| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Thoracic Imaging |
1 From the Departments of Diagnostic Radiology (G.C.O., P.L.K., L.J.Z.) and Medicine (J.C.M.H., B.L., K.W.T.T.), University of Hong Kong, Queen Mary Hospital, Rm 405, Block K, Pokfulam Rd, Hong Kong, Special Administrative Region, China; Department of Radiology, Queen Mary Hospital, Hong Kong, Special Administrative Region, China (W.C.Y.); and Department of Radiology, Vancouver General Hospital, University of British Columbia, Canada (N.L.M., S.N.). Received June 2, 2003; revision requested June 23; final revision received September 8; accepted September 16. Address correspondence to G.C.O. (e-mail: cgcooi@hkucc.hku.hk).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Serial thin-section CT scans in 30 patients (17 men, aged 42.5 years ± 12.2 [SD]) with SARS were reviewed by two radiologists together for predominant patterns of lung abnormalities: ground-glass opacities, ground-glass opacities with superimposed linear opacities, consolidation, reticular pattern, and mixed pattern (consolidation, ground-glass opacities, and reticular pattern). Scans were classified according to duration in weeks after symptom onset. Longitudinal changes of specific abnormalities were documented in 17 patients with serial scans obtained during 3 weeks. Each lung was divided into three zones; each zone was evaluated for percentage of lung involvement. Summation of scores from all six lung zones provided overall CT score (maximal CT score, 24).
RESULTS: Median CT scores increased from 1 in the 1st week to 12.5 in the 2nd week. Ground-glass opacities with or without smooth interlobular septal thickening and consolidation were predominant patterns found during the 1st week. Ground-glass opacities with superimposed irregular reticular opacities, mixed pattern, and reticular opacities were noted from the 2nd week and peaked at or after the 4th week. After the 4th week, 12 (55%) of 22 patients had irregular linear opacities with or without associated ground-glass opacities and CT scores greater than 5; five of these patients had bronchial dilatation. When specific opacities were analyzed in 17 patients, consolidation generally resolved completely (n = 4) or to minimal residual opacities; six (55%) of 11 patients with ground-glass opacities had substantial residual disease (CT scores > 5) on final scans.
CONCLUSION: There is a temporal pattern of lung abnormalities at thin-section CT in SARS. Predominant findings at presentation are ground-glass opacities and consolidation. Reticulation is evident after the 2nd week and persists in half of all patients evaluated after 4 weeks. Long-term follow-up is required to determine whether the reticulation represents irreversible fibrosis.
© RSNA, 2004
Index terms: Lung, CT, 60.12118 • Lung, infection, 60.21 • Severe acute respiratory syndrome (SARS)
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
The computed tomographic (CT) findings at presentation usually include unilateral or bilateral ground-glass opacities or areas of consolidation (5,6,8,9). The utility of thin-section CT in the documentation of parenchymal abnormalities in SARS when chest radiographs appear normal or show only questionable abnormalities has also been established (8,9). In their initial description of the epidemiologic, clinical, and radiologic features of the first 10 SARS patients, including the index case for Hong Kong, Tsang et al (5) alluded to radiologic features that suggest the development of fibrosis. Thin-section CT evidence of fibrosis has also recently been reported in SARS patients who have been discharged after treatment (9). The aim of this study, therefore, was to evaluate lung abnormalities on serial thin-section CT scans in patients with SARS during the acute and convalescent periods of the illness.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Thin-Section CT Scans
Thin-section CT scans were obtained with the patients in the supine position, and scanning was performed at end inspiration. One unit (HiSpeed Advantage; GE Medical Systems, Milwaukee, Wis) was used in 28 patients with the following parameters: 1.0-mm section thickness, 10-mm gap, 1- or 2-second scanning time per section, 120 kV, and 150 mA. Another unit (Lightspeed plus; GE Medical Systems) was used in one patient with the following parameters: 1.3-mm section thickness, 10-mm gap, 1-second scanning time per section, 120 kV, and 180 mA. A third unit (Toshiba Asteion; Toshiba, Tokyo, Japan) was used in one patient with the following parameters: 1.0-mm section thickness, 10-mm gap, 1-second scanning time per section, 120 kV, and 180 mA. All follow-up scans were obtained by using the same scanner as that used to obtain the initial scans. In the patient who underwent thin-section CT with 1.3-mm section thickness, the initial scans were spiral. Spiral scans were obtained with the same scanner and the following parameters: 5-mm collimation, 120 kV, 181 mA, pitch of 0.75, scanning interval of 2.5 mm, and table feed of 11.25 mm. The thin-section CT findings at presentation in five patients in the current study were previously reported (9). In 14 of 30 patients, two CT scans were obtained; in 10 patients, three scans; in five patients, four scans; and in one patient, five scans. The images were photographed at lung (window width, 1,000–1,500 HU; window level, -700 HU) and mediastinal (window width, 350 HU; window level, 35–40 HU) settings.
