| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Thoracic Imaging |
1 From the Department of Diagnostic Radiology, Ajou University Medical Center, Suwon, South Korea (K.J.P.); and the Departments of Radiology (C.J.B.) and Pulmonary Medicine (J.L.C.), University of California Medical Center, San Diego. Received January 2, 1998; revision requested March 27; final revision received August 11; accepted November 5. Supported in part by National Institutes of Health grants R29-HL48854 and MO1 RR00827. Address reprint requests to C.J.B., Department of Radiology, Auckland Hospital, Park Road, Grafton, Auckland, New Zealand.
 |
Abstract |
|---|
 TOP Abstract IntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
MATERIALS AND METHODS: Thoracic helical CT scans were obtained in 60 adult patients (35 with no visual evidence of emphysema and 25 with emphysema). The lungs were reconstructed as a 3D model on a commercial workstation, with a threshold of -600 HU. By analysis of histograms, the proportions of lung volumes with attenuation values below -950, -910, and -900 HU were measured, in addition to mean lung attenuation. These values were compared with lung densitometric results obtained from 2D CT images, visual emphysema scores, and data from pulmonary function tests.
RESULTS: Quantitation of emphysema with 3D reconstruction was efficient and accurate. Correlation was good among densitometric quantitation with 3D analysis, that obtained with 2D analysis (r = 0.98–0.99), and visual scoring (r = 0.74–0.82). Correlation was reasonable between 3D densitometric quantitation and the diffusing capacity of the lungs for carbon monoxide (DLCO) (r = -0.57 to -0.64), total lung capacity (r = 0.62–0.71), forced expiratory volume in 1 second (FEV1) (r = -0.57 to -0.60), and the ratio of FEV1 to forced vital capacity (FVC) (r = -0.75 to -0.82). The visual CT quantitation of emphysema correlated best with DLCO (r = -0.82) and FEV1/FVC (r = -0.89).
CONCLUSION: Lung densitometry with 3D reconstruction of helical CT data is a fast and accurate method for quantifying emphysema.
Index terms: Computed tomography (CT), three-dimensional, 60.12117 • Emphysema, pulmonary, 60.751, 60.7511, 60.7512, 60.7513 • Lung, CT, 60.751 • Lung, density, 60.751
|
Introduction |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
First applied to two-dimensional (2D) CT sections, the "density mask" technique was shown to represent accurately the morphologic extent of emphysema (11), but this method is time-consuming to perform on multiple sections. Various automated and semiautomated techniques subsequently have been developed to increase the speed with which emphysema in the whole lung can be quantitated (5–7,12–14), but such software has not been universally available.
With the advent of helical CT technology, volumetric data can now be acquired and manipulated with the use of three-dimensional (3D) lung reconstruction on software that is available on most commercial scanners. To our knowledge, measurements of lung attenuation obtained from 3D reconstruction have not been compared with those from 2D analysis or with visual assessment. The objectives of this study were to compare lung densitometric quantitation of emphysema obtained with the use of a 3D lung model reconstructed from helical CT data with densitometric measurements from analysis of 2D CT images, visual emphysema scores, and data from pulmonary function tests.
|
MATERIALS AND METHODS |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Patient data are summarized in Table 1. The patients included 37 men and 23 women whose ages ranged from 18 to 90 years (mean, 53.9 years). Clinical indications for the CT examinations were staging or metastatic work-up of malignant disease in 22 patients; further evaluation of a chest radiographic abnormality in 17 patients; assessment for pulmonary symptoms, including dyspnea and hemoptysis, in nine patients; evaluation of emphysema in eight patients; and suspected mediastinal abnormality in four patients.
|
1-antitrypsin deficiency. The group with no clinical or CT evidence of emphysema included 10 smokers and 22 nonsmokers. Smoking histories were unavailable in five other patients.
CT Technique
CT examinations were performed with a commercially available scanner (HiSpeed Advantage; GE Medical Systems, Milwaukee, Wis) in the helical mode without intravascular contrast material. With the patient in the supine position, scans were obtained during full inspiration, with the following parameters: 120 kVp, 250–280 mA, 7-mm (56 patients) or 5-mm (four patients) collimation, and a pitch of 1.5. At our institution, 7-mm collimation is used for most thoracic helical CT scans. Thinner collimation was used to investigate hemoptysis or for better evaluation of the tracheobronchial tree. Inspiratory scans were selected because of their reported greater accuracy in depicting anatomic emphysema (15). Scan volumes extended from the thoracic inlet to the lung base and were acquired in two breath-hold periods. Lung images were reconstructed with the lung algorithm.
