| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Respiratory Department, Taragi Municipal Hospital, Kumamoto, Japan.
Correspondence to: Kosuke Kashiwabara, MD, Respiratory Department, Taragi Municipal Hospital, 4210 Taragi-machi, Kuma-gun, Kumamoto, 868-0501, Japan; e-mail: cbi54000{at}pop02.odn.ne.jp
Abstract
Study objectives: To evaluate characteristics or disease activity of early interstitial lung disease (ILD) in subjects with true parenchymal abnormalities in the posterior subpleural aspect of the lung.
Patients and methods: This study enrolled 14 subjects with dependent densities that disappeared on helical CT obtained with the subject prone (control group) and 7 subjects with true parenchymal abnormalities that remained unchanged on prone CT image but were not detectable on chest radiographs (true abnormalities group). Pulmonary function tests and serum markers for idiopathic lung fibrosis as KL-6, surfactant protein D (SP-D), and surfactant protein A (SP-A) in the two groups were evaluated.
Results: In the true abnormalities group, curvilinear subpleural lines or thickened interlobular and intralobular lines were observed more frequently in the lower lung fields. Diffusing capacities of the lung for carbon monoxide (15.3 ± 3.5 mL/min/mm Hg vs 18.8 ± 3.7 mL/min/mm Hg, p = 0.0493) were lower, and KL-6 (607 ± 297 U/mL vs 318 ± 143 U/mL, p = 0.0090), SP-A (59 ± 24 ng/mL vs 34 ± 12 ng/mL, p = 0.0207), and SP-D (112 ± 54 ng/mL vs 42 ± 24 ng/mL, p = 0.0028) were higher in the true abnormalities group than in the control group (± SD).
Conclusion: True parenchymal abnormalities in the posterior subpleural aspect of the lung may indicate early ILD activity.
Key Words: annual health examinations • curvilinear subpleural lines • KL-6 • subpleural ground-glass opacities • surfactant protein A • surfactant protein D • thickened interlobular and intralobular lines
We have occasionally encountered asymptomatic subjects with parenchymal abnormalities in the posterior subpleural aspect of the lung that were visualized on helical CT but were not detectable on chest radiographs. Although one problem is that these opacities closely mimic the appearance of one of the idiopathic interstitial pneumonias,123 most of them are gravity-related physiologic phenomena because these opacities disappear on helical CT obtained with the subject prone, resulting in a "dependent density" as described by Webb et al.1 However, true parenchymal abnormalities that did not disappear on prone CT image are suspected of early interstitial lung disease (ILD) including idiopathic pulmonary fibrosis (IPF). This disease is characterized on high-resolution CT by the presence of reticulation or honeycombing involving predominantly the subpleural lung lesions and the lower lung fields,4567 but there has not been any report on the appearance of early ILD not detectable on chest radiographs. In addition, it is not clear whether or not these parenchymal abnormalities in the posterior subpleural aspect of the lung have characteristics or disease activity for ILD. The purpose for this study was to evaluate pulmonary function tests and serum-sensitive markers for IPF in subjects with dependent densities that disappear on prone CT imaging and in subjects with true parenchymal abnormalities that remain unchanged on prone CT imaging.
Materials and Methods
During the 4-year period from April 1, 1999, to March 31, 2003, helical CT screening was performed as an optional test in 1,385 subjects who received annual health examinations. This screening was done at the end of two breath holds and on supine imaging (ProSeed Accell CT system; GE Yokokawa Medical System; Tokyo, Japan), and examination parameters were 120 kilovolts, 140 mA, 10-mm collimation, and 10 mm-per-rotation table speed. All evaluations in the present study were made at the same window setting (1,500 Hounsfield units, – 600 Hounsfield units). The CT images were assessed independently by both a radiologist and a chest specialist without reference to clinical data.
