| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Thoracic Imaging |
1 From the Dept of Radiology, Osaka Univ Graduate School of Medicine, 2-2 Yamadaoka, Suita, Osaka 565-0825, Japan (T.J., M. Koyama, H.N.); Dept of Radiology, Univ of British Columbia and Vancouver Hosp and Health Sciences Centre, Canada (N.L.M.); Dept of Pathology and Laboratory Medicine, Mayo Clinic Scottsdale, Ariz (T.V.C.); First Dept of Internal Medicine, Kumamoto Univ School of Medicine, Japan (K.I., M.S., M.A.); Dept of Respiratory Medicine, Tosei General Hosp, Aichi, Japan (H.T., Y.K.); Depts of Radiology (K.F.) and Internal Medicine (M. Kinoshita), Kurume Univ School of Medicine, Fukuoka, Japan; Dept of Radiology, Dokkyo Univ School of Medicine, Tochigi, Japan (H.A.); and Dept of Internal Medicine, St Marianna Univ School of Medicine, Kanagawa, Japan (H.Y.). Received Sept 19, 2001; revision requested Nov 23; final revision received Mar 1, 2002; accepted Mar 28. Address correspondence to T.J. (e-mail: johkoh@radiol.med.osaka-u.ac.jp).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: The study included 55 cases of pathologically proven NIP. The 55 cases were categorized histologically into four grades: grade 1, interstitial inflammation without fibrosis (n = 6); grade 2, interstitial inflammation predominating over fibrosis (n = 16); grade 3, fibrosis predominating over inflammation (n = 5); and grade 4, fibrosis only (n = 28). Two independent observers evaluated the presence, extent, and distribution of various CT findings. Thin-section CT findings and histologic grades were compared by using the Spearman rank correlation coefficient. Observer agreement was assessed.
RESULTS: Areas with ground-glass attenuation and architectural distortion were present in all 55 patients. Traction bronchiectasis and intralobular reticular opacities were seen in 52 and 48 patients, respectively. The extent of traction bronchiectasis (r = 0.68; P < .001) and intralobular reticular opacities (r = 0.35; P < .05) correlated with the histologic grade. Honeycombing was seen in 12 (43%) of 28 patients with grade 4 NIP and in three (11%) of the remaining 27 patients (
2 test, P < .001). There was good agreement between the observers for the presence (
= 0.7–1.0) and extent (Spearman rank correlation; r = 0.87–0.98; P < .001) of various abnormalities.
CONCLUSION: The extent of traction bronchiectasis and intralobular reticulation at thin-section CT correlates with increased fibrosis in NIP. Honeycombing is seen almost exclusively in patients with fibrotic NIP.
© RSNA, 2002
Index terms: Computed tomography (CT), thin-section, 68.12118 • Pneumonia, nonspecific interstitial and fibrosis, 68.213
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Results of initial studies, based on relatively small numbers of cases, suggest that the computed tomographic (CT) findings of NIP are relatively homogeneous, consisting mainly of ground-glass attenuation and reticulation involving the peripheral regions of the lower lung zones (5,6). In contrast, results of more recent studies demonstrate a more heterogeneous pattern of CT findings of NIP, including consolidation, centrilobular nodules, and honeycombing (7,8). Kim et al (9) compared CT findings with histologic findings in 23 patients. The results of their study showed a potential role for CT in assessing the presence of fibrosis. However, the small number of patients did not allow any definite conclusions to be reached about the value of thin-section CT in distinguishing between the various subtypes of NIP. The aim of this study was to characterize the thin-section CT findings of the pathologic subgroups of NIP in a sizable number of patients.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Images and Review
All patients underwent thin-section CT of the chest. The CT scans consisted of 1.0- to 1.5-mm collimation sections, reconstructed by using a high-spatial-frequency algorithm. The protocols consisted of thin sections obtained at 1.0-cm (45 patients) or 1.5-cm (10 patients) intervals. The CT scans were performed with a variety of scanners. None of the patients received intravenous contrast medium.
The CT images were reviewed independently by two observers (K.I., M.K.). These observers were different from the five authors who reviewed clinical histories, and they did not know any clinical information, other than the fact that the patients had NIP. The lungs were divided into three zones (upper, middle, and lower); each zone was evaluated separately. Each of the three zones corresponded to approximately one-third of the images from the lung apex to 1 cm below the domes of the diaphragm.
