. It is clear that central pulmonary emboli can be identified with CT to the segmental arterial level (Fig. 10-10). Mayo and co-workers (1997) found a higher sensitivity with this test than with. [V with dot above]/[Q with dot above] scanning. However, as Goodman (1995)

P.158

As with angiography, findings on CT in patients with chronic pulmonary thromboembolic disease are different from those in patients with acute pulmonary emboli. In this disorder, acute emboli fail to undergo normal lysis, become organized and fibrotic, and may ultimately become incorporated within the walls of the pulmonary arteries. Such chronic thromboemboli may lead to the development of pulmonary arterial hypertension, which may be treated with surgical thromboendarterectomy. Only patients with clots in the proximal pulmonary arteries are candidates for surgery. Remy-Jardin and associates (1992) have shown that the presence and extent of such central mural clots may be underestimated on conventional pulmonary arteriograms, and CT may demonstrate them without the risks of the pulmonary arteriogram.

FINDINGS ON COMPUTED TOMOGRAPHY: PLEURAL PROCESSES

Much smaller abnormalities of the pleura are likely to be visible on CT compared with other imaging modalities, particularly when there are abnormalities within the lung parenchyma that may mask a pleural process. In the case of pleural effusions, this is particularly true in patients who are too ill to cooperate with lateral decubitus radiographs. Pneumothoraces are similarly more difficult to detect in patients in whom erect radiographs cannot be obtained.

Ascites Versus Pleural Effusion

Distinction of pleural fluid from ascites is not always straightforward with CT, and a number of CT signs have been described to help with this differentiation. Specifically, the following criteria suggest ascites: (a) fluid is inside the curved lines created by axial sections through the diaphragm and therefore below the diaphragm; (b) fluid does not elevate the crus of the diaphragm; (c) the interface between the fluid and liver or spleen is sharply defined (implying the absence of intervening diaphragmatic tissue); and (d) fluid is not posterior or medial to the liver at the level of the bare area. If the alternate conditions are met, the fluid is probably within the pleural space. Used individually, each of these signs may be indeterminate or misleading; however, if all four criteria for either pleural fluid or ascites are met, accurate distinction of one from the other may be made. The distinction of pleural fluid from ascites has been discussed by Dwyer (1978), Griffin (1984), Halvorsen (1986), Naidich (1983), and Teplick (1982) and their colleagues. With helical technique, sagittal or coronal reconstructions may also occasionally be used to help distinguish ascites from pleural fluid.

Although the distinction of exudative effusion from transudative effusion is definitively made only by thoracentesis, several CT signs have been reported to suggest that an effusion is exudative. Pleural thickening suggests an exudative effusion, with a specificity of 96% in a series reported by Aquino and co-workers (1994); however, it is much less sensitive than specific, particularly in the setting of malignancy. As McLoud and Flower (1991) note, acute hemorrhage in the pleural space can sometimes be identified on CT by high attenuation within the pleural space or by the fluid-hematocrit level.

Parapneumonic Effusion Versus Empyema

Pleural fluid seen in a patient with pneumonia may represent a simple parapneumonic effusion, a complicated parapneumonic effusion, or empyema. The distinction is made by sampling the pleural fluid. However, some CT features may suggest that an effusion does not represent a simple parapneumonic effusion. Specifically, thickening of the visceral and parietal pleura, enhancement of these surfaces with intravenous contrast administration, and inflammatory stranding of the extrapleural fat may suggest either a complicated effusion or frank infection. The split-pleura sign strongly suggests an exudative effusion, which is likely to be either empyema or complicated effusion in the setting of infection. This sign, as described by Stark and associates (1983), consists of demonstration of fluid, of relatively low attenuation, between enhancing visceral and parietal pleura.

The presence of air in the pleural space is most commonly caused by recent instrumentation or by bronchopleural fistula. Theoretically, infection of the pleural space with gas-forming organisms can also lead to gas in the pleural space; however, this is exceedingly rare.

Bronchopleural Fistula

The term bronchopleural fistula indicates a communication between the lung parenchyma or airways and the pleural space. The presence of a bronchopleural fistula is often inferred on imaging studies from the presence of an

P.159

As Hsu and colleagues (1972) have indicated, bronchopleural fistulae can be divided into central fistulae, which most commonly occur in the postoperative setting, particularly following lobectomy or pneumonectomy, and peripheral fistulae. According to McManigle and co-workers (1990), central bronchopleural fistulae arise from the large airways and are best evaluated and characterized with bronchoscopy.

