| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Department of Radiology, Kurume University School of Medicine, 67 Asahimachi,
Kurume, Fukuoka 830-0011, Japan.
2 Department of Radiology, Vancouver General Hospital, University of British
Columbia, Vancouver, BC, V5Z 1M9 Canada.
3 First Department of Internal Medicine, Kurume University School of Medicine,
Kurume, Fukuoka 830-0011, Japan.
Received July 4, 2004; accepted after revision October 15, 2004.
Address correspondence to H. Terasaki
(terra{at}med.kurume-u.ac.jp).
OBJECTIVE. The purpose of our study was to compare the high-resolution CT and pulmonary function test findings of smokers and nonsmokers with pulmonary sarcoidosis.
MATERIALS AND METHODS. Full inspiratory and expiratory high-resolution CT of the thorax and pulmonary function tests were performed in 46 patients (23 smokers and 23 lifelong nonsmokers) with histologically proven sarcoidosis. The median interval between high-resolution CT and pulmonary function tests was 8 days (range, 0–27 days). High-resolution CT findings were categorized into six patterns, and the overall extent of each pattern was scored independently (high-resolution CT score). Correlation between each high-resolution CT score with each pulmonary functional parameter was performed using Spearman's rank correlation and stepwise multiple regression analysis.
RESULTS. Air trapping on expiration (45/46 patients, 98%) and small nodules on inspiration (all 46 patients, 100%) were the most common findings. Smokers had a greater extent of emphysema than nonsmokers (p = 0.002). No significant difference was seen in the extent of air trapping, consolidation, ground-glass attenuation, reticular opacities, or small nodules between smokers and nonsmokers. On Spearman's rank correlation, the extent of air trapping negatively correlated with forced vital capacity in smokers (p < 0.05) but not in nonsmokers. The extent of small nodules negatively correlated with forced vital capacity and positively correlated with the ratio of forced expiratory volume in 1 sec to forced vital capacity in nonsmokers (p < 0.05, both) but not in smokers, respectively. On stepwise multiple regression analysis, the extent of air trapping on CT was independently associated with decreased forced vital capacity (p < 0.05), and cigarette smoking was the main determinant of decrease in maximum midexpiratory flow and forced expiratory flow at 50% of vital capacity (p < 0.01).
CONCLUSION. Cigarette smoking confounds the correlation between the CT and pulmonary function test findings in patients with sarcoidosis. Therefore, smoking history must be taken into account when correlating the extent of parenchymal sarcoidosis on CT with functional impairment.
Sarcoidosis is a systemic disease of unknown cause characterized by the presence of noncaseating granulomas. Pulmonary manifestations occur in 90% of patients [1]. Pulmonary parenchymal involvement can result in functional impairment, with reduction in lung volumes, impairment in gas exchange, and respiratory failure [2]. The presence and extent of parenchymal abnormalities on CT have been shown to correlate with the severity of functional impairment [3–5]. Although interstitial fibrosis typically results in restrictive lung function, several studies have shown that patients with sarcoidosis can have a combination of restrictive and obstructive lung function or mainly airway obstruction [6]. The clinical and functional abnormalities in these patients can resemble more common obstructive airway diseases, such as asthma and chronic bronchitis [6]. Air trapping is also commonly seen on expiratory high-resolution CT [7–9]. The presence of air trapping on high-resolution CT has been shown to correlate with functional parameters of air trapping and small-airway obstruction on pulmonary function tests [7, 9, 10]. However, limited information exists regarding the influence of cigarette smoking on the correlation between the extent of abnormalities on high-resolution CT and functional impairment in patients with sarcoidosis.
