| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 All authors: Department of Radiology, Langenbeckstr. 1, 55131 Mainz, Germany.
Received October 12, 2001;
accepted after revision April 11, 2002.
Address correspondence to S. Ley.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. Seventeen patients were included in this
study. MR phase-contrast flow measurements were used to calculate the net
forward volumes in the right and left pulmonary arteries and in the ascending
aorta. Single-detector helical CT scans were assessed for the presence of
dilated bronchial arteries that could be delineated from the descending aorta
to the mainstem bronchi. Their perpendicular cross-sectional area at the level
of the main bronchi was measured using a double-threshold region of interest
(
100-3072 H).
RESULTS. The mean net forward volume in the aorta was 44.6 mL per heartbeat (R-R interval) and in the pulmonary arteries, 30 mL per R-R interval. Thus, the mean difference was 14.6 mL per R-R interval; this value represents the shunt volume between the systemic arterial and pulmonary venous circulations. On CT, dilated bronchial arteries were depicted in all patients (mean, three arteries per patient). The mean cross-sectional area of the bronchial arteries was 0.19 cm2. Pearson's correlation coefficient (r) between cross-sectional area and shunt volume was 0.86 (p < 0.01).
CONCLUSION. MR imaging was able to reveal substantial differences in flow between the systemic arterial and pulmonary venous circulations in patients with chronic thromboembolic pulmonary hypertension. These differences correlated well with the diameters of the bronchial arteries seen on helical CT. Furthermore, these differences resolved after pulmonary thromboendarterectomy. MR imaging enables the accurate estimation of flow in the bronchial arteries in patients with chronic thromboembolic pulmonary hypertension.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Because the embolic material is endothelialized and incorporated into the intimal layer of the pulmonary vessels, treatment with pharmaceuticals is not possible. Surgical intervention, called pulmonary thromboendarterectomy, is the only treatment option. Thromboendarterectomy is technically feasible if the obstruction begins at or proximal to the level of the lobar pulmonary arteries [3].
To determine where the obstruction begins in a fast and noninvasive manner, we perform single-detector helical CT angiography during the normal preoperative workup for patients undergoing pulmonary thromboendarterectomy. In several studies, helical CT has proven to be the cross-sectional modality of choice for direct visualization of thrombotic material [4,5,6]. In addition, hemodynamic [7] and ancillary parenchymal changes can be observed on helical CT [8, 9].
Besides the pulmonary arteries, the lung parenchyma is maintained by the bronchial arteries—the vasa vasorum of the lung. The bronchial arteries arise from the aorta and drain into pulmonary veins [10]. Normally, the pulmonary arteries only supply nutrition and do not take part in gas exchange [11]. During this baseline condition, these arteries have a maximum diameter of 1.5 mm and are rarely seen on helical CT [12]. However, in patients with pathologic conditions (e.g., occlusion of one main pulmonary artery), the pulmonary arteries participate in blood oxygenation; up to 25% of the bronchial circulation can do so [13]. In patients with chronic thromboembolic pulmonary hypertension, flow through the bronchial arteries increases and these arteries become visible on helical CT angiography [14] because they are dilated. Another important finding is that dilated bronchial arteries are positively correlated with a lower mortality rate after pulmonary thromboendarterectomy [12].
The collateral flow of the bronchial arteries can be measured by an invasive contrast medium—dilution technique [15]. Another approach is to measure the right and left cardiac outputs and to calculate the shunt volume between the systemic circulation and the pulmonary circulation. These values can be obtained noninvasively using MR imaging—based flow measurements [16].
Our hypothesis was that the shunt volumes determined by MR phase-contrast flow measurements correlate with the cross-sectional areas of the dilated bronchial arteries found on CT angiography.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
All patients underwent pulmonary thromboendarterectomy and thus could be examined during the postoperative course 14 days after surgery (range, 8-45 days).
To validate MR flow measurements and obtain normal values, we examined a control group composed of 10 healthy volunteers (10 men; age range, 23-27 years; mean age, 24 years).
