State of the Art Computed Tomography of Chronic Diffuse Infiltrative Lung Disease Part 2

1,2

NESTOR L. MOLLER and ROBERTA R. MILLER Contents

Introduction Technique Normal Anatomy Irregular Linear Pattern Idiopathic Pulmonary Fibrosis Pulmonary Lymphangitic Carcinomatosis Asbestosis Cystic Pattern Lymphangioleiomyomatosis Nodular Pattern Silicosis and Coal Workers' Pneumoconioses Sarcoidosis Histiocytosis X Extrinsic Allergic Alveolitis Ground-Glass Pattern Alveolar Proteinosis Chronic Eosinophilic Pneumonia Bronchiolitis Obliterans Conclusions Nodular Pattern

Silicosis and Coal Workers' Pneumoconiosis Silicosis is defined as a fibrotic disease of the lungs caused by inhalation of dust containing crystallized silicon dioxide (67). The diagnosis of silicosis is dependent upon a history of exposure and confirmed by consistent radiographic and physiologic abnormalities (68). Estimates of disease severity conventionally involve a combination of radiographic and pulmonary functional assessment (69). Chest radiography is used to detect lymphadenopathy and the characteristic interstitial changes of silicosis. Pulmonary function tests are used to detect associated functional impairment. On computed tomography (CT), as on the radiograph, the most characteristic feature is the presence of nodules. In patients with mild silicosis, these nodules may be seen only in the upper lobes. They vary in size but are less than 1 cm in diameter. A posterior predominance of 1440

nodules is seen on CT (figure 9). More severe silicosis is characterized on CT by an increase in the number and size of nodules. CT is not superior to the standard chest radiograph for early detection of small opacities in workers exposed to silica. Indeed, mild disease may be easier to detect on the radiograph than on CT (70, 71). However, CT provides significant additional information on the stage ofthe disease because it detects coalescence of nodules that may not be apparent on the radiograph (70, 71). Early recognition of coalescence of silicotic lesions is important because such coalescence is associated with the appearance of respiratory symptoms and leads to deterioration of lung function (72). This complicated form of silicosis also has a worse prognosis than does simple silicosis (73). We compared CT in the qualitative and quantitative assessment of silicosis with chest radiographs and pulmonary function tests in 17 patients with silicosis and six control subjects (71). CT scans were graded for extent of silicosis visually and using mean attenuation values. The extent of associated emphysema was also determined. Significant correlation was found between both the mean attenuation values (r > 0.62, p < 0.001) and the visual CT scores (r > 0.84, p < 0.001) compared with the ILO category of profusion on the radiograph. There was poor correlation between the pulmonary function tests and the nodular profusion on the chest radiograph and on CT (r < 0.50). Correlation was significant, however, between the CT emphysema score and both the FEV 1070 predicted (r > 0.66, p < 0.001) and the carbon monoxide diffusing capacity (r > 0.71, p < 0.001). Attenuation values were less reliable and their use is not recommended as an inde-

pendent assessment of disease severity. The reduced levels of lung function in these patients correlated with superimposed emphysema rather than with the nodular profusion. Emphysema associated with silicosis was easily detected on CT but not on the radiograph. On CT, emphysema is characterized by the presence of areas of abnormally low attenuation. Several studies have shown a high sensitivity and specificity of CT in the detection of emphysema and a good correlation between the CT and pathologic emphysema scores (74-78). Kinsella and colleagues (79) reviewed pulmonary function testing and chest CT scans in 30 subjects with silicosis. Eighteen of them were either current or exsmokers, and 12 were nonsmokers. The extent of emphysema was the strongest independent predictor of pulmonary function impairment; extent of silicosis was also an independent predictor, albeit a weaker one. In those subjects who did not have evidence of progressive massive fibrosis, smokers had worse emphysema than did nonsmokers. In progressive massive fibrosis, the severity of emphysema in smokers and nonsmokers was not statistically different. These data suggest that silicosis does not cause significant emphysema in the absence of progressive

This is Part 2 of two parts; the first part appeared in the last issue ofthe Review (Vol. 142, Number 5). 1 From the Departments of Radiology and Pathology, University of British Columbia and Vancouver General Hospital, Vancouver, British Columbia, Canada. 2 Correspondence and requests for reprints should be addressed to Nestor L. Muller, M.D., Department of Radiology, University of British Columbia, 855 W. 12th Avenue, Vancouver, BC V5Z IM9, Canada.

