| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Thoracic Imaging |
1 From the Departments of Clinical Radiology (K.N., S.O.S., U.A., J.S., O.D., A.H., M.F.R.), Nuclear Medicine (F.R.), and Internal Medicine I (H.L., R.B., J.B.), Ludwig-Maximilians-University Munich, Klinikum Grosshadern, Marchioninistr 15, 81377 Munich, Germany; Siemens Medical Solutions, Erlangen, Germany (B.K.); and Radiologic Center Munich-Pasing, Munich, Germany (J.S.). From the 2002 RSNA Annual Meeting. Received March 16, 2004; revision requested May 25; revision received August 11; accepted September 8. Address correspondence to K.N. (e-mail: konstantin.nikolaou{at}med.uni-muenchen.de).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
MATERIALS AND METHODS: The study was approved by the institution's internal review board, and all patients gave written consent prior to participation. A total of 29 patients (16 women; mean age, 54 years ± 17 [± standard deviation]; 13 men; mean age, 57 years ± 15) with known pulmonary hypertension were examined with a 1.5-T MR imager. MR perfusion imaging (temporal resolution, 1.1 seconds per phase) and MR angiography (matrix, 512; voxel size, 1.0 x 0.7 x 1.6 mm) were performed with parallel acquisition techniques. Dynamic perfusion images and reformatted three-dimensional MR angiograms were analyzed for occlusive and nonocclusive changes of the pulmonary arteries, including perfusion defects, caliber irregularities, and intravascular thrombi. MR perfusion imaging results were compared with those of radionuclide perfusion scintigraphy, and MR angiography results were compared with those of digital subtraction angiography (DSA) and/or contrast material–enhanced multi–detector row computed tomography (CT). Sensitivity, specificity, and diagnostic accuracy of MR perfusion imaging and MR angiography were calculated. Receiver operator characteristic analyses were performed to compare the diagnostic value of MR angiography, MR perfusion imaging, and both modalities combined. For MR angiography and MR perfusion imaging, 
values were used to assess interobserver agreement.
RESULTS: A correct diagnosis was made in 26 (90%) of 29 patients by using this comprehensive MR imaging protocol. Results of MR perfusion imaging demonstrated 79% agreement (ie, identical diagnosis on a per-patient basis) with those of perfusion scintigraphy, and results of MR angiography demonstrated 86% agreement with those of DSA and/or CT angiography. Interobserver agreement was good for both MR perfusion imaging and MR angiography ( = 0.63 and 0.70, respectively).
CONCLUSION: The combination of fast MR perfusion imaging and high-spatial-resolution MR angiography with parallel acquisition techniques enables the differentiation of PPH from CTEPH with high accuracy.
© RSNA, 2005
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
A prerequisite for the correct and reliable diagnosis of CTEPH is the depiction of occluding thrombotic material and concomitant perfusion defects. Until recently, pulmonary perfusion could be assessed only by using radionuclide perfusion scintigraphy and conventional pulmonary angiography. While the former has substantial limitations with respect to spatial and temporal resolution, the latter requires invasive catheterization of the right side of the heart. Also, conventional angiography is limited to two-dimensional projection images. Pulmonary magnetic resonance (MR) angiography is a promising noninvasive imaging technique. Authors of initial studies have reported convincing sensitivities and specificities in the diagnosis of pulmonary embolism without the need for ionizing radiation or iodinated contrast material (4–7). Experience on the usefulness of MR angiography in the diagnostic work-up of other diseases involving the pulmonary arterial circulation is still limited. Additionally, time-resolved MR perfusion imaging can be carried out, thereby allowing the assessment of pulmonary perfusion defects (8).
Detection of chronic occlusive and nonocclusive changes in the pulmonary arteries at the segmental or subsegmental level requires high spatial resolution, which is limited by the imaging time of a single breath hold. Recently, parallel imaging techniques have become widely available. With these techniques, the temporal and spatial resolution can be improved substantially (9,10), which could give MR imaging an important role in the comprehensive assessment of the pulmonary circulation.