Densitometry
The 3D models of the lungs were reconstructed with the Advantage Windows 3D Analysis Package (GE Medical Systems). The time for reconstruction of the 3D model was reduced considerably by not using the high-resolution mode that manipulates a voxel matrix of 512 x 512 pixels. Threshold limits of -600 to -1,024 HU were applied to exclude soft tissue surrounding the lung and large vessels within the lung. The 3D model was viewed as a shaded surface display at multiple angles to ensure that the model was valid (Fig 1). The trachea, main-stem bronchi, and gastrointestinal structures were selectively removed from the model.
|
|
|
With this program, 2D axial images can also be analyzed with the same pixel attenuation range as that chosen for the voxel attenuation values on the 3D lung model. We measured the cross-sectional area of the lung, the mean lung attenuation, and the percentage of the cross-sectional area with lower attenuation (lower attenuation area) at thresholds of -950, -910, and -900 HU. The 2D densitometry was performed on axial images selected from three levels: the top of the aortic arch, through the carina, and at the level of the left atrium. The mean values for the lower attenuation area for these three sections were calculated for each patient to provide the mean 2D estimate of the extent of emphysema.
Visual Emphysema Score
On hard-copy CT images photographed at a window width of 1,500 HU and centered at a level of -600 HU, visual emphysema scores for all axial sections were determined for each patient. Emphysema was identified as areas of hypovascular low attenuation and was graded with a five-point scale based on the percentage of lung involved: 0, no emphysema; 1, up to 25% of lung parenchyma involved; 2, between 26% and 50% of lung parenchyma involved; 3, between 51% and 75% of lung parenchyma involved; and 4, between 76% and 100% of lung parenchyma involved. Grades for the axial images of each lung were added and divided by the number of images evaluated to yield emphysema scores that ranged from 0 to 4.
Pulmonary Function Tests
The results of pulmonary function tests were available for 26 patients. The interval between pulmonary function tests and CT examinations was less than 8 months for all but two patients. In one patient with panlobular emphysema, the interval was 3
years, and in a patient with centrilobular emphysema, the interval was 15 months. The measures of pulmonary function tested included the forced expiratory volume in 1 second (FEV1), forced vital capacity (FVC), total lung capacity (TLC), and diffusing capacity of the lung for carbon monoxide (DLCO). Values for FEV1, TLC, and DLCO were expressed as the percentages of predicted values (percentage of predicted).
Spirometry was performed in accord with the 1987 American Thoracic Society standards (16), and single-breath DLCO measurements were obtained in accord with the 1987 American Thoracic Society recommendations (17). DLCO values were adjusted for the most recently measured hemoglobin concentration in accord with the American Thoracic Society standards (17). TLC values were measured with a variable-pressure body plethysmograph. All pulmonary function tests were performed with a pneumotachograph-based system (Master Lab models; Erich Jaeger, Würzburg, Germany). The predicted normal values used were those of Crapo et al (18) for spirometric data, those of Miller et al (19) for DLCO, and those of Goldman and Becklake (20) for TLC.
Statistical Analysis
The Student t test was used to compare the two groups of patients (with and without visual evidence of emphysema) with regard to the mean values of (a) lung attenuation; (b) lower attenuation volumes at thresholds of -950, -910, and -900 HU; and (c) lower attenuation areas at thresholds of -950, -910, and -900 HU.
Pearson correlation was used to compare data from pulmonary function tests with densitometric values obtained from both 3D and 2D analyses that included mean attenuation and lower attenuation volume at -950, -910, and -900 HU. Pearson correlation was also used to compare measures of lower attenuation volume at -950, -910, and -900 HU obtained with 3D densitometry with those obtained with 2D densitometry. The relationships between visual emphysema scores and measurements of lung attenuation were evaluated with Spearman rank correlation. A commercially available computer program (PC!INFO; Retriever Data Systems, Seattle, Wash) was used for statistical analysis.
|
RESULTS |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Because the 2D sections were selected after postprocessing removal of the esophagus and tracheobronchial structures, 2D analysis also took approximately 10 minutes. The time for visual inspection of all sections in each patient varied from 10 to 15 minutes.
