| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Thoracic Imaging |
1 From the Departments of Radiology (R.H.B., M.N., H.H., P.M.B.), Pulmonary Medicine (D.F.K., A.E.), and Anesthesia (S.H.L.), Beth Israel Deaconess Medical Center, Harvard Medical School, 330 Brookline Avenue, Boston, MA 02215. Received February 19, 2004; revision requested April 29; revision received June 22; accepted July 27. Address correspondence to P.M.B. (e-mail: pboisell@caregroup.harvard.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Institutional review board approval was obtained, and informed consent was not needed. Retrospective review was performed of all patients with a CT diagnosis of TBM in a 10-month period (n = 34) who underwent evaluation of airway disease by means of three different sequences at multi–detector row CT: end inspiration, dynamic expiration, and end expiration (the latter was performed only at the levels of the aortic arch, carina, and bronchus intermedius). Fourteen patients (11 men, three women; age range, 19–79 years) who had comparable images obtained with all three sequences at any of these three levels were included in the study. The degree of airway collapse was measured by two thoracic radiologists in consensus by calculating the percentage change in the area of the airway between inspiratory and expiratory scanning. Statistical analysis was performed by using the paired t test.
RESULTS: Dynamic expiratory CT elicited a significantly greater degree of airway collapse than end-expiratory CT at all three levels (P < .005). The mean percentages of airway collapse at each of the three levels were as follows: aortic arch, 53.9% with dynamic expiration versus 35.7% with end expiration (P = .0046); carina, 53.6% with dynamic expiration versus 30.9% with end expiration (P < .0001); and bronchus intermedius, 57.5% with dynamic expiration versus 28.6% with end expiration (P = .0022).
CONCLUSION: Dynamic expiratory CT elicits a significantly greater degree of airway collapse than standard end-expiratory CT in patients with TBM.
© RSNA, 2005
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Of importance, the acquired form of TBM has been increasingly recognized as a relatively common cause of chronic respiratory symptoms in adults. For example, bronchoscopy series results have revealed an incidence of TBM varying from 5% to 23% (5,6), and results of a recent study showed TBM to be the cause of chronic cough in 14% of nonsmoking patients (7). Other symptoms associated with this disorder include dyspnea and recurrent infections (6,8).
It is generally accepted that the diagnosis of TBM can be established by the identification of a reduction in cross-sectional area of the airway greater than or equal to 50% at expiration or during coughing (6,9–11). Although bronchoscopy has been considered the standard diagnostic method, its intrinsic invasiveness and other relative limitations have led to an increasing use of expiratory computed tomography (CT) as a major tool for evaluation of suspected cases and for planning of therapeutic procedures. To date, two different CT methods have been employed for the diagnosis of TBM: end-expiratory imaging (at suspended end expiration) and dynamic expiratory imaging (during the expiratory phase of respiration) (11–15). The recent advent of faster multi–detector row CT scanners, which can image the entire central airways in less than 5 seconds, has facilitated the ability to perform dynamic expiratory imaging (13). Indeed, results of recent CT studies performed with dynamic expiratory imaging have shown an accuracy comparable to that of bronchoscopy for diagnosing TBM (14,15).
It is known that positive pleural (intrathoracic) pressure worsens the dynamic collapse of the airway in patients with TBM (8,10). Because dynamic expiration produces a higher level of intrathoracic-extratracheal pressure than does end expiration, it should theoretically elicit a greater degree of central airway collapse. To date, however, these methods have not been compared in the same patient population and, thus, their differences have not been quantified. The purpose of this study was to compare dynamic expiratory and end-expiratory CT for depicting central airway collapse in patients with acquired TBM.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Because it was necessary to have accurate measurements of the airway lumen at the same anatomic level (aortic arch, carina, or bronchus intermedius) during the three different respiratory maneuvers for comparison purposes, the images for all 34 patients were initially reviewed by a radiologist (R.H.B., 5 years of experience with thoracic CT) for the following exclusion criteria: (a) absence of comparable images from all three sequences on at least one anatomic level (n = 6); (b) excessive respiratory motion during one or more sequences at the levels of analysis, leading to image degradation and precluding precise airway measurement (n = 2); (c) presence of a stent at the levels of analysis, which would prevent the ability to determine the degree of airway collapse (n = 4); (d) failure of the patient to cooperate with inspiratory or expiratory breathing instructions (n = 3); (e) signs of malacia restricted to areas of the airway other than the levels of analysis (n = 4); or (f) absence of end-expiratory images because of protocol error (n = 1).
