HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Fink, C. Articles by Kauczor, H.-U. Search for Related Content PubMed PubMed Citation Articles by Fink, C. Articles by Kauczor, H.-U.

Regional Lung Perfusion: Assessment with Partially Parallel Three-dimensional MR Imaging¹

¹ From the Department of Radiology (C.F., M.P., S.D., H.U.K.), Department of Medical Physics in Radiology (M.B.), and Department of Radiation Therapy (I.Z.), Deutsches Krebsforschungszentrum, Innovative Krebsdiagnostik und Therapie, Im Neuenheimer Feld 280, 69120 Heidelberg, Germany; Bracco Altana Pharma, Konstanz, Germany (K.P.L.); and Thoraxklinik Heidelberg, Germany (A.S.). Received February 5, 2003; revision requested April 23; final revision received July 22; accepted August 29. Address correspondence to C.F. (e-mail: c.fink@dkfz.de).

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
PURPOSE: To evaluate partially parallel three-dimensional (3D) magnetic resonance (MR) imaging for assessment of regional lung perfusion in healthy volunteers and patients suspected of having lung cancer or metastasis.

MATERIALS AND METHODS: Seven healthy volunteers and 20 patients suspected of having lung cancer or metastasis were examined with 3D gradient-echo MR imaging with partially parallel image acquisitions (fast low-angle shot 3D imaging; repetition time msec/echo time msec, 1.9/0.8; flip angle, 40°; acceleration factor, two; number of reference k-space lines for calibration, 24; field of view, 500 x 440 mm; matrix, 256 x 123; slab thickness, 160 mm; number of partitions, 32; voxel size, 3.6 x 2.0 x 5.0 mm³; acquisition time, 1.5 seconds) after administration of 0.1 mmol/kg of gadobenate dimeglumine. In volunteers, 3D MR perfusion data sets were assessed for topographic and temporal distribution of regional lung perfusion. Sensitivity, specificity, accuracy, and positive and negative predictive values for perfusion MR imaging for detecting perfusion abnormalities in patients were calculated, with conventional radionuclide perfusion scintigraphy as the standard of reference. Interobserver and intermodality agreement was determined by using statistics.

RESULTS: Topographic analysis of lung perfusion in volunteers revealed a significantly higher signal-to-noise ratio (SNR) of up to 327% in gravity-dependent lung areas. Temporal analysis similarly revealed much shorter lag time to peak enhancement in gravity-dependent lung areas. In patients, perfusion MR imaging achieved high sensitivity (88%–94%), specificity (100%), and accuracy (90%–95%) for detection of perfusion abnormalities. Interobserver agreement ( = 0.86) was very good and intermodality agreement ( = 0.69–0.83) was good to very good for detection of perfusion defects. A significant difference (P < .0001) in SNR was observed between normally perfused lung (14 ± 7 [SD]) and perfusion defects (7 ± 4) in patients.

CONCLUSION: Partially parallel MR imaging with high spatial and temporal resolution allows assessment of regional lung perfusion and has high diagnostic accuracy for detecting perfusion abnormalities.

© RSNA, 2004

Index terms: Lung, MR, 68.12143, 68.12144 • Lung, perfusion • Lung neoplasms, MR, 68.12143, 68.12144 • Lung neoplasms, radionuclide studies, 68.12166 • Magnetic resonance (MR), perfusion study, 68.12143, 68.12144

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Changes in regional lung perfusion can be observed in a variety of lung diseases, including pulmonary embolism, chronic obstructive pulmonary disease, and lung cancer (1). In clinical practice, regional lung perfusion is usually assessed with radionuclide perfusion scintigraphy, which offers limited spatial and temporal resolution.

More recently, contrast agent–enhanced magnetic resonance (MR) imaging has been proposed for the assessment of lung perfusion (2–9). The technique is based on the contrast enhancement of the lung parenchyma during the first pass of a bolus of contrast agent through the pulmonary circulation. Because the lung, unlike other organs, is characterized by a very short circulation time, perfusion MR imaging requires rapid imaging to enable visualization of the peak enhancement of the lung parenchyma. Inaddition, MR imaging of lung perfusion also requires high spatial resolution and large anatomic coverage to enable visualization of small perfusion defects at the lung periphery. Despite major improvements in gradient hardware in the past few years, previous studies involving MR imaging of lung perfusion mostly have been limited to the investigation of single-section or projection two-dimensional (2D) MR imaging (2–6). Whenever three-dimensional (3D) MR imaging has been attempted, the temporal and/or spatial resolution was insufficient (6–9). As a consequence, perfusion MR imaging of the lungs has not yet become established in clinical practice.