Image Interpretation
Two experienced radiologists at each center (G.C.O. and P.L.K.; N.L.M. and S.N.) reviewed the thin-section CT images on hard copies and reached a decision in consensus. Three of these radiologists (N.L.M., G.C.O., S.N.) had 19, 8, and 6 years of experience in thoracic radiology, respectively, and one (P.L.K.) had 11 years of experience in radiology. The observers categorized the predominant pattern on CT scans as ground-glass opacification (hazy areas of increased attenuation without obscuration of the underlying vessels), consolidation (homogeneous opacification of the parenchyma with obscuration of the underlying vessels), reticular pattern, mixed pattern (combination of consolidation, ground glass opacities, and reticular opacities in the presence of architectural distortion), and honeycomb pattern. Presence of smooth interlobular septal thickening, intralobular lines (lacy pattern within the lobule), and irregular lines and interfaces with architectural distortion superimposed on ground-glass opacities were also noted. Reticular pattern consisted of either coarse linear or curvilinear opacities or fine subpleural reticulation without substantial ground-glass opacities. Presence of irregular or corkscrew bronchial or bronchiolar dilatation associated with any of the previously mentioned findings was noted. On the scans, presence of mediastinal lymphadenopathy (defined as a lymph node 
1 cm in short-axis diameter), pneumothorax, pneumomediastinum, and pleural effusion was also noted.
The distribution of opacities was also noted as being predominantly subpleural (involving mainly the peripheral one-third of the lung), random (without predilection for subpleural or central regions), or diffuse (continuous involvement without respect to lung segments). After evaluation, the scans were categorized according to the time between the date of onset of symptoms and the date on which the scan was obtained at 1, 2, 3, 4, and longer than 4 weeks after onset of symptoms. Patients with initial scans obtained during the first 2 weeks of the illness and with follow-up scans obtained after a minimum of 3 weeks were selected for further analysis to determine longitudinal changes of main lung abnormalities that were detected.
The extent of disease at thin-section CT was also evaluated. Each lung was divided into three lung zones: upper (above the carina), middle (below the carina up to the inferior pulmonary vein), and lower (below the inferior pulmonary vein) zones. Each lung zone (total of six lung zones) was assigned a score that was based on the following: score 0, 0% involvement; score 1, less than 25% involvement; score 2, 25% to less than 50% involvement; score 3, 50% to less than 75% involvement; and score 4, 75% or greater involvement. Summation of scores provided overall lung involvement (maximal CT score for both lungs was 24).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Mean alanine aminotransferase and aspartate aminotransaminase levels, total white blood cell count, and lymphocyte count at presentation for the 30 patients were 49.82 U/L ± 44.2 (normal range, 6–53 U/L), 67.5 U/L ± 66.3 (normal range, 13–33 U/L), 5.968 x 109/L ± 1.73 (normal range, 4–11 x 109/L), and 0.70 x 109/L ± 0.3 (normal range, 1.5–400.0 x 109/L), respectively. There was only one patient with a premorbid lung condition; this patient had childhood asthma. Four others had acute myelogenous leukemia treated with bone marrow transplantation, hypertension, resected renal cell carcinoma, and thyrotoxicosis controlled with medication. The other 25 patients were healthy and did not have a medical history of note prior to contraction of SARS. Coronavirus infection was confirmed in all patients from Hong Kong by means of a fourfold or greater increase in anti–SARS-CoV antibody (n = 28) and/or reverse transcriptase polymerase chain reaction positive for SARS-CoV RNA detected in nasopharyngeal aspirates and/or stool specimens (n = 24). In the two patients from Vancouver, coronavirus was isolated from respiratory secretions (n = 1) or fecal material (n = 1).
