| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Vascular and Interventional Radiology |
1 From the Department of Radiology and Center for Imaging Science (Y.C.Y., K.S.L., Y.J.J., S.W.S., M.J.C.) and Department of Medicine, Division of Pulmonary and Critical Care Medicine (O.J.K.), Samsung Medical Center, Sungkyunkwan University School of Medicine, 50 Ilwon-Dong, Kangnam-Ku, Seoul 135–710, Korea. Received December 22, 2003; revision requested March 2, 2004; revision received March 11; accepted April 8; updated April 27. Address correspondence to K.S.L. (e-mail: kyungs.lee@samsung.com).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Institutional review board approval was obtained, and informed consent was not required. Sixteen–detector row helical CT and conventional angiography of the thorax were performed in 22 patients (16 men, six women; age range, 18–75 years; mean age, 50 years) with hemoptysis. Three observers in consensus analyzed retrospectively transverse, multiplanar reconstruction, or three-dimensional CT images for visibility, traceability of bronchial arteries from their origin at the aorta or aortic branches to the hilum, and presence of nonbronchial systemic arteries. CT and angiographic findings of bronchial and nonbronchial systemic arteries causing hemoptysis were compared by two radiologists in consensus. Differences in visibility, traceability, and diameter of bronchial arteries causing and those not causing hemoptysis were tested by using generalized estimating equation method or the mixed model.
RESULTS: Fifty-two (30 right and 22 left) bronchial arteries and 33 nonbronchial systemic arteries were visible at CT. Thirty-four (20 right and 14 left) of 52 bronchial arteries were traceable from their origins to the hilum. Thirty-one (16 right and 15 left) of 46 (27 right and 19 left) bronchial arteries and 26 of 64 nonbronchial systemic arteries evaluated at angiography were causing hemoptysis. Forty (87%, 23 right and 17 left) of 46 bronchial arteries seen at angiography were also detected at CT. All 31 bronchial arteries and sixteen (62%) of 26 nonbronchial systemic arteries causing hemoptysis were detected at CT. Twenty-three (74%) of 31 bronchial arteries causing hemoptysis were traceable from their origins to the hilum, and one (11%) of nine bronchial arteries not causing hemoptysis was traceable (P = .002).
CONCLUSION: Sixteen–detector row CT provides depiction and traceability of the bronchial arteries in patients with hemoptysis, and in most patients it enables detection of the bronchial and nonbronchial arteries causing hemoptysis.
© RSNA, 2004
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Computed tomography (CT) is now considered a primary noninvasive imaging modality in the evaluation of patients with hemoptysis (6,11,12), and it might also serve as aguide for other diagnostic or therapeutic procedures (13). Multi–detector row CT allows rapid scanning throughout the thorax, multiplanar reconstruction (MPR) without loss of z-axis information owing to near isotropic voxel size, and acquisition of various three-dimensional images and, therefore, visualization of the bronchial and nonbronchial systemic feeder vessels (7). Thus, the purpose of our study was to retrospectively evaluate 16–detector row CT compared with conventional angiography in the depiction of bronchial and nonbronchial systemic arteries in patients with hemoptysis.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
CT Scanning and Image Analysis
A 16–detector row helical CT scanner (LightSpeed Ultra 16; GE Medical Systems, Milwaukee, Wis) was used to evaluate the thorax and the upper abdomen from the level of the supraclavicular area to the level of the upper pole of the right kidney. Imaging parameters were a 10-mm beam width, beam pitch of 1.375, 1.25-mm reconstruction thickness, 120 kV, and 180 mA. All patients underwent craniocaudal scanning in a supine position and at end-inspiratory suspension during a single breath hold. All patients received 120 mL (total 36 g) of nonionic contrast material with 300 mg of iodine per milliliter (iohexol, Omnipaque 300; Nycomed, Princeton, NJ), which was administered intravenously into the antecubital vein through an 18-gauge catheter at a rate of 3 mL/sec by using an automated injector (Envision CT; Medrad, Pittsburgh, Pa). Scan delay time was 20 seconds, and scan time was 12–15 seconds. A 52-cm field of view and 512 x 512 matrix size were used.