Recently, parallel imaging techniques (eg, simultaneous acquisition of spatial harmonics, or SMASH, and sensitivity encoding, or SENSE) have become clinically available. With these techniques, the acquisition time of MR imaging is substantially reduced because the spatial information inherent in the geometry of surface coil arrays is exploited (10,11). In practice, parallel imaging techniques allow substantial improvement in temporal and/or spatial resolution, which might facilitate improved visualization of lung perfusion with MR imaging (12). The aim of this study was to evaluate the use of partially parallel 3D MR imaging for assessment of regional lung perfusion in healthy volunteers, as well as in patients suspected of having lung cancer or metastasis, and to assess the diagnostic accuracy of perfusion MR imaging for the detection of perfusion defects as compared with the accuracy of the standard of reference, radionuclide perfusion scintigraphy.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Volunteers and Patients
Prior to subject enrollment, the investigation protocol was approved by the institutional human research ethics committee of the Deutsches Krebsforschungszentrum and the University of Heidelberg. Written informed consent was acquired from all volunteers and patients after the nature of the procedure had been fully explained.

Over a 4-month period (from September 2002 to December 2002), seven healthy volunteers with no evidence or history of pulmonary disease (mean age, 27 years; age range, 19–32 years) and 20 consecutive patients suspected of having lung cancer or metastasis (mean age, 62 years; age range, 39–81 years) were included in this study. There were six male volunteers (mean age, 27 years; age range, 19–32 years) and one female volunteer (age, 27 years) and 13 male patients (mean age, 64 years; age range, 55–81 years) and seven female patients (mean age, 59 years; age range, 39–79 years). All patients had previously been examined with conventional radionuclide perfusion scintigraphy as part of a preoperative work-up (mean interval between radionuclide scintigraphy and MR imaging, 2.4 days; range, 0–7 days).

MR Imaging
All examinations were performed with a clinical 1.5-T whole-body MR imaging unit (Magnetom Symphony; Siemens Medical Systems, Erlangen, Germany) equipped with eight receiver channels and a high performance gradient system offering a maximum gradient strength of 30 mT/m and a slew rate of 125 T/m/sec. For signal reception, a combination of two spine array coils with two four-element body phased-array coils was used. To fully exploit the benefits of partially parallel MR imaging, the placement of the coils was dictated by the fact that the projection of the sensitivity profiles of the different coils in the phase-encoding direction is required to be noncoincident (Fig 1).

View larger version (93K): [in this window] [in a new window] [Download PPT slide]   Figure 1a. (a) Positioning and (b) sensitivity profiles of receiver coils in phase-encoding direction in relation to a patient’s anatomy. Two spine array coils (C1) are combined with two four-element body phased-array coils (C2 and C3).

View larger version (11K): [in this window] [in a new window] [Download PPT slide]   Figure 1b. (a) Positioning and (b) sensitivity profiles of receiver coils in phase-encoding direction in relation to a patient’s anatomy. Two spine array coils (C1) are combined with two four-element body phased-array coils (C2 and C3).

A total of eight consecutive data sets were acquired in a coronal orientation, starting 2 seconds after the beginning of the contrast agent injection. All subjects were asked to hold their breath for as long as possible during the data acquisition. All injections were performed with an automatic power injector (Tomojet; Bruker Amersham Health, Ismaning, Germany) by using a dose of 0.1 mmol of gadobenate dimeglumine (Multihance; Bracco, Milan, Italy) per kilogram of body weight followed by a 30-mL saline flush. To ensure an optimized bolus profile, a high injection rate of 5 mL/sec was used in combination with 16-gauge intravenous catheters, which were placed into an antecubital vein (13).

In addition to 3D lung perfusion MR imaging, a half-Fourier single-shot turbo spin-echo pulse sequence (600/28) was performed in all subjects to assess lung morphology. In patients, high-spatial-resolution contrast-enhanced 3D MR angiography (fast low-angle shot 3D sequence; 2.8/1.1; flip angle, 40°) for the assessment of vascular pathology was also performed.

Radionuclide Perfusion Scintigraphy
Radionuclide lung perfusion scintigraphy was performed in all patients for the preoperative evaluation of lung function (14). Imaging was performed with a double-head unit (Axis; Philips Medical Systems, Best, the Netherlands) and 80–120 MBq of technetium 99m–marked macroaggregated albumin (MAA Sol; Amersham Health), which equaled an amount of about 200,000 particles. Imaging was performed with a field of view of 52 x 39 cm and an image matrix of 128 x 128 and was initiated immediately after injection of the tracer into a cubital vein.