Peripheral bronchopleural fistulae are most frequently seen in the settings of tumor, infection, or bronchiectasis. CT can be used to characterize the abnormality in the underlying lung that might cause the fistula. Additionally, Westcott and Volpe (1995) demonstrated that, if the monitoring physician is alerted to the specific clinical question, thin-section CT through an area of abnormality may be able to demonstrate the communication or fistula between the abnormal lung and pleural space.

Abscess Versus Empyema

Sometimes, it can be difficult to distinguish a lung parenchymal abscess from an empyema with a bronchopleural fistula. Both abnormalities may occur in the setting of pneumonia and will demonstrate an air fluid level on plain radiographs and CT. Shape and location should help distinguish between the two entities.

Empyemas are enclosed within the pleural cavity and conform to the shape of the chest wall, whereas lung abscesses arise within the lung parenchyma. CT, by giving greater spatial information, may help to make this distinction. Empyemas are more commonly elliptical and usually have thin, smooth walls, especially along their inner margins (Fig. 10-11). Abscesses, in contrast, because they originate within the pulmonary parenchyma and destroy adjacent lung, have a more spherical shape. Most abscesses tend to have thickened irregular walls and margins. As Stark and co-workers (1983) indicated, compression and crowding of the adjacent pulmonary parenchymal markings also suggests a pleural process rather than a lung abscess. In the setting of a lung abscess, in contrast, the bronchi and vessels of the adjacent lung appear to be truncated abruptly at the edge of the abscess at the margins of the destroyed lung.

Pleural Soft Tissue Processes

CT is superior to plain film when attempting to distinguish a pleural mass from a subpleural mass; however, even with CT, this distinction is not always easy to make. Soft tissue processes that can involve the pleura include neoplastic processes such as metastatic disease, malignant mesothelioma, fibrous tumors, and lipomas. Nonneoplastic soft tissue processes that can be seen in the pleural space include asbestos-related pleural plaques, postinflammatory or posttraumatic pleural thickening, and subpleural fat deposition.

When a new pleural effusion of unclear etiology is discovered, the effusion is often suspected to be malignant. The definitive diagnosis of a malignant pleural effusion requires a pathologic diagnosis. CT findings that suggest that an effusion is malignant include irregular pleural thickening and, more specifically, enhancement of pleural nodules or masses (Fig. 10-12). According to Kuhlman (1997), other features that have been described in association with malignant effusions include pleural thickening that exceeds 1 cm and involvement of the mediastinal pleura. Malignant pleural effusions with pleural thickening or masses may be seen with pleural metastases from a wide variety of tumors, or in malignant mesothelioma.

Malignant mesothelioma usually appears on CT as a thick rind of soft tissue encasing the lung (Fig. 10-13). A variable quantity of fluid, usually loculated, is generally present. The density of this fluid is usually less than that of the rind of soft tissue, and this difference will be increased if intravenous contrast is administered. Malignant mesothelioma may spread directly to involve the mediastinum, pericardium, contralateral lung, or chest wall. It may also spread through the diaphragm to involve the abdominal and retroperitoneal viscera. Mediastinal lymphadenopathy is common. CT is generally considered the optimal study to identify the extent of involvement with mesothelioma,

P.160

Only two types of benign tumors commonly involve the pleura: pleural fibromas (which may be benign or malignant, and were formerly known as benign mesotheliomas) and lipomas. Pleural fibromas appear as a mass of soft tissue attenuation, similar to muscle, without an associated pleural effusion. A diagnosis of pleural lipoma can be made with confidence when a mass of uniformly fatty density is seen at CT, as reported by McLoud and Flower (1991) (Fig. 10-14).

Asbestos-related pleural plaques appear as plateau-shaped pleural soft tissue masses. These are usually bilateral and are predominantly located in the middle to lower chest and over the diaphragms. They are frequently calcified, which adds to the confidence of the diagnosis. It may be difficult to distinguish noncalcified plaques from pulmonary nodules. Occasionally, asbestos-related plaques may simulate a pulmonary nodule on plain radiographs; if oblique radiographs are not reassuring, CT may demonstrate the absence of nodules as well as showing the plaques that explain the radiographic abnormality.