For the retrospective analysis of CT findings, the stored raw data of the 1.25-mm-thick transverse CT scans were transferred to a workstation (MxView; Philips Medical Systems, Best, the Netherlands). Three chest radiologists (Y.C.Y., K.S.L., Y.J.J., with 3 years, 14 years, and 1 year of experience, respectively, in interpretation of chest CT images), who were blinded to angiographic findings, analyzed the CT images. Final decisions regarding the findings were determined in consensus.
First, 1.25-mm-thick transverse images with mediastinal window settings (window width, 400 HU; window level, 20 HU) were evaluated on a full screen and in cine mode. Minor alterations of window settings were allowed. Second, MPR images were obtained parallel to the axis of the bronchial artery origin for the confirmation of the level of origin and for the measurement of the diameter. In addition, MPR images were obtained at various angles to evaluate the mediastinal course and the traceability of bronchial arteries to the hilum. Nodular and tubular enhancing structures in the mediastinum or around the central airways that were connected to the aorta or its major branches at CT were regarded as bronchial arteries (3,15).
The visibility, level of origin, side (right or left), diameter, and traceability (depiction of the vessels from their origin at the aorta or its major branches to the hilum without partial volume averaging on a single or consecutive scans) of the bronchial arteries were evaluated on 1.25-mm-thick transverse and 1.0-mm-thick MPR images. The diameter of each bronchial artery was measured three times at the nearest point to its origin on transverse or MPR images, which allowed the display of the artery in tubular shape. The average value was chosen as its diameter. Finally, three-dimensional volume-rendered and maximum intensity projection images were acquired to display the arteries as a whole on a single image. The thickness and the rotating angle of the volume of interest for the three-dimensional images were determined to demonstrate the mediastinal course of each bronchial artery.
When enlarged vascular structures within extrapleural fat were demonstrated in association with pleural thickening (3 mm), the observers regarded these vessels as nonbronchial systemic arteries causing hemoptysis (6). A systemic artery that did not have vascular enlargement and an associated pleural thickening but coursed to the lungs was regarded at CT as a nonbronchial systemic artery not causing hemoptysis.
Angiography and Interpretation
Conventional angiography was performed in all 22 patients by an interventional radiologist (S.W.S.) with 3 years of experience in performing bronchial angiography. The interval between CT and conventional angiography ranged from 0 to 11 days (mean, 1.3 days). In one of the patients who underwent conventional angiography 11 days after CT, pulmonary sequestration was diagnosed. His hemoptysis was controlled with medical therapy, but angiography was performed for the diagnosis and treatment of pulmonary sequestration. The interval in the remaining 21 patients ranged from 0 to 3 days (mean, 0.8 day) after CT.
In all patients, conventional angiography was performed with a digital subtraction technique by using a transfemoral approach and with the Seldinger technique. Various types of angiographic catheters were used to select different arteries. The most commonly used catheter to select bronchial arteries was HS 1 (Cook, Bloomington, Ind). The bronchial arteries or nonbronchial systemic arteries to be evaluated at angiography were chosen on the basis of the presence of bronchopulmonary lesions on CT scans or both CT and bronchoscopic findings. Bilateral bronchial arteriography was performed in 16 patients, and unilateral bronchial arteriography was performed in four. In the remaining two patients, in whom bronchial arteries were not identified, selective angiography of only the intercostal artery was performed.
An experienced interventional radiologist (S.W.S.) reviewed retrospectively all of the angiographic images. The angiographic criteria for arteries causing hemoptysis were as follows: enlarged and tortuous bronchial or nonbronchial systemic arteries, neovascularity, hypervascularity, shunting into the pulmonary artery or vein, extravasation of contrast medium, and bronchial artery aneurysm (7). Both bronchial arteries and nonbronchial systemic arteries causing hemoptysis were recorded for each patient.