Perfusion images were acquired with patients in the supine position in the following eight projections: anterior, posterior, left anterior oblique, right anterior oblique, left posterior oblique, right posterior oblique, and both lateral projections. In each projection, 400,000 counts were registered. The location and number of hypoperfused areas or defects were recorded by a senior nuclear medicine physician (A.S.) with 9 years of experience in the field of perfusion scintigraphy. For this image evaluation, all clinical information for all subjects was supplied.

Data Analysis
The analysis of the MR images was performed independently by two radiologists (C.F., M.P.) with 2 and 4 years of experience in the field of pulmonary MR imaging, respectively. Blinded to the subjects’ identities and clinical histories, the readers performed the image analysis by using the software included with the MR imaging unit.

For volunteers, both readers independently assessed the topographic and temporal distribution of regional lung perfusion. For this purpose, six circular regions of interest (ROIs) of at least 300 pixels (range, 300–456 pixels) were placed over the cranial, middle, and lower levels of both lungs in ventral, middle, and dorsal partitions of the 3D data sets and the peak signal-to-noise ratio (SNR) of the lung parenchyma was calculated for each lung region. In addition, the lag time between the peak enhancement in the main pulmonary artery and the peak enhancement of the lung parenchyma was calculated for each lung region.

In the next step, signal intensity–time curves for the lung parenchyma were calculated for each subject by using ROI analysis. For this purpose, circular ROIs of at least 300 pixels (range, 300–956 pixels) were placed in the central partition of the 3D data sets over the middle aspects of both lungs. From these signal intensity–time curves, the data set showing the peak parenchymal enhancement was determined, and 3D perfusion-weighted data sets were obtained by subtracting the corresponding precontrast data sets. In selected cases, additional ROIs were placed over the central pulmonary arteries and veins or left atrium so that we could get an impression of the temporal relationship of contrast enhancement of lung vessels and lung tissue. The sizes of these ROIs were adjusted to cover as much of the vascular lumen as possible.

Finally, readers were asked to evaluate the perfusion-weighted data sets for the presence of perfusion abnormalities. In case of perfusion abnormalities, the contrast between normal lung and perfusion defects was quantified by calculating the SNRs in the perfusion defects and in the adjacent normally perfused lung. As usual, image noise () was defined as = SD_ROI/0.66, where SD_ROI is the SD of the image noise in an ROI placed outside the patient (15).

Statistical Analysis
The statistical analyses of the data were performed with SAS software (SAS Institute, Cary, NC) and nQuery Advisor software (Statistical Solutions, Saugus, Mass). For assessment of the topographic and temporal distribution of lung perfusion in volunteers, the data of both readers were pooled after significant differences had been excluded. This resulted in a sample size of 14. For this sample size, a power of 80% was achieved (for SD/mean < 1.23) to detect a difference in means by using a paired t test with a two-sided significance level of .05. Bonferroni corrections were performed for the multiple testing problems. For the detection of perfusion abnormalities in patients, the sensitivity, specificity, accuracy, and positive and negative predictive values of perfusion MR imaging were calculated by using conventional radionuclide perfusion scintigraphy as the standard of reference. The interobserver agreement was quantified by using statistics and interpreted as follows: A value less than 0.20 was considered to indicate poor agreement; a value of 0.21–0.40, fair agreement; a value of 0.41–0.60, moderate agreement; a value of 0.61–0.80, good agreement; and a value of 0.81–1.00, very good agreement (16).

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The 3D perfusion MR images of the lungs acquired with partially parallel MR techniques were interpretable and of high quality in all subjects.

As expected, signal intensity–time curves showed a rapid transit of the contrast agent bolus through the pulmonary circulation. Typically, the peak enhancement of the lung parenchyma coincided with the peak enhancement of the lung vessels (Fig 2).

View larger version (146K): [in this window] [in a new window] [Download PPT slide]   Figure 2a. Images and graph illustrate the process of temporal analysis of the signal intensity of the lung parenchyma and vessels in a volunteer. (a) ROIs are placed over both lungs (white circles), the main pulmonary artery (small black circle), and the left atrium (large black circle) on a coronal 3D perfusion-weighted MR image (fast low-angle shot 3D sequence; 1.9/0.8; flip angle, 40°). (b) Signal intensity-time curves show that peak enhancement of the lung parenchyma largely coincides with peak enhancement of the lung vessels. Only the high temporal resolution achieved with partially parallel MR imaging allows visualization of lung perfusion. LA = left atrium, PA = pulmonary artery, SI (a.u.) = signal intensity in arbitrary units. (c) Source image and (d) transverse and (e) sagittal reconstructions of a coronal 3D perfusion-weighted MR imaging data set (fast low-angle shot 3D sequence; 1.9/0.8; flip angle, 40°). Higher signal intensity can be observed in gravity-dependent lung areas (arrows).