P.161

COMPUTED TOMOGRAPHY IN THE IDENTIFICATION OF CHEST WALL PATHOLOGY

Abnormalities of the chest wall for which cross-sectional imaging may be indicated include benign and mesenchymal tumors, primary and secondary malignancies, and inflammatory and infectious diseases. As with lung carcinoma, MR imaging may also be used for evaluating abnormalities near the lung apex. This particularly applies to lesions of the brachial plexus and abnormalities of the diaphragm.

As in the pleural space, demonstration of a mass with homogeneous low attenuation of -100 to -160 Hounsfield units allows for the diagnosis of a lipoma. As Kuhlman and co-workers (1994) indicated, liposarcoma will usually be inhomogeneous with areas of soft tissue attenuation as well as fat attenuation.

Masses of soft tissue density are much less likely to be specifically characterized with imaging, whether CT or MR. The primary utility of CT in other chest wall neoplasms is to assess the extent of disease. Masses may arise from any of the soft tissues that make up the chest wall, including fibrous connective tissue, nerves, muscles, blood vessels, breast tissue, lymphatics, cartilage, and bone. Associated bony changes are often seen with chest wall masses. Tumors of neural origin are very common and include neurofibromas, schwannomas, neurofibrosarcomas, and neuromas as well as neuroblastomas, ganglioneuromas, and ganglioneuroblastomas. Tumors of blood vessels include hemangiomas, a term that encompasses a variety of subtypes; focal calcifications within these, known as phleboliths, may suggest the diagnosis. Lymphangiomas are frequently low in attenuation at CT; however, differentiation from surrounding muscle may be very difficult, particularly if no intravenous contrast is administered, as reported by Kuhlman and co-workers (1994).

Of the malignant soft tissue tumors that may arise in the chest wall, breast carcinoma is by far the most common. CT has very little role to play in the diagnosis of breast carcinoma; however, advanced tumors may be incidentally identified, and CT can detect local recurrence following mastectomy (Fig. 10-15). Other malignancies that may arise in the chest wall include soft tissue sarcomas and lymphoma.

Tumors that are centered in bone on CT may include osseous metastases as well as tumors that are primary to cartilage or bone (Fig. 10-16). The most common primary tumor of bone in adults is multiple myeloma. Less common tumors include osteosarcoma, chondrosarcoma, and fibrosarcoma as well as benign tumors and tumor-like lesions, including fibrous dysplasia, enchondroma, and aneurysmal bone cyst. Although the appearance of all of these tumors on CT is likely to be fairly nonspecific, some of them will have characteristic findings on CT or radiography.

Infections in the chest wall are somewhat rare but may be life threatening. As Kuhlman and associates (1994) noted, organisms that may cause such infections include pyogenic bacteria, actinomycetes, blastomycetes, Nocardia and Aspergillus species, and Mycobacterium tuberculosis. Some of these agents tend to involve the chest wall through extension from an underlying pulmonary infection. Other mechanisms by which chest wall infection may arise include infection of a surgical incision or other direct trauma. CT may demonstrate areas of bone destruction as well as skin fistulae. In the setting of necrotizing fasciitis, CT is probably the most sensitive imaging technique for the detection of gas within the soft tissues (see Fig. 10-5).

P.162

Postoperative Evaluation and the Critically Ill Patient

The normal PA and lateral examination of the chest provides a great deal of spatial information about abnormalities within the chest because they can be localized in two planes. However, critically ill patients and many postoperative patients are unable to undergo a standard radiographic examination of the chest. Portable radiography is much more limited, especially when patients are imaged in the supine position.

In particular, pneumothoraces, pleural effusions, and parenchymal opacities at the lung bases may be very difficult to identify on supine radiographs (Fig. 10-17). Pleural fluid may be difficult to distinguish from parenchymal opacity at the bases on portable semierect films. Air fluid levels are also difficult to detect on portable films, and thus lung abscesses may also be missed on these radiographs. All of these abnormalities are easily detected on CT. Several series, including those reported by Mirvis (1987), Golding (1988), one of us (WTM) (1992), and Voggenreiter (2000) and colleagues, showed that CT contributed information affecting the management of selected patients when performed to assess unexplained clinical deterioration or failure to respond to therapy. Commonly detected abnormalities include unsuspected fluid collections, previously undetected foci of pneumonia, and malignancies.

