Radiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online before print July 12, 2006, 10.1148/radiol.2403051076
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Appendix
Right arrow All Versions of this Article:
2403051076v1
240/3/858    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Nael, K.
Right arrow Articles by Finn, J. P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Nael, K.
Right arrow Articles by Finn, J. P.
(Radiology 2006;240:858-868.)
© RSNA, 2006

Thoracic Imaging

Pulmonary Circulation: Contrast-enhanced 3.0-T MR Angiography—Initial Results1

Kambiz Nael, MD, Henrik J. Michaely, MD, Ulrich Kramer, MD, Margaret H. Lee, MD, Jonathan Goldin, MD, PhD, Gerhard Laub, PhD and J. Paul Finn, MD

1 From the Department of Radiological Sciences (K.N., H.J.M., U.K., M.H.L., J.G., J.P.F.), University of California Los Angeles, 10945 Le Conte Ave, Suite 3371, Los Angeles, CA 90095-7206; and Siemens Medical Solutions, Malvern, Pa (G.L.). Received June 27, 2005; revision requested September 14; revision received September 28; accepted October 26; final version accepted December 21. Address correspondence to K.N. (e-mail: nkambiz{at}mednet.ucla.edu).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 
Purpose: To prospectively evaluate the technical feasibility of both high-spatial-resolution and time-resolved contrast material–enhanced magnetic resonance (MR) angiography of the pulmonary circulation at 3.0 T.

Materials and Methods: All examinations were HIPAA compliant. After institutional review board approval and written informed consent, time-resolved and high-spatial-resolution three-dimensional contrast-enhanced MR angiography of the pulmonary circulation was performed with a 3.0-T MR system in 31 adults (13 men, 18 women; age range, 29–87 years old): 22 volunteers and nine patients (two with mediastinal masses, seven with pulmonary arterial hypertension [PAH]). The image quality of pulmonary arterial branches and parenchymal enhancement conspicuity were evaluated independently by two radiologists. The signal-to-noise ratio and quantitative analysis of perfusion parameters was performed. Statistical analysis of data was performed by using Wilcoxon rank sum test and two-sample Student t test, and interobserver variability was tested with {kappa} coefficient.

Results: Visualization up to fourth-order pulmonary arterial branches was observed on time-resolved MR angiograms and that up to fifth-order branches was observed on high-spatial-resolution MR angiograms, with diagnostic-quality blood vessel definition and good interobserver agreement. Evaluation of parenchymal enhancement and semiquantitative analysis of perfusion parameters yielded dynamic information in all subjects. Comparative analysis of definition scores for fourth- and fifth-order pulmonary arterial branches, parenchymal enhancement, the time lag between the pulmonary arterial and parenchymal enhancement, and all of the calculated perfusion indices in patients with PAH showed statistically significant differences from volunteers (P < .05).

Conclusion: Three-dimensional contrast-enhanced MR angiography of the pulmonary circulation was feasible at 3.0 T and provided high vascular morphologic detail and dynamic functional information. Clearly detectable abnormalities were present in patients with PAH.

Supplemental material: radiology.rsnajnls.org/cgi/content/full/2403051076/DC1

© RSNA, 2006


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 
Contrast material–enhanced magnetic resonance (MR) angiography has become established as a powerful noninvasive tool for use in most vascular territories (1–4). In the lungs, computed tomographic (CT) angiography is widely regarded as the technique of choice for the work-up of acute and chronic pulmonary vascular disease. The success of CT angiography in the thorax is due, in part, to the high spatial resolution of CT, and advances in multisection technology have dramatically increased the speed and simplicity of CT angiography (5–7). It seems likely, therefore, that CT angiography will continue to have a pivotal role in the evaluation of patients with pulmonary vascular disease. There will remain, however, a subset of patients who, whether because of renal impairment or contrast agent sensitivity, are not candidates for CT angiography. In other patients, the radiation burden of CT angiography may be considered undesirable, whether because of a young age or the prospect of multiple follow-up examinations. There is, therefore, a role for MR angiography in the examination of the lungs, but the quality of modern CT angiographic examinations has set a high standard for alternative imaging techniques.

Whole-body 3.0-T MR imaging systems have become available, with the promise of sufficient signal-to-noise ratio (SNR) to generate high-spatial-resolution images throughout the body. Also, because the longitudinal relaxation time (T1) of unenhanced blood increases with higher field strength (8), sensitivity to injected gadolinium-based agents for contrast-enhanced MR angiography may be heightened. There are, however, substantial technical challenges to large–field-of-view imaging at 3.0 T. Dielectric resonances and radiofrequency eddy currents are potentially troublesome at 3.0 T and may result in inhomogeneous radiofrequency excitation within the body (9–11). Also, the quadratic dependence of radiofrequency power deposition on field strength limits the maximal flip angles and minimal repetition times when 3.0-T imaging is compared with that at 1.5 T. The balance one can strike between increased SNR and radiofrequency limitations will ultimately determine whether contrast-enhanced MR angiography of the pulmonary circulation is feasible at 3.0 T. To our knowledge, there are no prior data about pulmonary contrast-enhanced MR angiography at 3.0 T, and in consideration of both the theoretic SNR advantages and the technical challenges, the purpose of our study was to prospectively evaluate the technical feasibility of high-spatial-resolution and time-resolved contrast-enhanced MR angiography of the pulmonary circulation at 3.0 T.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 
Our study was not directly industry supported, although our institution has a research collaboration with Siemens Medical Solutions, Malvern, Pa. Data and information submitted for publication were under the control of the authors who were not employees of Siemens Medical Solutions.