The study was approved by the institutional review board, and all patients completed informed consent forms. One hundred seven subjects (8%) had parenchymal abnormalities in the posterior subpleural aspect of the lung and were advised to receive additional CT scans in a prone position. Eighty-eight subjects underwent prone image CT scans (same parameters on supine image), but the other 19 subjects refused to undergo the test. In 78 subjects, the opacities disappeared on prone image helical CT and were diagnosed as dependent densities. Ten subjects had true parenchymal abnormalities that did not disappear on prone CT image. Three of them were excluded from this study because they had the appearance of honeycombing detectable on chest radiography. Twenty-one subjects underwent an additional high-resolution CT and blood examinations and pulmonary function tests. Patients were classified into two groups: patients with dependent density that disappeared on prone CT imaging (n = 14, control group); and patients with true parenchymal abnormalities that remained unchanged on prone CT imaging (n = 7, true abnormalities group).
The classification of Aberle et al2 was used to evaluate the radiologic findings of parenchymal abnormalities in the posterior subpleural aspect of the lung on helical CT. This classification consists of the following: (1) curvilinear subpleural lines, defined as linear densities within 1 cm of the pleura and parallel to the inner chest wall; (2) thickened interlobular and intralobular lines, defined as single or branching lines 1 to 2 cm in length, seen in the subpleural parenchyma, extending peripherally toward the pleural surface; (3) subpleural dependent density, defined as broad bands of featureless increased density marginated the dependent lung, varying in width from 2 to 20 mm and sufficiently opaque to obscure the morphologic structure of the underlying parenchyma; (4) parenchymal lines, defined as linear, nontapering densities 2 to 5 cm in length extended through the lung to contact the pleural surface; and (5) honeycombing, defined as small cyst-like spaces with thick walls were seen, most commonly in the subpleural regions of the lung. The term subpleural dependent density was not used in this study because this phrase is a generic term for gravity-related parenchymal abnormalities; instead, the expression subpleural ground-glass opacities was used. To evaluate the predominant distribution of parenchymal abnormalities on supine CT image, four levels of helical CT were stratified, as follows: apex levels (a horizontal level appeared adjacent to the upper margin of the aortic arch) representative of upper lung fields of the lung; hilum (a horizontal level appeared adjacent to the upper margin of trunk of the right pulmonary artery); heart (a horizontal 
appeared adjacent to the right inferior pulmonary vein connecting to the left atrium) levels representative of middle lung fields; and basal levels (a horizontal level appeared adjacent to he upper margin of spleen) representative of lower lung fields.
A hematologic analysis was performed (Sysmex SE-9000; Sysmex; Kobe, Japan), as was biochemical analysis (TBA-80FR NEO2; Toshiba Medical; Tokyo, Japan). Measurements of serum KL-6, surfactant protein A (SP-A), and surfactant protein D (SP-D) were performed using an electrochemiluminescence immunoassay kit (Sanko Junyaku; Kobe, Japan; reference range < 500 U/mL), an enzyme-linked immunoassay kit (Sysmex; reference range < 43.8 ng/mL), and an enzyme-linked immunoassay kit (Yamasa Shoyu; Chiba, Japan; reference range < 110 ng/mL) following the instructions of the manufacturer. Respiratory function measurements in all subjects were made (CHESTAC-33 spirometer; Chest; Tokyo, Japan). Measurements of vital capacity (VC); ratio of VC to predicted vital capacity (%VC); FVC; FEV1; ratio of FEV1 to FVC (FEV1%); total lung capacity (TLC); ratio of residual volume (RV) to TLC; diffusing capacity of lung for carbon monoxide (DLCO); and ratio of DLCO to predicted DLCO (%DLCO) were based on two satisfactory maximal forced expiratory maneuvers performed with the subjects in a standing position.
Statistical analysis was performed on a personal computer using statistical software (Stat View J 5.0; SAS Institute; Cary, CA). Differences between two independent samples were tested using the Mann-Whitney U test. Analysis of categorical data was performed with Fisher exact probability test; p < 0.05 was considered to indicate statistically significant difference.