Comparison of CT and Angiographic Findings
Comparisons of CT and selective angiographic findings of the bronchial and nonbronchial systemic arteries causing hemoptysis were performed by two radiologists (Y.C.Y., S.W.S.) in consensus by observing side-by-side display of images on a picture archiving and communication system monitor (Centricity, version 1.0; GE Medical Systems, Mt Prospect, Ill). Bronchial arteries seen at angiography were assessed at CT. Differences in the visibility of bronchial arteries at CT in terms of the side, the diameter, and the traceability of bronchial arteries causing and those not causing hemoptysis, as determined at angiography, were also evaluated.
Statistical Analyses
Statistical analyses were performed with commercially available software (SAS version 8.2; SAS Institute, Cary, NC). A generalized estimating equation was used to test the differences in the visibility of bronchial arteries causing and those not causing hemoptysis in terms of the side and traceability. The mixed model was used to test the differences in the diameter of bronchial arteries causing and those not causing hemoptysis. Statistical significance was achieved at P < .05.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
In eight (36%) of 22 patients, 33 nonbronchial systemic arteries causing hemoptysis (24 left intercostal arteries in five patients, three left inferior phrenic arteries in three, two right intercostal arteries in one, two right inferior phrenic arteries in two, and two left subclavian arteries in two) were detected on CT scans.
Angiographic Findings
In 20 of 22 patients, 46 bronchial arteries (27 right in 20 patients and 19 left in 16 patients) were evaluated with selective angiography. All of these bronchial arteries arose from the descending thoracic aorta. Twelve (44%) of 27 right bronchial arteries arose from intercostobronchial trunks. Nine pairs of both right (33%) and left (47%) bronchial arteries shared common bronchial trunks. Five (19%) right bronchial and nine (47%) left bronchial arteries arose directly from the descending thoracic aorta as an isolated branch. One pair of both right (4%, one of 27) and left (5%, one of 19) bronchial arteries arose from the common bronchial and intercostobronchial trunk. In 18 of 22 patients, 31 bronchial arteries (16 right in 16 patients and 15 left in 13 patients) were regarded as causing hemoptysis.
In 17 of 22 patients, 64 nonbronchial systemic arteries (30 left intercostal arteries in eight patients, 23 right intercostal arteries in nine, four left inferior phrenic arteries in four, three right inferior phrenic arteries in three, one right subclavian artery, one left subclavian artery, one right internal mammary artery, and one left internal mammary artery in each one) were evaluated with selective angiography. Of these, 26 arteries in nine patients (15 left intercostal arteries in five patients, four right intercostal arteries in two, three right inferior phrenic arteries in three, two left inferior phrenic arteries in two, one left subclavian artery, and one left internal mammary artery in each one) were regarded as causing hemoptysis, and the remaining 38 nonbronchial systemic arteries were regarded as not causing hemoptysis.
Comparison of CT and Angiography
Forty (87%) of 46 bronchial arteries (23 [85%] of 27 right and 17 [89%] of 19 left), which were evaluated with selective angiography, were detected at CT. All 31 bronchial arteries (16 right and 15 left), which were regarded at angiography as those causing hemoptysis, were detected at CT (Figs 1, 2). None of the six bronchial arteries, which were not detected at CT, were regarded at angiography as those causing hemoptysis. Four bronchial arteries that arose from the aorta as a common trunk, with a contralateral bronchial artery causing hemoptysis (Fig 2), and one pair of both right and left bronchial arteries that arose from the aorta as a common trunk were not visualized at CT.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
The normal bronchial artery is a small vessel that arises directly from the descending thoracic aorta and supplies blood to the airway of the lung, esophagus, and lymph nodes (3,8–10,15). Bronchial arteries show substantial anatomic variations in their origins, branching patterns, and courses (7,8,10). The right intercostobronchial trunk, which usually arises from the right posterolateral aspect of the thoracic aorta at the level of the T5 or T6 vertebra, is the most constant vessel (3,7–10). The left bronchial arteries usually originate from the anterior surface of the thoracic aorta or from the concavity of the aortic arch and pass forward beside the lateral wall of the esophagus and cross the peribronchial space from the level of the left main bronchus toward the hilum (3,8–10,15). In approximately 70% of cases, there are two left bronchial arteries, and the upper left bronchial artery appears more horizontal in course within the mediastinum (15). Right and left bronchial arteries that arise from the aorta as a common trunk are not unusual (7).