MR imaging was performed on a 1.5-T system (Magnetom Vision; Siemens Medical Systems, Erlangen, Germany) using a phased array body coil. To measure flow, we used an ECG-gated, velocity-encoded, k-space—segmented gradient-echo sequence (TReff/TE, 12/4.8; flip angle, 30°; velocity encoding, 150 cm/sec; slice thickness, 6 mm) with a temporal resolution of 110 msec. The imaging matrix was 100 x 256 pixels. With a field of view of 300 mm, the in-plane resolution was 2.63 x 1.37 mm2. Because k-space segmentation with a segmentation factor of 5 and 100 phase-encoding steps was used, 20 R-R intervals were required to obtain a complete sample of the k-space data.
To obtain images that were optimal for flow measurements, we chose a double oblique orientation. Flow was measured perpendicularly to the ascending aorta and to the right and left pulmonary arteries (Fig. 1A,1B,1C,1D) using dedicated software (Argus V2.3; Siemens Medical Systems). The vessels were segmented manually by an observer who drew a region of interest on all magnitude images. The observer was unaware of the results of the invasive measurements and the CT evaluation. The software automatically measured the flow in the corresponding area on the velocity-encoded images by providing the net forward volume in milliliters per heartbeat (R-R interval) after subtracting the reverse volume from the forward volume. The number of pixels per region of interest was 7 x 13 (horizontal x vertical).
|
|
|
|
The total pulmonary arterial flow was calculated by adding the flow volumes of the left and right pulmonary arteries. Because the flow in the ascending aorta was measured distal to the orifices of the coronary arteries, we had to add coronary blood volume to calculate the true stroke volume of the left ventricle. To correct for coronary blood flow, we used the percentage of aortic net forward volume to aortic forward volume obtained from measurements of the healthy volunteers [18].
Helical CT angiography was performed on a single-detector scanner (PQ 6000; Picker International, Cleveland, OH). The scanning parameters were a collimation of 3 mm, a pitch of 2, and an increment of 2 mm. The images were reconstructed onto the mediastinum using the zoom feature with a field of view of 200 mm (resulting pixel size, 0.38 mm2). We administered 120 mL of a nonionic contrast medium (Ultravist [iopromide 300]; Schering, Berlin, Germany) as an IV bolus with a flow rate of 3 mL/sec using a power injector followed by a 50-mL saline flush administered at the same rate (Fig. 2). The delay between the administration of the contrast agent and the start of helical scanning was 15 sec. The mean breath-hold time was 32 sec. For some subjects, an additional supply of oxygen was provided. To reduce the risk of breathing artifacts, we chose to scan in a caudocranial direction. Regardless, the posterior mediastinum containing the bronchial arteries is among the regions least affected by motion artifacts.
|
On axial slices, dilated bronchial arteries were shown (Fig. 3), and the cross-sectional areas of the vessels were measured. This evaluation was performed retrospectively in a consensus interpretation by two radiologists after the patients had undergone pulmonary thromboendarterectomy. These reviewers were unaware of the results of the invasive and MR flow measurements. Because bronchial arteries also nutrify the trachea and mediastinal structures [19] and because these branches do not participate in gas exchange, we measured the cross-sectional area of the bronchial arteries at the point where they run with the main bronchi to the distal parts of the lung. When measuring the cross-sectional area of a vessel, it is important to select a slice at the point where it is cut perpendicularly. Similar to how the cross-sectional area was measured on the axial slices, individual vessels were followed to the hili for measurement from an adequate position. A perpendicular cut was found for each vessel. This location was determined in consensus, and only one region per vessel was measured. For calculation of the precise cross-sectional area of the bronchial arteries, a region of interest with a double-threshold of from 100 to 3072 H was used. The cross-sectional areas of the different bronchial arteries were summed to yield the total cross-sectional area of the bronchial arteries of each individual subject. The results of a study comparing patients with chronic thromboembolic pulmonary hypertension and patients without pulmonary hypertension have already been reported [12]. In that study, the bronchial arteries of the patients without pulmonary hypertension were not dilated. Therefore, and because of ethical reasons, we performed CT studies only in the patients.
|
Statistical analyses were completed using software (SPSS version 9.0.1 for Windows [Microsoft, Redmond, WA]; SPSS, Chicago, IL). Pearson's correlation coefficient was used for correlation analysis. For comparison of aortic net forward volume and total pulmonary arterial flow in volunteers, Wilcoxon's signed rank test was used.