A summary of the densitometric data obtained from 3D models and 2D axial images, visual emphysema scores, and pulmonary function tests is shown in Table 1. Mean values of lung attenuation and of lower attenuation volume and lower attenuation area at thresholds of -950, -910, and -900 HU were significantly larger for patients with visual evidence of emphysema than for those in whom no visual evidence of emphysema was seen on the CT scans (P < .01 for all parameters).
Close correlation was observed between lung attenuation measurements obtained with the 3D model and those from the 2D axial images, with correlation coefficients (r) of 0.98 (-950 HU), 0.98 (-910 HU), and 0.99 (-900 HU). Attenuation measurements obtained from both the 3D model and from the 2D axial sections correlated well with visual emphysema scores, with the closest correlation demonstrated for lower attenuation volume at -950 HU and lower attenuation area at -950 HU (Table 2, Fig 2).
|
|
|
|
|
|
|
|
DISCUSSION |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
However, voxel attenuation values can now be measured without the need to transfer data to other computer systems. With software that is available on most modern commercial scanners, 3D lung models can be generated with volumetric data acquired in the helical mode. The lungs are easily separated from soft-tissue structures and from structures with lower attenuation, such as the trachea, main-stem bronchi, and esophagus, with minimal postprocessing on the shaded surface display. With this model, the proportion of voxels with attenuation values within the range of emphysema is readily determined by moving a boundary line to the defined threshold on a histogram of attenuation values that represent the whole lung.
Reproducibility of lung attenuation measurements is known to be affected by patient factors and by variation in CT scanning techniques (21,22). Enhancement with contrast media increases lung attenuation (23), and a change in section thickness may also affect measured Hounsfield numbers (23,24). Kilovoltage, positioning of the patient in the scanner, the reconstruction algorithm, and the type of scanner all have been described as affecting the reproducibility of CT densitometric measurements, although with modern scanners, these factors are considered to have minimal influence (24).
One of the most important factors that influence the reproducibility of lung attenuation measurements is the respiratory status of the patient. Spirometric gating has been used to control for depth of inspiration (14,25), but with helical CT, only one or two breath holds are necessary to scan the entire lungs, which greatly diminishes the problems of motion artifacts and respiratory misregistration.
We found that acquiring CT data in the helical mode and using the reconstructed 3D model improved our efficiency in quantitating emphysema and provided values that were very similar to those obtained from analysis of 2D images. Because of the well-known heterogeneous distribution of emphysema, we expected 3D analysis of the whole lung to reflect the extent of emphysema more accurately than evaluation of three axial sections but instead found very close correlation between the two techniques. Lung attenuation measurements obtained from voxels with the use of the 3D lung model correlated well with those obtained from 2D axial images and also with visual assessment.
In our study, visual inspection scores correlated more closely with DLCO (percentage of predicted) and FEV1/FVC than did either the 2D or 3D analyses. Müller et al (11) also found visual scoring of emphysema to be comparable in accuracy to the 2D density mask method correlated with pathologic scores. However, visual inspection is known to be affected by observer experience, by interobserver and intraobserver variability (26), and by technical factors such as window settings. Correlation between each of the lower attenuation volume thresholds and pulmonary function measurements was similar to lower attenuation areas measured by others with the density mask (2).
To distinguish patients with emphysema from those with no emphysema, we used visual inspection of CT images; this was based on the recognized superiority of CT scans, compared with pulmonary function tests, for use in identifying emphysema (3,21,27–29). Differences in mean pulmonary function data between these groups helped to confirm the visual impression that patients in the group with emphysema were physiologically different from those with no visual evidence of emphysema. The six patients with 1-antitrypsin deficiency likely contribute to the marked difference in mean values for lower attenuation volume, lower attenuation area, and pulmonary function data between these two groups; however, an additional 19 patients in the group with emphysema had visual CT evidence of centrilobular emphysema (centriacinar emphysema).
Some overlap in pulmonary function existed between the group with and the group without emphysema, particularly in measurements of TLC (percentage of predicted) and FEV1 (percentage of predicted). The broad range noted in patients with no visual CT evidence of emphysema is likely caused by airways disease not detected on CT images and emphasizes the limitations of pulmonary function tests in quantitating emphysema (3,27,28).