AM REV RESPIR DIS 1990; 142:1440-1448

1441

STATE OF THE ART: CT OF CILD

Fig. 9. CT scan in patient with silicosis shows numerous bilateral nodules. These are more numerous in the posterior half of the lungs. Early confluence of nodules is present. CT scan parameters: 10-mm collimation, 40-cm field-of-view, and standard reconstruction algorithm.

massive fibrosis and that it is primarily the degree of emphysema rather than of silicotic nodules that determines the level of pulmonary dysfunction in these subjects. Akira and colleagues (80) reviewed the high-resolution CT (HRCT) scans of 90 patients with pneumoconiosis who had small, rounded opacities on the radiographs. The 90 patients were divided into three groups based on the type of opacity seen on chest radiogaphs. The first group consisted of 55 patients whose radiographs showed predominantly type p-rounded opacities, including 32 patients with silicosis, six patients with coal workers' pneumoconiosis, and all patients with takosis, welder's lung, and graphite pneumoconiosis. The second group consisted of 29 patients whose radiographs showed predominantly type q-rounded opacities, including 23 patients with silicosis and six patients with coal workers' pneumoconiosis. The third group consisted of six patients with silicosis whose radiographs showed predominantly type r-rounded opacities. Radiographic type p pneumoconiosis was characterized on HRCT by tiny, branching structures or a gathering of a few dots. In 21 of the 55 patients, nonperipheral, small areas of low attenuation with a central dot were found. CTpathologic correlation in two postmortem specimens showed these tiny opacities and areas of low attenuation to correspond to irregular fibrosis around and along the respiratory bronchioles and to focal-duct emphysema, respectively. Focal emphysema was most commonly found in pneumoconiosis with type p

changes and could be identified with HRCT scanning. Opacities of the q and r type were characterized by sharply demarcated, rounded nodules or irregular, contracted nodules. CT appearances differed among the three types of opacities, but no differences were noted between the CT appearances of silicosis and the other pneumoconioses with the same opacity type.

Sarcoidosis Sarcoidosis is a systemic disorder of unknown cause characterized by noncaseating granulomata, which may resolve spontaneously or progress to fibrosis (54). It may involve almost any organ, but most of the morbidity and mortality is due to pulmonary disease (30). Pulmonary manifestations are present in 90070 of patients, 20 to 25% of whom have permanent functional impairment (30). Approximately 60 to 70% of patients with sarcoidosis have a characteristic radiologic appearance consisting of enlarged hilar and paratracheal lymph nodes with or without concomitant parenchymal changes (7, 81, 82). In 25 to 30% of cases, however, the radiologic findings are nonspecific or atypical, and in 5 to 10% of patients, the radiograph is normal (7, 30, 54, 81-83). The variable and often nonspecific radiographic findings are surprising given the characteristic pathologic appearance and distribution of sarcoidosis. Sarcoid granulomata, the hallmark of the disease, are distributed mainly along the lymphatics in the bronchovascular sheath and,

to a lesser extent, in the interlobular septa and pleura (54, 84, 85). This distribution is one of the most helpful features in recognizing sarcoidosis pathologically and is responsible for the high rate of success in diagnosis by bronchial and transbronchial biopsies (54). It can be clearly seen in illustrations of the macroscopic, pathologic appearance (54, 86) (figure 10). This distribution is difficult to appreciate on the radiograph because of the superimposition of the parenchymal shadows, but it is clearly seen on CT (figure 11). The characteristic parenchymal abnormalities of sarcoidosis on CT are nodular opacities along the bronchovascular bundles, interlobular septa, major fissures, and subpleural regions (18, 22, 23, 87). In the majority of cases, the nodules have irregular margins. HRCT is superior to conventional CT in the assessment of subpleural nodules and irregular linear densities, but conventional CT is superior in the assessment of peribronchovascular nodules (18). Confluence of nodules may result in large opacities with ill-defined contours. These opacities may have air bronchograms and may occasionally cavitate (87). Depending on the stage of disease, 16 to 60% of patients with sarcoidosis have patchy areas of hazy increase in density or ground-glass opacities on HRCT. Preliminary data suggest that these may represent areas of active alveolitis as assessed by 67Ga scanning (22). As fibrosis develops, irregular linear opacities become a prominent feature. These, like the nodules, are situated mainly along the bronchovascular bundles (18, 88, 89). CT is superior to the radiograph in demonstrating early fibrosis and distortion.of the lung parenchyma (87). The most common early sign of distortion on CT was posterior displacement of the main and upper lobe bronchi, indicating loss of volume of the posterior segments of the upper lobes (87). CT is also helpful in assessing the presence and extent of complications of sarcoidosis (83). Whereas true cavitary sarcoidosis is rare, pseudocavities representing bullae or bronchiectasis are common in patients with extensive fibrosis. Superimposed bacterial infection and saprophytic fungal infection with mycetoma formation can be readily detected with CT (83). CT may show parenchymal abnormalities in patients with a normal chest radiograph and in patients with only hilar adenopathy apparent on the radiograph;

1442

MULLER AND MILLER

Fig. 10. Pathologic specimen cut in the transverse plane in a patient with sarcoidosis shows granulomata along the bronchovascular bundles (white arrows). interlobular septa (black arrows). and subpleural lung region (open arrows).