The final study population included 14 patients (11 men, three women), with a mean age of 53 years (range, 19–79 years). The various underlying respiratory or systemic diseases in these patients, along with presenting signs and symptoms and bronchoscopic results, are listed in Table 1. As shown in this Table, of the 14 patients in our study cohort, 12 (86%) had symptoms that could be associated with TBM, while the remaining two were asymptomatic but had a known risk factor for TBM (relapsing polychondritis) (16,17). Although bronchoscopic results are listed in this Table, we did not attempt to correlate CT and bronchoscopic findings because not all patients underwent bronchoscopy, and specific bronchoscopic maneuvers to assess for this disorder were not consistently performed.
|
End-inspiratory scanning was performed first in all cases (170 mAs, 120 kVp, 2.5-mm collimation, high speed mode, pitch equivalent of 1.5). For this sequence, patients were instructed to take a deep breath and hold it. After end-inspiratory scanning, patients were subsequently coached with instructions for the dynamic expiratory component of the study (40 mAs, 120 kVp, 2.5-mm collimation, high speed mode, pitch equivalent of 1.5). For this sequence, patients were instructed to take a deep breath and to blow it out during the image acquisition, which was coordinated to begin with the onset of the patient’s forced expiratory effort. Following this acquisition, patients were then coached with instructions for the transverse end-expiratory scanning (40 mAs, 120 kVp, 1.25-mm collimation), which was performed at three selected levels: the aortic arch, the carina, and approximately 2 cm below the carina (the latter corresponds to the level of the bronchus intermedius). For each image acquisition of this sequence, the patients were instructed to take a deep breath, blow it all the way out, and hold it.
To minimize radiation exposure for this three-phase study, a low-dose technique (40 mAs) was employed for both dynamic expiratory and end-expiratory sequences. The use of a low-dose technique for expiratory imaging of the central airways has been validated by Zhang et al (15). An injection of iodinated contrast material was administered in six (43%) of 14 patients. Although intravenous contrast material is not necessary in evaluation for airway malacia, it was administered in these cases because of suspicion of thoracic malignancy or other extratracheal mediastinal abnormality at either clinical or radiographic evaluation.
Image Review and Measurements
The CT images in the 14 patients in the study cohort were reviewed by a thoracic radiologist (R.H.B.) who identified corresponding images at the same anatomic levels (aortic arch, carina, and bronchus intermedius) for the three respiratory sequences. For each patient, only those levels with precise anatomic correlation among the images from the three sequences were analyzed (eg, if a subject had precise anatomic correlation at only the levels of the carina and the bronchus intermedius, only these two levels were evaluated for this patient).
To ensure that identical levels were selected for all three sequences, we relied both on vascular landmarks (such as the aortic arch, great vessels, and pulmonary arteries) and on the relative distance of each level to the carina, since vascular and airway structures can move at slightly different lengths during the phases of the respiratory cycle. The levels of the aortic arch, carina, and bronchus intermedius (the latter located approximately 2 cm below the carina) were chosen for analysis because these components of the airway are located in a plane relatively perpendicular to the transverse plane of image acquisition, which allowed more precise assessment of the cross-sectional area of these airway components compared with airways that run obliquely to the transverse plane.
A computerized tracing tool that is available as part of our hospital’s picture archiving and communications system (Centricity, version 2.0; GE Medical Systems) was used to hand trace the inner wall of the airway at the desired level and to calculate the cross-sectional area of the airway in square millimeters. This was performed in a similar manner for images from all three phases of respiration. For this analysis, images were viewed with bone window settings (width, 3077 HU; level, 570 HU) because we have previously found that this window display provides the best delineation of the interface between the tracheal wall and the tracheal lumen (Fig 1). For a given patient, all three sequences at each level were analyzed sequentially, without randomization, to ensure proper correlation of the images for each level.
|
Statistical Analysis
For each patient, the percentage of airway collapse achieved at each level of analysis was calculated by comparing the reduction in cross-sectional area between images from each expiratory sequence and the end-inspiratory sequence.