The low attenuation thresholds that have been used most widely to identify emphysema on conventional 10-mm-thick CT sections are -900 or -910 HU (2,5,6,11,13). In some studies, CT scans were obtained after intravenous administration of contrast material. Using thin-section CT scans at 1-mm collimation without intravenous administration of contrast material, Gevenois et al (30,31) found that a lower attenuation threshold of -950 HU correlated best with morphologic emphysema. No intravascular administration of contrast material was used in our study, in which we also found that the lower attenuation volume threshold of -950 HU obtained with both 3D and 2D analysis correlated marginally better with visual grades, DLCO (percentage of predicted), and FEV1 (percentage of predicted) than did either -910 or -900 HU but correlated less closely with TLC (percentage of predicted) and FEV1/FVC.
With the recent development of surgical and pharmacologic methods for treating patients with emphysema, accurate quantification of lung destruction in this disease is becoming increasingly relevant. Lung-volume reduction surgery is being performed more frequently to improve pulmonary function and exercise tolerance in patients with severe emphysema, and CT quantification of emphysema has been used both for patient selection and to evaluate surgical outcome in these patients (6–10). For purposes of patient selection or surgical planning, the distribution of emphysema is better appreciated by viewing the axial CT images; however, emphysema distribution can be quantitated by reconstructing the upper and lower halves of the lungs separately in three dimensions (9,10,32).
Antielastolytic agents are being evaluated for potential roles in preventing progressive lung destruction in patients with panlobular emphysema caused by 1-antitrypsin deficiency (33), but the feasibility of studies to assess the efficacy of these new therapeutic options is markedly limited by the relative insensitivity of tools for assessing outcome (34). Efficient quantitation of emphysema in these patients may help both in the selection of patients for drug therapy and in the assessment of the therapeutic results.
In conclusion, we found that lung attenuation measurements obtained from 3D lung models that were reconstructed from volumetric helical CT data accurately distinguished patients with emphysema from those with normal lungs, correlated closely with both 2D analysis and visual quantitation, and showed significant correlation with measurements of DLCO (percentage of predicted), TLC (percentage of predicted), FEV1 (percentage of predicted), and FEV1/FVC. Lung densitometry derived from 3D lung models was easily and efficiently performed and provides a widely available alternative to 2D methods for quantifying emphysema.
|
Footnotes |
|---|
Author contributions: Guarantor of integrity of entire study, C.J.B.; study concepts and design, C.J.B.; literature research, K.J.P.; clinical studies, K.J.P., J.L.C.; data acquisition, K.J.P., C.J.B., J.L.C.; data analysis, K.J.P., C.J.B.; statistical analysis, K.J.P.; manuscript preparation, K.J.P.; manuscript editing, K.J.P., C.J.B.; manuscript review, J.L.C.
|
References |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  P. J. Friedman Imaging Studies in Emphysema Proceedings of the ATS, May 1, 2008; 5(4): 494 - 500. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Matsuoka, Y. Kurihara, K. Yagihashi, M. Hoshino, N. Watanabe, and Y. Nakajima Quantitative Assessment of Air Trapping in Chronic Obstructive Pulmonary Disease Using Inspiratory and Expiratory Volumetric MDCT Am. J. Roentgenol., March 1, 2008; 190(3): 762 - 769. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Cazzola, W. MacNee, F. J. Martinez, K. F. Rabe, L. G. Franciosi, P. J. Barnes, V. Brusasco, P. S. Burge, P. M. A. Calverley, B. R. Celli, et al. Outcomes for COPD pharmacological trials: from lung function to biomarkers Eur. Respir. J., February 1, 2008; 31(2): 416 - 469. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. H. Groenewegen, D. S. Postma, W. C. J. Hop, P. L. M. L. Wielders, N. J. J. Schlosser, E. F. M. Wouters, and for the COSMIC Study Group Increased Systemic Inflammation Is a Risk Factor for COPD Exacerbations Chest, February 1, 2008; 133(2): 350 - 357. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. A. Gietema, A. M. Schilham, B. van Ginneken, R. J. van Klaveren, J. W. J. Lammers, and M. Prokop Monitoring of Smoking-induced Emphysema with CT in a Lung Cancer Screening Setting: Detection of Real Increase in Extent of Emphysema Radiology, September 1, 2007; 244(3): 890 - 897. [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]