however, CT may fail to show any evidence of sarcoidosis in patients with pulmonary involvement proved by transbronchial biopsy or lobectomy (18, 90). Although CT gives a superior pictorial assessment of disease pattern and distribution, it is controversial whether it correlates better than the radiograph with the clinical and functional impairment in sarcoidosis. In a review of 27 patients with sarcoidosis, Muller and coworkers (91) demonstrated that CT and radiographic assessment of disease extent had similar correlations with the severity of dyspnea (r = 0.61 and 0.58, respectively;

p < 0.001), with TLC (r = -0.54 and - 0.62, respectively; p < 0.01), and in gas transfer as assessed by the carbon monoxide diffusing capacity (r = - 0.62 and -0.52, respectively; p < 0.01). In a prospective study of 44 patients using only HRCT, Brauner and associates (87) found that the CT visual score had a lower correlation than did the radiographic score with TLC (r = -0.30 and -0.49, respectively), FEV! (r = -0.41 and -0.40, respectively) and carbon monoxide diffusing capacity (r = -0.41 and - 0.46, respectively). Bergin and colleagues (89), on the other hand, found

that the CT scores correlated better with functional impairment (all r >0.49) than did the radiographic scores (all r < 0.15). The discrepancy between the different results may be due to a difference in patient groups as well as a difference in data analysis. Carrington and coworkers (84) suggested that the poor correlation between the radiographic severity of disease and the functional impairment in patients with sarcoidosis may be due to the fact that the nodular lesions, although easily seen and quantitated, cause minimal dysfunction. This situation is similar to that in silicosis in which the severity of interstitial fibrosis rather than the number or size of nodules is responsible for the impaired function (84). Indeed, in the study by Muller and associates (91), patients with predominantly irregular opacities had more severe dyspnea and lower lung volumes than did patients with predominantly nodular opacities (p < 0.05). The distribution of sarcoid granulomata along the lymphatics is similar to that seen with pulmonary lymphangitic carcinomatosis. Both conditions may cause a beaded appearance of the bronchovascular bundles and of the interlobular septa (19,22,41). However, in sarcoidosis, the septal thickening is usually less extensive than that seen in pulmonary lymphangitic carcinomatosis, and often associated with distortion of the lobular architecture (18, 20). The nodules in sarcoidosis often have irregular margins; those in lymphangitic carcinomatosis are smooth. Irregular linear opacities and evidence of fibrosis are also more suggestive of sarcoidosis. In some patients, however, the pattern of parenchymal involvement of sarcoidosis may be identical to that of lymphatic spread of tumor (18, 25).

Histiocytosis X Pulmonary histiocytosis X is an uncommon disorder of unknown cause characterized in its early stages by nodular collections of Langerhans histiocytes and eosinophils, and in its later stages by replacement of cellular elements with fibrosis. When confined to the lungs, it is also known as eosinophilic granuloma of the lung. Gaensler and Carrington (92) found it in only 3.4070 of 502 patients who underwent open lung biopsy for chronic diffuse infiltrative lung disease. Most patients with pulmonary histiocytosis X are young or middle-aged adults (93); the average patient age is 32 yr (94). Most, but

1443

STATE OF THE ART: CT OF CILD

~

~

~

A

•

~

~

Fig. 11. A. Conventional CT of a 57-yr-old woman with sarcoidosis scan at level of right upper lobe bronchus shows small irregular nodules predominantly along the branchovascular bundles (arrows). CT scan parameters: 10-mm collimation, 40-cm field-of-view, and standard reconstruction algorithm. B. High-resolution CT through the right lung at the same level also demonstrates thickening of the interlobular septa (arrows). CT scan parameters: 1.5-mm collimation, 25-cm field-of-view, and high-resolution reconstruction algorithm.

not all, series report a male predominance (94-98). Cigarette smoking is reported in over 90070 of patients (96, 99, 100). Pathologically, early lesions are cellular nodules with some tendency to be juxtabronchiolar (54). In the early stages, the intervening lung tissue may be normal and no zonal predominance apparent. With disease progression and chronicity, acellular, inactive, stellate scars and finally severe honeycomb fibrosis ensues. In severe, endstage disease, upper zone honeycombing is very striking, with obvious sparing of the bases and lingular/middle lobe tips (figure 12). There is no particular subpleural accentuation or sparing in this disease. Nodules, if grossly discernible at all in the late stages, are likely to be discrete stellate scars in the bases. This pathologic variability over the course of the disease is reflected in the variability of radiologic and CT findings. The radiographic abnormalities consist of reticular, nodular, reticulonodular, and cystic abnormalities, often in combination (93, 95, 96, 101). The dis-

ease is usually bilateral and diffuse, with relative sparing of the costophrenic angles (95, 96). Predominance of the disease in the upper lung has been reported (96) but not in all series (95, 97). Lung volumes are characteristically normal or increased. Moore and colleagues (65) evaluated retrospectively radiographs, CT scans, and results of pulmonary function tests for 17 patients with biopsy-proved pulmonary histiocytosis X. In 11 patients, HRCT was used. CT was superior to the chest radiograph in showing the morphology and distribution of lung abnormalities. In 12 patients, CT demonstrated cystic airspaces, usually less than 10 mm in diameter (figure 13). In three of these 12, cysts were the only abnormality, but in six others, nodules (usually less than 5 mm in diameter) were also present. CT showed that many lesions that appeared reticular on plain radiographs were actually cysts. CT showed no central or peripheral predominance of lesions, but it did reveal that many small nodules were distributed in the centers

of secondary lobules around small airways. Extent of disease on CT correlated better with the diffusing capacity (r = 0.71) than did the plain radiographic findings (r = - 0.57). In the 14 patients with nodules, the abundance of zero to 2 mm and 2 to 5 mm nodules was similar and much greater than that of larger nodules. The number of nodules varied from single to myriad. The nodules were usually solid but sometimes had lucent centers, presumably corresponding to small "cavities" within the inflltrates, well-described pathologically (54). Their margins were usually indistinct and often irregular, particularly when there was surrounding cystic or reticular disease. In many patients with cysts or nodules, the intervening lung parenchyma was normal on CT.