View larger version (20K): [in this window] [in a new window] [Download PPT slide]   Figure 2b. Images and graph illustrate the process of temporal analysis of the signal intensity of the lung parenchyma and vessels in a volunteer. (a) ROIs are placed over both lungs (white circles), the main pulmonary artery (small black circle), and the left atrium (large black circle) on a coronal 3D perfusion-weighted MR image (fast low-angle shot 3D sequence; 1.9/0.8; flip angle, 40°). (b) Signal intensity-time curves show that peak enhancement of the lung parenchyma largely coincides with peak enhancement of the lung vessels. Only the high temporal resolution achieved with partially parallel MR imaging allows visualization of lung perfusion. LA = left atrium, PA = pulmonary artery, SI (a.u.) = signal intensity in arbitrary units. (c) Source image and (d) transverse and (e) sagittal reconstructions of a coronal 3D perfusion-weighted MR imaging data set (fast low-angle shot 3D sequence; 1.9/0.8; flip angle, 40°). Higher signal intensity can be observed in gravity-dependent lung areas (arrows).

View larger version (153K): [in this window] [in a new window] [Download PPT slide]   Figure 2c. Images and graph illustrate the process of temporal analysis of the signal intensity of the lung parenchyma and vessels in a volunteer. (a) ROIs are placed over both lungs (white circles), the main pulmonary artery (small black circle), and the left atrium (large black circle) on a coronal 3D perfusion-weighted MR image (fast low-angle shot 3D sequence; 1.9/0.8; flip angle, 40°). (b) Signal intensity-time curves show that peak enhancement of the lung parenchyma largely coincides with peak enhancement of the lung vessels. Only the high temporal resolution achieved with partially parallel MR imaging allows visualization of lung perfusion. LA = left atrium, PA = pulmonary artery, SI (a.u.) = signal intensity in arbitrary units. (c) Source image and (d) transverse and (e) sagittal reconstructions of a coronal 3D perfusion-weighted MR imaging data set (fast low-angle shot 3D sequence; 1.9/0.8; flip angle, 40°). Higher signal intensity can be observed in gravity-dependent lung areas (arrows).

View larger version (97K): [in this window] [in a new window] [Download PPT slide]   Figure 2d. Images and graph illustrate the process of temporal analysis of the signal intensity of the lung parenchyma and vessels in a volunteer. (a) ROIs are placed over both lungs (white circles), the main pulmonary artery (small black circle), and the left atrium (large black circle) on a coronal 3D perfusion-weighted MR image (fast low-angle shot 3D sequence; 1.9/0.8; flip angle, 40°). (b) Signal intensity-time curves show that peak enhancement of the lung parenchyma largely coincides with peak enhancement of the lung vessels. Only the high temporal resolution achieved with partially parallel MR imaging allows visualization of lung perfusion. LA = left atrium, PA = pulmonary artery, SI (a.u.) = signal intensity in arbitrary units. (c) Source image and (d) transverse and (e) sagittal reconstructions of a coronal 3D perfusion-weighted MR imaging data set (fast low-angle shot 3D sequence; 1.9/0.8; flip angle, 40°). Higher signal intensity can be observed in gravity-dependent lung areas (arrows).

View larger version (99K): [in this window] [in a new window] [Download PPT slide]   Figure 2e. Images and graph illustrate the process of temporal analysis of the signal intensity of the lung parenchyma and vessels in a volunteer. (a) ROIs are placed over both lungs (white circles), the main pulmonary artery (small black circle), and the left atrium (large black circle) on a coronal 3D perfusion-weighted MR image (fast low-angle shot 3D sequence; 1.9/0.8; flip angle, 40°). (b) Signal intensity-time curves show that peak enhancement of the lung parenchyma largely coincides with peak enhancement of the lung vessels. Only the high temporal resolution achieved with partially parallel MR imaging allows visualization of lung perfusion. LA = left atrium, PA = pulmonary artery, SI (a.u.) = signal intensity in arbitrary units. (c) Source image and (d) transverse and (e) sagittal reconstructions of a coronal 3D perfusion-weighted MR imaging data set (fast low-angle shot 3D sequence; 1.9/0.8; flip angle, 40°). Higher signal intensity can be observed in gravity-dependent lung areas (arrows).

