| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Thoracic Imaging |
1 From the Department of Radiology, Hospital Calmette, University Center of Lille, Boulevard Jules Leclerc, 59037 Lille, France (M.R.J., I.M., C.Z., J.R.); and Environmental and Occupational Health and Ergonomics Research Center (A.S.) and Department of Medical Statistics (A.D.), University of Lille, Lille, France. Received July 23, 2003; revision requested October 3; final revision received February 9, 2004; accepted March 23. Address correspondence to M.R.J. (e-mail: mremy-jardin@chru-lille.fr).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Eighty-three male workers with a mean duration of occupational exposure to asbestos of 18 years underwent CT as part of a medicolegal investigation. CT examination included low-dose multi–detector row spiral CT of the entire thorax, with reconstruction of contiguous 5-mm-thick images, and thin-section CT, which served as the reference standard for the detection of pleural and parenchymal asbestos-related abnormalities. Two main groups of abnormalities were identified: (a) pleural plaques and diffuse pleural thickening and (b) thickened interstitial short lines, curvilinear subpleural lines, ground-glass opacity with or without bronchiectasis, and honeycombing. The frequencies of the depiction of these abnormalities on the low-dose multi–detector row images and the thin-section images were compared by using the McNemar test.
RESULTS: No significant differences were observed between the low-dose and thin-section CT images in the depiction of either (a) parietal pleural fibrosis consisting of pleural plaques (identified in 67 [81%] vs 65 [78%] workers, P = .157), which appeared mainly as thick, calcified pleural linear structures; or (b) features of parenchymal fibrosis, which consisted of various combinations of intralobular and septal lines (identified in 12 [14%] vs 13 [16%] workers, P = .564), subpleural curvilinear lines (identified in 10 [12%] vs eight [10%] workers, P = .157), and ground-glass opacity with (identified in six [7%] vs six [7%] workers) or without (identified in five [6%] vs three [4%] workers, P = .317) traction bronchiectasis. A honeycombing pattern was depicted on only the thin-section CT images (P < .001). Emphysema (identified in 26 [31%] vs 14 [17%] workers at low-dose and thin-section CT, respectively; P < .001) and noncalcified nodules (identified in 18 [22%] workers vs one [1%] worker, P < .001) were depicted significantly more frequently on the low-dose images than on the thin-section images.
CONCLUSION: Low-dose multi–detector row spiral CT accurately depicts asbestos-related disease.
© RSNA, 2004
Index terms: Asbestos • Computed tomography (CT), comparative studies, 60.12111, 60.12115, 60.12118 • Lung, CT, 60.12111, 60.12115, 60.12118 • Lung, diseases, 60.742, 60.75, 60.76, 70.773, 60.774 • Pneumoconiosis, 70.773, 60.774
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Although thin-section computed tomography (CT) is more sensitive than radiography in the detection of early asbestos-related pleural and parenchymal changes and the correlation between thin-section CT findings and pathologic findings has been established (3), chest radiography remains the main radiologic tool for the detection of these lesions, with CT reserved for problem solving (1). Both the expense and the time required to perform CT of the entire thorax have made this examination impractical for examining large asbestos-exposed populations (4). In addition, typical chest CT protocols have been associated with relatively high radiation doses to patients, which have raised concern about the potential for induced malignant disease, particularly in screening settings.
The introduction of low-radiation-dose scanning techniques has facilitated renewed interest in the potential usefulness of CT as a first-line imaging tool. Possible adjustments to the scanning parameters selected at the console, as well as automatic dose modulation systems, are important in this respect and are being used so that the radiation dose is as low as possible yet sufficient to yield the required image quality. With use of such approaches, dose reductions during thoracic CT have been focused on mainly at low-dose thin-section CT examinations (5–9). The usefulness of low-dose single-section spiral CT of the lung parenchyma in lung cancer screening programs has been emphasized (10–12), and the capability of this modality to depict mediastinal structures and abnormalities has been reported in a few studies (13,14). The advent of multi–detector row spiral CT has generated additional possibilities for screening in terms of both low radiation dose requirements and high spatial resolution. The purpose of this study was to evaluate the depiction of lung and pleural asbestos-related lesions with low-dose four–detector row spiral CT.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The study group included 83 consecutive male subjects who were referred to the Department of Radiology of Hospital Calmette for the evaluation of asbestos-related disease between April 2001 and March 2002. For all subjects, 57 of whom were retired and 26 of whom were actively working, CT was indicated as part of the medicolegal follow-up, which also included the acquiring of a detailed medical history and the performing of a physical examination, pulmonary function tests, and posteroanterior chest radiography. The histories of occupational exposure to asbestos fibers included work involving the manufacturing of insulation products, mainly for gas-liquefying machines or shipbuilding, and the production of friction materials used for automotive products.
