| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Department of Radiology, St. Marianna University School of Medicine, 2-16-1
Sugao, Miyamae-Ku, Kawasaki City, Kanagawa 216-8511, Japan.
2 Present address: Department of Radiology, Brigham and Women's Hospital,
Harvard Medical School, 75 Francis St., Boston, MA 02115.
3 Division of Respiratory and Infectious Diseases, Department of Internal
Medicine, St. Marianna University School of Medicine, Kanagawa, Japan.
Received July 4, 2007;
accepted after revision October 3, 2007.
Address correspondence to S. Matsuoka
(shin4114{at}mac.com).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. This study included 36 patients with COPD who underwent 64-MDCT. The entire lung volume with attenuation between –500 and –1,024 H was segmented as whole lung. The lung volume with attenuation between –500 and –950 H was segmented as limited lung, while the lung volume of less than –950 H was segmented as emphysema and eliminated. The relative volumes for limited lung (relative volumen–950) with attenuation values below thresholds (n) ranging from –850 to –950 H, and relative volume for whole lung (relative volume<n) were obtained on inspiratory and expiratory CT. Then the differences of relative volumes after expiration in whole lung (relative volume change<n) and limited lung (relative volume changen–950) were calculated. Patients were classified into two groups according to mean relative volume less than –950 H. Correlations between densitometry parameters and pulmonary function tests (PFTs) reflecting airway dysfunction were evaluated.
RESULTS. The highest correlation with PFTs was observed at the upper threshold of –860 H. In the moderate to severe emphysema group (relative volume<–950 > 15%), relative volume change860–950 significantly correlated with the results of PFTs, whereas no significant correlations were seen between relative volume change<–860 and PFTs. In the minimal or mild emphysema group (inspiratory relative volume<–950 < 15%), all densitometric parameters correlated with PFTs.
CONCLUSION. The densitometric parameter of relative volume change calculated on paired inspiratory and expiratory MDCT using the threshold of –860 H in limited lung correlated closely with airway dysfunction in COPD regardless of the degree of emphysema.
Keywords: air trapping • chronic obstructive pulmonary disease • CT • emphysema • lung
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Meanwhile, quantification of small airways disease using CT is less well developed than that of emphysema because recent CT techniques have not yet allowed direct morphologic assessment of small airways. However, some studies have shown that the densitometric parameters of lung calculated on paired inspiratory and expiratory CT scans allowed indirect evaluation of airway obstruction in several obstructive lung diseases [8–12]. These quantitative methods are based on the detection of air trapping. Air trapping reflects the retention of excess gas in all or part of the lung and is detected as decreased attenuation on expiratory CT as compared with the corresponding inspiratory images. However, in the case of COPD, the area of decreased attenuation includes not only air trapping but also emphysema. Thus, this method would be influenced by the extent of emphysema.
Recently, Matsuoka et al. [13] investigated the percentage change of relative area with attenuation values from –900 to –950 H between inspiratory and expiratory CT scans in patients with COPD. The lower threshold of –950 H was adopted to eliminate the influence of emphysema, and the upper threshold value of –900 H was adopted because the relative area in lung less than –900 H on expira-tory CT scans correlated to the degree of air trapping [8]. They found that the change in relative area with attenuation values from –900 to –950 H after expiration decreased with the deterioration of the pulmonary function tests (PFTs), reflecting airway obstruction, and that change correlated more closely with airway obstruction than change of relative area with attenuation values less than –900 H, including emphysema. Those authors concluded that their method using paired inspira-tory and expiratory CT could be useful for the quantitative evaluation of air trapping in COPD without the influence of emphysema.