The statistics showed very good interobserver agreement ( = 0.86) as well as good to very good intermodality agreement ( = 0.69–0.83) in the detection of perfusion defects. Neither reader identified a small perfusion defect observed at radionuclide perfusion imaging in the right upper lobe of a patient with a peripheral T1 lung cancer (Fig 3). For another patient with small cell lung cancer of the left lower lobe, reader 2 did not identify subtle perfusion abnormalities of the laterobasal segment of the left lower lobe, but instead misinterpreted them as the lack of perfusion signal caused by coexistent left-sided elevation of the diaphragm (Fig 4). At comparison of the mean SNRs of normally perfused lung parenchyma (14 ± 7) and perfusion defects (7 ± 4), a significant difference (P < .0001) was found; this enabled sharp delineation of perfusion defects (Fig 5).

View larger version (160K): [in this window] [in a new window] [Download PPT slide]   Figure 3a. Images in an 81-year-old man with a T1N0 non-small cell lung cancer of the right upper lobe. (a) Coronal half-Fourier single-shot turbo spin-echo MR image (600/28) shows a small hyperintense tumor (arrow) in the right upper lobe. (b) Conventional radionuclide perfusion image reveals a small peripheral perfusion defect (arrow) in right upper lobe. (c) Coronal perfusion MR image (fast low-angle shot 3D sequence; 1.9/0.8; flip angle, 40°) does not show any perfusion abnormalities.

View larger version (150K): [in this window] [in a new window] [Download PPT slide]   Figure 3b. Images in an 81-year-old man with a T1N0 non-small cell lung cancer of the right upper lobe. (a) Coronal half-Fourier single-shot turbo spin-echo MR image (600/28) shows a small hyperintense tumor (arrow) in the right upper lobe. (b) Conventional radionuclide perfusion image reveals a small peripheral perfusion defect (arrow) in right upper lobe. (c) Coronal perfusion MR image (fast low-angle shot 3D sequence; 1.9/0.8; flip angle, 40°) does not show any perfusion abnormalities.

View larger version (119K): [in this window] [in a new window] [Download PPT slide]   Figure 3c. Images in an 81-year-old man with a T1N0 non-small cell lung cancer of the right upper lobe. (a) Coronal half-Fourier single-shot turbo spin-echo MR image (600/28) shows a small hyperintense tumor (arrow) in the right upper lobe. (b) Conventional radionuclide perfusion image reveals a small peripheral perfusion defect (arrow) in right upper lobe. (c) Coronal perfusion MR image (fast low-angle shot 3D sequence; 1.9/0.8; flip angle, 40°) does not show any perfusion abnormalities.

View larger version (144K): [in this window] [in a new window] [Download PPT slide]   Figure 4a. Images in a 49-year-old woman with limited small cell lung cancer in the left lower lobe. (a) Coronal half-Fourier single-shot turbo spin-echo MR image (600/28) shows a hyperintense tumor (arrow) in the left lower lobe and an elevated left hemidiaphragm (arrowhead). (b) Conventional radionuclide perfusion image reveals an area of subtle hypoperfusion (arrow) in the laterobasal segment of the left lower lobe, as well as a lack of perfusion (arrowhead) due to an elevated left hemidiaphragm. (c) Coronal 3D perfusion MR image (fast low-angle shot 3D sequence; 1.9/0.8; flip angle, 40°) similarly shows an area of subtle hypoperfusion (arrow) in the laterobasal segment of the left lower lobe and a lack of perfusion signal (arrowhead) due to an elevated left hemidiaphragm.

View larger version (69K): [in this window] [in a new window] [Download PPT slide]   Figure 4b. Images in a 49-year-old woman with limited small cell lung cancer in the left lower lobe. (a) Coronal half-Fourier single-shot turbo spin-echo MR image (600/28) shows a hyperintense tumor (arrow) in the left lower lobe and an elevated left hemidiaphragm (arrowhead). (b) Conventional radionuclide perfusion image reveals an area of subtle hypoperfusion (arrow) in the laterobasal segment of the left lower lobe, as well as a lack of perfusion (arrowhead) due to an elevated left hemidiaphragm. (c) Coronal 3D perfusion MR image (fast low-angle shot 3D sequence; 1.9/0.8; flip angle, 40°) similarly shows an area of subtle hypoperfusion (arrow) in the laterobasal segment of the left lower lobe and a lack of perfusion signal (arrowhead) due to an elevated left hemidiaphragm.