Volunteers and Patients
Thirty-one subjects, including 22 adult volunteers (10 men, 12 women; mean age, 46.8 years; range, 31–67 years) and nine patients (three men, six women; mean age, 56 years; range, 29–87 years) were prospectively enrolled in this study. Volunteers had no history of cardiopulmonary disease and no cough or shortness of breath at the time of the study. A history of cardiopulmonary disease and/or contraindications to MR imaging were exclusion criteria for healthy volunteers. Indications for clinical examinations in the nine consecutive patients included evaluation of three-dimensional (3D) vascular anatomy in seven patients with pulmonary arterial hypertension (PAH) and preoperative assessment in two subjects with mediastinal masses. All examinations were Health Insurance Portability and Accountability Act compliant and were performed in accordance with institutional review board guidelines with an approved protocol. Prospective written informed consent for study participation was obtained from all subjects after the nature of the procedure had been fully explained. The study was not industry supported, and none of the authors who had control over the inclusion of patients were employees of the industry.

Image Acquisition
All examinations were performed with a 3.0-T whole-body imaging system (Magnetom Trio; Siemens Medical Solutions, Malvern, Pa), which was equipped with eight receiver channels and a fast three-axis gradient system characterized by a peak gradient amplitude of 40 mT/m and a maximal slew rate of 200 mT · m · msec–1 on each physical axis.

Subjects were positioned supine on the imaging table and were entered headfirst into the magnet. A 20-gauge cannula was sited in an antecubital vein and connected to an electronic power injector (MR Spectris; Medrad, Pittsburgh, Pa). The arms were positioned above the head, and an eight-element wraparound torso-array coil was positioned over the anterior and posterior thorax for signal reception.

After scout images were obtained, a timing bolus of 2 mL of gadodiamide (Omniscan; Amersham Health, Princeton, NJ) was used to measure transit time by using a multiphase sagittal T1-weighted gradient-echo timing sequence. The test bolus was administered at a flow rate of 1.5 mL/sec and was followed by a 20-mL saline flush administered at the same flow rate. The same contrast agent injection rate was used subsequently for high-spatial-resolution contrast-enhanced MR angiography, as will be described later. The mean delay for the transit of the contrast material to the main pulmonary artery was 7 seconds (range, 5–14 seconds).

On the basis of the localizing images, breath-hold time-resolved MR angiography was implemented in the coronal plane by using a 3D fast Fourier transform gradient-echo sequence. Temporal echo sharing, with application of three k-space segments of equal size (12), was employed, as was generalized autocalibrating partially parallel acquisitions with an acceleration factor of two and 24 reference lines in the left-to-right phase-encoding direction (13). The 3D data sets were updated every 1.5 seconds, for a total of 12–14 sequential measurements. Pulse sequence imaging parameters are detailed in Table 1.


View this table:
[in this window]
[in a new window]

  		Table 1. Imaging Parameters for Pulmonary Contrast-enhanced MR Angiography

 
A fixed dose of 6 mL of gadodiamide (corresponding to 0.035–0.055 mmol/kg) was injected at 4 mL/sec, followed by a 20-mL saline flush at the same rate. Subjects were requested to suspend respiration from the start of data acquisition, which coincided with the start of the contrast agent injection. In the image domain, magnitude subtraction of the first (unenhanced) data set from all subsequent data sets was performed online, as was on-axis full-thickness maximum intensity projection (MIP) reconstruction.

Subsequently, high-spatial-resolution contrast-enhanced MR angiography was performed in the coronal plane by using a fast spoiled gradient-echo sequence, which was integrated with generalized autocalibrating partially parallel acquisitions (Table 1). With 88 partitions, partition thickness was adapted individually to ensure coverage of the entire vascular territory and ranged between 1.4 and 2.0 mm. The combination of a 640 x 374 matrix with a 390–351-mm field of view resulted in voxel dimensions of 0.6 x 0.9 x 1–1.4 mm3.

The contrast agent injection protocol involved 0.2 mmol/kg of gadodiamide administered at a rate of 1.5 mL/sec, followed by a flush with 30 mL of saline at the same flow rate. Subjects were requested to hyperventilate before MR angiography and to hold their breath during image acquisition, which was timed to coincide with the arrival of the contrast material bolus to the main pulmonary artery.

For both time-resolved and high-spatial-resolution MR angiographic sequences, an asymmetric k-space sampling scheme was applied in all three planes to minimize the echo time and the acquisition time. Zero interpolation was performed as appropriate to the nearest binary exponent to facilitate fast Fourier transform.

All examinations were performed successfully and without complications. None of the examinations had to be repeated because of technical problems, and all subjects were able to cooperate with breath-hold instructions. The examination time, which was defined as the time from patient entry into the MR imaging suite until patient exit from the suite, was 43 minutes (range, 24–52 minutes). The average examination time was 18 minutes after the subjects were advanced into the magnet bore.

Image Processing
After data acquisition, image processing was performed at a 3D workstation (Leonardo; Siemens Medical Solutions, Malvern, Pa) with standard commercial software. The study coordinator (K.N.), who had 4 years of experience in 3D reconstruction techniques but was not involved in the subsequent image analysis, performed all reconstructions. For analysis of time-resolved MR angiograms, optimum pulmonary arterial and parenchymal enhancement phases were selected from the time series. These data sets, in addition to the entire 3D volume from high-spatial-resolution MR angiography, were reconstructed in thin MIP subvolumes that were 10 mm thick and that overlapped by 9 mm. In addition, 36 full-thickness rotational MIPs, which included 360° of rotation in 10° increments, were also reconstructed for high-spatial-resolution data. Postprocessing time was approximately 15 minutes per subject.