Thin-Section CT Findings
There was a marked increase in extent of disease during the 2nd week of illness, and the median CT score was 12.5 (range, 2–24). After that time, the extent decreased slowly to a median CT score of 8 (range, 0–24) during week 5, and this decrease reflected the presence of residual disease (Fig 1). The predominant patterns of abnormality changed over time (Fig 2). Within the 1st week after onset of symptoms, the main abnormalities included ground-glass opacities (10 [56%] of 18) and consolidation (eight [44%] of 18). The frequency of ground-glass opacities was highest in the 2nd week (10 [62%] of 16), and it decreased thereafter (Fig 2). Ground-glass opacities alone or with superimposed interlobular septal thickening were most commonly found in the 1st week after onset of symptoms (Figs 3, 4). At the 2nd week, smooth intralobular lines and other superimposed reticular opacities were noted in association with ground-glass opacities (Figs 3, 4). Three of 10 scans with ground-glass opacities in the 2nd week showed a combination of superimposed septal and reticular opacities. Irregular linear opacities and interfaces superimposed on ground-glass opacities accounted for 90% (nine of 10) of all ground-glass opacification found after 4 weeks (Figs 3, 4). The prevalence of consolidation as the predominant abnormality was highest within the 1st week of symptoms and decreased thereafter (Fig 2).
|
|
|
|
|
|
In 24 patients, final thin-section CT scans were obtained at 4 weeks or longer after onset of symptoms. Of these patients, 12 had substantial residual disease (score > 5) and a mean score of 10.7 ± 5.4 (median, 9; range, 5–24), seven had minimal residual opacities (score 
3) and a mean score of 2.21 ± 0.86 (median, 2.5; range, 1–3), and five had normal scans. In those with substantial residual disease, a reticular pattern was noted in five patients, a mixed pattern in two, consolidation in one, and ground-glass opacities with superimposed irregular linear opacities in four. Bronchial and/or bronchiolar dilatation was a feature in five patients. In the seven patients with minimal residual opacities, a subpleural reticular pattern was found in five patients, and small areas of ground-glass opacities with irregular linear opacities were found in two patients. There was no zonal predominance in the distribution of the residual changes.
Longitudinal Changes of Specific Thin-Section CT Features
The Table summarizes the longitudinal changes in 17 patients in whom initial scans were obtained during the first 2 weeks of illness, and final scans were obtained at least 3 weeks later. In these patients, the initial CT scans demonstrated predominant ground-glass opacities without reticulation (eight scans), ground-glass opacities with reticulation (two scans), and consolidation (seven scans).
|
|
|
|
|
|
|
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
The extent of parenchymal abnormalities in our cohort of SARS patients increased markedly between the 1st and 2nd weeks after the onset of symptoms. This was followed by a slow decline in scores that reflected residual lung abnormalities at 4 weeks in 50% (12 of 24) of patients. In concordance with initial reports about lung changes in SARS, ground-glass opacities and consolidation were the predominant abnormalities found on thin-section CT scans in our cohort (5–8). These were predominantly subpleural (13 [72%] of 18) in the initial week of illness and became more diffuse (five [50%] of 10) at the 3rd week, after which opacities became distributed either in the subpleural regions of the lungs or were diffuse.