View larger version (140K): [in this window] [in a new window] [Download PPT slide]   Figure 4c. Images in a 49-year-old woman with limited small cell lung cancer in the left lower lobe. (a) Coronal half-Fourier single-shot turbo spin-echo MR image (600/28) shows a hyperintense tumor (arrow) in the left lower lobe and an elevated left hemidiaphragm (arrowhead). (b) Conventional radionuclide perfusion image reveals an area of subtle hypoperfusion (arrow) in the laterobasal segment of the left lower lobe, as well as a lack of perfusion (arrowhead) due to an elevated left hemidiaphragm. (c) Coronal 3D perfusion MR image (fast low-angle shot 3D sequence; 1.9/0.8; flip angle, 40°) similarly shows an area of subtle hypoperfusion (arrow) in the laterobasal segment of the left lower lobe and a lack of perfusion signal (arrowhead) due to an elevated left hemidiaphragm.

View larger version (136K): [in this window] [in a new window] [Download PPT slide]   Figure 5a. Images in a 75-year-old woman with limited small cell lung cancer in the right lower lobe. (a) Coronal half-Fourier single-shot turbo spin-echo MR image (600/28) shows a hyperintense tumor (arrow) in the right lower lobe. (b) Conventional radionuclide perfusion image reveals a peripheral perfusion defect (arrow) in the right lower lobe. (c, d) Coronal 3D lung perfusion MR images (fast low-angle shot 3D sequence; 1.9/0.8; flip angle, 40°) show a much sharper delineation of the segmental perfusion defect (arrow) in the right lower lobe from adjacent normally perfused lung than does the 2D projection radionuclide perfusion image (b).

View larger version (84K): [in this window] [in a new window] [Download PPT slide]   Figure 5b. Images in a 75-year-old woman with limited small cell lung cancer in the right lower lobe. (a) Coronal half-Fourier single-shot turbo spin-echo MR image (600/28) shows a hyperintense tumor (arrow) in the right lower lobe. (b) Conventional radionuclide perfusion image reveals a peripheral perfusion defect (arrow) in the right lower lobe. (c, d) Coronal 3D lung perfusion MR images (fast low-angle shot 3D sequence; 1.9/0.8; flip angle, 40°) show a much sharper delineation of the segmental perfusion defect (arrow) in the right lower lobe from adjacent normally perfused lung than does the 2D projection radionuclide perfusion image (b).

View larger version (148K): [in this window] [in a new window] [Download PPT slide]   Figure 5c. Images in a 75-year-old woman with limited small cell lung cancer in the right lower lobe. (a) Coronal half-Fourier single-shot turbo spin-echo MR image (600/28) shows a hyperintense tumor (arrow) in the right lower lobe. (b) Conventional radionuclide perfusion image reveals a peripheral perfusion defect (arrow) in the right lower lobe. (c, d) Coronal 3D lung perfusion MR images (fast low-angle shot 3D sequence; 1.9/0.8; flip angle, 40°) show a much sharper delineation of the segmental perfusion defect (arrow) in the right lower lobe from adjacent normally perfused lung than does the 2D projection radionuclide perfusion image (b).

View larger version (159K): [in this window] [in a new window] [Download PPT slide]   Figure 5d. Images in a 75-year-old woman with limited small cell lung cancer in the right lower lobe. (a) Coronal half-Fourier single-shot turbo spin-echo MR image (600/28) shows a hyperintense tumor (arrow) in the right lower lobe. (b) Conventional radionuclide perfusion image reveals a peripheral perfusion defect (arrow) in the right lower lobe. (c, d) Coronal 3D lung perfusion MR images (fast low-angle shot 3D sequence; 1.9/0.8; flip angle, 40°) show a much sharper delineation of the segmental perfusion defect (arrow) in the right lower lobe from adjacent normally perfused lung than does the 2D projection radionuclide perfusion image (b).

View larger version (26K): [in this window] [in a new window] [Download PPT slide]   Figure 6a. (a) For topographic and temporal analysis of lung perfusion in a volunteer, cranial, middle, and caudal ROIs were placed over both lungs in ventral, middle, and dorsal partitions of the 3D data set. (b) Signal intensity-time curves in ventral, middle, and dorsal lung regions. Dotted lines represent the cranial level; continuous lines, the middle level; and dashed lines, the caudal level. A substantially higher peak signal intensity and shorter lag time in gravity-dependent lung areas can be observed. SI (a.u.) = signal intensity in arbitrary units.