We received approval from the institutional review board of the University Center Hospital of Lille to include low-dose CT scanning as part of the medicolegal radiologic evaluation of asbestos-exposed workers who were referred to us from the Environmental and Health Department of our institution, which also agreed with the decision to include low-dose CT. The investigated protocol was expected to yield more diagnostic information than chest radiography—the only radiologic examination routinely performed in asbestos-exposed workers—with use of an overall radiation dose that was lower than that delivered during routine conventional CT examinations of asbestos-exposed patients. At our institution, these examinations include the acquisition of both conventional spiral CT images of the entire thorax and sequential thin-section CT images in 20-mm intervals from the apices to the bases of the lungs during the same session. For our present study, the protocol included low-dose spiral CT instead of the conventional spiral CT examination. Thin-section CT was also performed, as it is in our routine protocol. Informed consent was systematically obtained from all subjects who were included in our study.
The mean duration of occupational exposure to asbestos was 18 years ± 9.6 (standard deviation) (range, 3–40 years), and the mean latency time between the first exposure and the time of the study was 46.29 years ± 7.7 (range, 31–59 years). Twenty-three subjects (28%) received state compensation, whereas 60 (72%) subjects had not been previously recognized as being eligible for compensation for asbestos-related disease. The mean age of the study group was 67.1 years ± 7.3 (range, 49–81 years). Twenty-six of the 83 subjects were smokers (mean cigarette consumption, 28 pack-years; range, 2–70 pack-years), 30 had not smoked for at least 1 year (mean cigarette consumption, 22 pack-years; range, 2–70 pack-years), and 27 were nonsmokers. At the time of their enrollment in the present investigation, 16 (19%) workers reported having dyspnea, whereas 67 (81%) workers were asymptomatic.
CT Examinations
The CT examinations were performed with a four–detector row scanner (Sensation 4; Siemens, Forchheim, Germany) and consisted of two successive acquisitions performed during the same session. For the first acquisition, low-dose spiral CT of the entire thorax was performed with the subject in the supine position and at full inspiration by using 2.5-mm collimated sections acquired with four detectors (ie, 4 x 2.5-mm collimation), a pitch of 1.75, and a scanning time of 0.5 second. The scanning parameters were systematically adjusted to the subject’s body size. For subjects weighing 70 kg or less, 120 kVp and between 60 and 80 mAs per section were used. For subjects weighing more than 70 kg, 120 kVp and between 80 and 100 mAs per section were used. In addition, all of these examinations were performed while a radiation dose reduction system based on online tube current controls (Siemens Care Dose System; Siemens) with prospective dose saving was applied. This dose-reduction system adapted the maximum value of the tube current in addition to the inherent tube current modulation. The system was installed on the Sensation 4 scanner as a work-in-progress option; the minimum dose saving was set at 32%. From each data set, two series of images—5-mm-thick lung and mediastinal scans—were systematically reconstructed by using high-spatial-frequency and soft algorithms, respectively.
For the second acquisition, thin-section CT was performed with the subject in the prone position immediately after spiral CT scanning and served as the reference standard for the diagnosis of pleural and parenchymal asbestos-related abnormalities. These CT images were obtained from the level of the carina to the costophrenic angles, 30 mm apart, by using 1.0-mm collimation (two 0.5-mm acquisitions), a 0.75-second rotation time, 140 kV, and between 60 and 100 mAs per section according to the subject’s body habitus. So that the low-dose images could be compared with the high-quality thin-section reference images (ie, our routine protocol images), no dose-reduction system was applied for the thin-section CT examinations.
