| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Innovation |
1 All authors: Center for Airway Imaging, Department of Radiology, Harvard Medical School, Beth Israel Deaconess Medical Center, 330 Brookline Ave., Boston, MA 02215.
Received August 19, 2005;
accepted after revision December 7, 2005.
Address correspondence to P. M. Boiselle
(pboisell{at}bidmc.harvard.edu).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CONCLUSION. This method is technically feasible and offers a promising alternative to previous cine CT methods for diagnosing tracheomalacia.
Keywords: cine imaging • CT • trachea • tracheomalacia
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Because tracheomalacia escapes detection on routine end-inspiratory imaging studies, it is widely considered an underdiagnosed condition [1-5]. Its detection requires imaging during maneuvers such as forced exhalation and coughing, which are known to increase the intrathoracic-extratracheal pressure. Because coughing elicits a higher level of intrathoracic-extratracheal pressure than forced exhalation or end-exhalation, this maneuver is considered the most sensitive method for eliciting tracheal collapse [1, 4].
Prior studies using electron beam CT and single-detector helical CT to image the trachea in cine fashion during functional maneuvers have shown promising results [6, 7]. However, these methods were limited to imaging a single slice of the trachea during each cine sequence. This resulted in a sampling of the trachea at selected levels and was associated with a relatively high radiation exposure and long examination time [5, 6].
64-MDCT is advantageous for functional imaging because of its high spatial and temporal resolution, combined with its large anatomic coverage capability [4]. During a single cine acquisition, it provides anatomic coverage of more than 3 cm in the z-axis. The purpose of this article is twofold: to describe the technical aspects of using 64-MDCT for cine CT during coughing and to share our preliminary clinical experience using this method.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
A consecutive series of patients referred for CT of known or suspected benign central airways disease during a 3-month period underwent imaging on a 64-MDCT scanner (Aquilion, Toshiba America Medical Systems) with the following parameters: detector collimation = 0.5 mm x 64; mA = 80; kVp = 120; gantry rotation = 0.4 seconds. The use of a low-dose (80 mA) technique for evaluating the tracheal lumen has been previously validated [5].
An initial scout topographic image was obtained to determine the area of coverage, which extended 3.2 cm in craniocaudad length. The inferior aspect of the acquisition was set at the level of the carina, and the superior aspect of the acquisition was set at 3.2 cm above this level, which corresponded to approximately the level of the aortic arch. A 7.2-second acquisition was acquired in cine mode beginning at end-inspiration and followed by repeated coughing maneuvers.
Images were reconstructed at 8-mm collimation in a standard algorithm, creating four contiguous cine data sets from a single acquisition. The CT data were transferred to a workstation where a commercial software program (Analyze 6.0, Analyze Direct) was used for automated measurement of changes in tracheal lumen cross-sectional area values during the cine sequence. Figure 1 shows the display of this data in graphic format. Tracheomalacia was defined according to standard criteria as more than 50% reduction in cross-sectional area lumen during coughing [1-4] (Figs. 2A and 2B).
|
|
|
The CT data were also transferred to a PACS workstation (Centricity 2.0, GE Healthcare) for visual assessment by a thoracic radiologist. On PACS, the scans were reviewed in both a cine loop and freeze-frame fashion. Cine data sets of subjects with and without tracheomalacia are shown in Figures S3 and S4, respectively, at www.ajronline.org. For each patient, the CT data were evaluated for coverage of the intended area of interest and for the presence of motion artifacts involving the tracheal walls or lumen (e.g., blurring or doubling of airway walls and lumen).
CT findings were correlated with results of bronchoscopy in the subset of patients who underwent this procedure. Patients with stents or fixed airway stenoses were excluded from analysis.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Most patients completed two or three coughs during the 7.2-second acquisition time. However, one patient with a rapid coughing mechanism completed seven coughs. Figure 1 shows the graphic data for this patient.
Twelve (75%) of 16 patients met CT criteria for tracheomalacia (Figs. 2A and 2B). The high percentage of positive cases reflects the fact that our hospital is a referral center for patients with this disorder. In this subset, the mean percentage collapse of the tracheal lumen was 80% (range, 69-100%).
Bronchoscopy was performed in six (38%) of 16 patients. CT findings were concordant with bronchoscopy for the presence or absence of malacia (four positive for malacia, two negative) in all six patients.