View larger version (43K): [in this window] [in a new window] [Download PPT slide]   Figure 6b. (a) For topographic and temporal analysis of lung perfusion in a volunteer, cranial, middle, and caudal ROIs were placed over both lungs in ventral, middle, and dorsal partitions of the 3D data set. (b) Signal intensity-time curves in ventral, middle, and dorsal lung regions. Dotted lines represent the cranial level; continuous lines, the middle level; and dashed lines, the caudal level. A substantially higher peak signal intensity and shorter lag time in gravity-dependent lung areas can be observed. SI (a.u.) = signal intensity in arbitrary units.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

In addition to achieving a high temporal resolution by using partially parallel MR imaging, we were able to achieve spatial resolution and anatomic coverage that were superior to those observed in previous studies involving lung perfusion MR imaging (2–9). The combination of high spatial and temporal resolution allowed us to reveal differences in the topographic and temporal distribution of regional lung perfusion. This included the observation of a significantly higher peak SNR in gravity-dependent lung areas that represented a posture-related distribution of regional lung perfusion, a phenomenon that has been described in previous studies involving 2D perfusion MR imaging, conventional angiography, single photon emission computed tomography, and positron emission tomography (5,17–19). Unlike the investigators in those studies, we were also able to demonstrate significant differences in the temporal distribution of lung perfusion. Future studies are needed to further investigate these findings in a larger series, as well as in the setting of pulmonary disease.

The effect of the results of our study is further limited, insofar as the potential for absolute quantification of lung perfusion by applying curve fitting with a gamma-variate function, which has previously been described by Hatabu et al (2) for 2D perfusion MR imaging, was not evaluated in our study. However, it is assumed that, in contrast to previous studies of 3D lung perfusion MR (8), the temporal resolution of partially parallel MR imaging may have been sufficient for absolute quantification of lung perfusion.

One limitation of the use of partially parallel MR imaging is the associated increase in image noise. In the processing of perfusion MR images, image noise is further increased by image subtraction techniques that are applied to enable visualization of lung perfusion. A potential solution to this problem might be the application of a correlation analysis that has been previously proposed for the reduction of image noise in the processing of time-resolved MR angiographic data (13). Another advantage of this technique would be the suppression of the signal from vascular components, which would improve the differentiation of lung tissue from vascular structures.

To compensate for the low SNR of the lungs, we optimized the bolus profile by using a high injection rate of 5 mL/sec and a contrast agent (gadobenate dimeglumine) with a high T1 relaxivity. We assumed that, as in previous studies in which gadobenate dimeglumine was compared with gadopentetate dimeglumine for contrast-enhanced MR angiography (20–22), gadobenate dimeglumine would achieve contrast characteristics superior to those of gadopentetate dimeglumine. Unlike previous investigators of contrast-enhanced perfusion MR imaging of the lungs, we did not use a fixed volume of contrast agent, but instead used the recommended dose of gadobenate dimeglumine for MR angiography (ie, 0.1 mmol/kg) (20). Higher doses were not evaluated in the present study, since the imaging protocol included a second contrast agent injection for high-spatial-resolution contrast-enhanced MR angiography. Future studies are needed to assess whether higher doses of gadobenate dimeglumine might further improve the SNR in perfusion MR imaging of the lungs.

As in previous studies (7–10), in our study, results of perfusion MR imaging had high diagnostic accuracy in comparison with results of the standard of reference, radionuclide perfusion scintigraphy. Perfusion MR imaging offers higher in-plane spatial resolution than radionuclide perfusion scintigraphy. Specifically, we achieved an in-plane resolution of 3.6 x 2.0 mm² with parallel MR imaging, versus an in-plane resolution of 4.1 x 3.1 mm² at radionuclide perfusion imaging. Furthermore, perfusion MR imaging offers the advantage of a 3D data set that can be reconstructed in any imaging plane desired.

Nevertheless, MR imaging did not depict all perfusion defects that were revealed at perfusion scintigraphy in this study. In a patient with a peripheral T1 lung cancer of the right upper lobe, radionuclide perfusion imaging revealed a small perfusion defect that was, even retrospectively, not visible on the MR perfusion images. A potential explanation might be the presence of a process that might have temporarily altered lung perfusion (eg, inflammation or hypoventilation) and had resolved by the time of MR imaging. This assumption is further supported by the fact that in this patient, MR imaging was performed 4 days after radionuclide perfusion imaging—a longer interval than that for most other patients (mean interval, 2.4 days). In another case, one reader did not identify an area of subtle hypoperfusion in the laterobasal segment of the left lobe but instead misinterpreted it as a lack of perfusion signal caused by a coexistent left-sided elevation of the diaphragm. However, if the perfusion MR images had been viewed together with the morphologic MR images, which were concealed from the readers to avoid any bias, it would have been easier to differentiate the perfusion defect from the lack of perfusion signal caused by the elevated diaphragm.