|
|
|
|
|
|
D. A. Torigian, W. B. Gefter, J. D. Affuso, K. Emami, and L. Dougherty Application of an Optical Flow Method to Inspiratory and Expiratory Lung MDCT to Assess Regional Air Trapping: A Feasibility Study Am. J. Roentgenol., March 1, 2007; 188(3): W276 - W280. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Ueda, M. Jinbo, T.-S. Li, T. Yagi, K. Suga, and K. Hamano Computed Tomography-Diagnosed Emphysema, Not Airway Obstruction, Is Associated with the Prognostic Outcome of Early-Stage Lung Cancer. Clin. Cancer Res., November 15, 2006; 12(22): 6730 - 6736. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Ueda, M. Sudoh, M. Jinbo, T.-S. Li, K. Suga, and K. Hamano Physiological rehabilitation after video-assisted lung lobectomy for cancer: a prospective study of measuring daily exercise and oxygenation capacity. Eur. J. Cardiothorac. Surg., September 1, 2006; 30(3): 533 - 537. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. B. Fain, S. R. Panth, M. D. Evans, A. L. Wentland, J. H. Holmes, F. R. Korosec, M. J. O'Brien, H. Fountaine, and T. M. Grist Early Emphysematous Changes in Asymptomatic Smokers: Detection with 3He MR Imaging. Radiology, June 1, 2006; 239(3): 875 - 883. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Sudoh, K. Ueda, Y. Kaneda, J. Mitsutaka, T.-S. Li, K. Suga, Y. Kawakami, and K. Hamano Breath-hold single-photon emission tomography and computed tomography for predicting residual pulmonary function in patients with lung cancer J. Thorac. Cardiovasc. Surg., May 1, 2006; 131(5): 994 - 1001. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Ueda, Y. Kaneda, M. Sudo, J. Mitsutaka, T.-S. Li, K. Suga, N. Tanaka, and K. Hamano Quantitative Computed Tomography Versus Spirometry in Predicting Air Leak Duration After Major Lung Resection for Cancer Ann. Thorac. Surg., November 1, 2005; 80(5): 1853 - 1858. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Ueda, Y. Kaneda, M. Sudoh, J. Mitsutaka, N. Tanaka, K. Suga, and K. Hamano Role of Quantitative CT in Predicting Hypoxemia and Complications After Lung Lobectomy for Cancer, With Special Reference to Area of Emphysema Chest, November 1, 2005; 128(5): 3500 - 3506. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Chaouat, A.-S. Bugnet, N. Kadaoui, R. Schott, I. Enache, A. Ducolone, M. Ehrhart, R. Kessler, and E. Weitzenblum Severe Pulmonary Hypertension and Chronic Obstructive Pulmonary Disease Am. J. Respir. Crit. Care Med., July 15, 2005; 172(2): 189 - 194. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Ueda, Y. Kaneda, M. Sudou, M. Jinbo, T.-S. Li, K. Suga, N. Tanaka, and K. Hamano Prediction of hypoxemia after lung resection surgery Interactive CardioVascular and Thoracic Surgery, April 1, 2005; 4(2): 85 - 89. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C Hill and E J R van Beek MRI of the chest: present and future Imaging, October 1, 2004; 16(1): 61 - 70. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J.D. Newell Jr, J.C. Hogg, and G.L. Snider Report of a workshop: quantitative computed tomography scanning in longitudinal studies of emphysema Eur. Respir. J., May 1, 2004; 23(5): 769 - 775. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Boschetto, M. Miniati, D. Miotto, F. Braccioni, E. De Rosa, I. Bononi, A. Papi, M. Saetta, L.M. Fabbri, and C.E. Mapp Predominant emphysema phenotype in chronic obstructive pulmonary disease patients Eur. Respir. J., March 1, 2003; 21(3): 450 - 454. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N L Muller and H Coxson Chronic obstructive pulmonary disease * 4: Imaging the lungs in patients with chronic obstructive pulmonary disease Thorax, November 1, 2002; 57(11): 982 - 985. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Madani, C. Keyzer, and P.A. Gevenois Quantitative computed tomography assessment of lung structure and function in pulmonary emphysema Eur. Respir. J., October 1, 2001; 18(4): 720 - 730. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G.R. Ferretti, I. Bricault, and M. Coulomb Virtual tools for imaging of the thorax Eur. Respir. J., August 1, 2001; 18(2): 381 - 392. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. R. Flaherty, E. A. Kazerooni, J. L. Curtis, M. Iannettoni, L. Lange, M. A. Schork, and F. J. Martinez Short-term and Long-term Outcomes After Bilateral Lung Volume Reduction Surgery : Prediction by Quantitative CT Chest, May 1, 2001; 119(5): 1337 - 1346. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. D. Maki, W. B. Gefter, and A. Alavi Recent Advances in Pulmonary Imaging Chest, November 1, 1999; 116(5): 1388 - 1402. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||