The intent of this study was to determine prospectively the accuracy of an MR perfusion imaging and MR angiography protocol with parallel acquisition techniques for the differentiation of CTEPH and PPH.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Final Reference Diagnosis
A final reference diagnosis of PPH or CTEPH was made by an attending senior pulmonologist (J.B., 11 years of experience) according to the classifications of Rich (1) and Simonneau et al (11). For this reference diagnosis, the results of reference imaging, as well as those of catheterization of the right side of the heart and patient history, were taken into account. A diagnosis of CTEPH was made if the results of any of the reference modalities (ie, perfusion scintigraphy, DSA, or CT angiography) were indicative of occlusive disease and if the findings agreed with the catheterization results and patient history. Otherwise, a diagnosis of PPH was made.
Diagnostic Reference Imaging
Pulmonary perfusion scintigraphy.—For perfusion scintigraphy, an 80-MBq dose of technetium 99m (99mTc)-labeled macroaggregated albumin (MAASOL; Amersham Health, Vienna, Austria) was injected into an antecubital vein while the patient was in a supine position. After tracer administration, lung scanning was performed by using a gamma camera system (Multispect 3; Siemens Medical Solutions, Erlangen, Germany) that was equipped with a high-spatial-resolution collimator. Image acquisition parameters included a 128 x 128 matrix and a 15% main window centered at the 99mTc photopeak (140 keV). Four views—that is, anterior, posterior, anterior oblique, and posterior oblique—were obtained. Each view was acquired for 750 000 counts.
Pulmonary DSA.—Pulmonary DSA was performed by a fellow of radiology (A.H., 7 years of experience) by using a multiplanar angiography system (Multistar T.O.P.; Siemens Medical Solutions) and the transfemoral venous approach (Seldinger technique). The main pulmonary artery (pulmonary trunk) was selectively catheterized by using a 4.0-F catheter (Altaflow Premium; Optimed, Ettlingen, Germany). Three injection series were performed by administering 30 mL of iodinated contrast material (iopromide, Ultravist 300; Schering, Berlin, Germany) at a flow rate of 18 mL/sec each. In this way, three projections—anterior-posterior, 30° right anterior oblique, and 60° left anterior oblique—were obtained. Arteriograms were acquired at 10 frames per second. Additional selective segmental arteriography was performed when thrombi or occlusions were suggested on the overview images. In such cases, magnified views of the corresponding lung segment were acquired.
CT angiography.—All spiral CT angiograms were obtained by using a 16–detector row CT scanner (Sensation 16; Siemens Medical Solutions) with the following parameters: scanning direction, craniocaudal; 100 kV; 200 mAs; section thickness, 0.75 mm; table feed, 12 mm per rotation (24 mm/sec); rotation time, 0.5 second (acquired during suspended breathing at deep inspiration); and mean acquisition time, 13 seconds. A 100-mL injection of iopromide was administered through an antecubital vein at a flow rate of 5 mL/sec. Delay time between start of contrast material administration and start of scanning was obtained by using a bolus-tracking technique (CARE-Bolus software; Siemens Medical Solutions); measurements were obtained at the origin of the pulmonary artery, and a delay of 6 seconds was added to the circulation time before scanning. For image analysis, transverse images with a section thickness of 3 mm and coronal maximum intensity projections (MIPs) with a section thickness of 10 mm and an overlap of 5 mm were reconstructed.
Pulmonary MR Imaging
MR perfusion imaging.—All MR perfusion images were obtained by using a 1.5-T MR imager (Magnetom Sonata, Maestro class; Siemens Medical Solutions) with high performance gradients (40 mT, 200 [T · m–1]/sec slew rate, and 200-µsec rise time) and eight receiver channels. Patients were positioned head first in the supine position. A special 12-channel receiver coil that was dedicated for parallel imaging was used. With this coil, eight of the 12 receiver channels are combined in pairs of two to fit the eight receiver channels of the MR imager. The receiver channels are arranged circularly around the patient's chest to optimize the spatial sensitivity profiles of the receiver coils for parallel imaging techniques. After performing initial localizer sequences, a test bolus sequence was performed by injecting 2 mL of 1.0-mol/L gadolinium chelate (gadobutrol, Gadovist; Schering) at a flow rate of 5 mL/sec and by acquiring a series of coronal images of the right side of the heart.