When CT is performed in postoperative patients, trauma patients, or other critically ill patients, thoracostomy tube position can be assessed. Cameron and co-workers (1997) have shown that a number of abnormalities in thoracostomy drain placement are visible on CT. These drains may terminate within the lung parenchyma, indicating that the pleura has been violated. Intraparenchymal placement may, however, be difficult to distinguish from placement within a fissure. Drains may also be placed with their tips abutting the mediastinum, which over time may lead to erosion of mediastinal structures or to contusion. Occasionally, the tubes may be placed outside the thoracic cavity, either in the chest wall or into the abdomen.

Templeton and Fishman (1992) indicated that CT is an important part of evaluation of patients with suspected sternal wound infections following median sternotomy. CT can demonstrate fluid collections, which may be seen postoperatively both superficial and deep to the sternum. The identification of a fluid collection is not alone diagnostic of abscess; collections may be sterile or infected, and the fluid must be sampled for definitive diagnosis. If there are large amounts of gas in a collection, this increases the probability that infection is present, but small amounts of gas are normal in at least the first postoperative week. Bony destruction is very suggestive of sternal wound infection (Fig. 10-18). Mediastinitis, a life-threatening postoperative complication, can be difficult to diagnose. Soft tissue density obliterating the normal mediastinal fat can suggest this diagnosis, and gas collections are also suggestive; however, these also may be normal postoperative findings, and the temporal relationship to surgery must be considered.

As Templeton and Fishman (1992) also noted, CT has only a small role to play in the evaluation of sternal dehiscence, which is usually detectable on clinical grounds. Nevertheless, rotation of the sternal wires, displacement or fracture, and apparent widening of the lucency between the sternal fragments may be seen in such a dehiscence.

Kopka and co-workers (1996) have noted that retained surgical sponges are not always easily detected on CT. The dense markers on surgical sponges may be mistaken for calcifications, and small gas bubbles are frequently, but not always, seen. Small gas bubbles may represent retained air in these sponges and should not be considered a sign of infection.

P.163

CONCLUSION

Although chest radiographs remain the primary means of imaging the thorax, chest CT has become an integral part of the evaluation of thoracic diseases. It provides increased detail regarding the contents of the thorax, particularly the mediastinum, pleura, and chest wall, and is more sensitive than radiography for a variety of thoracic diseases, including diffuse lung diseases. In patients in whom two-view radiographs cannot be performed, CT may give even more additional information. The additional information obtained from CT can improve accuracy in the differential diagnosis and staging of thoracic diseases. CT also has an important role to play in surgical planning. It is important for referring clinicians to understand the capabilities and limitations of CT so that CT may be protocoled and performed correctly, maximizing the diagnostic yield of these important studies.

REFERENCES

Aberle DR, et al: A consensus statement of the Society of Thoracic Radiology: screening for lung cancer with helical computed tomography. J Thorac Imaging 16:65, 2001.

Aquino SL, Webb, WR, Gushiken BJ: Pleural exudates and transudates: diagnosis with contrast-enhanced CT. Radiology 192:803, 1994.

Arita T, et al: Bronchogenic carcinoma: incidence of metastases to normal sized lymph nodes. Thorax 50:1267, 1995.

Austin JH, et al: Glossary of terms for CT of the lungs: recommendations of the Nomenclature Committee of the Fleischner Society. Radiology 200:327, 1996.

Berland LL, Smith JK: Multidetector-array CT: once again, technology creates new opportunities. Radiology 209:327, 1998.

Bilbey JH, et al: MR imaging of adrenal masses: value of chemical-shift imaging for distinguishing adenomas from other tumors. AJR Am J Roentgenol 164:637, 1995.

Boland GW, et al: Adrenal masses: characterization with delayed contrast-enhanced CT. Radiology 202:693, 1997.

Brodeur FJ Jr, Kazerooni EA: Delayed enhanced CT: a method of differentiating adrenal adenomas from nonadenomas [Abstract]. Radiology 197(P):185, 1995.

Cameron EW, et al: Computed tomography of malpositioned thoracostomy drains: a pictorial essay. Clin Radiol 52:187, 1997.

Caoili EM, et al: Adrenal masses: characterization with combined unenhanced and delayed enhanced CT. Radiology 222:629, 2002.