Ground-glass opacities with or without septal thickening or reticular opacities were the commonest pattern during the first 2 weeks of illness. This finding is similar to that of Wong et al (8), who evaluated thin-section CT manifestations at presentation in patients exposed to or with SARS. In that study, the main thin-section CT features in 36 of the 40 patients who had symptoms of SARS or who were clinically suspected of having SARS were predominantly ground-glass opacities alone or in combination with consolidation. Interlobular septal thickening and intralobular lines were present in 22 and 26, respectively, of their 40 patients. In our study, the number of scans with ground-glass opacities was highest in the 2nd week (10 [63%] of 16) and decreased in the following 2 weeks, with a small upsurge in numbers after the 4th week. This upsurge was accounted for by an increase in ground-glass opacities with superimposed irregular linear opacities and interfaces.
A predominant pattern of consolidation was most common in the first 2 weeks of the illness and was not seen after the 4th week. The areas of consolidation generally either resolved completely or to small areas of fine reticulation. The reticular pattern associated with architectural distortion and bronchial or bronchiolar dilatation was noted to increase progressively from the 3rd week. In addition to this finding, 12 (50%) of 24 final scans obtained 4 weeks or longer after onset of symptoms showed substantial residual disease, consisting predominantly of a reticular pattern or of ground-glass opacities with a superimposed reticular pattern. Bronchial or bronchiolar dilatation was an associated feature in five (42%) of these 12 scans. When specific opacities in a subgroup of 17 patients were analyzed, consolidation in general either resolved completely or decreased to minimal areas of residual opacities. However, six (55%) of 11 scans with ground-glass opacities showed a substantial amount of residual disease after 3 weeks or longer.
These findings indicate that a proportion of patients develop persistent lung changes that may suggest the development of fibrosis, although ground-glass opacities are largely reversible in SARS (12). Antonio et al (9) studied 24 SARS patients discharged from the hospital after treatment who had undergone thin-section CT; they observed that 15 (62%) of the patients had evidence of fibrosis. These were described to be parenchymal bands, traction bronchiectasis, and irregular interfaces. However, since the natural history of SARS is as yet uncharted, it may be too early to label the lung abnormalities found as irreversible fibrosis. Features that were reversible were bandlike consolidation and parenchymal bands, terms that generally described the mixed pattern found in our patients. The parenchymal bands probably represent subsegmental atelectasis that was reversed with resolution of inflammation with reexpansion of alveoli. Similarly, resolution of interstitial edema and cellular infiltration, particularly of interlobular septa, could also explain resolution of subpleural curvilinear lines noted.
In a recent publication (13) about postmortem results in six fatal cases of SARS, with mean disease duration of 16.8 days ± 5.3 (median, 17 days; range, 8–24 days), fibrosis was not observed in the lungs, although diffuse alveolar damage was noted in cases with duration of illness longer than 10 days. Macrophages, however, featured strongly as the main neutrophil infiltrate in the alveoli and interstitium of these fatal cases, even in those with early disease, which suggests that proinflammatory cytokines released by macrophages may underlie the pathogenesis of SARS (13). The macrophage and cellular infiltrates may also explain the predominant pattern of ground-glass opacities noted in the lungs at thin-section CT in our patients.
Thin-section CT features of SARS are not specific and could be found in other airspace diseases such as bronchiolitis obliterans organizing pneumonia, particularly in the initial phase of the illness when the ground-glass opacities and consolidation are primarily subpleural (14,15). When diffuse change develops in SARS, the thin-section CT and radiographic appearances may be indistinguishable from acute respiratory distress syndrome (5,16). Similarly the presence of smooth interlobular septal thickening superimposed on extensive ground-glass opacities can also be found in other conditions—primarily alveolar proteinosis, pulmonary edema, and hemorrhage—and also in other types of bacterial or viral pneumonia, such as herpes simplex and cytomegalovirus infections (17–20).