For dynamic perfusion imaging, a turbo fast low-angle shot gradient-echo sequence (repetition time msec/echo time msec, 1.7/0.6; flip angle, 25°) was performed by using an enhanced k-space-based reconstruction sequence (generalized autocalibrating partially parallel acquisition technique, or GRAPPA; Siemens Medical Solutions) with an acceleration factor of two (10). Further sequence parameters for MR perfusion imaging were as follows: readout bandwidth, 1220 Hz/pixel; number of reference k-space lines for calibration, 24; section thickness, 4.0 mm; number of sections, 24; resulting slab thickness, 96 mm; acquisition matrix, 133 x 256 (52% phase resolution); field of view, 400 mm; and resulting spatial resolution, 2.9 x 1.5 x 4.0 mm. The k-space readout scheme was centrically reordered, and acquisition time for one three-dimensional data set of 24 sections was 1.1 seconds.
A total of 24 slabs were acquired for each perfusion series, resulting in a total breath-hold time of 26 seconds (inspiration). Image acquisition was started 4 seconds before the estimated arrival time of the contrast material so that unenhanced images could be acquired for subtraction purposes. All injections were performed by using a power injection system (Spectris Solaris; Medrad, Indianola, Pa) with 0.1 mmol gadobutrol per kilogram body weight, followed by a 25-mL saline flush. To ensure an optimized bolus profile, a high injection rate of 5 mL/sec was used in combination with a 16-gauge intravenous catheter that was placed into an antecubital vein. Unenhanced images were subtracted from the contrast-enhanced images. For each phase (n = 25), MIPs for all 24 sections were acquired for three-dimensional display.
MR angiography.—Because different flow rates of contrast material were used for high-spatial-resolution MR angiography, test bolus injections were repeated to determine the exact circulation time of the contrast material bolus from the injection site to the pulmonary arteries. During the second test bolus for MR angiography, identical sequence parameters, plane orientation, and contrast agent (2.0 mL of gadobutrol) were used. A flow rate of 2 mL/sec was used to ensure a longer imaging time and more homogeneous contrast enhancement at MR angiography. Breath-hold contrast-enhanced three-dimensional fast low-angle shot (2.9/1.2; flip angle, 25°) MR imaging was subsequently performed in the coronal orientation by using the same parallel imaging technique used for MR perfusion imaging (enhanced k-space-based reconstruction sequence with an acceleration factor of two) (10). Further sequence parameters for the MR angiography protocol were as follows: readout bandwidth, 650 Hz/pixel; number of reference k-space lines for calibration, 24; section thickness, 1.6 mm; number of sections, 88; resulting slab thickness, 144 mm; acquisition matrix, 410 x 512 (80% phase resolution); field of view, 400 mm; and resulting spatial resolution, 1.0 x 0.7 x 1.6 mm.
The acquisition time for one three-dimensional data set of 88 sections was 22 seconds. The data set was acquired during breath hold (inspiration), and the k-space readout scheme was centrically reordered. Because of the asymmetric sequential k-space readout scheme, image acquisition was initiated 5 seconds before the estimated arrival of the contrast material bolus. Unenhanced MR angiograms, which were acquired at the beginning of the examination before the administration of any contrast material, were subtracted from the contrast-enhanced images, and for image analysis, single coronal sections and MIP reconstruction of the complete data set (88 sections) were stored. Table 1 shows the typical imaging parameters for conventional MR angiography and MR perfusion imaging and the improved imaging parameters for MR angiography and MR perfusion imaging with implementation of parallel acquisition techniques.
|
value) and subsequently made a final diagnosis during a consensus reading of the MR data. Diagnostic criteria for perfusion imaging in the differential diagnostic work-up of PPH and CTEPH were modified from earlier reports on perfusion scintigraphy (12). During diagnostic work-up with perfusion imaging (both perfusion scintigraphy and MR perfusion imaging), two decisions had to be made by the readers. The first decision was whether a perfusion defect was present. This decision was used to compare the results of MR perfusion imaging with those of perfusion scintigraphy and to assess the intermodality agreement between data obtained with these two modalities. This primary decision did not yet include the differential diagnostic aspects of differentiating PPH and CTEPH.