Image Analysis
Qualitative analysis.—MR angiograms were interpreted independently by two fellowship-trained thoracic radiologists (J.G. and M.H.L., with 10 and 5 years of experience, respectively) who were both blinded to the subjects' identities and clinical data. Separate image reading sessions were organized for both readers by the study coordinator (K.N.), who attended all reading sessions. The readers were instructed to use the postprocessed data in a first step and, if necessary, to use the source data for interactive reformatting in a second step.

Pulmonary arterial branches observed at high-spatial-resolution MR angiography and during the pulmonary arterial phase at time-resolved MR angiography were evaluated in each subject in the upper and lower lobes bilaterally to the highest branch order visualized (main right pulmonary artery or left pulmonary artery, first division of the right pulmonary artery or the left pulmonary artery, second division of the right pulmonary artery or the left pulmonary artery [segmental branches], third division of the right pulmonary artery or the left pulmonary artery [subsegmental branches], fourth division [subsubsegmental branches], and fifth division; hereafter, these divisions will be referred to as first-, second-, third-, fourth-, and fifth-order branches, respectively).

Each observer evaluated time-resolved and high-spatial-resolution MR angiograms in separate reading sessions, in a blinded manner, and in random order. Each visualized branch was analyzed for image quality with regard to how clearly the vessel was defined by using a scale of 1–4 (score 1, poor image quality and blurring of the arterial segment; score 2, fair image quality but inadequate arterial enhancement for confident diagnosis; score 3, good image quality and arterial enhancement and adequate definition for confident diagnosis; and score 4, excellent image quality and arterial enhancement for highly confident diagnosis).

Image quality of an arterial segment was rated to be diagnostic (score of ≥3) if the reviewers were confident that clinically relevant diagnostic information was visible with clear discrimination between the blood vessel and background tissue. Image quality was considered nondiagnostic (score of ≤2) if the vessel was blurred or there was inadequate vessel enhancement. The presence of any vascular abnormality, artifact, or incidental finding was recorded for each individual.

Parenchymal enhancement was assigned a score on a scale of 1–4 as follows: score 1, no clear delineation of lung borders from adjacent chest wall; score 2, delineation of lung parenchyma from adjacent chest wall but reader would not be confident with the diagnosis of a perfusion defect; score 3, parenchymal enhancement that was believed to be adequate for confident diagnosis of perfusion defects; and score 4, very sharp definition of lung fissures that was believed to be adequate for very confident diagnosis of perfusion defects.

Quantitative analysis.—For SNR evaluation, quantitative analysis of signal intensity (SI) was performed on both time-resolved and high-spatial-resolution MR angiograms by one observer (K.N.).

On time-resolved MR angiograms, by selecting the temporal MIPs, regions of interest (ROIs) of at least 200 pixels (range, 200–400 pixels) were placed over the main pulmonary artery, and ROIs of at least 3000 pixels (range, 3000–6000 pixels) were placed over both lung fields. The SI values were measured for each region. In addition, the time between peak enhancement of the main pulmonary artery and that of the parenchyma was calculated (Fig 1).


Figure 1
View larger version (133K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1a: (a) Coronal image and (b) graph illustrate the process of temporal analysis of the SI values of the lung parenchyma and pulmonary artery (PA). ROIs are placed over both lungs and main pulmonary artery on a coronal MIP image. SI-time curve shows the peak enhancement of the pulmonary artery, which was immediately followed by the peak enhancement of the lung parenchyma. Lt = left, Rt = right.

 

Figure 1
View larger version (24K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1b: (a) Coronal image and (b) graph illustrate the process of temporal analysis of the SI values of the lung parenchyma and pulmonary artery (PA). ROIs are placed over both lungs and main pulmonary artery on a coronal MIP image. SI-time curve shows the peak enhancement of the pulmonary artery, which was immediately followed by the peak enhancement of the lung parenchyma. Lt = left, Rt = right.

 
In high-spatial-resolution MR angiography, ROIs were placed at the following intravascular sites: first-, second-, third-, fourth-, and fifth-order pulmonary branches in upper and lower lobes. The SI values were calculated. The sizes of these ROIs were adjusted to include as much of the vascular lumen as possible.

The calculated SI values were divided by background noise to measure the SNR for each region. In an effort to minimize the nonuniformity of noise distribution across the field of view as a potential result of parallel image acquisition (14), evaluation of background noise was performed as follows: Six ROIs were placed on the extracorporeal background (two at the top, two at the middle, and two at the bottom of the field of view), and the mean SI ± standard deviation of noise was measured in these six regions. The SNR values were calculated by dividing the SI values of ROIs by the mean standard deviation of the background noise (mean standard deviation of these six regions).

For evaluation of dynamic enhancement, dynamic analysis was performed by one observer (H.J.M.), with 3 years of experience, by using prerelease analytic software (MERZ, research version 0.92; Siemens Medical Solutions, Erlangen, Germany). This software allows an ROI-based evaluation of contrast agent enhancement on a pixel-by-pixel basis, and such an evaluation results in semiquantitative analysis of perfusion indices that include the following: time to peak, mean transit time, maximal SI, and maximal upslope of the curve (Appendix E1 [radiology.rsnajnls.org/cgi/content/full/2403051076/DC1]).

All quantitative results of perfusion indices were plotted as the mean ± standard deviation. Pulmonary perfusion parameters were displayed as color-coded maps for visual assessment of regional perfusion. To compensate for intraobserver variability, the mean of two ROIs of the pulmonary periphery was used for further analysis.