Unfortunately, their study had several limitations: First, the validity of using a threshold of –900 H for the evaluation of air trapping was not confirmed. In fact, the appropriate attenuation threshold value for the quantification of air trapping on paired inspiratory and expiratory CT has not been clarified. Second, they obtained densitometric parameters from only six slices using a 2D analysis. Thus, the misregistration of CT slices between inspiration and expiration might have influenced their results. In addition, both emphysema and air trapping are heterogeneously distributed throughout the lung in COPD. Only six slices would not reflect morphologic and functional abnormalities in the whole lung. Those authors should have assessed whole-lung volumetric data. Third, most of their subjects had severe emphysema. The adaptability of this method to the patient without severe emphysema was not confirmed. Thus, the validity of this method is expected to improve by solving these limitations.
The first aim of our study was to determine the attenuation threshold value for the detection and quantification of air trapping using paired inspiratory and expiratory volumetric MDCT scans. The second aim of this study was to assess whether the densitometric parameter can be used for the quantification of airway dysfunction in COPD without being influenced by the degree of emphysema.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
MDCT
All patients were scanned with a 64-MDCT scanner (Aquilion-64, Toshiba
Medical Systems). The scanner was calibrated regularly with air and a water
phantom to allow reliable measurements. CT was performed during deep
inspiratory and expiratory breath-holding with the patient in the supine
position. Every patient was carefully instructed how to breathe before the
study and again right before the scanning. MDCT parameters for both scans were
as follows: collimation, 0.5 mm; 120 kV; 200 mA; gantry rotation time, 0.5
second; beam pitch, 53/64. All images were reconstructed using a standard
reconstruction algorithm with a slice thickness of 1 mm and a reconstruction
interval of 0.5 mm.
Quantitative Assessment of Lung Attenuation
The reconstruction images were transferred to a workstation (Ziostation,
Ziosoft). This workstation uses a semiautomatic threshold technique to isolate
lung volume from other tissues and structures using CT attenuation values of
–500 to –1,024 H; the volume of the entire lung was calculated by
summing the voxels in those attenuation values and was defined as the whole
lung. Minimal user intervention by one radiologist was required to exclude
nonlung structures that satisfied the threshold criteria, such as the trachea
and large bronchi near the hilum. Then the lung volumes with attenuation
values lower than thresholds ranging from –850 to –950 H
(–850, –860, –870, –880, –890, –900,
–910, –920, –930, and –950 H) were obtained on both
inspiratory and expiratory scans. The percentage of the lung volume with
attenuation values lower than each threshold value for the whole lung were
calculated as relative volume of the whole lung on both inspiratory and
expiratory scans (relative volume<n [%] =
volume with attenuation values less than each threshold/volume with
attenuation value of –500 to –1,024 H; for inspir atory relative
volume<n and expiratory relative
volume<n, n = each selected attenuation
threshold value between –850 and –950 H) (Fig.
1A,
1B,
1C,
1D,
1E,
1F).
|
|
|
|
|
|
To evaluate the change of relative volume after expiration, the difference between the relative volumes on expiratory CT and inspiratory CT in the whole lung (relative volume change<n) and the limited lung (relative volume changen–950) were calculated using the following formulas: relative volume change<n (%) = expiratory relative volume<n – inspiratory relative volume<n; relative volume changen–950 (%) = expiratory relative volumen–950 – inspiratory relative volumen–950; n = each selected attenuation threshold value between –850 and –930 H.
Pulmonary Function Tests
PFTs were performed within 2 weeks of obtaining thin-section CT scans.
PFTs, including spirometry and measurement of diffusing capacity for carbon
monoxide (DLco), were performed. Forced expiratory volume in 1
second (FEV1) and forced vital capacity (FVC) were measured
according to standard techniques, and the ratio of FEV1 to the
forced vital capacity (FEV1/FVC) and mid expiratory phase of the
forced expiratory flow (FEF25–75%) were obtained. The lung
volume subdivisions of functional residual capacity (FRC), residual volume
(RV), and total lung capacity (TLC) were measured with the helium dilution
method. Values for each PFT, except for RV/TLC and FEV1/FVC, were
expressed as percentages of predicted values according to the prediction
equations described previously
[15]. DLco was
measured by the single-breath method, and the predicted values for
DLco were determined as described previously
[15].