Extrinsic Allergic Alveolitis Extrinsic allergic alveolitis (EAA) is an allergic disease of the lungs caused by inhalation of antigens contained in a variety of organic dusts (102). The radiologic appearance varies with the different

1444

MOLLER AND MILLER

Fig. 12. Autopsy lung specimen in patient with pulmonary histiocytosis X shows fine honeycombing, involving the upper and middle lung zones, and relatively sparing of the lower lung zone, including lingular tip.

stages of the disease. Acutely, heavy exposure to the inciting antigen causes diffuse airspace consolidation (103). The consolidation resolves within a few days

to reveal a fine nodular or reticulon()dular pattern, which is characteristic of the subacute phase (103, 104). The changes in the subacute phase may be complete-

ly reversible (104, -105). The chronic stage of hypersensitivity pneumonitis is characterized by the presence of fibrosis, which may occur months to years after the initial exposure (105). Repeated exposure to the antigen may lead to acute and subacute changes superimposed on chronic fibrosis. Recent studies have described the CT appearance of EAA in a small number of patients (25, 106, 107). The largest series is by Silver and colleagues (107), who described the CT features in 11 patients with EAA. 1\vo patients had clinically acute disease; both had bilateral airspace consolidation and small, rounded opacities on the radiograph and on CT. In these two patients, the CT findings were identicai to the radiograph, and CT added no further information. Comparison of the CT findings with .the histology suggested that the opacities represented active granulomata in conjunction with confluent collections of intraalveolar histiocytes. The consolidation may also be caused by pulmonary edema (107). Three patients had clinically subacute EAA at the time of their CT scan. Their symptoms had been present for 3 to 7 months before assessment. These three patients had small, rounded opacities seen on their chest radiographs and CT scans. The CT scans also showed patchy bilateral areas of airspace opacification

.... Fig. 13 (left). High-resolution CT through right lung apex in patient with endstage histiocytosis X shows diffuse fine honeycombing. CT scan parameters: 1.5-mm collimation,23-cm field-of-view, and high-resolution reconstruction algorithm. Fig. 14 (right). Thin-section CT through lung bases in patient with extrinsic allergic alveolitis shows areas of airspace opacification (white arrows) that do not obscure the underlying lung markings and ill-defined small nodular densities (black arrows). CT scan parameters: 1.5-mm collimation, 4Q-cm field-of-view, and standard reconstruction algorithm.

1445

STATE OF THE ART: CT OF CILD

Fig. 15. CT scan of a 44-yr-old man with alveolar proteinosis shows extensive bilateral areas of airspace opacification. This pattern is nonspecific. Also noted is thickening of the interlobular septa, leading to the appearance of polygonal lines (arrows). CT scan parameters: 1.5-mm collimation, 40-cm field-of-view, and standard reconstruction algorithm.

that were not apparent on the radiographs. This airspace opacification did not obscure the underlying vascular markings and was seen in the same distribution as were the small, rounded opacities (figure 14). The remaining six patients had subacute symptoms superimposed on chronic EAA. In these patients, symptoms had been present for 1 to 6 yr. The chest radiographs and CT scans from this group showed irregular linear opacities, presumably representing fibrosis. The CT scans also showed patchy bilateral areas of airspace opacification and scattered, small, rounded opacities that were apparent on the chest radiograph in only one of these six patients. Pathologically, these changes were due to subacute changes characteristic of EAA, namely, mononuclear cell bronchiolitis, mononuclear cell interstitial infiltrates, and scattered, non-necrotizing granulomata (108, 109).

In all cases, HRCT was superior to conventional CT in delineating fine parenchymal detail and in distinguishing normal from abnormal parenchyma. In particular, the HRCT scans better delineated the areas of patchy airspace opacification from adjacent normal lung. The presence of small, rounded opacities « 5 mm diameter) amid patchy areas of airspace opacification appears to be characteristic of the subacute stage of EAA. Similar findings were found by P.