Chorost MI, et al: The management of the unknown primary. J Am Coll Surg 193:666, 2001.

Curry TS III, Dowdey JE, Murry RC Jr: Christensen's Physics of Diagnostic Radiology. 4th Ed. Philadelphia: Lea&Febiger, 1990.

D'Amico TA, et al: Impact of computed tomography-positron emission tomography fusion in staging patients with thoracic malignancies. Ann Thorac Surg 74:160, 2002.

Davis SD, et al: Screening for pulmonary metastases. American College of Radiology. ACR Appropriateness Criteria. Radiology 215:655, 2000.

Dewan NA, et al: Diagnostic efficacy of PET-FDG imaging in solitary pulmonary nodules. Potential role in evaluation and management. Chest 104:997, 1993.

Dwyer A: The displaced crus: a sign for distinguishing between pleural fluid and ascites on computed tomography. J Comput Assist Tomogr 2:598, 1978.

Epstein DM, Gefter WB, Miller WT: Lobar bronchioloalveolar cell carcinoma. AJR Am J Roentgenol 139:463, 1982.

Flaherty KR, et al: Histopathologic variability in usual and nonspecific interstitial pneumonias. Am J Respir Crit Care Med 164:1722, 2001.

Fraser RS, et al (eds): Synopsis of Diseases of the Chest. 2nd Ed. Philadelphia: WB Saunders, 1994.

Gefter WB: The spectrum of pulmonary aspergillosis. J Thorac Imaging 7:56, 1992.

Giron J, et al: Interventional chest radiology. Eur J Radiol 23:58, 1996.

Glazer HS, et al: Pleural and chest wall invasion in bronchogenic carcinoma: CT evaluation. Radiology 157:191, 1985.

Goldberg SN, et al: Pulmonary embolism: prediction of the usefulness of initial ventilation-perfusion scanning with chest radiographic findings. Radiology 193:801, 1994.

Golding RP, et al: Computed tomography as an adjunct to chest x-rays of intensive care unit patients. Crit Care Med 16:211, 1988.

Goodman LR, et al: Detection of pulmonary embolism in patients with unresolved clinical and scintigraphic diagnosis: helical CT versus angiography. AJR Am J Roentgenol 164:1369, 1995.

Griffin DJ, et al: Observations on CT differentiation of pleural and peritoneal fluid. J Comput Assist Tomogr 8:24, 1984.

Gupta NC, et al: Solitary pulmonary nodules: detection of malignancy with PET with 2-[F-18]-fluoro-2-deoxy-D-glucose. Radiology 184:441, 1992.

Hagberg RC, et al: Characterization of pulmonary nodules and mediastinal staging of bronchogenic carcinoma with F-18 fluorodeoxyglucose positron emission tomography. Eur J Cardiothorac Surg 12:92, 1997.

Halvorsen RA, et al: Ascites or pleural effusion? CT differentiation: four useful criteria. Radiographics 6:135. 1986.

Hansell DM, Kerr IH: The role of high resolution computed tomography in the diagnosis of interstitial lung disease. Thorax 46:77, 1991.

Henschke CI, Mateescu I, Yankelevitz DF: Changing practice patterns in the workup of pulmonary embolism. Chest 107:940, 1995.

Herman SJ, et al: Mediastinal invasion by bronchogenic carcinoma: CT signs. Radiology 190:841, 1994.

Hsu JT, Bennett GM, Wolff E: Radiologic assessment of bronchopleural fistula with empyema. Radiology 103:41, 1972.

P.164

Hunninghake GW, et al: Utility of a lung biopsy for the diagnosis of idiopathic pulmonary fibrosis. Am J Respir Crit Care Med 164:193, 2001.

Hunsaker AR, et al: Lung volume reduction surgery for emphysema: correlation of CT and [V with dot above]/[V with dot above] imaging with physiologic mechanisms of improvement in lung function. Radiology 222:491, 2002.

Huston J III, Muhm JR: Solitary pulmonary opacities: plain tomography. Radiology 163:481, 1987.

Kalender WA, et al: Spiral volumetric CT with single-breath-hold technique, continuous transport, and continuous scanner rotation. Radiology 176:181, 1990.

Katagiri H, et al: Determining the site of the primary cancer in patients with skeletal metastasis of unknown origin: a retrospective study. Cancer 86:533, 1999.