In the group of patients evaluated at each particular airway level, the mean percentage of cross-sectional area reduction achieved at dynamic expiratory imaging was compared with the mean percentage achieved at end-expiratory imaging. Statistical analysis was performed by using a paired two-tailed t test of difference versus no difference (JMP, version 4.0.4; SAS Institute, Cary, NC) to compare these mean percentages. A P value of less than .05 was considered to show a statistically significant difference.
By using the criterion of 50% or more reduction in cross-sectional luminal area as a positive result for the diagnosis of TBM, we also calculated the percentage of false-negative diagnoses for each series (dynamic expiratory and end expiratory). For example, if a patient had 50% or more reduction in luminal area at end-expiratory imaging but less than 50% reduction at dynamic expiratory imaging, the case was considered to have a false-negative diagnosis with the dynamic expiratory sequence.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
|
|
|
|
|
|
|
|
|
|
|
|
With the previously established CT criterion of 50% or more reduction of the cross-sectional area as diagnostic of TBM, the use of only end-expiratory imaging would have resulted in a false-negative diagnosis of TBM in five (56%) of nine patients at the level of the aortic arch, in six (60%) of 10 patients at the level of the carina, and in six (86%) of seven patients at the level of the bronchus intermedius. Conversely, we did not observe any false-negative results for TBM by using solely dynamic expiratory imaging.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
A variety of noninvasive imaging tools have been applied to the diagnosis of TBM during the past few decades. Historically, cinefluoroscopy has been used in the evaluation of this disease, with a degree of collapse equal to or greater than 50% considered positive for diagnosis (1). However, several factors limit the use of this imaging method, including a relatively poor display of anatomic detail (especially in larger patients), the subjective and operator-dependent nature of the procedure, the inability to simultaneously display the anteroposterior and lateral walls of the trachea, and a tendency to underestimate the degree of collapse in comparison with bronchoscopy (13,18). With the advent of CT, it became possible to obtain an objective and reproducible display of the airways and adjacent structures, including the ability to quantitatively measure the degree of collapse (13). With the use of helical scanners, image quality was further improved but single–detector row scanners were still limited by a relatively long acquisition time. The ability to obtain dynamic cross-sectional images of the airways became possible with the development of electron-beam CT, also referred to as "cine CT." This technique allowed the acquisition of images with a temporal resolution of only 50–100 msec per section, so real-time scanning during dynamic breathing and coughing was possible (6,9–11,19). This method was limited, however, by its relatively low spatial resolution and restricted availability.
More recently, the advent of multi–detector row helical CT has overcome the limitations of single–detector row helical CT by providing faster speed, greater coverage, and improved spatial resolution (20–24). One of the greatest benefits of this technology for the diagnosis of TBM relates to the ability to acquire a volumetric data set of the entire central airways in only a few seconds. Thus, the entire central airways can be imaged during one dynamic expiratory sequence. Volumetric data acquisition also allows for the creation of three-dimensional reconstructions and multiplanar reformations, which have the potential to aid diagnosis and preoperative planning (25). Even though these reconstructions were performed routinely in our patient population, we emphasize that we relied on transverse images alone for analysis in this study. Future studies are necessary to evaluate the contribution of three-dimensional and multiplanar images in this setting.
Gilkeson et al (14) published results from a series of patients with TBM who were imaged by using dynamic airway evaluation with multi–detector row CT. These authors studied 13 patients suspected of having TBM, all of whom demonstrated greater than 50% airway collapse during dynamic expiratory imaging. The CT findings correlated well with those at bronchoscopy; however, the authors of that study did not compare dynamic expiratory imaging with the more commonly used end-expiratory imaging.