The purpose of this study was to compare the inspiratory and expiratory high-resolution CT and pulmonary function test findings of lifelong nonsmokers and smokers with sarcoidosis.
|
|
|
|
Patients
We retrospectively reviewed the medical records of all patients who had a
diagnosis of sarcoidosis and who had undergone high-resolution CT at our
hospital between 1998 and 2002. We selected all patients with the diagnosis of
sarcoidosis who had pulmonary function tests performed within 1 month of CT
and whose medical records revealed a clear description of a history of
smoking. Fifty-one of 137 patients who had undergone CT fulfilled these
criteria. Five of the 51 patients were excluded because the diagnosis had been
based only on clinical and radiologic findings. The 46 patients included in
the study had a diagnosis of sarcoidosis proven histologically on the basis of
specimens obtained at transbronchial lung biopsy (n = 37),
mediastinal node biopsy (n = 2), or extrathoracic site biopsy
(n =7). The 46 patients ranged in age from 20 to 75 years (median, 49
years); 17 were men and 29 were women. Twenty-three of 46 patients were
current smokers (cigarette use: median, 12.0 pack-years; range, 1.3–70.0
pack-years) and 23 patients were lifelong nonsmokers.
Our institutional review board does not require its approval or informed consent for review of a patient's records, files, and images.
High-Resolution CT Examination and Evaluation
High-resolution CT was performed on all patients using an X-Vigor scanner
(Toshiba Medical) or a HiSpeed Advantage scanner (GE Healthcare). The scans
were obtained using 1-mm collimation at 10-mm intervals for full inspiratory
scans and at 20-mm intervals for full expiratory scans (120 kVp and 200 mAs,
from the lung apices to below the costophrenic angles). The images were
reconstructed using a high-spatial-resolution algorithm and were photographed
at lung window settings (window level, –650 H; window width, 1,600
H).
High-resolution CT findings were categorized into six patterns [11, 12]: air trapping (on expiratory high-resolution CT, Fig. 1A, 1B), consolidation, emphysema (Fig. 2A, 2B), ground-glass attenuation, reticular opacities (including honeycombing, Fig. 3A, 3B), and small nodules (Fig. 4A, 4B). Air trapping was defined as persistent radiolucent areas within the lung on expiratory CT images. Emphysema was defined as the presence of localized areas of decreased attenuation with permeative destruction of the lung parenchyma and associated distortion of the pulmonary vasculature. Consolidation was defined as an area of homogeneous increased attenuation with obscuration of the vessels. Ground-glass attenuation was defined as hazy areas of increased attenuation in which the vessels remained visible. Reticular opacities were defined as interlacing linear opacities, including thickening of interlobular septa and honeycombing. Small nodules were defined as rounded opacities of less than 1 cm in diameter. The presence and extent of these six findings in each lobe were evaluated using a semiquantitative grading system by Hansell et al. [8] and were scored as follows: grade 0 (no involvement), grade 1 (1–25% of the lobe involved), grade 2 (26–50% of the lobe involved), grade 3 (51–75% of the lobe involved), and grade 4 (76–100% of the lobe involved). The total score of each pattern was determined by adding the scores in the six lobes (the lingula was considered a separate lobe). The high-resolution CT findings were reviewed independently by two experienced chest radiologists who were not aware of the clinical or pulmonary function test results.
|
|
|
|
Pulmonary Function Tests
The median interval between high-resolution CT and pulmonary function tests
was 8 days (range, 0–27 days). The spirometric measurements were made on
a pneumotachograph (Chestac-33, Chest Co.). At least three reproducible
measurements were performed on all patients, and the best of the three curves
was selected. The following parameters were recorded: forced vital capacity
(FVC); forced expiratory volume in 1 sec (FEV1); maximum mid
expiratory flow (MMEF); forced expiratory flow at 75%, 50%, and 25% of vital
capacity (FEF75, FEF50, and FEF25); and
forced expiratory volume in 1 sec–forced vital capacity ratio
(FEV1/FVC). All measured results were calculated as percentages of
values predicted from the patient's age, sex, and height
[13–16].