The observers assessed the presence, extent, and distribution of areas with ground-glass attenuation, areas of air-space consolidation, parenchymal nodules, intralobular reticular opacities, traction bronchiectasis, interlobular septal thickening, and thickening of bronchovascular bundles. The presence, extent, and distribution of associated findings, such as honeycombing, emphysema, architectural distortion, lymph node enlargement, and pleural effusion, were also assessed. Ground-glass attenuation was defined as an area of hazy increased attenuation without obscuration of underlying vascular markings. Air-space consolidation was considered present when the opacity obscured the underlying vessels. Traction bronchiectasis was defined as irregular bronchial dilatation within areas with parenchymal abnormality. Lymphadenopathy was considered present when the short-axis diameter of the nodes was greater than 1 cm.
The anatomic distribution was noted to be central if there was a predominance of abnormalities in the inner two-thirds of the lung, peripheral if there was a predominance of abnormalities in the outer third of the lung, dorsal if there was predilection for the dependent portion, peribronchovascular if there was predilection for the peribronchovascular areas, and random if there was no predominance. Predominance was assessed as being upper lung zone, lower lung zone, or random. Upper lung zone predominance was considered present when most of the abnormalities were above the level of the tracheal carina, and lower zone predominance was considered present when most of the abnormalities were below this level.
The extent of involvement of each abnormality was assessed independently for each of three zones of each lung. The CT score in upper, middle, and lower lung zones was determined by visually estimating the extent of disease in each zone. The score was based on the percentage of lung parenchyma that showed evidence of abnormality and was estimated to the nearest 10% of parenchymal involvement. Overall percentage of involvement was obtained by averaging the six lung zones. The extent of traction bronchiectasis was evaluated by counting the number of segments or subsegments that showed traction bronchiectasis. The following 18 segments or subsegments were evaluated: right apical upper, right anterior upper, right posterior upper, right lateral middle, right medial middle, right superior upper, right medial basal, right anterior basal, right lateral basal, right posterior basal, left apicoposterior upper, left anterior upper, left superior lingular, left inferior lingular, left superior lower, left anteromedial basal, left lateral basal, and left posterior basal.
The extent of traction bronchiectasis was also quantified by assessing the generations of bronchial divisions involved. Traction bronchiectasis was scored as follows: 0, no bronchial dilatation; 1, dilatation limited to bronchi distal to the sixth-generation bronchi, the main bronchi being considered the first generation; 2, bronchial dilatation involving the fifth-generation bronchi; 3, bronchial dilatation involving the fourth generation; 4, bronchial dilatation involving the third generation; and 5, bronchial dilatation involving bronchi proximal to the second generation.
These 55 cases were categorized into pathologic subgroups according to the classification of Katzenstein and Fiorelli (1): group 1, inflammation without fibrosis; group 2, mixture of inflammation and fibrosis; and group 3, fibrosis alone. Group 2 was further categorized as cellular or fibrotic, according to the classification of Nagai et al (2) and Travis et al (3). Thus, the 55 cases were categorized conceptually for both of the above classifications into four grades: grade 1, interstitial inflammation without fibrosis (n = 6); grade 2, interstitial inflammation predominating over fibrosis (n = 16); grade 3, fibrosis predominating over inflammation (n = 5); and grade 4, only fibrosis (n = 28).
Statistical Evaluation
The pathologic grade and the extent of each finding were compared by using the Spearman rank correlation coefficient. The difference of prevalence of each abnormality in each subgroup was analyzed with the 2 test. The interobserver variation of the extent of various abnormalities was evaluated with the Spearman rank correlation coefficient. A P value of less than .05 was considered to indicate a statistically significant difference. The interobserver variation of the existence of predominant distribution, architectural distortion, lymph node enlargement, and pleural effusion was analyzed by using statistics. The interobserver agreement was classified as follows: poor, = 0–0.20; fair, = 0.21–0.40; moderate, = 0.41–0.60; good, = 0.61–0.80; and excellent, = 0.81–1.00.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
= 0.7–1.0). Therefore, the description of the CT findings and the correlation with the pathologic grades is based on the average of the observations made by the two observers. CT findings of NIP are summarized in the Table. Areas with ground-glass attenuation were present in all 55 patients (Figs 1–4). The areas with ground-glass attenuation involved 29% ± 14 (mean ± SD) of the lung parenchyma. Areas of air-space consolidation were seen in 54 patients (98%) and involved 11% ± 8.6 of the lung parenchyma (Fig 2). Nodules were found in 53 patients (96%) and involved 9.1% ± 4.3 of the lung parenchyma. Traction bronchiectasis was identified in 52 patients (95%) (Figs 2–4) and involved in 12 ± 0.9 segments and 3.0 ± 1.4 generations of bronchial divisions. Architectural distortion also was found in 50 patients (91%) (Figs 2–4). Intralobular reticular opacities were present in 48 patients (87%) and involved 9.8% ± 7.6 of the lung parenchyma (Fig 3). This finding, when present, was always superimposed on areas of ground-glass attenuation. Mild interlobular septal thickening was seen in 39 patients (71%).