Kazerooni EA, et al: Thin-section CT obtained at 10-mm increments versus limited three-level thin-section CT for idiopathic pulmonary fibrosis: correlation with pathologic scoring. AJR Am J Roentgenol 169:977, 1997.

Klein JS, Salomon G, Stewart EA: Transthoracic needle biopsy with a coaxially placed 20-gauge automated cutting needle: results in 122 patients. Radiology 198:715, 1996.

Klein JS, Schultz S, Heffner JE: Interventional radiology of the chest: image-guided percutaneous drainage of pleural effusions, lung abscess, and pneumothorax. AJR Am J Roentgenol 164:581, 1995.

Kopka L, et al: CT of retained surgical sponges (textilomas) pitfalls in detection and evaluation. J Comput Assist Tomogr 20:919, 1996.

Korobkin M, et al: Differentiation of adrenal adenomas from nonadenomas using CT attenuation values. AJR Am J Roentgenol 166:531, 1996.

Kuhlman JE: Complex disease of the pleural space: the 10 questions most frequently asked of the radiologist new approaches to their answers with CT and MR imaging. Radiographics 17:1043, 1997.

Kuhlman JE, et al: Invasive pulmonary aspergillosis in acute leukemia: characteristic findings on CT, the CT halo sign, and the role of CT in early diagnosis. Radiology 157:611, 1985.

Kuhlman JE, et al: CT and MR imaging evaluation of chest wall disorders. Radiographics 14:571, 1994.

Lee MJ, et al: Benign and malignant adrenal masses: CT distinction with attenuation coefficients, size and observer analysis. Radiology 179:415, 1991.

Lesser BA, et al: The diagnosis of acute pulmonary embolism in patients with chronic obstructive pulmonary disease. Chest 102:17, 1992.

Leung AN, Miller RR, Muller NL: Parenchymal opacification in chronic infiltrative lung diseases: CT-pathologic correlation. Radiology 188: 209, 1993.

Leung AN, Staples CA, Muller NL: Chronic diffuse infiltrative lung disease: comparison of diagnostic accuracy of high-resolution and conventional CT. AJR Am J Roentgenol 157:693, 1991.

Libshitz HI, et al: Mediastinal evaluation in lung cancer. Radiology 151:295, 1984.

Mayo JR, et al: Pulmonary embolism: prospective comparison of spiral CT with ventilation-perfusion scintigraphy. Radiology 205:447, 1997.

McLoud TC, Flower CDR: Imaging the pleura: sonography, CT, and MR imaging. AJR Am J Roentgenol 156:1145. 1991.

McManigle JE, Fletcher GL, Tenholder MF: Bronchoscopy in the management of bronchopleural fistula. Chest 97:1235, 1990.

Miller WT Jr, et al: Thoracic CT in the intensive care unit: assessment of clinical usefulness. Radiology 209:491, 1998.

Mirvis SE, et al: Thoracic CT in detecting occult disease in critically ill patients. AJR Am J Roentgenol 148:685, 1987.

Mitchell DG, et al: Benign adrenocortical masses: diagnosis with chemical-shift imaging. Radiology 185:345, 1992.

Munden RT, et al: Small pulmonary lesions detected at CT: clinical importance. Radiology 202:105, 1997.

Naidich DP, et al: Computed tomography of the diaphragm: peridiaphragmatic fluid localization. J Comput Assist Tomogr 7:641, 1983.

Naidich DP, et al: Computed tomography of the pulmonary parenchyma. Part I. Distal air-space disease. J Thorac Imaging 1:39, 1985.

Naidich, DP, et al: Hemoptysis: CT-bronchoscopic correlations in 58 cases. Radiology 177:357, 1990.

Onitsuka H, et al: Differentiation of central lung tumor from postobstructive lobar collapse by rapid sequence computed tomography. J Thorac Imaging 6:28, 1991.

Patz EF Jr, et al: Focal pulmonary abnormalities: evaluation with F-18 fluorodeoxyglucose PET scanning. Radiology 188:487, 1993.

Patz EF Jr, et al: Lung cancer staging and management: comparison of contrast-enhanced and nonenhanced helical CT of the thorax. Radiology 212:56, 1999.

Patz EF Jr, et al: CT screening for lung cancer: not ready for routine practice. Radiology 221:587, 2001.