Kullnig (personal communication) in six patients. Ground-Glass Pattern

infiltration of the lungs with eosinophils. It is usually associated with an increased number of eosinophils in the circulating blood, although this may be transient. Radiologically, chronic eosinophilic pneumonia is characterized by the presence of homogeneous peripheral airspace consolidation. The pattern remains unchanged for days or weeks unless steroid therapy is given. The combination of blood eosinophilia, peripheral infiltrates on the radiograph, and rapid response to steroid therapy are often sufficiently characteristic to obviate the need for lung biopsy (111, 112). The diagnosis may be difficult in patients with minimal blood eosinophilia or in whom the peripheral distribution of infiltrates is not apparent on the plain film. A recent study indicates that the classic radiologic picture may be seen in fewer than 50070 of cases (113). Mayo and coworkers (114) reviewed the chest radiographs and CT scans in six patients \lTith chronic eosinophilic pneumonia. Each patient had patchy, airspace consolidation, and in five of six cases, the consolidation was most marked in the middle and upper lung zones. In only one patient was the classic pattern of airspace consolidation confined to the outer third of the lungs, readily apparent on the radiograph. In three patients, the consolidation appeared to be diffuse, although a slight peripheral predominance was present. In two patients, a peripheral predominance was diffiwlt to appreciate even in retrospect. 1ft aU :six cases, a characteristic peripheral cm~ consolidation was clearly seen on CT. This study suggests that CT may be helpful in the diagnosis of chronic eosinophilic pneumonia when the clinical findings are suggestive but the radiographic pattern is nonspecific.

Alveolar Proteinosis Godwin and associates (110) reported CT findings in nine patients with pulmonary alveolar proteinosis. CT showed the airspace disease to have a variable appearance ranging from ill-defined nodular densities to large areas of confluent airspace consolidation. The airspace consolidation in some patients was sharply demarcated from surrounding normal parenchyma, without any apparent anatomic reason accounting for this sharp edge. CT, and particularly HRCT, showed smooth thickening of the interlobular septa, which was not as apparent on the radiograph (figure 15). The pathologic explanation for this finding is not entirely clear. The CT findings in alveolar proteinosis are nonspecific; however, CT may demonstrate focal pneumonia that is not apparent on the radiograph (110). Therefore, CT is probably indicated in patients with alveolar proteinosis, particularly immunocompromised patients in whom secondary infection is especially common, when the radiographic findings are not helpful.

Bronchiolitis Obliterans Bronchiolitis obliterans is a lung disease characterized by the presence of gran'lllation tissue polyps within the lumen of bronchioles, alveolar ducts, and alveoli. It may lead to extensive obliteration of the small airways by scarring. The inflammatory process may involve the surrounding alveoli, resulting in a component of organizing pneumonia. Bronchiolitis obliterans is a nonspecific reaction that may be caused by a variety of insults. It may be classified into:

Chronic Eosinophilic Pneumonia Chronic eosinophilic pneumonia is an often idiopathic condition characterized by

(1) Toxic-fume bronchiolitis obliterans, resulting from exposure to gases such as nitrogen dioxide (silo-filler's), sulfur di-

1446

oxide, ammonia, chlorine, phosgene, and ozone (115-119); (2) postinfectious bronchiolitis obliterans, due to bacterial, mycoplasma, or viral infection (116, 120, 121); (3) bronchiolitis obliterans associated with connective tissue diseases, particularly rheumatoid arthritis and polymyositis (122-125); (4) Bronchiolitis obliterans as a complication of bone-marrow and heart-lung transplantation (126-129); (5) localized lesion with bronchiolitis obliterans, a focal organizing pneumonia presenting as a lobar pneumonia or an irregular nodule; (6) idiopathic bronchiolitis obliterans with organizing pneumonia. The radiologic manifestation of the various forms of bronchiolitis obliterans was described by Gosink and associates (116) and recently reviewed by McLoud and coworkers (130). The most important form in the setting of chronic infiltrative lung disease (CILD) is idiopathic bronchiolitis obliterans with organizing pneumonia (BOOP). Clinically, patients with BOOP usually present with 1 to 6 months history of

MULLER AND MILLER

nonproductive cough and malaise (131, 132). Approximately 50070 of the patients have low-grade fever and 50% have mild dyspnea (131, 132). The most common findings on physical examination are rales. Clubbing and cyanosis are unusual. Pulmonary function tests characteristically show a restrictive pattern with decreased lung volumes and impaired gas exchange. Clinically and functionally, the findings may be similar to usual interstitial pneumonia, although the duration of symptoms is usually shorter. Systemic symptoms are more common and finger clubbing is rarely seen. The most common radiologic findings of BOOP are patchy, nonsegmental, unilateral or asymmetric, bilateral areas of airspace consolidation (131-133). Small nodular and irregular linear opacities may be seen in up to 50% of patients. Although there have been several recent reports on the manifestations of BOOP, little has been written about the CT findings. Muller and colleagues (132) described the CT appearance in two patients with BOOP. In one patient, the distribution was patchy, and in the other the consoli9ation was almost entirely sub-

r'"

pleural, similar to that seen in chronic eosinophilic pneumonia. Miki and associates (134) reported a case of BOOP studied using HRCT. CT showed bilateral, segmental airspace consolidation with air bronchograms. HRCT was used to select a site for open lung biopsy. Followup CT after steroid therapy showed almost complete resolution. Mayo and colleagues (114) reported a patient with both BOOP and chronic eosinophilic pneumonia on open lung biopsy. The patient had mild peripheral eosinophilia. The CT showed asymmetric, predominantly lower lobe airspace consolidation in a predominantly subpleural distribution. We recently reviewed the chest radiograph, CT scans, and biopsy specimens in 14 patients with BOOP (135). Six patients were immunocompromised due to leukemia or bone-marrow transplantation. The main CT findings consisted of patchy unilateral or bilateral airspace consolidation and small nodular opacities. A predominantly subpleural distribution of the airspace consolidation was apparent on CT in six patients and on the radiograph in two (figure 16). In three patients, the consolidation was almost

"

•

Fig. 16. A. Chest radiograph of a 64-yr-old woman with bronchiolitis obliterans organizing pneumonia shows extensive airspace consolidation in the lung bases. A SUbpleural predominance is not readily apparent. B. CT scan demonstrates peripheral distribution of the airspace consolidation. CT scan parameters: 10-mm collimation, 40-em field-ofview, and standard reconstruction algorithm.