|
|
|
|
|
|
|
|
|
Lower lung zone predominance was present in 52 patients (95%), and random (no zonal) predominance was present in three (5%). Peribronchovascular distribution was evident in 31 patients (56%), peripheral distribution in 21 (38%), and random distribution in 13 (24%).
The correlation between pathologic grade and extent of each CT finding also is summarized in the Table. Numbers of segments showing traction bronchiectasis (r = 0.68; P < .001) and the generations of bronchial divisions involved by traction bronchiectasis (r = 0.72; P < .001) correlated significantly with pathologic grades. Extent of areas with ground-glass attenuation (r = 0.32; P = .018), intralobular reticular opacities (r = 0.35; P = .011), and honeycombing (r = 0.31; P = .022) showed significant but weak correlation with pathologic grades. No other correlation between extent of CT findings and pathologic grades was found.
Honeycombing was seen in 12 (43%) of 28 patients with grade 4 NIP and in three (11%) of the remaining 27 patients (2 test, P < .001). There was no other significant difference in frequency of findings between the four subgroups of NIP.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Results of early studies, based on relatively small numbers of cases, suggest that NIP had a fairly uniform appearance at thin-section CT (5,6). Park et al (5), on the basis of a review of findings in seven patients, reported that the most common abnormalities at thin-section CT consist of patchy bilateral areas of ground-glass attenuation with or without associated areas of consolidation or irregular lines. Cottin et al (6), according to their evaluation of 12 patients, concluded that the characteristic CT findings of NIP consist of ground-glass attenuation, areas of air-space consolidation, and thickening of interlobular septa. In contrast, more recent studies, which have included greater numbers of patients, demonstrate a more heterogeneous appearance of NIP (7,8). Hartman et al (8) reported that 39 (78%) of 50 patients with NIP had CT findings that were not compatible with previous descriptions of NIP. Atypical findings included isolated air-space consolidation, centrilobular nodules, and subpleural honeycombing. These authors, however, did not compare CT findings with histologic findings. Kim et al (9) compared CT and histologic findings in 23 patients. They concluded that the most common abnormality in NIP consisted of areas of ground-glass attenuation and irregular lines or bronchial dilatation and that these findings reflect the presence of interstitial inflammation, fibrosis, or both. The small number of patients in their study, however, did not allow adequate assessment of the ability to distinguish between the various histologic subgroups of NIP on the basis of the CT findings.
In the current study, honeycombing was seen in approximately one-third of patients with NIP, and the presence of honeycombing on CT images was highly suggestive of fibrotic NIP. These patients, therefore, are unlikely to respond to corticosteroid therapy. Honeycombing indicates the presence of end-stage pulmonary fibrosis (10,11). The frequency of honeycombing in the present study (27%) was similar to that of two other studies (26% and 30%, respectively) (7,8) and higher than those reported in initial studies (0%–1%) (5,6,9).
The remaining abnormalities seen on thin-section CT images were less helpful than honeycombing in distinguishing histologic subtypes in any given patient. As a group, the findings that correlated best with pathologic grades of NIP were extent of traction bronchiectasis (r = 0.68; P < .001) and extent of reticulation (r = 0.35; P < .05). Traction bronchiectasis has been shown to be a reliable indicator for the presence of fibrosis in other interstitial lung diseases, particularly UIP and acute interstitial pneumonia (12–15).