Statistical Analysis
To obtain the attenuation threshold value for the detection and
quantification of air trapping, we calculated Spearman's correlation
coefficients between each relative volume
changen–950 and the results of PFTs
(FEV1, FEV1/FVC, FEF25–75%, and
RV/TLC). Next, all patients were classified into two groups according to the
extent of emphysema obtained on the basis of the mean value of inspiratory
relative volume<–950. Using that threshold value
(n*) obtained from this study, Spearman's correlation coefficients
between the expiratory relative volume<n*, expiratory
relative volumen*–950, relative volume
change<n*, or relative volume
changen*–950 and the results of PFTs
(FEV1, FEV1/FVC, FEF25–75%, FVC,
RV/TLC, and DLco) in both groups were obtained. Comparisons of
expiratory relative volume<n*,
expiratory relative volumen*–950,
relative volume change<n*, or
relative volume changen*–950 between the
moderate to severe emphysema group and the minimal or mild emphysema group
were done using the Wilcoxon's signed rank test. All statistical analyses were
performed using JMP 5.0.1 software (SAS Institute). Data are expressed as mean
± SD. For all statistical analyses, a p value of less than
0.05 was considered significant.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
|
Densitometric Parameters in Whole and Limited Lung
Mean value of the inspiratory relative volume<–950 was
15.1% ± 14.3%. According to the mean value of the inspiratory relative
volume<–950, all patients were divided into two groups as
follows: moderate to severe emphysema group (inspiratory relative
volume<–950 > 15%): 14 patients (13 men and one woman;
mean age, 72.1 years; mean inspiratory relative
volume<–950, 23.7% ± 12.3%), minimal or mild
emphysema group (inspiratory relative volume<–950 <
15%): 22 patients (18 men and four women; mean age, 70.4 years; mean
inspiratory relative volume<–950, 5.0% ± 4.6%). The
results of densitometric parameters are shown in
Table 3. Significant
differences were seen in expiratory relative volume<–860,
expiratory relative volume860–950, relative volume
change<–860, or relative volume
change860–950 between the moderate to severe emphysema group
and the minimal or mild emphysema group (p < 0.001,
respectively).
|
In the moderate to severe emphysema group, relative volume change860–950 significantly correlated with results of PFTs that associate with airway dysfunction (r = –0.76, p = 0.002 for FEV1; r = –0.64, p = 0.013 for FEV1/FVC; r = –0.61, p = 0.02 for FEF25–75%; and r = 0.79, p < 0.001 for RV/TLC). No significant correlations were seen between relative volume change<–860, expiratory relative volume860–950, or expiratory relative volume<–860 and results of PFTs.
In the minimal or mild emphysema group, relative volume change860–950 significantly correlated with results of PFTs that associate with airway dysfunction (r = –0.56, p = 0.007 for FEV1; r = –0.56, p = 0.006 for FEV1/FVC; r = –0.55, p = 0.008 for FEF25–75%; and r = 0.50, p = 0.001 for RV/TLC). The correlation coefficients between relative volume change<–860 and the results of PFTs were the same as the correlation between relative volume change860–950 and the results of PFTs that associate with airway dysfunction. No significant correlations were seen between relative volume change860–950, relative volume change<–860, expiratory relative volume860–950, or the expiratory relative volume<–860 and the DLco in the minimal or mild emphysema group. The correlation between the densito-metric parameters and the results of PFTs are shown in Table 4.
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Although emphysema is now detectable with the use of CT, it is difficult to quantify airway obstruction and air trapping in patients with COPD. However, several researchers have tried to quantify the degree of air trapping using densitometric techniques on expiratory or paired inspiratory and expiratory CT in various obstructive lung diseases [8–12]. It has been reported that the area of air trapping does not show an increase in CT attenuation and remains more radiolucent than the surrounding normal pulmonary tissue [16–18]. Consequently, the degree of the change of the lung attenuation value after expiration has been quantified using some densitometric parameters such as the relative area below a certain threshold value or the ratio of mean lung attenuation value of inspiration and expiration. However, these densitometric parameters were calculated from lung attenuation including values less than –950 H that reflect the extent of emphysema. Moreover, the relative area with attenuation values less than –950 H is not appreciably changed after expiration as compared with the relative area of decreased attenuation with values more than –950 H [12, 13, 19].