To our knowledge, ours is the first study performed to directly compare dynamic expiratory imaging with end-expiratory imaging in the same patient population. Our finding that end-expiratory imaging is relatively insensitive in the detection of TBM is supported by results from prior studies performed with electron-beam CT and magnetic resonance (MR) imaging, in which end-expiratory imaging was compared with imaging during a coughing maneuver. In a study of eight adult patients suspected of having TBM who were evaluated at electron-beam CT, Hein et al (6) obtained a much higher average percentage of collapse during coughing (71%) than at end expiration (36%). Similarly, Heussel et al (11), in a study of 29 adult patients suspected of having or verified to have tracheal stenosis or collapse at single–detector row helical CT scanning, obtained a significantly higher (P < .0001) degree of collapse with dynamic electron-beam CT (77%) than with paired end-inspiratory and end-expiratory helical CT (59%). However, electron-beam CT scanning in that study was performed with continuous gantry rotation and a stationary table only at a few selected levels. The latter differs substantially from our protocol, in which dynamic expiratory scanning covered the entire central airways in a single volumetric acquisition. A benefit of volumetric imaging is that it ensures that a focal area of malacia is not overlooked.
MR imaging has also been employed in the study of patients suspected of having TBM, since fast techniques such as T1-weighted turbo fast low-angle shot MR imaging provide good contrast resolution with a temporal resolution as low as 150 msec per section. By performing this technique in six patients suspected of having TBM, Suto and Tanabe (26) obtained a substantially higher degree of tracheal collapse during coughing (75% ± 12 [± standard deviation]) than during forced expiration (50% ± 15).
Although coughing produces extremely high intrathoracic pressure and thus elicits a high degree of airway collapse, patient movement during coughing has the potential to generate motion artifacts, which may interfere with the ability to obtain accurate measurements of the airway lumen. Future studies are necessary to directly compare dynamic expiratory imaging with imaging during coughing maneuvers.
Even though the majority of study results reported in the literature support the use of a threshold of greater than or equal to 50% collapse as diagnostic of TBM (1,6,9–11,13–15,19,26), it is important to note that authors of two studies have advocated the adoption of different threshold values. Stern et al (27) obtained a degree of tracheal collapse greater than 50% at end expiration in four of 10 healthy young adult male volunteers who underwent electron-beam CT. On the basis of their findings, these authors recommended a more conservative threshold of 70% collapse as indicative of TBM. Additional data from a larger group of subjects of varying age and sex would be helpful to further characterize the normal range of airway collapsibility. On the other hand, Aquino et al (18) studied 23 healthy subjects and 10 patients with bronchoscopically proved acquired TBM by using end-expiratory CT scans and obtained a positive predictive value of 89%–100% for TBM by using a threshold of more than 18% collapse for the upper trachea and of more than 28% for the midtrachea. Their recommendation of using a relatively low threshold when imaging at end expiration fits well with our finding that this method elicits a significantly lesser degree of airway collapse than the dynamic expiratory method.
Our finding that dynamic expiration is better than end expiration for eliciting central airway collapse is supported by the mechanics of respiration. Changes in the size of malacic trachea and bronchi depend on the difference between the intraluminal pressure inside the airways and the pleural (intrathoracic) pressure outside. Pleural pressure depends mostly on respiratory muscles and is high during expiratory efforts. In contrast, intraluminal pressure is highly variable and depends on airflow. When airflow is zero, intraluminal pressure equals alveolar pressure and differs from pleural pressure only by the elastic recoil pressure of the lung, which depends on lung volume. At maximal lung volume with no flow (end inspiration), the intraluminal pressure is 20–30 cm H2O greater than pleural pressure, and the pressure difference expands the trachea. At low lung volumes with no flow (end expiration), the intraluminal pressure is nearly equal to the pleural pressure and the trachea is unstressed. The trachea is most compressed during coughing and dynamic expiration at low lung volume, when pleural pressure is high (approximately 100 cm H2O), and expiratory flow limitation in the small airways prevents transmission of the high alveolar pressures to the central airways. Under these conditions, intraluminal pressure is nearly atmospheric, and the large amount of transmural pressure causes tracheal collapse (10,28).