The estimated dose-length product (DLP) for the cine CT acquisition (as recorded from the CT display monitor for a 70-kg patient) is 220.8 mGy · cm. This compares to a reference standard of 650 mGy · cm for a standard chest CT [8]. In comparison, the estimated total DLP for a dualphase helical (noncine) standard-dose inspiratory phase and low-dose expiratory CT scan of the trachea [9] for a 70-kg patient at our institution is 508.1 mGy · cm.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Coughing is a powerful functional maneuver for eliciting collapse in patients with tracheomalacia. For example, a previous study of eight patients with suspected tracheomalacia by Hein et al. [6] using electron beam CT showed that coughing elicits a much higher average percentage of collapse (71%) compared with imaging at end-expiration (36%) [5]. A previous study by Heussel et al. [7] showed the potential benefits of cine CT using single-detector helical CT. These authors obtained a significantly higher degree of collapse with dynamic continuous respiration cine CT (77%) than with paired end-inspiratory/end-expiratory helical CT (59%) in a group of 29 patients [6]. Notably, however, these methods were limited to imaging a single slice of the trachea during each cine sequence. This resulted in a "sampling" of the trachea, with the need for repeated acquisitions, resulting in a relatively high radiation exposure and long examination time [6].
64-MDCT is advantageous for functional imaging because of its high spatial and temporal resolution, combined with its large anatomic coverage capability. The latter feature provides coverage of more than 3 cm in the z-axis during a single cine acquisition. Moreover, depending on the CT manufacturer, up to 4 cm (0.625-mm detector width x 64 detectors) in length can be covered in one cine acquisition. Thus, by providing greater anatomic coverage during a single acquisition, 64-MDCT overcomes the limitations of electron beam CT and single-detector helical CT.
We emphasize that although a reduced-dose technique (80 mA) was used, the overall dose for the 3.2-cm volume cine acquisition was approximately a third of the dose of the reference standard for a full chest CT scan [8]. Because the dose of the cine acquisition is directly proportional to the duration of the scan, a 50% reduction in dose can be achieved by reducing the total acquisition time in half. In retrospect, on review of our data, a shorter acquisition would have been sufficient for diagnostic purposes. We have thus recently reduced the acquisition time to approximately 3 seconds for this sequence. Given the high inherent contrast between the air-filled tracheal lumen and adjacent soft-tissue density of the mediastinum, we anticipate that further dose reductions may be possible by reducing the kVp and mA values. We plan to address this in a future study. Although the 3.2-cm volume acquisition in our protocol is superior to single-slice-level imaging, it does not image the entire intrathoracic trachea. Because most cases of acquired tracheomalacia are diffuse [1], such coverage is likely adequate for screening purposes. However, it is possible that tracheomalacia could be isolated to a region above the scanning acquisition level. To reduce sampling error, two separate approximately 3-second acquisitions could be performed, increasing the length of coverage to 6.4 cm (or 8 cm, depending on the CT manufacturer). Fortunately, technical advances in CT, including 256-detector scanners and flat panel scanners, will soon allow cine imaging of the entire trachea during a single acquisition.
Although CT and bronchoscopy were concordant in the patients who underwent both procedures, we acknowledge that less than half of the patients in our cohort underwent bronchoscopy. We emphasize that the major aim of this preliminary study was to determine the technical feasibility of this method. Future aims include correlation with bronchoscopy in a larger cohort and direct comparison with dynamic expiratory helical CT of the trachea, another promising CT method for evaluating tracheomalacia [9].
We used a standard criterion of more than 50% collapse during coughing as diagnostic of tracheomalacia. However, it is interesting to note that the positive cases in this study showed values considerably higher than this threshold (mean percentage collapse, 80%; range, 69-100%). Notably, there is a need to establish the normal range of tracheal collapse during coughing among a group of healthy volunteers of varying ages, ethnicity, and both sexes. In the future, we plan a prospective study to establish normative data for tracheal dynamics during coughing. In the meantime, we suggest that careful correlation with clinical symptoms may be helpful in determining the clinical significance of patients who show tracheal collapse between 50% and 69% during coughing.
In summary, cine CT using 64-MDCT is technically feasible. By overcoming the limitations of previous cine imaging methods, it has the potential to make significant contributions to the noninvasive diagnosis of tracheomalacia.
Acknowledgments
The authors gratefully acknowledge Milliam Kataoka for assistance with
creating electronic media files of the coughing sequences and Pei-Jan Paul Lin
and Carol Wilcox for assistance with CT radiation dose estimates.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  E. Y. Lee and P. M. Boiselle Tracheobronchomalacia in Infants and Children: Multidetector CT Evaluation Radiology, July 1, 2009; 252(1): 7 - 22. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. McDermott, S. C. Barry, E. E. Judge, S. Collins, P. A. de Jong, H. A. W. M. Tiddens, E. F. McKone, C. C. Gallagher, and J. D. Dodd Tracheomalacia in Adults with Cystic Fibrosis: Determination of Prevalence and Severity with Dynamic Cine CT Radiology, June 9, 2009; (2009) 2522081956. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