To compare the shunt volumes of the volunteers and patients, we used the Mann-Whitney rank sum test; to correlate pre- and postoperative data, we used Wilcoxon's signed rank test. A p value of less than 0.05 was considered to indicate a statistically significant difference.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The MR flow measurements in volunteers revealed an aortic net forward volume of 70.2 mL per R-R interval (range, 36.1-102.7 mL per R-R interval). This flow volume was measured distally to the origin of the coronary vessels. The forward volume in the aorta was 74 mL per R-R interval (range, 41.3-109.7 mL per R-R interval). Because no signs of aortic valve insufficiency were present, the reverse volume was considered to represent coronary blood flow. Based on the MR phase-contrast flow measurements, the correction for coronary blood flow was 5%.
After the correction for coronary blood flow was added, the aortic net forward volume was 73.7 mL per R-R interval. Flow measurements in the pulmonary arteries revealed 36.5 mL per R-R interval for the right (range, 16.5-48.9 mL per R-R interval) and 36.7 mL per R-R interval for the left (range, 13.5-59.7 mL per R-R interval). These findings resulted in a total pulmonary arterial blood flow of 73.2 mL per R-R interval. Thus, the difference between aortic net forward volume and total pulmonary arterial flow (0.5 mL per R-R interval), referred to as the shunt volume, was not significant (p = 0.45).
Before undergoing surgery, patients with chronic thromboembolic pulmonary hypertension showed a mean aortic net forward volume of 44.6 mL per R-R interval (range, 26.8-70.3 mL per R-R interval). After the correction for coronary blood flow of 5% was added, the aortic net forward volume was 46.9 mL per R-R interval. The values of the pulmonary arteries were as follows. The right pulmonary artery showed a mean net forward volume of 14.2 mL per R-R interval (range, 1.7-29.5 mL per R-R interval). The blood flow of the left pulmonary artery was calculated as 15.7 mL per R-R interval (range, 7.2-29.2 mL per R-R interval). The total pulmonary arterial flow was 29.9 mL per R-R interval. These findings resulted in a mean bronchosystemic shunt volume of 17 mL per R-R interval (p < 0.001).
After undergoing surgery, patients with chronic thromboembolic pulmonary hypertension had a mean flow in the ascending aorta of 39.2 mL per R-R interval (range, 6.7-78.2 mL per R-R interval); with 5% added for coronary blood flow, the value was 41.2 mL per R-R interval. This value did not differ significantly (p = 0.5) from the preoperative value. The flow in the right pulmonary artery was 25.5 mL per R-R interval (range, 2.5-55.7 mL per R-R interval), and the flow in the left pulmonary artery was 14.8 mL per R-R interval (range, 7.9-27.2 mL per R-R interval), resulting in a total pulmonary arterial blood flow of 40.3 mL per R-R interval. The shunt volume (0.9 mL per R-R interval) revealed that the difference between pulmonary and systemic blood flow was not significant (p = 0.76). Table 1 summarizes the hemodynamic results as determined by MR flow measurements in the healthy volunteers and in patients before and after surgery. Shunt volumes were calculated by subtracting systemic and pulmonary blood flow. All values are given in milliliters per heartbeat.
|
The difference between the shunt volumes of volunteers and patients before undergoing surgery was significant (Mann-Whitney test, p < 0.001), and no difference between the shunt volumes of volunteers and patients after undergoing surgery was detected (p = 0.53). Accordingly, the shunt volume in patients after surgery had significantly decreased compared with the shunt volume in patients before surgery (Wilcoxon's test, p = 0.005).
All CT examinations were high quality, thus allowing quantitative evaluation. No artifacts caused by respiration were noted. On CT angiography, we detected an average of three dilated bronchial arteries per patient (range, from one to five arteries). The mean total cross-sectional area was 19.4 mm2 (range, 3.3-37.5 mm2). Pearson's correlation coefficient (r) between the mean total cross-sectional area and the shunt volume (i.e., the difference between aortic net forward volume and total pulmonary arterial flow) was 0.86 (p < 0.001) (Fig. 4).