Statistical Analysis
Normally distributed data are described using mean ± SD, and skewed
data are summarized using median (interquartile range). The correlations
between each high-resolution CT score and results of pulmonary function tests
were respectively performed and analyzed using Spearman's rank correlation in
all patients, smokers, and nonsmokers. Stepwise multiple regression analysis
was performed to correlate the high-resolution CT scores of each of the six
patterns of abnormality and cigarette smoking with each of the pulmonary
function tests. Statistical comparisons of high-resolution CT findings,
pulmonary function test results, and the clinical parameters between smokers
and nonsmokers were analyzed using the Mann-Whitney U test and a
Student's t test. A p value of less than 0.05 was considered
statistically significant.
Results
Frequency of High-Resolution CT Findings
The frequency and extent of each high-resolution CT finding are summarized
in Table 1. Small nodules on
inspiratory scans (all 46 cases) and air trapping on expiratory scans (45/46
cases) were the most common findings. The extent of emphysema was
significantly greater in smokers than in nonsmokers (p = 0.002). No
other significant difference was seen between smokers and nonsmokers for any
of the other high-resolution CT findings.
|
Pulmonary Function Tests
The mean ± SD of each pulmonary function test result calculated as
percentages of predictive value are summarized in
Table 2. No significant
difference was seen in any of the pulmonary function test values between
smokers and nonsmokers. The percentages of predicted FEF25 and
FEF50 were low in both groups of patients.
|
Comparison of High-Resolution CT Scores and Pulmonary Function Test Correlations Between Smokers and Nonsmokers
To determine whether smoking affected the correlations between
high-resolution CT and pulmonary function test findings, we evaluated these
relations using Spearman's rank correlation in all patients, smokers, and
nonsmokers, respectively. The Spearman's rank correlation in all patients
revealed several significant correlations between pulmonary function tests and
high-resolution CT scores (Table
3). The extent of air trapping was negatively correlated with FVC
when all patients were assessed as a group (p = 0.014) and in all
smokers as a group (p = 0.036) but not in nonsmokers (p
=0.103). The extent of small nodules was negatively correlated with FVC
(p = 0.013), positively correlated with FEV1/FVC
(p =0.037), and positively correlated with FEF75
(p = 0.041) in nonsmokers but not in smokers. The extent of reticular
opacities was negatively correlated with FEV1/FVC (p =
0.007), MMEF (p =0.006), FEF50 (p = 0.012), and
FEF25 (p = 0.022) in smokers but not in nonsmokers. The
extent of ground-glass attenuation was also negatively correlated with
FEV1/FVC (p =0.002), MMEF (p =0.005),
FEF50 (p = 0.025), and FEF25 (p
=0.007) in smokers but not in nonsmokers.
|
Correlation Between High-Resolution CT Scores, Cigarette Smoking, and Pulmonary Function Tests
The results of stepwise multiple regression analysis performed to correlate
the high-resolution CT scores of each of the six patterns of abnormality and
cigarette smoking with each of the pulmonary function test results are
summarized in Table 4. The
extent of air trapping was independently associated with FVC (p =
0.041). The extent of emphysema was independently associated with
FEV1/FVC (p = 0.002) and FEF75 (p =
0.017), respectively. The extent of small nodules and cigarette smoking were
independently associated with MMEF (p =0.025 and p = 0.003,
respectively) and with FEF50 (p =0.025 and p =
0.005, respectively).
|
Discussion
Several studies have shown a correlation between the high-resolution CT and pulmonary function test findings in patients with sarcoidosis [3–5]. However, these studies did not take into account a patient's smoking habit as a potential cause of airflow obstruction. In this study, we showed statistically significant differences between smokers and nonsmokers. In the group of smokers, some of the high-resolution CT scores were associated with obstructive and restrictive lung function, whereas high-resolution CT scores in nonsmokers were associated only with restrictive lung function. Furthermore, smoking history and emphysema were shown to be the main determinants of airflow obstruction on multivariate analysis. These results suggest that a smoking habit affects the morphologic and functional correlations in pulmonary sarcoidosis.