All CT scans—both the low-dose spiral images and the thin-section images—were obtained at two window settings that were appropriate for viewing the lung parenchyma (window width, 1600 HU; window level, –600 HU) and the mediastinum (window width, 350 HU; window level, 30 HU).
Interpretation of CT Images
Conditions of image interpretation.—For the first reading, the low-dose and thin-section CT images were interpreted separately, several weeks apart. Each CT image was viewed simultaneously by two experienced chest radiologists (I.M. with 3 years of experience and M.R.J. with 15 years of experience), who reached a consensus regarding the findings. A second and simultaneous reading of the low-dose and thin-section CT images obtained in each subject was undertaken several weeks later to evaluate the concordant or discordant detection and characterization of pleuroparenchymal abnormalities and to search for differences in the appearance of the lung based on the two scanning positions (ie, prone and supine).
Evaluation of low-dose spiral CT image quality.—Evaluation of image quality included the search for respiratory motion–induced artifacts, which were judged to be absent (grade 0), minimal (grade 1), or altering image interpretation (grade 2). The location of artifacts in the upper, middle, or lower lung zone was systematically recorded: The upper lung zone extended from the apex to the level of the carina; the middle lung zone, from the level of the carina to the level of the inferior pulmonary veins; and the lower lung zone, from the level of the inferior pulmonary veins down to the diaphragm. The criteria for motion artifacts included (a) double-imaged lung structures (ie, bronchial walls, vessels, fissures), (b) vessel distortion resulting in star artifacts or low-attenuating areas adjacent to vessels, and (c) linear streak artifacts in the lung parenchyma. The most severe form of motion artifact was always taken into consideration for the purpose of grading the overall image quality. Interpolation artifacts were searched for at the level of the ribs, where they were seen as hypoattenuating areas around the bone sections, and within the lung parenchyma, where they were seen as nonhomogeneously attenuating regions (15).
Assessment of pleural and parenchymal abnormalities.—According to established criteria (16–19), the following pleural and parenchymal abnormalities were recorded as present or absent on each side (left and right lungs) on the low-dose and thin-section CT images:
1. Asbestos-related pleural abnormalities were grouped into two categories: (a) Pleural plaques—either parietal or fissural—were defined as circumscribed, pleural areas of opacity with well-demarcated edges. The presence of calcification, the pleural plaque thickness (ie, 
2 mm or >2 mm), and the smooth or nodular aspect of the pleural plaque surface were systematically recorded. (b) Diffuse pleural thickening was defined as a contiguous sheet of pleural thickening that was more than 5 cm in extent along the pleural surface on transverse CT images, more than 8 cm in extent on craniocaudal CT images, and more than 3 mm thick.
2. CT features of asbestosis included five major abnormalities that were depicted when the subject was scanned while in the prone position: (a) Thickened interstitial short lines were seen in the subpleural region as septal and intralobular lines. Septal lines are short and discrete nonbranching lines, whereas intralobular lines appear as Y-shaped branching structures. Both line types were seen in the subpleural parenchyma. (b) Curvilinear subpleural lines were defined as linear areas of opacity within 1 cm of the pleura and parallel to the inner chest wall. (c) Areas of ground-glass opacity were defined as areas of increased attenuation in which the vessels and the bronchial walls remained visible. (d) Areas of ground-glass opacity could be accompanied by bronchiectasis—that is, abnormally depicted airways either within 1 cm of the parietal pleura or abutting the mediastinal pleura. (e) Honeycombing was defined as an area of lung containing cystlike spaces with thickened walls.
In addition to analyzing the CT features of asbestosis, we systematically analyzed the distribution of these features in the upper, middle, and lower lung zones, as well as in the central and peripheral portions of the lungs. The peripheral part of each lung corresponded to a 3-cm band of subpleural lung parenchyma on transverse images, whereas the remaining lung surface was considered to represent the central portion of the corresponding lung.