If a perfusion defect was present, the second decision that had to be made was whether the defect was (a) patchy and/or diffuse (indicative of PPH) or (b) segmental and/or circumscribed (indicative for CTEPH). Interobserver agreement for both imaging criteria—that is, for the presence of perfusion defect and for the type of perfusion defect—was then analyzed.
Analysis of angiograms.—DSA and/or CT angiography were used as a reference standard for MR angiography. Images obtained at DSA, CT angiography, and MR angiography were first read independently by the same two readers (S.O.S., K.N.) who interpreted the MR perfusion images at two different time points. A consensus reading was subsequently performed to determine the final diagnosis at MR angiography. To avoid any bias in methods, MR perfusion images, MR angiograms, CT angiograms, and DSA images were read on two different days, with a time interval of at least 7 days between readings. Readers were blinded to patient names, and the images were presented in random order.
Two categories of imaging criteria were evaluated during the review of data obtained at DSA, CT angiography, and MR angiography. The first category, termed nonoccluding imaging criteria, included (a) dilatation of the pulmonary arterial main stem (diameter, >3 cm), (b) proximal caliber changes (pruned tree sign), (c) peripheral vessel reduction, (d) focal vessel ectasia, and (e) presence of the corkscrew phenomenon (ie, a distinctive tortuous course of peripheral pulmonary arteries). These nonoccluding imaging criteria were used to compare the results of DSA with those of CT angiography and to assess the intermodality agreement of these findings. Nonoccluding imaging criteria were not used for the differential diagnosis of PPH and CTEPH because these criteria can be noted for both entities. No further ranking of these nonoccluding imaging criteria concerning their relative value was applied.
The second category, termed occluding imaging criteria, included (a) complete vessel occlusion, (b) free-floating thrombus, (c) wall-adherent or endothelialized thrombus, and (d) webs and bands. These occluding imaging criteria were considered proof of chronic thrombembolic disease, and thus, if one or several of these criteria were found, CTEPH was determined to be the final diagnosis. No further ranking of these occluding image criteria concerning their relative value was applied. Both readers made a final diagnosis of PPH or CTEPH by using the MR angiographic data alone. Interobserver agreement was analyzed for all nonoccluding and occluding imaging criteria.
Combined analysis of MR angiograms and MR perfusion images.—After analyzing the MR perfusion images and MR angiograms separately, both readers (S.O.S., K.N.) evaluated the images together to assess the diagnostic value of a combined MR perfusion imaging and MR angiographic approach to the differentiation of PPH and CTEPH. Both readers made a final diagnosis of PPH or CTEPH by using the combined MR perfusion imaging and MR angiographic data. Again, the two readers first read the combined MR perfusion images and MR angiograms independently (for value assessment) and subsequently made a final diagnosis during a consensus reading of the complete MR data set. These diagnoses were then compared with the final reference diagnoses.
Statistical Analysis
For statistical analysis, two software products were used (MedCalc, version 7.0.0.2, MedCalc software, Mariakerke, Belgium and SPSS 12.0.1, SPSS, Chicago, Ill). For all statistical tests, P <.05 was considered to indicate a statistically significant result. For all single imaging parameters (occluding and nonoccluding imaging criteria), comparisons were made between each imaging modality and its corresponding reference standard—that is, between MR angiography and DSA and/or CT angiography and between MR perfusion imaging and perfusion scintigraphy. These comparisons were made to determine the sensitivity, specificity, and accuracy of MR angiography and MR perfusion imaging for the detection of single imaging parameters. Intermodality agreement (ie, congruent diagnosis on a per-patient basis) was quantified as a percentage that compared the diagnoses obtained at MR perfusion imaging and MR angiography with the diagnoses obtained at reference imaging. Interobserver agreement was quantified by using values and was interpreted as follows: 0.20 or less, poor agreement; 0.21–0.40, fair agreement; 0.41–0.60, moderate agreement; 0.61–0.80, good agreement, and 0.81–1.00, very good agreement.