Therefore, the exclusion of the pixels less than –950 H on both inspiratory and expira-tory CT is desirable for the quantification of air trapping without influence of the extent of emphysema. Actually, in the moderate to severe emphysema group, relative volume change860–950 was the only parameter that related to the result of PFTs associated with airway obstruction and air trapping. Meanwhile, no significant correlations were seen between the results of PFTs reflecting airway dysfunction and relative volume change<–860, which is the parameter including voxels with attenuation values less than –950 H regarding the extent of emphysema. In addition, no significant correlations were found between airway dysfunction and the densitometric parameters obtained on only expiratory CT such as expiratory relative volume<–860 or expiratory relative volume860–950 in the emphysema-dominant group. These results could support the necessity of using paired inspira-tory and expiratory CT for the quantification of airway dysfunction in COPD with emphysema.
Although the value of –950 H is recognized as an acceptable cutoff for segmentation of emphysema, emphysema also exists in areas having lung attenuation greater than –950 H [20]. Furthermore, Gevenois et al. [21] showed that the threshold of –910 H on expiratory CT scans correlated better with the macroscopic assessment of emphysema. Therefore, the emphysematous lesions cannot be completely excluded using the cutoff value of –950 H, and they contribute to airway dysfunction to some degree. However, in our study and in a previous study [13], densitometric parameters with attenuation less than –950 H are not strongly related to airway dysfunction. Therefore, although airflow limitation in COPD is a dynamic phenomenon related to both small airways disease [1–3] and an increase in lung compliance due to emphysematous lung destruction [4], our results suggest that the extent of emphysema is not the major cause of airflow limitation in COPD.
The threshold value that regulates the extent and degree of air trapping also has not sufficiently been clarified. In our study, the upper attenuation threshold value of –860 H had the highest correlation with results of FEF25–75% and RV/TLC, which indicate airway obstruction and air trapping. However, we cannot explain why the attenuation value of –860 H is the best threshold to quantify the extent of air trapping. Using lung volume data from all subjects in our study, the frequency distribution of pixels in the lung on both inspiration and expiration shows that the percentages of pixels at the attenuation value of –860 H on both inspiration and expiration are equivalent, and at more than –860 H, the percentage of pixels on expira-tory CT is greater than that on inspiratory CT (Fig. 3). Thus, differences in relative volume between inspiration and expiration could be detected effectively at the attenuation value of less than –860 H. Future evaluation of the relationship between the change in relative volume after expiration and the physiologic bases is required.
|
Meanwhile, in our study, although the upper attenuation threshold value of –860 H had the highest correlation with results of PFTs, most correlations with the results of PFTs are similar to each other, especially between –850 and –880 H. Actually, in both the moderate to severe emphysema group and the minimal or mild emphysema group, the correlation between the densitometry parameters with attenuation values of –850, –870, and –880 H were not quite different from the result of using a threshold value of –860 H (data not shown). Therefore, the optimal threshold could vary between –850 and –880 H because of variable conditions such as calibration of the CT scanner.
In this study, no significant correlation was seen between all densitometric parameters and DLco in the moderate to severe emphysema group. In contrast, the reduction of DLco correlated with the increased values of all densitometric parameters in the minimal or mild emphysema group. The decrease of DLco is probably the result of loss of alveolar surface area, such as occurs in emphysema [22, 23]. However, DLco alone is not specific for the diagnosis of emphysema [24]. Many studies have also found that measurement of DLco has a weak correlation with the pathologic assessment of emphysema [25]. At the same time, airflow obstruction, especially that induced by airway dysfunction, may enhance functional inhomogeneities and impairs DLco [26]. In the minimal or mild emphysema group, reduction of DLco might correlate with functional inhomogeneities due to airway obstruction. Hence, it could be reasonable to find correlations between relative volume change, which reflects air dysfunction, and the reduction of DLco in the minimal or mild emphysema group.