Bronchial arteries are identified in the posterior mediastinum as dots or lines of increased attenuation (15). The CT depiction of bronchial arteries is influenced by their size, course, and the scanning technique (9). Accurate identification is hampered by the azygos vein and retroaortic anastomoses of the azygos system, enhancing mediastinal lymph nodes, and esophageal wall with streak artifacts within the posterior mediastinum, which may also mimic bronchial arteries (9,15). Cardiac or respiratory motion artifacts may also preclude the visibility of the bronchial artery. In the current study, however, we were not confronted with these problems. Transverse images with 1.25-mm thickness and MPR images with 1-mm thickness at various angles allowed us to perform detailed anatomic evaluation of the bronchial artery.
Murayama et al (10) reported that 12 (75%) of 16 bronchial arteries seen at bronchial arteriography could be depicted from their origins at CT with use of the curved reformation technique. In the current study, 40 (87%) of 46 bronchial arteries and all of 30 bronchial arteries causing hemoptysis that were observed at selective angiography were also observed at CT. In addition, five bronchial arteries of aberrant origins that were not evaluated at angiography were seen at CT. Bronchial artery diameter greater than 2 mm has been considered abnormal (7,15). In our study, however, differences in diameters between bronchial arteries causing hemoptysis and those not causing hemoptysis were not statistically significant. At CT, eight bronchial arteries causing hemoptysis were smaller than 2 mm in diameter, and four bronchial arteries not causing hemoptysis were greater than 2 mm. However, differences in the percentage of traceability to the hilum of bronchial arteries causing hemoptysis and those not causing hemoptysis were statistically significant. Therefore, 16–detector row CT provides information as to which bronchial artery (arteries of traceability) should be selected in the interventional procedure.
In the presence of pleural thickening, nonbronchial systemic feeder vessels that arise from the various arteries (eg, the intercostal artery, branches of the subclavian and axillary arteries, internal mammary artery, or inferior phrenic artery) may develop along the pleural surface and become enlarged as a result of the inflammatory process, and they can be a major source of massive hemoptysis (6,7,17–21). In a previous study (6), the sensitivity, specificity, and accuracy of conventional CT in the depiction of a nonbronchial systemic artery were 80%, 84%, and 84%, respectively. In the current study, 16 (62%) of 26 nonbronchial systemic arteries regarded at angiography as causing hemoptysis were depicted at CT. Discrepancy between the findings of the two studies may be explained as owing to the use of a different evaluation method.
In the previous study (6), the authors calculated the sensitivity, specificity, and accuracy of CT on the basis of their classification of the nonbronchial systemic arteries according to the following anatomic location: superolateral (branches of the subclavian and axillary arteries at angiography and nonbronchial systemic arteries at the apex of the chest above the level of the aortic arch at CT), anteromedial (internal mammary artery and its branches at angiography and nonbronchial systemic arteries along the anterior and mediastinal pleura below the level of the aortic arch), and posterior (intercostal arteries at angiography and nonbronchial systemic arteries along the posterior pleura), regardless of the exact name of the artery. On the contrary, we directly correlated the CT and angiographic findings with each artery. Therefore, in our study, some intercostal arteries (eg, the third and fourth intercostal arteries) that were regarded at angiography as causing hemoptysis might not be correctly depicted at CT as arteries causing hemoptysis if the fifth and sixth intercostal arteries were regarded at CT as causing hemoptysis. Of 10 angiographically confirmed nonbronchial systemic arteries causing hemoptysis (seven intercostal arteries in six patients, two inferior phrenic arteries in two, and one internal mammary artery in one) that were not detected at CT, six might be regarded as arteries causing hemoptysis if the detectability was calculated according to the method of the previous study (6). In addition, not all systemic arteries that may be regarded at CT as a source of a nonbronchial systemic artery causing hemoptysis were evaluated at angiography because of the uncertainty in some cases of the presence of nonbronchial arteries causing hemoptysis on preinterventional CT scans, an overdose of contrast material, or a technical problem in a case of severe atherosclerosis.