In the present study, extent of reticular opacities had a significant correlation with the pathologic grade of fibrosis. Reticular opacities on thin-section CT images can be due to an alveolar filling process, an interstitial process, or a combination of alveolar and interstitial processes (16). In patients with UIP, this finding has histologically been shown to reflect the presence of intralobular interstitial thickening by inflammatory cells or by fibrosis (14,16). In the present study, the frequency of this finding in patients with cellular NIP was similar to that of patients with fibrotic NIP.
The presence and extent of ground-glass attenuation were not helpful in distinguishing between the various subtypes of NIP. Ground-glass attenuation indicates the presence of active inflammation and potentially treatable disease or the presence of fibrosis below the resolution of thin-section CT (17). Remy-Jardin et al (12) showed that, in patients with UIP, ground-glass attenuation should be considered to represent an active process only when there is no associated finding. In patients with UIP, findings indicative of fibrosis include intralobular reticular opacities, honeycombing, and traction bronchiectasis (12). In the current study, 48 of 55 patients had intralobular reticular opacities associated with ground-glass attenuation. Seventeen (77%) of the 22 patients with cellular NIP had intralobular reticular opacities associated with ground-glass attenuation, whereas 31 (94%) of the 33 patients with fibrotic NIP had this finding. In the seven patients who did not have intralobular reticular opacities, five had cellular NIP, and two had fibrotic NIP. Therefore, in patients with NIP, intralobular reticular opacities associated with ground-glass attenuation do not always indicate fibrosis.
Several investigators have emphasized the heterogeneous appearance of NIP on CT images (7,8). Hartman et al (8) reported that 16 (32%) of 50 patients had CT findings that were not compatible with previous descriptions of NIP. They suggested that the variability of findings is due to several factors: (a) There is no "classic appearance" of NIP at thin-section CT; (b) some cases are misdiagnosed as NIP because of sampling error; and (c) the diagnosis of NIP as currently applied by the pathologist might not define a uniform group of patients or a disease process. In the present study, pathologic diagnosis was assigned by both the regional pathologists and an external expert lung pathologist. The thin-section CT findings of patients with cellular NIP were ground-glass attenuation, air-space consolidation, smaller extent of traction bronchiectasis, and rare occurrence of honeycombing as compared with higher grades of NIP. Patients with fibrotic NIP had honeycombing, as well as greater extent of ground-glass attenuation, traction bronchiectasis, and intralobular reticular opacities than those with cellular NIP. Patients with mixed cellular pattern and fibrosis of various degrees had intermediate CT findings between the cellular and fibrotic subtypes. Overall, however, there was a less heterogeneous appearance than that suggested by Hartman and co-workers (8).
The current study has two main limitations. First, although it included a greater number of patients than did previous studies, the number of patients with purely cellular NIP was small. Second, owing to the retrospective nature of the study, it was not possible to perform a precise radiologic-pathologic correlation.
In conclusion, characteristic thin-section CT findings of cellular NIP consist of ground-glass attenuation and air-space consolidation, with relatively small extent of traction bronchiectasis and intralobular reticulation and no honeycombing. Patients with fibrotic NIP have ground-glass attenuation with extensive traction bronchiectasis and intralobular reticular opacities; approximately 40% have honeycombing. Patients with mixed cellular and fibrotic NIP have intermediate CT findings. Namely, the extent of traction bronchiectasis and intralobular reticulation at thin-section CT correlates with increased likelihood of fibrosis in NIP. Honeycombing is seen almost exclusively in patients with fibrotic NIP.