A receiver operating characteristic (ROC) analysis was performed to compare the diagnostic value of MR angiography alone, MR perfusion imaging alone, and MR angiography and MR perfusion imaging combined. This ROC analysis was performed for reader 1, reader 2, and readers 1 and 2 in consensus. An unpaired Student t test was performed to test for statistically significant differences between the mean age of women and that of men.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
|
= 0.63, good agreement). A detailed analysis of the data obtained at MR perfusion imaging can be found in Table 3. When the diagnosis at MR perfusion imaging was compared with the final reference diagnosis, a correct diagnosis of PPH or CTEPH was made in 20 (69%) of 29 patients (Table 4).
|
|
|
|
|
|
|
|
values for interobserver agreement ranged from 0.79 to 0.26, and intermodality agreement ranged from 97% to 72% (Table 5). The use of MR angiographic data alone enabled correct differentiation of PPH and CTEPH in 24 (83%) of 29 patients.
|
|
|
Results of ROC Analysis
Figure 4 shows the results of the ROC analysis comparing the diagnostic value of MR angiography and MR perfusion imaging alone with that of MR angiography and MR perfusion imaging combined. For reader 1, the area under the ROC curve was 0.800 (95% confidence interval [CI]: 0.601, 0.999) for MR angiography, 0.774 (95% CI: 0.571, 0.997) for MR perfusion imaging, and 0.850 (95% CI: 0.671, 1.029) for MR angiography and MR perfusion imaging combined. For reader 2, the area under the ROC curve was 0.824 (95% CI: 0.639, 1.009) for MR angiography, 0.797 (95% CI: 0.607, 0.988) for MR perfusion imaging, and 0.847 (95% CI: 0.680, 1.015) for MR angiography and MR perfusion imaging combined. Finally, the area under the ROC curve for the consensus reading was 0.850 (95% CI: 0.671, 1.029) for MR angiography, 0.824 (95% CI: 0.639, 1.009) for MR perfusion imaging, and 0.900 (95% CI: 0.748, 1.052) for MR angiography and MR perfusion imaging combined.
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Even with these improved imaging parameters, however, the diagnostic accuracy of high-spatial-resolution MR angiography was limited in demonstrating pathologic findings that affected the more peripheral territories of pulmonary vasculature and in facilitating the diagnosis of smaller and older organized thrombi. In particular, the sensitivity of MR angiography compared with that of DSA or CT angiography was rather low for the detection of the corkscrew phenomenon, as well as for the detection of webs and bands at the subsegmental level. A comparison between state-of-the-art MR angiography and multi–detector row CT angiography of the pulmonary vasculature shows that multi–detector row CT angiography still seems superior with regard to ease of use, short acquisition time, and high spatial resolution. These factors make CT angiography the modality of choice in acute settings like detection of pulmonary embolism. MR imaging, however, offers the great advantage of combining morphologic and functional information, which makes the use of MR imaging favorable in the comprehensive assessment of complex diseases like pulmonary arterial hypertension.
MR perfusion imaging.—When used during the clinical work-up, MR perfusion imaging adds valuable information; MR perfusion imaging can demonstrate the exact extent of perfusion defects and has increased diagnostic accuracy in the differentiation of various causes of pulmonary arterial hypertension, depending on the specific patterns of perfusion defects (14). The first studies on time-resolved dynamic imaging of the pulmonary circulation have shown promising results. MR perfusion imaging enables differentiation of regional differences in pulmonary perfusion (15), and temporal differentiation of pulmonary arterial and venous circulation becomes feasible (16). MR imaging of lung perfusion is performed with rapid imaging of the first pass of contrast material through the lungs after bolus injection into a peripheral vein. Despite substantial improvements in the gradient technique in the past few years, the three-dimensional imaging techniques that were used have been limited with regard to spatial and temporal resolution (17,18). A sufficient temporal resolution, however, is critical to visualizing peak enhancement of the lungs because the transit time of the contrast material bolus through the lungs is usually 3–4 seconds (19).