1447

STATE OF THE ART: CT OF CILD

entirely limited to the subpleural and peribronchovascular regions of the lungs. Four patients had diffuse consolidation with no subpleural predominance. There was no apparent difference between the CT or pathologic findings of BOOP in immunocompromised and nonimmunocompromised patients. The CT findings of bronchiolitis obliterans, like the chest radiograph, are nonspecific. We were unable to determine any single finding or combination of findings on CT that could be considered diagnostic of BOOP and that could not be appreciated on a radiograph. The main use of CT in this series was as a guide to the optimal biopsy site. Although CT gives a better assessment of disease pattern, most of the findings in BOOP are readily apparent on the chest radiograph. Conclusions It has been shown that CT is helpful in

the assessment of patients with CILD. In some patients, the characteristic CT appearance may eliminate the need for lung biopsy. More commonly, CT is helpful in determining the optimal biopsy site. Better visualization of the lung parenchyma, particularly using HRCT, has increased our understanding of the pattern and distribution of CILD. Preliminary results suggest that HRCT may be useful in assessing disease activity. Although encouraging, most of the CT studies are based on retrospective analysis of relatively small numbers of patients. Clearly, prospective multicenter studies are required to determine the role of CT in the initial assessment and follow-up of patients with CILD. Currently, indications for CT of the chest at our institution include assessment of patients with history or clinical findings suggestive of diffuse infiltrative lung disease but normal or nonspecific findings on the radiograph; further evaluation of the lung parenchyma in patients in whom the radiographic findings are not in keeping with the clinical history or symptomatology; as a guide to the optimal biopsy site in all patients undergoing open lung biopsy; follow-up of patients who show clinical or functional evidence of deterioration or who fail to respond to treatment. In these patients, CT may be helpful in the assessment of disease activity and progression of interstitial fibrosis and in the detection of suspected complications, e.g., infection. References 67. Seaton A. Silicosis. In: Morgan WKC, Sea-

ton A. Occupational lung diseases. 2nd ed. Philadelphia: W.B. Saunders, 1984; 250-94. 68. Lapp NL. Lung disease secondary to inhalation of nonfibrous minerals. Clin Chest Med 1981; 2:219-33. 69. Fraser RG, Pare JAP. The pneumoconioses and chemically induced lung diseases. In: Fraser RG, Pare JAP, ed. Diagnoses of diseases of the chest. Philadelphia: W.B. Saunders, 1979; 1475-570. 70. Begin R, Bergeron D, Samson L, Boctor M, Cantin A. CT assessment of silicosis in exposed workers. AJR 1987; 148:509-14. 71. Bergin CJ, Muller NL, Vedal S, Chan-Yeung M. CT in silicosis: correlation with plain films and pulmonary function tests. AJR 1986; 146:477-83. 72. Fraser RG, Pare JAP. Silicosis. In: Fraser RG, Pare JAP, ed. Diagnosis of diseases of the chest. Vol. 3. Philadelphia: W.B. Saunders, 1979; 1484-502. 73. Parkes WR. Diseases due to free silica. In: Parkes WR, ed. Occupational lung disorders. 2nd ed. London: Butterworth, 1982; 134-58. 74. Foster WL Jr, Pratt PC, Roggli VL, Godwin JD, Halvorsen RA Jr, Putman CEo Centrilobular emphysema: CT-pathologic correlation. Radiology 1986; 159:27-32. 75. Hruban RH, Meziane MA, Zerhouni EA, et al. High resolution computed tomography of inflation-fixed lungs: pathologic-radiologic correlation of centrilobular emphysema. Am Rev Respir Dis 1987; 136:935-40. 76. Kinsella M, Muller NL, Staples C, Vedal S, Chan-Yeung M. Hyperinflation in asthma and emphysema: assessment by pulmonary function and computed tomography. Chest 1988; 94:286-9. 77. Muller NL, Miller RR, Abboud RT. "Density mask": an objective method to quantitate emphysema using computed tomography. Chest 1988; 94:782-7. 78. Miller RR, Muller NL, Vedal S, Morrison Nl, Staples CA. Limitations of computed tomography in the assessment of emphysema. Am Rev Respir Dis 1989; 139:980-3. 79. Kinsella M, Muller NL, Vedal S, Staples C, Abboud RT, Chan-Yeung M. Emphysema in silicosis: a comparison of smokers with nonsmokers using pulmonary function testing and computed tomography. Am Rev Respir Dis 1990; 141:1497-500. 80. Akira M, Higashihara T, Yokoyama K, et al. Radiographic type p pneumoconiosis: high-resolution CT. Radiology 1987; 17l:ll7-23. 81. Scadding JG, Mitchell DN, eds. Sarcoidosis. London: Chapman and Hall Medical, 1985; 101-80. 82. Hillerdal G, Nou E, Osterman K, Schmekel B. Sarcoidosis: epidemiology and prognosis, a 15year European study. Am Rev Respir Dis 1984; 130:29-32. 83. Hamper UM, Fishman EK, Khouri NF, Johns Cl, Wang KP, Siegelman SS. Typical and atypical CT manifestations of pulmonary sarcoidosis. J Comput Assist Tomogr 1986; 10:928-36. 84. Carrington CB, Gaensler EA, Mikus JP, Schachter AW, Burke GW, Goff AM. Structure and function in sarcoidosis. Ann NY Acad Sci 1976; 278:265-83. 85. Thomas PD, Hunninghake GW. Current concepts of the pathogenesis of sarcoidosis. Am Rev Respir Dis 1987; 135:747-60. 86. Heitzman ER. Sarcoidosis. In: Heitzman ER, ed. The lung: radiologic-pathologic correlations. St. Louis: CY. Mosby, 1984; 294-310. 87. Brauner MW, Grenier P, Mompoint D, Lenoir S, de Cremoux H. Pulmonary sarcoidosis: evaluation with high-resolution CT. Radiology 1989; 172:467-71. 88. Bergin CJ, Muller NL. CT of interstitial lung disease: a diagnostic approach. AJR 1987; 148:8-15. 89. Bergin Cl, Bell DY, Coblentz CL, et al. Sarcoidosis: correlation of pulmonary parenchymal