Statistical Analysis
Image quality scores, as well as SNR values, were plotted as the mean ± standard deviation. Interobserver agreement for definition of pulmonary arterial branches within each imaging technique between the readers was determined by calculating {kappa} values with the {kappa} coefficient ({kappa} = 0, poor agreement; {kappa} = 0.01–0.20, slight agreement; {kappa} = 0.21–0.40, fair agreement; {kappa} = 0.41–0.60, moderate agreement; {kappa} = 0.61–0.80, good agreement; and {kappa} = 0.81–1.00, excellent agreement) (15). A Wilcoxon rank sum test was used to evaluate the significance of the differences in assigned scores of pulmonary arterial branches and parenchymal enhancement between volunteers and patients with PAH. A two-sample Student t test was used to evaluate the significance of differences of perfusion parameters and time lag from the peak enhancement of the main pulmonary artery and lung parenchyma between the volunteers and patients with PAH. A two-sided value of P < .05 was used as the criterion to indicate a statistically significant difference.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 
Time-resolved MR Angiography
The pulmonary arteries were routinely visualized up to fourth-order branches (Fig 2), with definition in the diagnostic range (Table 2). When the results of assignment of scores were evaluated by using the Wilcoxon rank sum test, there was no significant difference between the readers (P = .2). The {kappa} coefficient revealed good interobserver agreement ({kappa} = 0.74).


Figure 2
View larger version (79K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2: Coronal MIP images from 3D time-resolved MR angiography (2/0.84; flip angle, 17°), with a sample interval of 1.5 seconds, show sequential filling of pulmonary arteries, pulmonary parenchyma, and systemic arteries.

 

View this table:
[in this window]
[in a new window]

  		Table 2. Definition of Pulmonary Arterial Branches on Time-resolved MR Angiograms

 
Parenchymal enhancement was identified in all subjects, with scores for definition in the diagnostic range (range, 2–4; median, 3) for both readers. There was no significant difference between the readers when the results of assignment of scores were evaluated by using the Wilcoxon rank sum test (P = .78). The {kappa} coefficient revealed excellent interobserver agreement ({kappa} = 0.86). SNR values of the main pulmonary artery and the parenchymal perfusion phase of the lungs are shown in Table 3.


View this table:
[in this window]
[in a new window]

  		Table 3. SNR Analysis on Time-resolved MR Angiograms

 
High-Spatial-Resolution MR Angiography
On high-spatial-resolution MR angiograms (Fig 3), the pulmonary arteries were routinely identified up to fifth-order branches, with definition in the diagnostic range (Table 4). When the results of assignment of scores were evaluated by using the Wilcoxon rank sum test, there was no significant difference between the readers (P = .54). The {kappa} coefficient revealed good interobserver agreement ({kappa} = 0.70).


Figure 3
View larger version (115K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3a: (a) Coronal thin (20-mm-thick) (b) and full-thickness MIPs from high-spatial-resolution contrast-enhanced MR angiography (3/1.1; flip angle, 22°) in a volunteer reveal up to the fifth-order pulmonary branches with good definition. (c) Image shows the order (1st–5th) of the pulmonary arterial branches in left upper lobe.

 

Figure 3
View larger version (168K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3b: (a) Coronal thin (20-mm-thick) (b) and full-thickness MIPs from high-spatial-resolution contrast-enhanced MR angiography (3/1.1; flip angle, 22°) in a volunteer reveal up to the fifth-order pulmonary branches with good definition. (c) Image shows the order (1st–5th) of the pulmonary arterial branches in left upper lobe.

 

Figure 3
View larger version (131K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3c: (a) Coronal thin (20-mm-thick) (b) and full-thickness MIPs from high-spatial-resolution contrast-enhanced MR angiography (3/1.1; flip angle, 22°) in a volunteer reveal up to the fifth-order pulmonary branches with good definition. (c) Image shows the order (1st–5th) of the pulmonary arterial branches in left upper lobe.

 

View this table:
[in this window]
[in a new window]

  		Table 4. Definition of Pulmonary Arterial Branches on High-Spatial-Resolution MR Angiograms

 
The SNR values of the pulmonary arterial branch orders on high-spatial-resolution MR angiograms are summarized in Table 5. In two patients with mediastinal masses, pulmonary MR angiography revealed displacement of pulmonary vessels around the mass, without invasion (Fig 4).


View this table:
[in this window]
[in a new window]

  		Table 5. SNR Analysis on High-Spatial-Resolution MR Angiograms

 

Figure 4
View larger version (166K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4a: (a) Unenhanced coronal two-dimensional fat-suppressed T1-weighted spoiled gradient-echo (180/1.8; flip angle, 64°) image reveals an encapsulated well-defined right thyroid mass (arrow). (b) Coronal full-thickness MIP from high-spatial-resolution contrast-enhanced MR angiography (3/1.1; flip angle, 20°) shows that adjacent vascular structures are displaced and not invaded. Note the T2* artifact causing signal loss in the right subclavian artery.

 

Figure 4
View larger version (133K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4b: (a) Unenhanced coronal two-dimensional fat-suppressed T1-weighted spoiled gradient-echo (180/1.8; flip angle, 64°) image reveals an encapsulated well-defined right thyroid mass (arrow). (b) Coronal full-thickness MIP from high-spatial-resolution contrast-enhanced MR angiography (3/1.1; flip angle, 20°) shows that adjacent vascular structures are displaced and not invaded. Note the T2* artifact causing signal loss in the right subclavian artery.

 
A mild respiratory artifact was noticed in two subjects in whom breathing motion affected the quality of the fourth- and fifth-order branches in lower lobes, although high vessel definition in the upper lobes remained intact. In eight subjects, both reviewers noted some degree of signal loss in the subclavian artery on the side of the contrast agent injection.