Compared with conventional MR imaging techniques, parallel imaging techniques acquire only a fraction of the phase-encoding lines and reconstruct the missing information to a full-field-of-view image by using the inherent spatial encoding of the different receiver coils (10,20). In this way, as presented in this study, a temporal resolution of 1.1 seconds per phase was obtained. Even with these improvements in temporal resolution, the breath-hold time for acquisition of the complete data set for MR perfusion imaging was 26 seconds. In some instances, long breath-hold times may be difficult for patients with pulmonary disorders. In our patient cohort, however, no MR angiographic or MR perfusion imaging data were excluded owing to motion or breathing artifacts. This can be explained by two reasons. First, all patients were carefully instructed on how to observe the breathing commands and were allowed to start breathing out slowly at the end of imaging if the breath-hold time was getting too long. Second, the readout scheme of both MR angiography and MR perfusion imaging was centrically reordered (ie, the central lines of the k-space were read out at the start of imaging), which allowed sufficient contrast of the image. Also, the peripheral lines of the k-space were read out toward the end of imaging, which resulted in high resolution of the images. Choosing this way of collecting data from the k-space made our angiographic data acquisitions much more resistant to motion artifacts.
In all 29 patients, the quality of the MR angiographic and MR perfusion imaging data sets was sufficient for further evaluation. To maintain a sufficient signal-to-noise ratio in spite of an improved spatial resolution, the bolus timing had to be optimized. To achieve a distinct bolus profile for perfusion imaging, a high injection rate of 5 mL/sec was used. A further improvement in signal-to-noise ratio can be achieved through the use of contrast agents with higher gadolinium concentration (1.0 mol/L gadobutrol) (21,22). As described in previous articles (23–25), good agreement between MR perfusion imaging and conventional radionuclide scintigraphy was found for the detection of perfusion defects. In contrast to conventional radionuclide perfusion scintigraphy, MR imaging of lung perfusion has various advantages. Three-dimensional MR imaging offers a higher spatial resolution and better anatomic information and allows the reconstruction of data in any desired imaging plane. Additionally, MR imaging lacks radiation exposure.
Combination of MR angiography and MR perfusion imaging.—In our study, a combined approach for the correct differentiation of PPH and CTEPH was applied by performing time-resolved dynamic MR perfusion imaging and MR angiography with parallel imaging acquisition techniques at the same time. MR perfusion imaging and MR angiography alone showed good results in the differentiation of PPH and CTEPH, enabling the correct diagnosis of PPH and CTEPH in 69% and 83% of patients, respectively. The combination of data from both modalities also enhanced diagnostic accuracy, achieving a combined diagnostic accuracy of 90% (correct diagnosis in 26 of 29 patients). Traditionally, to differentiate between PPH and CTEPH, a whole array of diagnostic tests was necessary, with considerable costs and labor for patients. The correct differentiation of these causes of pulmonary arterial hypertension is crucial because therapeutic strategies are different. While PPH is typically treated by administering vasodilating drugs, CTEPH can potentially be treated by performing resection of intraarterial wall-adhesive material (26). Combining MR perfusion imaging and MR angiography in a comprehensive evaluation provides complementary diagnostic information, thereby increasing the diagnostic value of pulmonary MR imaging.
Limitations
We acknowledge certain limitations of our study. First, the number of patients with CTEPH was limited (n = 10) because CTEPH is a rather rare entity. Second, although this was a prospective study and the availability of data from reference imaging was one of the inclusion criteria, five of 29 patients had not undergone perfusion scintigraphy, which was used as a reference standard for MR perfusion imaging. Still, these patients were included in the study to maintain an acceptable total number of patients. In these five patients, the value of MR perfusion imaging was assessed by comparing the final MR perfusion imaging diagnosis with the final clinical reference diagnosis.