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The blood flow that originates from the systemic circulation and returns through the pulmonary veins to the left side of the heart is referred to as bronchial systemic to pulmonary flow [19]. As shown by Kauczor et al. [12], these dilated bronchial arteries are a good predictor of surgical outcome. An increase in bronchial collateralization represents a long-term adaptation to pulmonary vascular obstruction [20].
CT and Bronchial Arteries
In 1851, Virchow showed that after complete ligature of one pulmonary
artery, there was no necrosis of the ligated lung. This finding indicates that
normal blood flow through the bronchial arteries is able to sufficiently
supply lung tissue [21].
Anatomic studies have shown that two or three bronchial arteries run parallel
to the larger bronchi. These arteries follow the bronchial structures down to
the alveoli, form an anastomotic network with the pulmonary arterioles, and
drain via the pulmonary veins into the left atrium. Under normal conditions,
these anastomoses are functionally closed. However, under pathologic
conditions, they open and new anastomoses form
[22,23,24].
The resulting blood flow increase in the bronchial arteries can reach a factor
of 10, as shown in animal studies
[25], and can contribute up to
25% of pulmonary oxygenation
[13].
The bronchial arteries have two functions: first, they support the trachea and structures of the mediastinum; and second, they supply the bronchi where they have an anastomotic network with the pulmonary arterioles. To selectively determine the cross-sectional area of the vessels nutrifying the lung periphery, we measured the bronchial arteries at the point where they reached the main bronchi. A study showed that a tortuous course of bronchial arteries was found in 36% of patients with chronic thromboembolic pulmonary hypertension [12]. This minor drawback causes the perpendicular measurement of vessel area to be less precise. However, this drawback can be overcome using multidetector CT scanners, which can obtain images with isotropic voxels. Vessel areas can be determined using isotropic voxels on multiplanar reformations.
To our knowledge, no reports describe the normal tapering of bronchial arteries during their mediastinal course. Bronchial arteries with a diameter of more than 1.5 mm are considered to be enlarged [12]. We used a double-threshold region of interest. This procedure resulted in an area composed of voxels with a density higher than 100 H and an upper limit of 3072 H. We worked on contrast-enhanced scans and drew the region of interest in a soft-tissue region where there was no other structure of more than 100 H (mean ± SD for mediastinal and hilar soft tissues, 70 ± 10 H).
One other technique for diagnosing and evaluating surgical operability of chronic thromboembolic pulmonary hypertension is MR angiography. In one study, all segmental pulmonary arteries and 90% of the subsegmental pulmonary arteries could be visualized after the administration of 20 mL of gadopentetate dimeglumine on an acquisition obtained in less than 300 sec [26]. However, in that study, the contrast bolus was timed for optimal opacification of the pulmonary arteries. Therefore, bronchial arteries were not visualized on these examinations. Visualization of the bronchial arteries during MR angiography should be possible using time-resolved techniques. These techniques allow acquisition of late phase images that show contrast material filling the aorta and bronchial arteries.
Shunt and MR Measurements
The most reliable technique to measure the bronchopulmonary anastomotic
flow is to measure flow invasively. Baile et al.
[10] introduced artificial
circuitry during coronary artery bypass operations in 40 patients. This
procedure enabled researchers to measure precisely and separately the
bronchopulmonary shunt volume. In this setting, these researchers were able to
quantify the shunt volume as 3.2% of cardiac output. Because these experiments
were performed during an open chest technique, this method is not suitable for
routine clinical use.
Estimation of bronchopulmonary shunt volume is possible by subtracting the output of the right ventricle from the output of the left ventricle and has been done in dogs and humans [27, 28]. In our study, the blood flow volumes were measured noninvasively using MR imaging. The applied velocity-encoded gradient-echo sequence had a temporal resolution of 110 msec enabling acquisition of six to eight phases per cardiac cycle. This temporal resolution is considered to be sufficient to yield valid results [29]. Two other studies [16, 30] have shown the accuracy of this technique for the precise measurement of pulmonary and aortic blood flow.