Volunteers versus Patients with PAH
All seven patients with PAH were identified correctly by both observers at the review of both time-resolved and high-spatial-resolution contrast-enhanced MR angiograms. They noted dilatation of the central pulmonary arteries and peripheral pruning, as previously reported (16,17), and delayed pulmonary arteriovenous transit of contrast agent (Fig 5). In evaluation of time-resolved MR angiograms, fourth-order pulmonary arterial branches were assigned scores (range, 1–3; median, 2) that were significantly lower (P = .008) in patients with PAH in comparison with the scores assigned in volunteers. In evaluation of high-spatial-resolution MR angiograms, fourth- and fifth-order pulmonary arterial branches were assigned scores (range, 1–3; median, 2) that were significantly lower (P < .001) in patients with PAH in comparison with the scores assigned in volunteers.


Figure 5
View larger version (146K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5a: (a) Coronal thin (20-mm-thick) and (b) full-thickness MIPs from high-spatial-resolution pulmonary contrast-enhanced MR angiography (3/1.1; flip angle, 22°) show proximal dilatation of the central pulmonary arteries in a patient with PAH. Because the pulmonary transit time was prolonged, pulmonary venous enhancement was delayed, as shown in b. (c) Full-volume MIP shows comparative pulmonary transit time and pulmonary venous enhancement in a volunteer. Note hepatic venous reflux (arrow in a) in patient with PAH.

 

Figure 5
View larger version (156K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5b: (a) Coronal thin (20-mm-thick) and (b) full-thickness MIPs from high-spatial-resolution pulmonary contrast-enhanced MR angiography (3/1.1; flip angle, 22°) show proximal dilatation of the central pulmonary arteries in a patient with PAH. Because the pulmonary transit time was prolonged, pulmonary venous enhancement was delayed, as shown in b. (c) Full-volume MIP shows comparative pulmonary transit time and pulmonary venous enhancement in a volunteer. Note hepatic venous reflux (arrow in a) in patient with PAH.

 

Figure 5
View larger version (135K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5c: (a) Coronal thin (20-mm-thick) and (b) full-thickness MIPs from high-spatial-resolution pulmonary contrast-enhanced MR angiography (3/1.1; flip angle, 22°) show proximal dilatation of the central pulmonary arteries in a patient with PAH. Because the pulmonary transit time was prolonged, pulmonary venous enhancement was delayed, as shown in b. (c) Full-volume MIP shows comparative pulmonary transit time and pulmonary venous enhancement in a volunteer. Note hepatic venous reflux (arrow in a) in patient with PAH.

 
The definition of parenchymal enhancement was assigned consistently lower scores (range, 1–3; median, 2) in patients with PAH in comparison with the scores (range, 2–4; median, 3) assigned in the volunteers, with a statistically significant difference (P < .001).

SI time curves in volunteers showed rapid transit of the contrast agent bolus through the pulmonary circulation. Evaluation of the time lag between the peak enhancement of the main pulmonary artery and lung parenchyma in volunteers and in patients with PAH revealed that it was significantly longer in patients with PAH (P < .001) (Fig 6). Time to peak and mean transit time were significantly higher and maximal SI and maximal upslope of the curve were significantly lower in patients with PAH (P < .002) (Table 6). Figure 7 shows the maps of perfusion parameters in a patient with PAH.


Figure 6
View larger version (8K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 6: Graph shows comparison of perfusion indices in volunteers and patients with PAH. Mean values of time lags of peak enhancement of lung parenchyma after the peak enhancement of pulmonary artery, time to peak (TTP), and mean transit time (MTT). The error bars indicate the standard deviation. All values (*) were significantly longer in patients with PAH than they were in volunteers (P < .001).

 

View this table:
[in this window]
[in a new window]

  		Table 6. Semiquantitative Perfusion Indices in Volunteers and Patients with PAH

 

Figure 7
View larger version (80K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 7a: (a–d) Maps of pulmonary perfusion indices in a patient with PAH in right and left lungs, respectively: (a) time to peak (TTP), 8.1 and 7.8 seconds; (b) mean transit time (MTT), 10.8 and 11.1 seconds; (c) maximal SI (MSI), 167.7 and 153.3 arbitrary units; and (d) maximal upslope of the curve (MUS), 36.2 and 33.7 arbitrary units per second. Prolonged mean transit time and decreased maximal SI can be appreciated in heterogeneous pattern throughout the lung.

 

Figure 7
View larger version (80K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 7b: (a–d) Maps of pulmonary perfusion indices in a patient with PAH in right and left lungs, respectively: (a) time to peak (TTP), 8.1 and 7.8 seconds; (b) mean transit time (MTT), 10.8 and 11.1 seconds; (c) maximal SI (MSI), 167.7 and 153.3 arbitrary units; and (d) maximal upslope of the curve (MUS), 36.2 and 33.7 arbitrary units per second. Prolonged mean transit time and decreased maximal SI can be appreciated in heterogeneous pattern throughout the lung.

 

Figure 7
View larger version (80K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 7c: (a–d) Maps of pulmonary perfusion indices in a patient with PAH in right and left lungs, respectively: (a) time to peak (TTP), 8.1 and 7.8 seconds; (b) mean transit time (MTT), 10.8 and 11.1 seconds; (c) maximal SI (MSI), 167.7 and 153.3 arbitrary units; and (d) maximal upslope of the curve (MUS), 36.2 and 33.7 arbitrary units per second. Prolonged mean transit time and decreased maximal SI can be appreciated in heterogeneous pattern throughout the lung.