Pennes DR, et al: Chest wall invasion by lung cancer: limitations of CT evaluation. AJR Am J Roentgenol 144:507, 1985.

Ratto GB, et al: Chest wall involvement by lung cancer: computed tomographic detection and results of operation. Ann Thorac Surg 51:182, 1991.

Reinig JW, et al: Differentiation of adrenal masses with MR imaging: comparison of techniques. Radiology 192:41, 1994.

Remy J: Pulmonary arteriovenous malformations: evaluation with CT of the chest before and after treatment. Radiology 182:809, 1992.

Remy-Jardin M, et al: Central pulmonary thromboembolism: diagnosis with spiral volumetric CT with single breath-hold technique: comparison with pulmonary angiography. Radiology 185:381, 1992.

Remy-Jardin M, et al: Pulmonary nodules: detection with thick-section spiral CT versus conventional CT. Radiology 187:513, 1993.

Schneider HJ, Felson B, Gonzalez LL: Rounded atelectasis. AJR Am J Roentgenol 134:225, 1980.

Society for Computed Body Tomography: Special report. New indications for computed body tomography. AJR Am J Roentgenol 133:115, 1979.

Sone S, et al: Mass screening for lung cancer with mobile spiral computed tomography scanner. Lancet 351:1242, 1998.

Sostman HD: Use of pulmonary angiography for suspected pulmonary embolism: influence of scintigraphic diagnosis. AJR Am J Roentgenol 139:673, 1982.

Stark DD, et al: Differentiating lung abscess and empyema: radiography and computed tomography. AJR Am J Roentgenol 141:163, 1983.

Swensen SJ: CT screening for lung cancer. AJR Am J Roentgenol 179:833, 2002.

Swensen SJ, et al: CT evaluation of solitary pulmonary nodules: value of 185-H reference phantom. AJR Am J Roentgenol 156:925, 1991.

Swensen SJ, et al: Lung nodule enhancement at CT: multicenter study. Radiology 214:73, 2000.

Swensen SJ, et al: Screening for lung cancer with low-dose spiral computed tomography. Am J Respir Crit Care Med 165:508, 2002.

Teigen CL, et al: Pulmonary embolism: diagnosis with contrast-enhanced electron-beam CT and comparison with pulmonary angiography. Radiology 194:313, 1995.

Templeton PA, Fishman EK: CT evaluation of poststernotomy complications. AJR Am J Roentgenol 159:45, 1992.

Teplick JG, et al: The interface sign: a computed tomographic sign for distinguishing pleural and intra-abdominal fluid. Radiology 144:359, 1982.

Tsushima Y, Ishizaka H, Matsumoto M: Adrenal masses: differentiation with chemical-shift, fast low angle shot MR imaging. Radiology 186: 705, 1993.

Vock P, et al: Lung spiral volumetric CT with single breath-hold technique. Radiology 176:864, 1990.

Voggenreiter G, et al: Efficiency of chest computed tomography in critically ill patients with multiple traumas. Crit Care Med 28:1033, 2000.

von Haag DW, et al: Advantages of positron emission tomography over computed tomography in mediastinal staging of non-small cell lung cancer. J Surg Res 103:160, 2002.

Webb WR, Muller NL, Naidich DP: High-Resolution CT of the Lung. 3rd Ed. Philadelphia, Lippincott Williams&Wilkins, 2001.

Webb WR, et al: CT and MR imaging in staging non-small cell bronchogenic carcinoma: report of the Radiologic Diagnostic Oncology Group. Radiology 178:705, 1991.

Westcott JL, Volpe JP: Peripheral bronchopleural fistula: CT evaluation in 20 patients with pneumonia, empyema or postoperative air leak. Radiology 196:175, 1995.

Westcott J, et al: Rib fractures. American College of Radiology. ACR Appropriateness Criteria. Radiology 215:637, 2000.

Winer-Muram HT, et al: Volumetric growth rate of stage in lung cancer prior to treatment: serial CT scanning. Radiology 223:798, 2002.

Yankelevitz D, Henschke CI: Does 2-year stability imply that pulmonary nodules are benign? AJR Am J Roentgenol 168:325, 1997.

Zerhouni EA, et al: A standard phantom for quantitative CT analysis of pulmonary nodules. Radiology. 149:767, 1983.