There were no patients with peribronchiolar consolidation, bronchopneumonia, and nodules—particularly centrilobular and tree-in-bud opacities—in our cohort. The absence of these thin-section CT features may serve to distinguish SARS from other types of atypical pneumonia, particularly pneumonia of viral or Mycoplasma origins, and infectious bronchiolitis (17–21). The predominance of ground-glass opacities over consolidation may allow differentiation of SARS from bacterial pneumonia, which characteristically manifests as consolidation in a segmental or lobular distribution with ground-glass opacities found just around the consolidation rather than with the extensive involvement seen in SARS (18,21). Our study findings have also further reinforced the absence of mediastinal lymph nodes or substantial effusions in SARS, which may be used as additional helpful diagnostic signs.
It should be noted, however, that our patients probably represent those at the severe end of the disease spectrum, primarily hospitalized patients, who also have developed appropriate indications or complications to justify serial evaluations with thin-section CT. Hence, natural selection bias may have been introduced into the study design. The absence of scans obtained in the prone position is another limitation of the study, particularly with reference to bandlike opacities in the lung bases, which could have represented atelectasis that may have been observed as reversible on scans obtained in the prone position had they been obtained. Other limitations include the nonuniform scanning intervals among all patients caused by the retrospective nature of this study. However, because SARS is highly infectious, it would have been inappropriate and against institutional infection-control policy to systematically evaluate all SARS patients. Thin-section CT should be reserved for patients in whom a diagnosis is uncertain or for patients with clinical deterioration in whom further evaluation of lung abnormalities is required.
We conclude that there is a temporal pattern of lung abnormalities in SARS, and the abnormalities increased considerably at the 2nd week of illness and were observed at thin-section CT; these abnormalities resulted in substantial residual disease in 50% (12 of 24) of patients at 4 weeks or longer after disease onset. Of these 24 patients, 21% (five) had normal scans and 29% (seven) had minimal residual abnormalities. The residual abnormalities most commonly consisted of a reticular pattern with or without associated ground-glass opacities and bronchial or bronchiolar dilatation. Long-term follow-up with thin-section CT and concomitant functional studies are required to determine the long-term pulmonary sequelae of SARS.
|
FOOTNOTES |
|---|
Author contributions: Guarantors of integrity of entire study, G.C.O., K.W.T.T.; study concepts and design, G.C.O., P.L.K., N.L.M., K.W.T.T.; literature research, G.C.O., P.L.K.; clinical studies, J.C.M.H., B.L., K.W.T.T.; data acquisition, W.C.Y., J.C.M.H., B.L.; data analysis/interpretation, G.C.O., P.L.K., L.J.Z., S.N., N.L.M.; statistical analysis, G.C.O.; manuscript preparation, G.C.O., P.L.K., N.L.M.; manuscript definition of intellectual content, G.C.O., P.L.K., N.L.M., K.W.T.T.; manuscript editing, G.C.O., P.L.K., N.L.M.; manuscript revision/review and final version approval, all authors
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  Hui Li, Shaohui Du, Lina Yang, Yangyan Chen, Wei Huang, Rong Zhang, Yinghai Cui, Jun Yang, Dongfeng Chen, Yiwei Li, et al. Rapid pulmonary fibrosis induced by acute lung injury via a lipopolysaccharide three-hit regimen Innate Immunity, June 1, 2009; 15(3): 143 - 154. [Abstract] [PDF]
|
|
|
|
 |
|
 W. C. W. Chu, A. M. Li, A. W. H. Ng, H.-k. So, W. W. M. Lam, K. L. Lo, M.-c. A. Yeung, Y.-s. Yau, W.-k. Chiu, C.-w. Leung, et al. Thin-Section CT 12 Months After the Diagnosis of Severe Acute Respiratory Syndrome in Pediatric Patients. Am. J. Roentgenol., June 1, 2006; 186(6): 1707 - 1714. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y.-C. Chang, C.-J. Yu, S.-C. Chang, J. R. Galvin, H.-M. Liu, C.-H. Hsiao, P.-H. Kuo, K.-Y. Chen, T. J. Franks, K.-M. Huang, et al. Pulmonary Sequelae in Convalescent Patients after Severe Acute Respiratory Syndrome: Evaluation with Thin-Section CT Radiology, September 1, 2005; 236(3): 1067 - 1075. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