|
FOOTNOTES |
|---|
Author contributions: Guarantors of integrity of entire study, T.J., N.L.M.; study concepts, T.J., N.L.M., H.T.; study design, T.J., K.I., Y.K., H.A.; literature research, K.F., T.J., K.I.; clinical studies, M. Kinoshita, M. Koyama, H.Y.; data acquisition, T.J., K.I.; data analysis/interpretation, T.J., M. Koyama; statistical analysis, T.J., M. Koyama; manuscript preparation, T.J., N.L.M., H.N.; manuscript definition of intellectual content, T.J., N.L.M., M.A.; manuscript editing, N.L.M., M.S., H.T.; manuscript revision/review, H.A., N.L.M., H.N.; manuscript final version approval, T.J., N.L.M.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  W. D. Travis, G. Hunninghake, T. E. King Jr., D. A. Lynch, T. V. Colby, J. R. Galvin, K. K. Brown, M. P. Chung, J.-F. Cordier, R. M. du Bois, et al. Idiopathic Nonspecific Interstitial Pneumonia: Report of an American Thoracic Society Project Am. J. Respir. Crit. Care Med., June 15, 2008; 177(12): 1338 - 1347. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Sumikawa, T. Johkoh, T. V. Colby, K. Ichikado, M. Suga, H. Taniguchi, Y. Kondoh, T. Ogura, H. Arakawa, K. Fujimoto, et al. Computed Tomography Findings in Pathological Usual Interstitial Pneumonia: Relationship to Survival Am. J. Respir. Crit. Care Med., February 15, 2008; 177(4): 433 - 439. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. I. S. Silva, N. L. Muller, D. A. Lynch, D. Curran-Everett, K. K. Brown, K. S. Lee, M. P. Chung, and A. Churg Chronic Hypersensitivity Pneumonitis: Differentiation from Idiopathic Pulmonary Fibrosis and Nonspecific Interstitial Pneumonia by Using Thin-Section CT Radiology, January 1, 2008; 246(1): 288 - 297. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. B Gotway, M. M Freemer, and T. E King Jr Challenges in pulmonary fibrosis {middle dot} 1: Use of high resolution CT scanning of the lung for the evaluation of patients with idiopathic interstitial pneumonias Thorax, June 1, 2007; 62(6): 546 - 553. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Mueller-Mang, C. Grosse, K. Schmid, L. Stiebellehner, and A. A. Bankier What Every Radiologist Should Know about Idiopathic Interstitial Pneumonias RadioGraphics, May 1, 2007; 27(3): 595 - 615. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. A. Lynch Quantitative CT of Fibrotic Interstitial Lung Disease Chest, March 1, 2007; 131(3): 643 - 644. [Full Text] [PDF]
|
|
|
|
|
|
H. Sumikawa, T. Johkoh, K. Ichikado, H. Taniguchi, Y. Kondoh, K. Fujimoto, U. Tateishi, T. Hiramatsu, A. Inoue, J. Natsag, et al. Usual Interstitial Pneumonia and Chronic Idiopathic Interstitial Pneumonia: Analysis of CT Appearance in 92 Patients Radiology, October 1, 2006; 241(1): 258 - 266. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Misumi and D. A. Lynch Idiopathic pulmonary fibrosis/usual interstitial pneumonia: imaging diagnosis, spectrum of abnormalities, and temporal progression. Proceedings of the ATS, January 1, 2006; 3(4): 307 - 314. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. R. Flaherty, T. E. King Jr., G. Raghu, J. P. Lynch III, T. V. Colby, W. D. Travis, B. H. Gross, E. A. Kazerooni, G. B. Toews, Q. Long, et al. Idiopathic Interstitial Pneumonia: What Is the Effect of a Multidisciplinary Approach to Diagnosis? Am. J. Respir. Crit. Care Med., October 15, 2004; 170(8): 904 - 910. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T Fischer, J H Reynolds, and S E Trotter The idiopathic interstitial pneumonias: a beginner's guide Imaging, October 1, 2004; 16(1): 37 - 49. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Ujita, E. A. Renzoni, S. Veeraraghavan, A. U. Wells, and D. M. Hansell Organizing Pneumonia: Perilobular Pattern at Thin-Section CT Radiology, September 1, 2004; 232(3): 757 - 761. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Kaminski, J. A. Belperio, P. B. Bitterman, L. Chen, S. W. Chensue, A. M.K. Choi, S. Dacic, J. H. Dauber, R. M. du Bois, J. J. Enghild, et al. Idiopathic Pulmonary Fibrosis Am. J. Respir. Cell Mol. Biol., September 1, 2003; 29(3): S1 - 105. [Full Text] [PDF]
|
|
|
|
 |
|
 V. Cottin, F. Thivolet-Bejui, M. Reynaud-Gaubert, J. Cadranel, P. Delaval, P-J. Ternamian, J-F. Cordier, and and the Groupe d'Etudes et de Recherche sur les Ma Interstitial lung disease in amyopathic dermatomyositis, dermatomyositis and polymyositis Eur. Respir. J., August 1, 2003; 22(2): 245 - 250. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