Several studies have addressed the ability of 3D volumetric data to accurately quantify the extent and severity of emphysema [27–29]. In previous studies, comparing only a few single inspiratory and expiratory image pairs, the misregistrations of CT slices between inspiration and expiration might be due to disturbances of accurate evaluation of airway dysfunction. Because MDCT has the major advantage that the entire thorax is imaged during a single breath-hold, the disadvantage of using single-detector CT has been overcome. Meanwhile, expiratory CT does expose patients to additional radiation, and multidetector technology can further increase the delivered dose. Therefore, further research is needed to optimize radiation dose for the quantification of airway dysfunction.
In the past few years, much attention has been paid to therapeutic agents that specifically address airflow obstruction in patients with COPD [30–32]. Although emphysematous lesions are irreversible, there is a good chance for treatment and prevention of airway obstruction in patients with COPD. Quantitative CT analysis has been used to assess the relative efficacy of drugs on small airway hyperreactivity and regional air trapping [30]. The evaluation of drug efficacy in the peripheral airways is important, and progress toward specific treatment for COPD might be accelerated by moving beyond the measurement of airflow limitation to the precise diagnosis of the specific targets responsible for the airflow limitation. The efficacy of various drugs might be best assessed using the paired inspiratory and expiratory CT analysis in the limited lung, especially in COPD patients, because this method allows a more accurate assessment of peripheral airway obstruction without the influence of the extent of emphysema.
Our study has several limitations. First, the number of patients was relatively small. To prove the validity of our method for the quantification of air trapping, this method should be applied to a prospective larger set of patients with various degrees of emphysema. Second, the concept and definition of air trapping are still confusing and not agreed upon. On CT images, it is generally accepted that the area of air trapping does not show a significant increase in CT attenuation. According to this concept, to quantify air trapping, we must match each lung area on inspiratory and expiratory CT and compare the CT density between inspiratory and expira-tory scans. However, we did not perform pixel-by-pixel comparison. Thus, some areas having a value less than –950 H on inspiration CT may have the value of –860 to –950 H on expiration CT, which may affect the results to some degree. Moreover, we decided on the optimal upper threshold value of –860 H depending on correlations with the results of FEF25–75% and RV/TLC. However, in the strict sense, FEF25–75% and RV/TLC are not entirely representative of small airway obstruction and air trapping. Therefore, relative volume change860–950 might reflect not only a limitation in airflow caused by small airways disease but also other pathophysiologic conditions. Third, CT densitometry is influenced by the level of inspiration during CT [33]. A spirometrically controlled CT technique has been developed, offering the opportunity to obtain CT scans at defined levels of inspiration [34, 35].
In conclusion, using paired inspiratory and expiratory MDCT, the change in relative lung volume with attenuation values from –860 to –950 H after expiration correlated closely with results of PFTs, reflecting the severity of airway dysfunction. Furthermore, the result of the correlations of densitometry parameters and PFTs indicated that the relative volume change860–950 was the only parameter that could reflect airway dysfunction in both the moderate to severe emphysema group and the minimal or mild emphysema group. Thus, the densitometry parameter of relative volume change860–950 can be used for the quantification of air trapping in patients with COPD regardless of the degree of emphysema.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  M. Akira, K. Toyokawa, Y. Inoue, and T. Arai Quantitative CT in Chronic Obstructive Pulmonary Disease: Inspiratory and Expiratory Assessment Am. J. Roentgenol., January 1, 2009; 192(1): 267 - 272. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) CT in this study. Percentages of pixels at
attenuation of –860 H on both inspiratory and expiratory CT are
equivalent; at more than –860 H, percentage of pixels on expiratory CT
is greater than on inspiratory CT.