3. Additional CT features were grouped into two categories. The features in the first category were directly related to asbestos exposure and included (a) parenchymal bands, which were defined as linear, 2–5-cm-long areas of opacity extending through the lung to contact the pleural surface; these bands represented fibrosis along bronchovascular sheaths or interlobular septa, which were generally related to moderate pleural fibrosis; and (b) round atelectasis, which was defined as a mass near an area of pleural thickening, with a partial interposition of lung parenchyma between the pleura and the mass and a visible "comet tail" of vessels and bronchi sweeping into the lateral aspect or the medial and lateral aspects of the mass. The second category of additional findings consisted of smoking-induced abnormalities and included emphysema, bronchial wall thickening, and noncalcified lung nodules.
Evaluation of Radiation Exposure
At the initiation of this study, two common CT radiation dose parameters—the weighted CT dose index and the dose length product—were not routine settings on CT monitors and thus could not be measured and recorded on an individual basis. For each subject, the total milliampere-second product per image section was systematically and prospectively recorded after each CT examination. After the examinations were completed, theoretic estimates of the weighted CT dose index and the dose length product associated with the investigated low-dose scanning protocol were calculated as follows: (a) The mean value of the total milliampere-second product per image section, determined by using the acquisition parameters described earlier, enabled us to estimate the weighted CT dose index value for a standard low-dose CT examination. (The theoretic calculation was available on the user interface of the CT scanner, as requested according to the international standards for CT systems.) (b) The mean length of the region scanned multiplied by the weighted CT dose index value for a standard low-dose examination yielded the dose length product value for a standard low-dose CT examination.
Statistical Analysis
Statistical analysis was performed by using commercially available software (SAS Institute, Cary, NC). The frequencies of the appearance of abnormalities on the low-dose and thin-section CT images were compared by using the McNemar test. P < .05 was considered to indicate statistical significance.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Detection of Asbestos-related Pleuroparenchymal Abnormalities
Asbestos-related pleural abnormalities.—The frequencies with which asbestos-related pleural abnormalities were identified are summarized in Tables 1 and 2. No significant difference in the frequency of parietal pleural plaques was observed between the low-dose and thin-section CT images: These abnormalities were identified on the low-dose images obtained in 67 (81%) subjects and on the thin-section images obtained in 65 (78%) subjects (P = .157) (Table 1, Fig 1). Fissural pleural plaques were observed significantly more frequently (in 10 [12%] subjects) on the low-dose images than on the thin-section images (in three [4%] subjects) (P = .008). Fissural plaques coexisted with parietal plaques in all but one subject, who had a right fissural plaque and no parietal plaques. Pleural plaques were mainly observed as bilateral findings on both the low-dose images (in 56 [68%] subjects) and the thin-section images (in 54 [65%] subjects).
|
|
|
|
Asbestos-related parenchymal abnormalities.—We observed no significant difference in the detection of CT features of asbestosis between the low-dose and thin-section CT images (Table 1, Figs 2–4). Honeycombing was seen on the thin-section CT images obtained in one subject; this CT feature was not depicted on the corresponding low-dose images. This apparent discrepancy resulted because the honeycombing that was seen on the thin-section CT images was recorded as a pattern of ground-glass opacity on the low-dose images. Table 3 summarizes the distribution of the CT features of asbestosis, which were mainly observed in the lower lung zones and the subpleural regions. In all but one subject, the subpleural abnormalities seen on the low-dose images obtained with the subjects in the supine position were also identified on the thin-section images obtained with the subjects in the prone position.
|
|
|
|
|
|
|
Additional abnormalities.—Emphysema and lung nodules were depicted with a significantly higher frequency on the low-dose CT images, as compared with the frequency with which they were depicted on the thin-section CT images (Table 1). A total of 32 noncalcified and small (ie, <10 mm in diameter) lung nodules were depicted on the low-dose images obtained in 18 (22%) subjects; one of these nodules was seen on the thin-section images. The number of nodules per subject varied between one and six: 12 subjects had one nodule each, two subjects had two nodules each, two subjects had three nodules each, one subject had four nodules, and one subject had six nodules. The presence of regular contours and the absence of massive calcifications led the readers to classify these nodules as indeterminate.