Future Perspectives
Apart from being used for the distinct determination of the peak enhancement of the lung tissue, the temporal information obtained by using parallel imaging techniques could also be used for the temporal analysis of tissue perfusion and ultimately for the calculation of semiquantitative or quantitative parameter maps of parenchymal perfusion (27). Furthermore, MR angiography of the pulmonary vasculature could be complemented by additional functional measurements, such as flow measurements, and by assessment of cardiac function in the right side of the heart (28,29). Another major effect might occur from the combination of pulmonary MR angiographic techniques and functional imaging of lung ventilation by using either hyperpolarized gases, such as helium 3 (30), or oxygen (31). Further potential clinical applications of pulmonary MR imaging in the diagnosis of pulmonary hypertension could include an assessment of the treatment effects of vasodilative drug therapy in patients with PPH.
Conclusions
In patients with pulmonary hypertension, MR imaging of the pulmonary circulation by using the combination of MR angiography and MR perfusion imaging with implemented parallel acquisition techniques provides an accuracy of 90% in the differentiation of PPH from CTEPH. Additional larger sample studies will be needed for validation.
|
FOOTNOTES |
|---|
Abbreviations: CI = confidence interval • CTEPH = chronic thromboembolic pulmonary arterial hypertension • DSA = digital subtraction angiography • MIP = maximum intensity projection • PPH = primary pulmonary hypertension • ROC = receiver operating characteristic
Author contributions: Guarantors of integrity of entire study, K.N., S.O.S.; study concepts, K.N., S.O.S., J.S., A.H.; study design, K.N., S.O.S., F.R.; literature research, K.N., U.A.; clinical studies, K.N., S.O.S., U.A.; experimental studies, K.N., U.A., O.D.; data acquisition, K.N., S.O.S., U.A., O.D., B.K., F.R., A.H., H.L., R.B., J.B.; data analysis/interpretation, K.N., S.O.S., U.A., J.S., H.L., R.B., J.B.; statistical analysis, K.N.; manuscript preparation, K.N., S.O.S., F.R.; manuscript definition of intellectual content, K.N., S.O.S., M.F.R.; manuscript editing, K.N., U.A.; manuscript revision/review, K.N., S.O.S., J.S., O.D., B.K., A.H., H.L., J.B., M.F.R.; manuscript final version approval, K.N., S.O.S., J.S., F.R., M.F.R., J.B.
|
References |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. F. Thieme, T. R. C. Johnson, C. Lee, J. McWilliams, C. R. Becker, M. F. Reiser, and K. Nikolaou Dual-Energy CT for the Assessment of Contrast Material Distribution in the Pulmonary Parenchyma Am. J. Roentgenol., July 1, 2009; 193(1): 144 - 149. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Sanz, M. Kariisa, S. Dellegrottaglie, S. Prat-Gonzalez, M. J. Garcia, V. Fuster, and S. Rajagopalan Evaluation of pulmonary artery stiffness in pulmonary hypertension with cardiac magnetic resonance. J. Am. Coll. Cardiol. Img., March 1, 2009; 2(3): 286 - 295. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Benza, R. Biederman, S. Murali, and H. Gupta Role of Cardiac Magnetic Resonance Imaging in the Management of Patients With Pulmonary Arterial Hypertension J. Am. Coll. Cardiol., November 18, 2008; 52(21): 1683 - 1692. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Dehnert, F. Risse, S. Ley, T. A. Kuder, R. Buhmann, M. Puderbach, E. Menold, D. Mereles, H.-U. Kauczor, P. Bartsch, et al. Magnetic Resonance Imaging of Uneven Pulmonary Perfusion in Hypoxia in Humans Am. J. Respir. Crit. Care Med., November 15, 2006; 174(10): 1132 - 1138. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Roberts, I. Preston, and N. S. Hill Pulmonary hypertension trials: current end points are flawed, but what are the alternatives? Chest, October 1, 2006; 130(4): 934 - 936. [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]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