Cigarette smoking is the cause of chronic bronchitis and chronic obstructive pulmonary disease. In severe chronic obstructive pulmonary disease, emphysema is usually associated with airflow obstruction [17]. Comparison of expiratory flow rates in smokers and nonsmokers showed an increase in airway obstruction in smokers, with a dose–response relationship between the degree of airflow obstruction and cigarette smoking [18]. In our study, a significant correlation was seen between airflow obstruction and severity of CT findings in smokers but not in nonsmokers. The extent of emphysema in smokers was variable and the range of cigarette smoking was wide (1.3–70.0 pack-years). However, both the extent of emphysema and the amount of cigarette smoking were shown to be important determinants of airflow obstruction on multivariate analysis.
Previous studies have shown that air trapping in pulmonary sarcoidosis is commonly seen on expiratory high-resolution CT and that it correlates with the severity of small-airway obstruction on pulmonary function tests [7, 9]. The air trapping in sarcoidosis is believed to be the result of small-airway obstruction by peribronchiolar or intraluminal granulomas [7]. Recent studies have shown that air trapping is positively correlated with the ratio of residual volume to total lung capacity [10, 19] and negatively correlated with MMEF [19]. However, these previous studies did not closely analyze the influence of a history of smoking. Unfortunately, static lung volumes are not routinely obtained at our institution in patients with sarcoidosis. Therefore, we could not assess static lung volumes and residual volume to total lung capacity ratio. However, in our study, air trapping on CT correlated with decreased FVC in smokers but did not show significant correlation with any of the pulmonary function test results in nonsmokers.
Pulmonary sarcoidosis usually produces a restrictive impairment with reduction in lung volume, and several studies have shown a significant but modest correlation between extent of parenchymal abnormalities on high-resolution CT and FVC [3–5, 20]. In our study, the extent of small nodules negatively correlated with FVC, positively correlated with FEV1/FVC, and positively correlated with FEF75 in nonsmokers but not in smokers. These results suggest that small nodules result in restrictive lung function. Pathologically, pulmonary nodules in sarcoidosis correspond to aggregates of granulomas [21], and widespread granulomas in sarcoidosis are characteristically associated with restrictive lung function and a decrease in diffusing capacity [22]. The presence of restriction in pulmonary functional measurement is commonly inferred from a decreased VC; however, VC may also be reduced in the presence of airflow obstruction [23]. Therefore, the relation between small nodules and FVC may include the influence of airflow obstruction.
The extent of reticular opacities and the extent of ground-glass attenuation correlated with airflow obstruction on univariate analysis. However, the correlations were significant only in smokers. Neither reticular opacities nor ground-glass attenuation correlated with any of the pulmonary function tests on multivariate analysis. Hansell et al. [8] showed that the extent of reticulation was independently negatively correlated with airflow obstruction. Our results partially agree with those of Hansell et al. but also show that smoking influences these results.
Our study has several limitations. It was retrospective, and it included only a small number of patients with severe fibrosis. Patients with extensive fibrosis evident radiologically and on CT show restrictive lung function [2, 3, 5].
We conclude that the common high-resolution CT findings of pulmonary sarcoidosis are small nodules on inspiration and air trapping on expiration, which were detected in almost all patients. The extent of small nodules correlated with restrictive lung function in nonsmokers but not in smokers. The extent of air trapping correlated with a decrease in FVC in smokers but not in nonsmokers. Pulmonary functional parameters of airflow obstruction correlated most closely with cigarette smoking in patients rather than sarcoidosis. A history of smoking must be taken into account when correlating the CT findings with pulmonary function tests.
References
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  J. P. Lynch III Idiopathic Pulmonary Fibrosis, Nonspecific Interstitial Pneumonia/Fibrosis, and Sarcoidosis ACCP Pulmonary Med Brd Rev, January 1, 2009; 25(0): 635 - 686. [Full Text] [PDF]
|
|
|
|
 |
|
 E. Mireles-Cabodevila, H. Sahi, C. Farver, T.-L. Mohammed, and D. A. Culver A Young Patient With a Minimal Smoking History Presents With Bullous Emphysema and Recurrent Pneumothorax Chest, July 1, 2007; 132(1): 338 - 343. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