Radiation Dose at Spiral CT
The characteristics of the low-dose CT examinations, in which the dose modulation system was always applied, can be summarized as follows: The mean milliampere-second value per image section ± the standard deviation was 50 mAs ± 6 (range, 32–69 mAs), the mean extent of the surveyed volume was 303 mm ± 21 (range, 245–360 mm), the mean weighted CT dose index was 4.7 mGy, and the mean dose length product was 142.7 mGy · cm. The morphologic characteristics of the population scanned included a mean body weight of 79.9 kg ± 13.2 (range, 50–128 kg) and a mean body height of 172.1 cm ± 6 (range, 160–185 cm).
Although the effective radiation dose was not individually calculated in the examined population, an estimate of the effective dose delivered with the investigated CT scanning protocol was calculated for a thoracic region of 30 cm in male and female patients. According to the manufacturer’s data, (a) the estimated effective dose for the investigated low-dose CT protocol is 1.9 mSv for a male patient and 2.4 mSv for a female patient, and (b) when surveying the same volume of interest by using the same CT scanner, the estimated effective dose for the standard spiral lung CT protocol is 3.4 mSv for a male patient and 4.4 mSv for a female patient.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
With regard to the detection of pleural abnormalities, we observed no significant difference in the frequency of the depiction of parietal pleural plaques, which were identified on the low-dose CT images obtained in 67 (81%) subjects and on the thin-section CT images obtained in 65 (78%) subjects. The only significant difference observed was that in the depiction of fissural pleural plaques, a well-known feature of visceral pleural fibrosis, which were seen more frequently on the low-dose CT images (in 10 [12%] subjects) than on the thin-section CT images (in three [4%] subjects). This apparent discrepancy was due to the sequential mode of scanning used in the thin-section CT examinations, at which some lesions were "missed" because of gaps between the two successive image acquisitions. These findings are in agreement with those in previous reports, which revealed that thin-section CT is insufficient for excluding the presence of pleural plaques (20).
When we compared low-dose CT images with thin-section CT images (the reference standard for the detection of interstitial lung diseases) obtained during the same session, we found no significant difference in the depiction of CT features compatible with the presence of lung parenchymal asbestosis. Nevertheless, at simultaneous analysis of the low-dose and thin-section CT images obtained in each subject, we noted the depiction of asbestos-related parenchymal abnormalities exclusively on the low-dose scans obtained in four subjects. This exclusive depiction of these abnormalities was due to differences in tissue sampling between the two CT techniques, as previously noted with regard to the depiction of fissural pleural plaques.
Our lung parenchyma analysis was based on the evaluation of five main abnormalities—thickened interstitial short lines, which were seen as subpleural septal and intralobular lines; curvilinear subpleural lines; honeycombing; ground-glass opacity; and ground-glass opacity in conjunction with traction bronchiectasis (16,18).
Although in many thin-section CT studies of asbestos-related disease, parenchymal bands have been considered a sign of asbestosis (16,21), we did not include this sign among the CT features of asbestosis. In a study in which meticulous pathologic-CT correlations in workers exposed to asbestos were reported, Akira et al (18) could not relate parenchymal bands to lung fibrosis because none of the seven workers with asbestosis who had been exposed to asbestos and had been considered in the in vitro pathologic-CT correlation had this CT feature.
In a more recent study (22) in which the various patterns of response to asbestos exposure were evaluated, the results of a cluster analysis suggested that parenchymal bands are more likely to reflect visceral pleural fibrosis than to reflect interstitial fibrosis. Last, the empiric finding that this abnormality is never seen with other chronic infiltrative lung diseases that do not have primary visceral pleural involvement reinforces the difference between asbestos-induced parenchymal abnormalities and the thin-section CT features that are more directly related to interstitial fibrosis.