We prefer to measure total pulmonary arterial blood flow in the left and right pulmonary arteries rather than in the pulmonary trunk because there are fewer cardiac motion artifacts in the left and right pulmonary arteries. Moreover, measuring arterial blood flow in the left and right pulmonary arteries is easier to plan. Caputo et al. [31] found no statistical difference between the measurement of flow in the main pulmonary artery (mean ± SD, 64 ± 16 mL) and the sum of flow volumes in the right and left pulmonary arteries (mean ± SD, 63 ± 15 mL).
In general, we observed that the net forward volume in the aorta was diminished in the patients with chronic thromboembolic pulmonary hypertension compared with the healthy volunteers (47 vs 74 mL, respectively). This difference can be explained in two ways. First, the left ventricle is compressed during systole by the paradoxical septal motion [32]. Second, right heart failure is caused by high pressure in the pulmonary vasculature.
Coronary Blood Flow
Because aortic flow is measured distal to the coronary arteries, this small
volume of blood is not included in the systemic flow volume measurements. In
the literature, left cardiac output was measured as coronary blood flow under
resting conditions and ranged from a mean ± SD of 5.2% ± 1.6% in
one study [33] to a mean of
6.3% (range, 2.5-14%) in another study
[18]. Bogren et al.
[18] calculated the retrograde
flow in the aorta with MR flow measurements in the ascending aorta. These
researchers assumed a relationship between retrograde flow and coronary blood
flow, but they were unable to give definite numbers. With our technique and
the software we used, we were able to quantify the reverse volume in the
ascending aorta and calculate the coronary blood flow. Our value of 5% of
coronary blood flow is in the range of the values yielded using an invasive
measurement technique [33].
Because our volunteers and patients had to lie calmly for both examinations,
MR imaging and CT, we can assume the same resting myocardial perfusion.
In healthy volunteers, we found a difference between the aortic and pulmonary flow of 0.6% of systemic blood flow. This difference is less than the difference known from a physiologic study. In patients without lung disease, blood flow through the bronchial arteries was determined to be 1-3% of cardiac output [11]. One reason our finding differs from the reported range could be that the temporal resolution of our flow measurements did not cover the entire diastolic phase of the cardiac cycle.
Before undergoing surgery, our patients had a mean shunt volume of 36.2%, which was significantly higher than that of the volunteers (p < 0.001). Endrys et al. [15] examined eight patients with chronic thromboembolic pulmonary hypertension in whom pulmonary blood flow and systemic blood flow were measured using the contrast medium—dilution method after direct catheterization of the right and left ventricles of the heart. These researchers found a mean ± SD shunt volume of 29.8% ± 18.6% of systemic blood flow. Additionally, they observed dilated bronchial arteries in all patients with chronic thromboembolic pulmonary hypertension. These results agree with our findings of a mean of three dilated bronchial arteries per patient (range, from one to five).
In another study, decreased partial pressure of alveolar carbon dioxide and increased bronchial—pulmonary pressure gradient are discussed as reasons for increased circulation in the bronchial arteries and the opening of anastomoses [34]. This hypothesis supports our data that an increased pressure gradient is associated with an increased shunt volume and that the bronchial pulmonary anastomoses were opened.
One group of researchers has postulated that increased shunt volumes will lead to an increased volume load on the left ventricle [11]. However, our flow measurements indicated blood flow through the pulmonary arterial system in our patients was significantly decreased, and even with the increased shunt volume, the output of the left ventricle was lower in the patients than in the healthy volunteers. Therefore, we conclude that there is no volume overload of the left ventricle in patients with chronic thromboembolic pulmonary hypertension.