pattern at CT with results of pulmonary function tests. Radiology 1989; 171:619-24. 90. Nakata H, Kimoto T, Nakayama T, Kido M, Miyazaki N, Harada S. Diffuse peripheral lung disease: evaluation by high-resolution computed tomography. Radiology 1985; 157:181-5. 91. Miiller NL, Mawson 18, Mathieson JR, Abboud R, Ostrow DN, Champion P. Sarcoidosis: correlation of extent of disease at CT with clinical, functional, and radiographic findings. Radiology 1989; 17l:613-8. 92. Gaensler EA, Carrington CG. Open biopsy for chronic diffuse infiltrative lung disease: clinical, roentgenographic, and physiologic correlations in 502 patients. Ann Thorac Surg 1980; 30:4ll-26. 93. Marcy TW, Reynolds HY. Pulmonary histiocytosis X. Lung 1985; 163:129-50. 94. Colby TV, Lombard C. Histiocytosis X in the lung. Hum Pathol 1983; 14:847-56. 95. Lacronique J, Roth C, Battesti JP, Basset F, Chretien J. Chest radiological features of pulmonary histiocytosis X: a report based on 50 adult cases. Thorax 1982; 37:104-9. 96. Friedman PJ, Liebow AA, Sokoloff J. Eosinophilic granuloma of lung: clinical aspects of primary pulmonary histiocytosis in the adult. Medicine 1981; 60:385-96. 97. Basset F, Corrin B, Spencer H, et al. Pulmonary histiocytosis X. Am Rev Respir Dis 1978; ll8:8ll-20. 98. Lewis JG. Eosinophilic granuloma and its variants with special reference to lung involvement: a report of 12 patients. Q J Med 1964; 33:337-59. 99. Hance AJ, Basset F, Saumon G, et al. Smoking and interstitial lung disease: the effect of cigarette smoking on the incidence of pulmonary histiocytosis X and sarcoidosis. Ann NY Acad Sci 1986; 465:643-56. 100. Winterbauer RH, Dreis DF, Jolly PC. Clinical correlation. In: Dail DH, Hammar SP, eds. Pulmonary pathology. New York: Springer-Verlag, 1988; ll48-51. 101. Prophet D. Primary pulmonary histiocytosisX. Clin Chest Med 1982; 3:643-53. 102. Chryssanthopoulos C, Fink IN. Hypersensitivity pneumonitis. J Asthma 1983; 20:285-96. 103. Fraser RG, Pare JAP. Diagnosis of diseases of the chest. 2nd ed. Philadelphia: W.B. Saunders, 1979; 957-74. 104. Cook PG, Wells IP, McGavin CR. The distribution of pulmonary shadowing in farmer's lung. Clin Radiol 1988; 39:21-7. 105. Hapke EJ, Seal RME, Thomas GO, Hayes M, Meek JC. Farmer's lung. A clinical, radiographic, functional and serologic correlation of acute and chronic stages. Thorax 1968; 23:451-68. 106. Bergin CJ, Muller NL. CT of interstitial lung disease: a diagnostic approach. AJR 1987; 148:8-15. 107. Silver SF, Muller NL, Miller RR, Lefcoe MS. Computed tomography in hypersensitivity pneumonitis. Radiology 1989; 173:441-5. 108. Coleman A, Colby TV. Histologic diagnosis of extrinsic allergic alveolitis. Am J Surg Pathol 1988; 12:514-8. 109. Reyes CN, Wenzel F1, Lawton BR, Emanuel DA. The pulmonary pathology of farmer's lung disease. Chest 1982; 81:142-6. llO. Godwin lD, Muller NL, Takasugi JE. Pulmonary alveolar proteinosis: CT findings. Radiology 1988; 169:609-13. lli. Pearson Dl, Rosenow EC. Chronic eosinophilic pneumonia (Carrington's). A follow-up study. Mayo Clin Proc 1978; 53:73-8. ll2. Dines DE. Chronic eosinophilic pneumonia: a roentgenographic diagnosis. Mayo Clin Proc 1978; 53:129-30. 113. Jederlinic PJ, Sicilian L, Gaensler EA. Chronic eosinophilic pneumonia. A report of 19 cases and a review of the literature. Medicine 1988;