 

Figure 7
View larger version (92K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 7d: (a–d) Maps of pulmonary perfusion indices in a patient with PAH in right and left lungs, respectively: (a) time to peak (TTP), 8.1 and 7.8 seconds; (b) mean transit time (MTT), 10.8 and 11.1 seconds; (c) maximal SI (MSI), 167.7 and 153.3 arbitrary units; and (d) maximal upslope of the curve (MUS), 36.2 and 33.7 arbitrary units per second. Prolonged mean transit time and decreased maximal SI can be appreciated in heterogeneous pattern throughout the lung.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
ADVANCES IN KNOWLEDGE
 References
 
To our knowledge, contrast-enhanced MR angiography of the pulmonary circulation at 3.0 T previously has not been described. The results of this preliminary study suggest strongly that pulmonary contrast-enhanced MR angiography at 3.0 T is feasible, can depict the pulmonary vasculature in encouraging detail, and can provide supplementary dynamic functional information. High-spatial-resolution MR angiography successfully showed up to fifth-order pulmonary arterial branches with high confidence, whereas time-resolved MR angiography repeatedly showed up to fourth-order pulmonary arterial branches and allowed visualization of the transient phase of pulmonary parenchymal enhancement. When combined with the higher baseline SNR, the parallel acquisition strategies implemented in the present study resulted in performance that was superior to that described in previous reports of imaging at 1.5 T (18–22).

Analysis of our perfusion data provided several parameters of physiologic relevance, including the time to peak, mean transit time, maximal SI, and maximal upslope of the curve, and our results are consistent with those in earlier reports of imaging at 1.5 T (23,24). The narrow range of values found in healthy subjects and the marked differences between the healthy subjects and patients with PAH suggest that these parameters may have useful discriminatory power for certain disease states. Further clinical evaluation, however, is necessary for confirmation.

Anticipated problems for large–field-of-view imaging at high magnetic field strength include radiofrequency nonuniformity and magnetic susceptibility effects, which may result in regional shading and focal T2*-induced signal loss, respectively. The maximal flip angles used in the current study for short–repetition time MR angiography were limited to relatively low values by Food and Drug Administration mandates on specific absorption rate. In turn, regional signal nonuniformity was not noticeable, a finding that is likely caused by the relatively small flip angles employed. Although T2*-induced signal loss caused artifact over the aortic arch and/or subclavian artery in eight subjects because of a locally high gadolinium concentration in the adjacent subclavian or innominate veins, visualization of even small pulmonary vascular branches was excellent, a finding that was likely caused by the very short echo time.

The main advantage of MR imaging at 3.0 T is the SNR gain that scales approximately linearly with field strength (25). Willinek et al (25) and Campeau et al (26) have shown that higher SNR at 3.0 T is an important element for increasing spatial resolution and would likely improve visualization of small blood vessel segments.

Several technical approaches have been described for acquiring time-resolved MR angiograms. By using sequences with a short repetition time, in combination with sparse k-space sampling, much faster imaging times can be achieved with fewer partitions to produce quasiprojectional 3D MR angiograms with frame durations on the order of 0.5 second (27) to 4.0 seconds (28,29). Another approach for faster image acquisition in time-resolved MR angiography can be achieved by reducing the size of k-space, which is regularly updated (12). A third approach to increasing data acquisition speed is to use parallel acquisition techniques (13,14,30). In our study, we combined all of these elements for time-resolved MR angiography of the pulmonary circulation.

Although parallel acquisition can improve both temporal and spatial resolution, SNR may become limiting, depending on the degree of k-space undersampling (30,31). SNR increases with increasing voxel size and the square root of the imaging time, so increasing spatial resolution and speeding image acquisition may push the limits of SNR. It seems logical that a higher field strength may be more tolerant of parallel acquisition, where SNR has a higher baseline level (25,26,32). Also, because the longitudinal relaxation time (T1) of unenhanced blood increases with higher field strength (8), sensitivity to injected gadolinium-based contrast agents for contrast-enhanced MR angiography may be heightened. A contrast agent dose as small as 0.03–0.07/kg has been used successfully for time-resolved MR angiography at 1.5 T (33). For accurate perfusion analysis of the first pass of contrast agent through the lungs, the signal response to tracer concentration should be linear. Therefore, a sufficiently small and rapid bolus should be used such that the signal change within a voxel remains proportional to the gadolinium concentration in the blood. A dose of 6 mL (approximately 0.035–0.055 mmol/kg), once distributed in the pulmonary blood pool, ensures that the concentration is well below the nonlinear range.

Uematsu et al (34) first reported high-spatial-resolution pulmonary MR angiography at high field strength (4 T), but they found that their image quality was inferior to that of pulmonary MR angiography at 1.5 T. One possible explanation for their results is signal loss due to magnetic susceptibility gradients of the lungs caused by multiple air–soft tissue interfaces. Since the susceptibility effect increases linearly with an increase in magnetic field, the T2* of the lung is expected to be smaller by a factor of two at 3.0 T compared with that at 1.5 T. To offset the effects of reduced T2* and the concomitant loss of signal, it is important to minimize the echo time of the gradient-echo sequence (18,35) and the voxel dimensions of the image. The sequence employed by Uematsu et al (34) included 6.1/1.3, with a matrix of 256 x 192. In our study, magnetic susceptibility gradients were not limiting, in part because of our relatively short echo time (0.84 and 1.13 msec for time-resolved and high-spatial-resolution MR angiography, respectively), and signal was well preserved in pulmonary arterial branches. In our study, although T2* effects did not degrade the quality of the pulmonary vessels, these effects did affect the evaluation of the subclavian artery in eight subjects on the side of the intravenous injection. Although this phenomenon also occurs at 1.5 T and has been addressed by decreasing the injection rate to decrease the gadolinium concentration (36), it is likely to be more prominent at 3.0 T for the reasons discussed previously.