Although low-dose CT accurately depicted the lung abnormalities, one should keep in mind that in two subjects, the characterization of abnormal infiltration differed between the low-dose and thin-section CT images. In both cases, thin-section CT depicted features of mild lung fibrosis, including the honeycomb pattern and the pattern of fine intralobular lines, both of which were erroneously recorded as areas of ground-glass opacity on the low-dose images. This limitation was inherent to the 5-mm thickness of the reconstructed images obtained with our scanning protocol; this thickness was subsequently responsible for the volume-averaging effects seen with the small abnormalities.
However, one should also note the limited number of discordant characterizations between the low-dose and thin-section images, despite there being a majority of mild forms of lung infiltration, which were mainly observed in the subpleural regions of the lower lung zones. Such a specific distribution raises additional questions about the effects of the subject’s positioning on the interpretation of lung abnormalities seen on low-dose and thin-section CT images obtained with the subject in the supine and prone positions, respectively.
The simultaneous reading of the two sets of lung images for each subject enabled us to observe similar findings on the low-dose and thin-section CT images obtained in all but one subject, in whom a complete clearing of dependent areas of opacity was observed on the prone thin-section scans. The low frequency with which dependent abnormalities were depicted might be explained by the high level of inspiration that was easily achieved by each individual before the acquisition of the low-dose image data. Another potential explanation could be the short duration of the data acquisitions, which may have made inspiratory scanning easier to perform compared with the inspiratory scanning involving the repetition of numerous inspiratory maneuvers that was performed in previous studies of sequential thin-section CT of the entire thorax.
In addition, the specificity of our study population, which consisted mainly of asymptomatic subjects, may explain the deeper inspiratory maneuvers and thus the lower frequency of dependent areas of opacity compared with those observed in previous studies involving not only patients with clinical asbestosis and more extensive lung infiltration (20) but also control subjects, in whom Aberle et al (16) reported a 35% occurrence of subpleural lines in the posterior regions of the lungs.
The subjects’ positioning for low-dose CT scanning was based on the fact that it is more comfortable for individuals to be scanned while they are supine rather than prone. In addition, in our clinical experience in scanning patients while they were prone, we observed a high number of cardiac motion artifacts in the lower lung zones, and this justified our a priori selection of the supine position.
In addition to enabling the accurate detection of pleuroparenchymal abnormalities, low-dose CT allowed us to identify additional findings, with a particular interest in those related to the smoking habits of the population scanned. With use of the volumetric data acquisition inherent to spiral CT scanning, the low-dose spiral CT protocol used in the present investigation can also be used as a tool to screen for malignant asbestos-related disease, lung carcinoma in particular. The individuals who are at high risk for lung carcinoma are those who have combined exposure to asbestos and cigarette smoke.
In a total of 18 subjects, low-dose CT depicted noncalcified nodules, which were missed on the thin-section CT scans obtained in all but one subject; these results led to the implementation of a specific follow-up program similar to that proposed for the detection of early lung cancer (11). To our knowledge, the use of spiral CT for the detection of early lung cancer in subjects exposed to asbestos has been evaluated in one study: Tiitola et al (23) used single-section spiral CT as a screening tool for the detection of early lung cancer in a group of patients with asbestos-related occupational disease who were followed up for 3 years. They screened 602 patients and found five lung cancers, three of which were curable.
With the systematic adjunct of a dose-modulation system, the multi–detector row CT images in the present investigation were obtained by using a mean of 50 mAs per section over a mean region volume of 303 mm, which resulted in a mean weighted CT dose index of 4.7 mGy and a mean dose length product of 142.7 mGy · cm. Although the effective dose was not individually calculated in the examined population, an estimate of the effective dose delivered with our low-dose CT scanning protocol involving 120 kV, 50 mAs per section, and 4 x 2.5-mm collimation can be calculated for a thoracic region of 30 cm in male and female patients. According to the manufacturer, the estimated effective dose is 1.9 mSv for a male patient and 2.4 mSv for a female patient, which are considerably smaller than the estimated doses reportedly required (according to the manufacturer) for a standard thin-section lung CT protocol to survey the same volume of interest with use of the same CT scanner: 3.4 mSv for a male patient and 4.4 mSv for a female patient.