MOLLER AND MILLER

1448 67:154-62. 114. Mayo JR, Muller NL, Road J, Sisler J, Lillington G. Chronic eosinophilic pneumonia: CT findings in six cases. AJR 1989; 153:727-30. 115. Epler GR, Colby TV. The spectrum of bronchiolitis obliterans. Chest 1983; 83:161-2. 116. Gosink BB, Friedman PJ, Liebow AA. Bronchiolitis obliterans: roentgenographic-pathologic correlation. Am J Roentgenol Radium Ther Nucl Med 1973; 117:816-32. 117. Lowry T, Schuman LM. "Silo-filler's disease": a syndrome caused by nitrogen dioxide. JAMA 1956; 162:153-8. 118. Cornelius EA, Betlach EH. Silo-filler's disease. Radiology 1960; 74:232-5. 119. Charan NB, Myers CG, Lakshminarayan S, Spencer TM. Pulmonary injuries associated with acute sulfur dioxide inhalation. Am Rev Respir Dis 1979; 119:555-60. 120. Laraya-Cuasay LR, DeForest A, Huff D, Lischner H, Huang NN. Chronic pulmonary complications of early influenza virus infection in children. Am Rev Respir Dis 1977; 116:617-25. 121. Nikki P, Meretoja 0, Valtonen V, et a/. Severe bronchiolitis probably caused by varicellazoster virus. Cdt Care Med 1982; 10:344-6. 122. Geddes DM, Corrin B, Brewerton DA, Da-

vies RJ, Turner-Warwick M. Progressive airway obliteration in adults and its association with rheumatoid disease. Q J Med 1977; 46:427-44. 123. Epler GR, Snider GL, Gaensler EA, Cathcart ES, Fitzgerald MK, Carrington CB. Bronchiolitis and bronchitis in connective tissue disease. JAMA 1979; 242:528-32. 124. Schwarz MI, Matthay RA, Sahn SA, et a/. Interstitial lung disease in polymyositis and dermatomyositis: an analysis of six cases and review of the literature. Medicine 1976; 55:89-104. 125. Herzog CA, Miller RR, Hoidal JR. Bronchiolitis and rheumatoid arthritis. Am Rev Respir Dis 1979; 119:555-60. 126. Ostrow D, Buskard N, Hill RS, Vickars L, Churg A. Bronchiolitis obliterans complicating bone marrow transplantation. Chest 1985; 87: 828-30. 127. RocaJ, GranenaA, Rodriguez-Roisin J, Alvarez P, Agusti-Vidal A, Rozman C. Fatal airway disease in an adult with chronic graft-versus-host disease. Thorax 1982; 37:77-8. 128. Stein-Streilen J, Lipscomb MF, Hart DA, Darden A. Graft-versus-host reaction in the lung. 1tansplantation 1981; 32:38-44. 129. Burke CM, Theodore J, Dawkins KD, et a/. Post transplant obliterative bronchiolitis and oth-

er late lung sequelae in human heart-lung transplantation. Chest 1984; 86:824-9. 130. McLoud TC, Epler GR, Colby TV, Gaensler EA, Carrington CB. Bronchiolitis obliterans. Radiology 1986; 159:1-8. 131. Epler GR, Colby TV, McLoud TC, et a/. Bronchiolitis obliterans organizing pneumonia. N Engl J Med 1985; 312:152-8. 132. Muller NL, Guerry-Force ML, Staples CA, et a/. Differential diagnosis of bronchiolitis obliterans with organizing pneumonia and usual interstitial pneumonia: clinical, functional, and radiologic findings. Radiology 1987; 162:151-6. 133. Chandler PW, Shin MS, Friedman SE, Myers JL, Katzenstein AL. Radiographic manifestations of bronchiolitis obliterans with organizing pneumonia versus usual interstitial pneumonia. AJR 1986; 147:899-906. 134. Miki Y, Hatabu H, Takahashi M, Sadatoh N, Kuroda Y. Computed tomographyofbronchiolitis obliterans. J Comput Assist Tomogr 1988; 12:512-4. 135. Muller NL, Staples CA, Miller RR. Bronchiolitis obliterans organizing pneumonia: CT features in 14 patients. AJR 1990; 154:983-7.