Results
Radiologic Findings on Supine CT Image
Eleven of 14 dependent densities (79%) and all true parenchymal abnormalities were located in the posterior subpleural aspect of the both lungs (Table 1
). Dependent densities were observed at the hilum or the heart levels, and true parenchymal abnormalities were observed at the basal levels as well as at the hilum or the heart levels. There was a difference between the percentages of subjects with parenchymal abnormalities at the basal levels of CT in the two groups (p = 0.0002). According to the appearance of these abnormalities on supine CT image (Fig 1
), the percentage of subjects with curvilinear subpleural lines was higher in the true abnormalities group than in the control group, but not with statistical significance (p = 0.0767). Thickened interlobular and intralobular lines were observed more frequently in the true abnormalities group than in the control group (p = 0.0182). A difference between the percentages of subpleural ground-glass opacities in the two groups was not found.
|
|
|
|
|
In subjects with true parenchymal abnormalities that did not disappear on prone CT image but were not apparent on chest radiographs, DLCO was decreased and serum levels of sensitive markers for IPF were increased compared with those in subjects with dependent densities. Curvilinear subpleural lines and thickened interlobular and intralobular lines were observed more frequently in subjects with true parenchymal abnormalities; however, parenchymal bands or honeycombing in either group were not found. Aberle et al2 reported that there were no subjects with parenchymal bands or honeycombing that disappeared on prone CT image. I suggest that parenchymal bands or honeycombing are not the manifestation of dependent density. The reason why these abnormalities are not found even in subjects with true parenchymal abnormalities may be that subjects with parenchymal bands or honeycombing were excluded from our study because they were visible on chest radiographs. Other explanation is that this could represent earlier disease, or also the disease does not represent IPF but one of the other idiopathic interstitial pneumonia. The predominant distribution of dependent densities was in the middle lung fields, while that of true parenchymal abnormalities was in the lower lung fields. It is not clear why dependent densities were predominantly found in the middle lung fields. As these opacities were gravity-related physiologic phenomena, one possible explanation for that may be that the vertical location of thoracic vertebrae of the middle lung fields are lower than that of the lower lung fields of CT on supine position. I suggest that curvilinear subpleural lines or thickened interlobular and intralobular lines observed in the lower lung fields are useful for distinguishing true parenchymal abnormalities from gravity-related physiologic phenomena.
KL-6 is a circulating high-molecular-weight glycoprotein classified in human as MUC1 mutin and expressed on type II pneumonocytes.9 It has been reported that KL-6 in serum or in BAL fluid might be a useful marker in the differential diagnosis of ILD, evaluation of disease activity, and prediction of disease outcome.10111213 SP-A and SP-D are collagenous glycoproteins belonging to the collectin subgroup of the C-type lectin superfamily and are important components of the antibody-independent pulmonary host defense system.14 SP-A and SP-D in sera or in BAL fluid have also been reported to be useful biomarkers of ILD.151617 In the present study, there was no difference between serum levels of LDH in the two groups, but average levels of KL-6, SP-A, and SP-D in sera were higher in the true abnormalities group. In addition, DLCO decreased in this group. There has not been any report in which parenchymal abnormalities in the posterior subpleural aspect of the lung that did not disappear on prone CT imaging were pathologically diagnosed as early IPF. Our results suggest that these opacities clinically correspond with the appearance of early ILD, although open-lung biopsy examinations were not performed. In conclusion, true parenchymal abnormalities in the posterior subpleural aspect of the lung may in fact indicate early ILD activity, and it is important to follow up the chronologic change of the appearance of these opacities because earlier histologic diagnosis would be advantageous and possibly more treatment responsive.
Acknowledgements
The author is indebted to Dr. Syu-ichi Kohshi of the Medical Health Care Center "COSMO" for his dedication in data collection, and Professor J. Patrick Barron of the International Medical Communications Center of Tokyo Medical University for review of the article.
Footnotes
Abbreviations: DLCO = diffusing capacity of lung for carbon monoxide; %DLCO = ratio of diffusing capacity of lung for carbon monoxide to predicted diffusing capacity of lung for carbon monoxide; FEV1% = ratio of FEV1 to FVC; ILD = interstitial lung disease; IPF = idiopathic pulmonary fibrosis; LDH = lactate dehydrogenase; SP-A = surfactant protein A; SP-D = surfactant protein D; TLC = total lung capacity; VC = vital capacity; %VC = ratio of vital capacity to predicted vital capacity
Received for publication April 19, 2005. Accepted for publication June 28, 2005.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