Postoperative Data
Our postoperative MR phase-contrast flow measurements did not reveal any
persistent substantial bronchosystemic shunt volume. Because only pulmonary
thromboendarterectomy was performed, these data strongly support the
hypothesis that the preoperative shunt volumes were exclusively caused by
increased broncho-systemic shunting. In healthy volunteers and in the patients
before surgery, blood was distributed equally to the left and right lungs. In
our patients, the flow through the left pulmonary artery after surgery equaled
that before surgery. However, flow in the right pulmonary artery increased
markedly. This finding can be explained by a higher surgical success rate on
the right side. On postoperative MR angiography of the pulmonary vasculature,
the number of visible (segmental) pulmonary arteries on the left side
increased by 20, whereas it increased by 50 on the right side
[26].
Limitations
We acknowledge several limitations of our study. It is a retrospective
study with a strong selection bias because only patients with chronic
thromboembolic pulmonary hypertension were included in the investigation. The
number of patients was limited because chronic thromboembolic pulmonary
hypertension is a rare entity.
Results of MR phase-contrast flow measurements were hampered by low temporal resolution especially during the diastolic cardiac phase. CT provided nonisotropic voxels, which prevented us from evaluating the cross-sectional areas using multiplanar reformations. If isotropic voxels had been generated and measurements had been performed using multiplanar reformations, we would have been able to more accurately determine cross-sectional areas.
Conclusion
Our study shows that there is a linear correlation between the
bronchosystemic shunt volume and the dilatation of the bronchial arteries in
patients with chronic thromboembolic pulmonary hypertension. This shunt volume
can be determined noninvasively with MR imaging—based flow measurements.
It is important to pay additional attention to the bronchial arteries when
interpreting helical CT examinations. With this additional information from
helical CT, estimation of the postoperative increase in pulmonary blood flow
is feasible.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  E. Castaner, X. Gallardo, E. Ballesteros, M. Andreu, Y. Pallardo, J. M. Mata, and L. Riera CT Diagnosis of Chronic Pulmonary Thromboembolism RadioGraphics, January 1, 2009; 29(1): 31 - 50. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Coulden State-of-the-Art Imaging Techniques in Chronic Thromboembolic Pulmonary Hypertension Proceedings of the ATS, September 1, 2006; 3(7): 577 - 583. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. M. Hoeper, E. Mayer, G. Simonneau, and L. J. Rubin Chronic Thromboembolic Pulmonary Hypertension Circulation, April 25, 2006; 113(16): 2011 - 2020. [Full Text] [PDF]
|
|
|
|
 |
|
 E. Fadel, E. Wijtenburg, R. Michel, J.-X. Mazoit, R. Bernatchez, B. Decante, E. Sage, M. Mazmanian, and P. Herve Regression of the Systemic Vasculature to the Lung after Removal of Pulmonary Artery Obstruction Am. J. Respir. Crit. Care Med., February 1, 2006; 173(3): 345 - 349. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Savai, J. C. Wolf, S. Greschus, B. G. Eul, R. T. Schermuly, J. Hanze, R. Voswinckel, A. C. Langheinrich, F. Grimminger, H. Traupe, et al. Analysis of Tumor Vessel Supply in Lewis Lung Carcinoma in Mice by Fluorescent Microsphere Distribution and Imaging with Micro- and Flat-Panel Computed Tomography Am. J. Pathol., October 1, 2005; 167(4): 937 - 946. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Remy-Jardin, A. Duhamel, V. Deken, N. Bouaziz, P. Dumont, and J. Remy Systemic Collateral Supply in Patients with Chronic Thromboembolic and Primary Pulmonary Hypertension: Assessment with Multi-Detector Row Helical CT Angiography Radiology, April 1, 2005; 235(1): 274 - 281. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K.-F. J. Kreitner, S. Ley, H.-U. Kauczor, E. Mayer, T. Kramm, M. B. Pitton, F. Krummenauer, and M. Thelen Chronic Thromboembolic Pulmonary Hypertension: Pre- and Postoperative Assessment with Breath-hold MR Imaging Techniques Radiology, August 1, 2004; 232(2): 535 - 543. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Hasegawa, P. M. Boiselle, and H. Hatabu Bronchial Artery Dilatation on MDCT Scans of Patients with Acute Pulmonary Embolism: Comparison with Chronic or Recurrent Pulmonary Embolism Am. J. Roentgenol., January 1, 2004; 182(1): 67 - 72. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