Several limitations of our study should be noted. First, we realize that this was a retrospective study with a small sample size, and, therefore, future prospective studies with a larger number of patients would be helpful to confirm our results. Future studies would also be helpful to correlate the degree of elicited airway collapse with clinical symptoms and results of pulmonary function tests.
Since the cross-sectional area measurements were manually obtained by a radiologist who was aware of the purpose of the study and the nature of each series, there was a potential for measurement bias. However, since all measurements were further reviewed by a second radiologist who was blinded to the nature of each series, we do not believe that this factor influenced our findings. In addition, it should be noted that we did not attempt to validate the cross-sectional area measurements of the airways obtained with our tracing software by comparing them with phantom measurements, because these methods have been previously validated by other authors using similar measurement software programs (6,19,27,29). In addition, although our method of patient selection was based on diagnosis at CT rather than a bronchoscopic diagnosis of malacia, we do not believe that this introduced a substantial selection bias. We emphasize that our purpose was to compare the degree of collapse elicited by two different expiratory methods of imaging in the same patient population rather than to compare CT with bronchoscopy.
With regard to bronchoscopy, it should be noted that three of the 14 patients included in the study were not evaluated with bronchoscopy. Moreover, in four patients with a positive diagnosis of TBM on the basis of CT criteria, there was no mention of airway collapse on their bronchoscopy reports, although specific maneuvers for eliciting TBM were not performed in these cases. It is thus uncertain whether this apparent discrepancy reflected an overdiagnosis at CT or an underdiagnosis at bronchoscopy. Regardless of the mechanism for this difference, we emphasize that our study focused solely on the comparison between two different expiratory CT methods and that we relied on CT criteria for establishing the diagnosis of TBM. Results of other studies have shown a good correlation between dynamic expiratory CT and bronchoscopy for the evaluation of central airway collapse in patients with TBM—specifically, in studies for which TBM was assessed by using specific maneuvers bronchoscopically. In a comparison of standard-dose and low-dose dynamic expiratory CT scans in 10 patients with TBM and 10 healthy control subjects, Zhang et al (15) observed concordant findings between dynamic expiratory CT and bronchoscopy in all 10 patients. Gilkeson et al (14), in the study with 13 patients suspected of having TBM on the basis of findings on dynamic expiratory CT scans, observed a good correlation between CT and bronchoscopy in five (83%) of the six patients in whom the latter procedure was performed.
In conclusion, results of our study demonstrate that dynamic expiratory CT elicits a significantly greater degree of airway collapse than standard end-expiratory CT in patients with TBM. These findings suggest that dynamic expiratory CT may be the preferred method of imaging in patients suspected of having TBM.
|
FOOTNOTES |
|---|
Abbreviation: TBM = tracheobronchomalacia
Authors stated no financial relationship to disclose.
Author contributions: Guarantors of integrity of entire study, R.H.B., P.M.B.; study concepts, R.H.B., P.M.B., A.E.; study design, R.H.B., P.M.B.; literature research, R.H.B., P.M.B., S.H.L.; clinical studies, R.H.B., P.M.B., A.E., D.F.K.; data acquisition, R.H.B., P.M.B., D.F.K., A.E.; data analysis/interpretation, R.H.B., P.M.B., H.H., M.N., S.H.L.; statistical analysis, H.H., M.N.; manuscript preparation, R.H.B., D.F.K., S.H.L., P.M.B.; manuscript definition of intellectual content, R.H.B., H.H., M.N., A.E., P.M.B.; manuscript editing, R.H.B., P.M.B.; manuscript revision/review and final version approval, all authors
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. H. Loring, C. R. O'Donnell, D. J. Feller-Kopman, and A. Ernst Central Airway Mechanics and Flow Limitation in Acquired Tracheobronchomalacia Chest, April 1, 2007; 131(4): 1118 - 1124. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T Stone, J H Reynolds, and H J Williams Imaging of large and small airway diseases Imaging, September 1, 2006; 18(3): 139 - 150. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. S. Lee, A. Ernst, D. E. Trentham, W. Lunn, D. J. Feller-Kopman, and P. M. Boiselle Relapsing Polychondritis: Prevalence of Expiratory CT Airway Abnormalities Radiology, August 1, 2006; 240(2): 565 - 573. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