Our study had several limitations. First, the study population comprised mainly volunteers and only nine clinical patients, so we were unable to address the accuracy of the technique for its clinical purposes at this time. We did not examine any patient who had a known pulmonary embolism or who was suspected of having a pulmonary embolism, which may be an important clinical application for pulmonary contrast-enhanced MR angiography. Second, the 20-second breath hold used in the current study may be limiting in patients with respiratory distress. Third, since our perfusion measurements are based on tracer kinetic and indicator dilution theory (18,37,38), assumptions were involved and they have limitations. As noted by Weisskoff et al (39), application of an indicator dilution method to the measurements of tissue perfusion provides only a semiquantitative rather than an absolute assessment. Depending on its size and location, a pixel may encompass more than one vascular compartment. We believe, however, that by achieving a higher in-plane resolution (pixel size, 1.6 x 1.2 mm2) in comparison with that in previous studies (21,23,40), the likely errors are small.

In conclusion, contrast-enhanced MR angiography of the lungs that incorporates both high-spatial-resolution and high-temporal-resolution protocols is feasible at 3.0 T. Quantitative analysis of dynamic enhancement data yielded multiple perfusion indices. Initial results show clearly detectable abnormalities in patients with PAH.


   ADVANCES IN KNOWLEDGE
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
References
 


   FOOTNOTES
 

Abbreviations: MIP = maximum intensity projection • PAH = pulmonary arterial hypertension • ROI = region of interest • SI = signal intensity • SNR = signal-to-noise ratio • 3D = three-dimensional

See Materials and Methods for pertinent disclosures.

Author contributions: Guarantors of integrity of entire study, K.N., J.P.F.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; manuscript final version approval, all authors; literature research, K.N., H.J.M.; clinical studies, K.N., H.J.M., U.K., M.H.L., J.G., G.L., J.P.F.; statistical analysis, K.N., H.J.M.; and manuscript editing, K.N., U.K., M.H.L., G.L., J.P.F.