There were some limitations of our study, such as the retrospective design and the small number of patients enrolled. Because selective angiography was performed in a limited number of patients, the sensitivities might be overestimated. To confirm our observations, a prospective study is needed in which interobserver variability, especially in defining bronchial arteries causing hemoptysis and those not causing hemoptysis, is evaluated with use of CT and angiography. Predetermined scan delay time of 20 seconds in our study may have caused a decreased degree of arterial enhancement. Intravenous injection of contrast medium at a rate of 3 mL/sec in our study may have limited the ability to visualize small vessels such as the bronchial arteries. Thus, diagnostic accuracy might be improved if the scan delay time is determined by using a bolus-tracking technique and if the injection rate of contrast medium is faster. One more limitation, which was unavoidable, was that the angiographic criteria we used for arteries causing hemoptysis were not perfect unless active extravasation of contrast medium was visualized. However, active extravasation is reported to be seen in only 4%–11% of patients with hemoptysis (7).
In conclusion, 16–detector row CT provides depiction and traceability of the bronchial arteries in patients with hemoptysis, and in most patients enables detection of the bronchial and nonbronchial arteries causing hemoptysis.
|
FOOTNOTES |
|---|
Authors stated no financial relationship to disclose.
Author contributions: Guarantor of integrity of entire study, K.S.L.; study concepts and design, Y.C.Y., K.S.L.; literature research, Y.C.Y., Y.J.J., M.J.C.; clinical studies, Y.C.Y., S.W.S., O.J.K.; data acquisition, Y.C.Y., K.S.L.; data analysis/interpretation, Y.C.Y., K.S.L., S.W.S.; statistical analysis, Y.C.Y., K.S.L.; manuscript preparation, Y.C.Y.; manuscript editing, Y.C.Y., K.S.L.; manuscript definition of intellectual content, revision/review, and final version approval, all authors
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  A. SIRAJUDDIN and T.-L. H. MOHAMMED A 44-year-old man with hemoptysis: A review of pertinent imaging studies and radiographic interventions Cleveland Clinic Journal of Medicine, August 1, 2008; 75(8): 601 - 607. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Savale, A. Parrot, A. Khalil, M. Antoine, J. Theodore, M.-F. Carette, C. Mayaud, and M. Fartoukh Cryptogenic Hemoptysis: From a Benign to a Life-threatening Pathologic Vascular Condition Am. J. Respir. Crit. Care Med., June 1, 2007; 175(11): 1181 - 1185. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Khalil, M. Fartoukh, M. Tassart, A. Parrot, C. Marsault, and M.-F. Carette Role of MDCT in Identification of the Bleeding Site and the Vessels Causing Hemoptysis Am. J. Roentgenol., February 1, 2007; 188(2): W117 - W125. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A Khalil, M Soussan, G Mangiapan, M Fartoukh, A Parrot, and M-F Carette Utility of high-resolution chest CT scan in the emergency management of haemoptysis in the intensive care unit: severity, localization and aetiology Br. J. Radiol., January 1, 2007; 80(949): 21 - 25. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. J. Chung, J. H. Lee, K. S. Lee, Y. C. Yoon, O J. Kwon, and T. S. Kim Bronchial and Nonbronchial Systemic Arteries in Patients with Hemoptysis: Depiction on MDCT Angiography. Am. J. Roentgenol., March 1, 2006; 186(3): 649 - 655. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