The radiation dose required for the low-dose CT protocol investigated in the present study can also be compared with the estimated doses for chest spiral CT examinations reported in the literature, which range from 5 to 10 mSv (24). As recently pointed out by McNitt-Gray (25), although these values may be typical, there have been considerable efforts to reduce the radiation dose at thoracic CT examinations, many of which have reportedly yielded substantial radiation dose savings with acceptable trade-offs in image quality (14,26–29). Jung et al (30) found that low-dose volumetric spiral CT performed with 40 mA, with its acceptable image quality and similar radiation dose, may offer better information in the evaluation of airway diseases than does thin-section CT. Our study results confirm the lack of image degradation: We found the image quality at low-dose CT to be acceptable across all weight categories of the population scanned.
With regard to the usefulness of low-dose CT in workers exposed to asbestos, one should keep in mind the specificity of this population. First, CT is useful for the detection of the fibrotic consequences of asbestos exposure. These features are poorly depicted on chest radiographs, which are also limited in the detection of asbestos-related pleural diseases. Moreover, individuals who have combined exposure to asbestos and cigarette smoke are known to be at high risk for lung carcinoma. In addition, carcinogenesis is one of the stochastic detrimental effects of ionizing radiation; this is one reason for the efforts to reduce radiation doses, especially in screening examinations (25).
Our study had several limitations. The first limitation was related to the sample size: Our study group consisted of only 83 subjects. A larger population of subjects with wider variations in body weight might have yielded different results. The second limitation was the small proportion of subjects in whom CT depicted features of asbestosis: 22 (26%) of 83 subjects. These results confirm the low frequency with which CT features of asbestosis are seen in asymptomatic patients. Using thin-section CT to detect benign asbestos-related diseases, Aberle et al (21) detected asbestosis-related changes in one-third of patients who had neither clinical nor chest radiographic evidence of asbestosis. A third limitation was that a large proportion of the subjects with lung infiltration had CT features of lung fibrosis that were limited to the lower lung zones. This finding may lead to the conclusion that our study was more targeted at the evaluation of low-dose CT in depicting subtle abnormalities than at the evaluation of this modality in depicting advanced disease.
Because use of a 5-mm-thick reconstructed CT image may hamper the detection of fine abnormalities, one might question why a 4 x 1.0-mm collimation (ie, 1.0-mm collimated sections acquired with four detectors) was not chosen. The subsequent advantage would have been the potential to reconstruct 1-mm-thick images and thus avoid the scan interpretation errors made in two cases owing to partial volume effects. The motivation for the investigated protocol was related to the radiation dose delivered during the examinations: The alternative choice of a 4 x 1.0-mm collimation would have increased the dose delivered to the subjects by 20%. The last limitation was related to our comparison of volumetric low-dose spiral CT images of the entire thorax with sequential thin-section CT images of only the lungs, the latter images being limited to a mean number of eight sections per subject.
On the basis of the present investigation results, we conclude that low-dose multi–detector row spiral CT enables the accurate detection of asbestos-related disease and the concurrent search for malignant asbestos-related processes of the lungs and pleura. With use of the described protocol, negative low-dose CT results were found to be sufficient to exclude asbestos-related pleuropulmonary diseases.
|
FOOTNOTES |
|---|
Authors stated no financial relationship to disclose.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  T. Kubo, P.-J. P. Lin, W. Stiller, M. Takahashi, H.-U. Kauczor, Y. Ohno, and H. Hatabu Radiation Dose Reduction in Chest CT: A Review Am. J. Roentgenol., February 1, 2008; 190(2): 335 - 343. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Madani, V. De Maertelaer, J. Zanen, and P. A. Gevenois Pulmonary Emphysema: Radiation Dose and Section Thickness at Multidetector CT Quantification--Comparison with Macroscopic and Microscopic Morphometry Radiology, April 1, 2007; 243(1): 250 - 257. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Vierikko, R. Jarvenpaa, T. Autti, P. Oksa, M. Huuskonen, S. Kaleva, J. Laurikka, S. Kajander, K. Paakkola, S. Saarelainen, et al. Chest CT screening of asbestos-exposed workers: lung lesions and incidental findings Eur. Respir. J., January 1, 2007; 29(1): 78 - 84. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