   References
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 

  1. Ho VB, Corse WR. MR angiography of the abdominal aorta and peripheral vessels. Radiol Clin North Am 2003;41:115–144.[CrossRef][Medline]
  2. Levy RA, Prince MR. Arterial-phase three-dimensional contrast-enhanced MR angiography of the carotid arteries. AJR Am J Roentgenol 1996;167:211–215.[Abstract/Free Full Text]
  3. Prince MR, Narasimham DL, Stanley JC, et al. Breath-hold gadolinium-enhanced MR angiography of the abdominal aorta and its major branches. Radiology 1995;197:785–792.[Abstract/Free Full Text]
  4. Prince MR, Narasimham DL, Jacoby WT, et al. Three-dimensional gadolinium-enhanced MR angiography of the thoracic aorta. AJR Am J Roentgenol 1996;166:1387–1397.[Abstract/Free Full Text]
  5. Alkadhi H, Wildermuth S, Desbiolles L, et al. Vascular emergencies of the thorax after blunt and iatrogenic trauma: multi-detector row CT and three-dimensional imaging. RadioGraphics 2004;24:1239–1255.[Abstract/Free Full Text]
  6. Remy-Jardin M, Bouaziz N, Dumont P, Brillet PY, Bruzzi J, Remy J. Bronchial and nonbronchial systemic arteries at multi-detector row CT angiography: comparison with conventional angiography. Radiology 2004;233:741–749.[Abstract/Free Full Text]
  7. Revel MP, Petrover D, Hernigou A, Lefort C, Meyer G, Frija G. Diagnosing pulmonary embolism with four-detector row helical CT: prospective evaluation of 216 outpatients and inpatients. Radiology 2005;234:265–273.[Abstract/Free Full Text]
  8. Rinck PA, Muller RN. Field strength and dose dependence of contrast enhancement by gadolinium-based MR contrast agents. Eur Radiol 1999;9:998–1004.[CrossRef][Medline]
  9. Robitaille PM. On RF power and dielectric resonances in UHF MRI. NMR Biomed 1999;12:318–319.[CrossRef][Medline]
  10. Kangarlu A, Baertlein BA, Lee R, et al. Dielectric resonance phenomena in ultra high field MRI. J Comput Assist Tomogr 1999;23:821–831.[CrossRef][Medline]
  11. Ibrahim TS, Lee R, Abduljalil AM, Baertlein BA, Robitaille PM. Dielectric resonances and B(1) field inhomogeneity in UHFMRI: computational analysis and experimental findings. Magn Reson Imaging 2001;19:219–226.[CrossRef][Medline]
  12. Korosec FR, Frayne R, Grist TM, Mistretta CA. Time-resolved contrast-enhanced 3D MR angiography. Magn Reson Med 1996;36:345–351.[Medline]
  13. Griswold MA, Jakob PM, Heidemann RM, et al. Generalized autocalibrating partially parallel acquisitions (GRAPPA). Magn Reson Med 2002;47:1202–1210.[CrossRef][Medline]
  14. Pruessmann KP, Weiger M, Scheidegger MB, Boesiger P. SENSE: sensitivity encoding for fast MRI. Magn Reson Med 1999;42:952–962.[CrossRef][Medline]
  15. Landis JR, Koch GG. An application of hierarchical kappa-type statistics in the assessment of majority agreement among multiple observers. Biometrics 1977;33:363–374.[CrossRef][Medline]
  16. Kruger S, Haage P, Hoffmann R, et al. Diagnosis of pulmonary arterial hypertension and pulmonary embolism with magnetic resonance angiography. Chest 2001;120:1556–1561.[Abstract/Free Full Text]
  17. Ley S, Kreitner KF, Fink C, Heussel CP, Borst MM, Kauczor HU. Assessment of pulmonary hypertension by CT and MR imaging. Eur Radiol 2004;14:359–368.[CrossRef][Medline]
  18. Hatabu H, Tadamura E, Levin DL, et al. Quantitative assessment of pulmonary perfusion with dynamic contrast-enhanced MRI. Magn Reson Med 1999;42:1033–1038.[CrossRef][Medline]
  19. Geva T, Greil GF, Marshall AC, Landzberg M, Powell AJ. Gadolinium-enhanced 3-dimensional magnetic resonance angiography of pulmonary blood supply in patients with complex pulmonary stenosis or atresia: comparison with x-ray angiography. Circulation 2002;106:473–478.[Abstract/Free Full Text]
  20. Hoffmann U, Schima W, Herold C. Pulmonary magnetic resonance angiography. Eur Radiol 1999;9:1745–1754.[CrossRef][Medline]
  21. Matsuoka S, Uchiyama K, Shima H, et al. Effect of the rate of gadolinium injection on magnetic resonance pulmonary perfusion imaging. J Magn Reson Imaging 2002;15:108–113.[CrossRef][Medline]
  22. Fink C, Bock M, Puderbach M, Schmahl A, Delorme S. Partially parallel three-dimensional magnetic resonance imaging for the assessment of lung perfusion: initial results. Invest Radiol 2003;38:482–488.[CrossRef][Medline]
  23. Fink C, Risse F, Buhmann R, et al. Quantitative analysis of pulmonary perfusion using time-resolved parallel 3D MRI: initial results. Rofo 2004;176:170–174.[Medline]
  24. Levin DL, Chen Q, Zhang M, Edelman RR, Hatabu H. Evaluation of regional pulmonary perfusion using ultrafast magnetic resonance imaging. Magn Reson Med 2001;46:166–171.[CrossRef][Medline]
  25. Willinek WA, Born M, Simon B, et al. Time-of-flight MR angiography: comparison of 3.0-T imaging and 1.5-T imaging—initial experience. Radiology 2003;229:913–920.[Abstract/Free Full Text]
  26. Campeau NG, Huston J 3rd, Bernstein MA, Lin C, Gibbs GF. Magnetic resonance angiography at 3.0 Tesla: initial clinical experience. Top Magn Reson Imaging 2001;12:183–204.[CrossRef][Medline]
  27. Finn JP, Baskaran V, Carr JC, et al. Thorax: low-dose contrast-enhanced three-dimensional MR angiography with subsecond temporal resolution—initial results. Radiology 2002;224:896–904.[Abstract/Free Full Text]
  28. Goyen M, Laub G, Ladd ME, et al. Dynamic 3D MR angiography of the pulmonary arteries in under four seconds. J Magn Reson Imaging 2001;13:372–377.[CrossRef][Medline]
  29. Carr JC, Laub G, Zheng J, Pereles FS, Finn JP. Time-resolved three-dimensional pulmonary MR angiography and perfusion imaging with ultrashort repetition time. Acad Radiol 2002;9:1407–1418.[CrossRef][Medline]
  30. Sodickson DK, McKenzie CA, Li W, Wolff S, Manning WJ, Edelman RR. Contrast-enhanced 3D MR angiography with simultaneous acquisition of spatial harmonics: a pilot study. Radiology 2000;217:284–289.[Abstract/Free Full Text]
  31. Ohliger MA, Grant AK, Sodickson DK. Ultimate intrinsic signal-to-noise ratio for parallel MRI: electromagnetic field considerations. Magn Reson Med 2003;50:1018–1030.[CrossRef][Medline]
  32. Pruessmann KP. Parallel imaging at high field strength: synergies and joint potential. Top Magn Reson Imaging 2004;15:237–244.[CrossRef][Medline]
  33. Frayne R, Grist TM, Swan JS, Peters DC, Korosec FR, Mistretta CA. 3D MR DSA: effects of injection protocol and image masking. J Magn Reson Imaging 2000;12:476–487.[CrossRef][Medline]
  34. Uematsu H, Dougherty L, Takahashi M, et al. Pulmonary MR angiography with contrast agent at 4 Tesla: a preliminary result. Magn Reson Med 2001;46:1028–1030.[CrossRef][Medline]
  35. Bergin CJ, Pauly JM, Macovski A. Lung parenchyma: projection reconstruction MR imaging. Radiology 1991;179:777–781.[Abstract/Free Full Text]
  36. Bernstein MA, Huston J 3rd, Lin C, Gibbs GF, Felmlee JP. High-resolution intracranial and cervical MRA at 3.0T: technical considerations and initial experience. Magn Reson Med 2001;46:955–962.[CrossRef][Medline]
  37. Tofts PS. Modeling tracer kinetics in dynamic Gd-DTPA MR imaging. J Magn Reson Imaging 1997;7:91–101.[Medline]
  38. Thompson HK Jr, Starmer CF, Whalen RE, McIntosh HD. Indicator transit time considered as a gamma variate. Circ Res 1964;14:502–515.[Free Full Text]
  39. Weisskoff RM, Chesler D, Boxerman JL, Rosen BR. Pitfalls in MR measurement of tissue blood flow with intravascular tracers: which mean transit time? Magn Reson Med 1993;29:553–558.[Medline]
  40. Matsuoka S, Uchiyama K, Shima H, et al. Detectability of pulmonary perfusion defect and influence of breath holding on contrast-enhanced thick-slice 2D and on 3D MR pulmonary perfusion images. J Magn Reson Imaging 2001;14:580–585.[CrossRef][Medline]




This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Appendix
Right arrow All Versions of this Article:
2403051076v1
240/3/858    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Nael, K.
Right arrow Articles by Finn, J. P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Nael, K.
Right arrow Articles by Finn, J. P.