In other patients, however, perfusion abnormalities could be delineated from normal lung parenchyma much better with 3D MR perfusion data than with 2D projection radionuclide perfusion data.

For the clinical assessment of pulmonary diseases, lung perfusion MR imaging, in addition to morphologic T1- and T2-weighted MR imaging, can easily be complemented by high-spatial-resolution 3D MR angiography, blood flow measurements, and functional measurements of lung ventilation (7,12,23–26).

There were several limitations to our study. Although perfusion MR imaging had a high diagnostic accuracy, the high prevalence of patients with perfusion abnormalities in our study sample must be taken into account. Therefore, further study of a larger patient sample is strongly warranted. Another limitation was the fact that only patients suspected of having lung malignancy were examined. Future studies are needed to assess the value of this technique in patients with different lung diseases such as pulmonary embolism.

In conclusion, partially parallel MR imaging allows the assessment of lung perfusion with high spatial and temporal resolution. Our results indicate that perfusion MR imaging has high diagnostic accuracy for the detection of perfusion abnormalities.

Future work is required to further assess the quantification of lung perfusion with 3D MR imaging and to evaluate the diagnostic accuracy of this technique for other pulmonary disorders.

	ACKNOWLEDGMENTS

 
The authors thank the MR Application Development Department of Siemens Medical Systems for providing the prototype parallel imaging sequence techniques and acknowledge the financial support of Bracco Altana Pharma.

	FOOTNOTES

 
Abbreviations: ROI = region of interest, SNR = signal-to-noise ratio, 3D = three-dimensional, 2D = two-dimensional

Author contributions: Guarantors of integrity of entire study, C.F., S.D., H.U.K.; study concepts and design, C.F., S.D., H.U.K.; literature research, C.F., M.P., M.B., K.P.L.; clinical studies, C.F., M.P., A.S., S.D., H.U.K.; data acquisition, C.F., M.P., A.S., S.D., H.U.K.; data analysis/interpretation, C.F., M.P., A.S., M.B.; statistical analysis, I.Z.; manuscript preparation, C.F., M.P., M.B., K.P.L., S.D., H.U.K.; manuscript definition of intellectual content, editing, and final version approval, C.F., M.P., M.B., S.D., H.U.K.; manuscript revision/review, C.F., M.P., M.B., S.D., H.U.K., K.P.L.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1. Fazio F, Wollmer P. Clinical ventilation-perfusion scintigraphy. Clin Physiol 1981; 1:323-337.[Medline]
2. 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]
3. Hatabu H, Gaa J, Kim D, Li W, Prasad PV, Edelman RR. Pulmonary perfusion: qualitative assessment with dynamic contrast-enhanced MRI using ultra-short TE and inversion recovery turbo FLASH. Magn Reson Med 1996; 36:503-508.[Medline]
4. Berthezene Y, Croisille P, Bertocchi M, Houzard C, Bendib K, Revel D. Lung perfusion demonstrated by contrast-enhanced dynamic magnetic resonance imaging: application to unilateral lung transplantation. Invest Radiol 1997; 32:351-356.[CrossRef][Medline]
5. Nakagawa T, Sakuma H, Murashima S, Ishida N, Matsumura K, Takeda K. Pulmonary ventilation-perfusion MR imaging in clinical patients. J Magn Reson Imaging 2001; 14:419-424.[CrossRef][Medline]
6. Iwasawa T, Saito K, Ogawa N, Ishiwa N, Kurihara H. Prediction of postoperative pulmonary function using perfusion magnetic resonance imaging of the lung. J Magn Reson Imaging 2002; 15:685-692.[CrossRef][Medline]
7. 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]
8. 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]
9. Stock KW, Chen Q, Levin D, Hatabu H, Edelman RR. Demonstration of gravity-dependent lung perfusion with contrast-enhanced magnetic resonance imaging. J Magn Reson Imaging 1999; 9:557-561.[CrossRef][Medline]
10. Sodickson DK, Manning WJ. Simultaneous acquisition of spatial harmonics (SMASH): fast imaging with radiofrequency coil arrays. Magn Reson Med 1997; 38:591-603.[Medline]
11. Griswold MA, Jakob PM, Heidemann RM, et al. Generalized autocalibrating partially parallel acquisitions (GRAPPA). Magn Reson Med 2002; 47:1202-1210.[CrossRef][Medline]
12. Nikolaou K, Schoenberg SO, Nittka M, et al. Magnetic resonance imaging in the diagnosis of pulmonary arterial hypertension: high resolution angiography and fast perfusion imaging using intelligent parallel acquisition techniques (IPAT) (abstr). Radiology 2002; 225(P):473.
13. Bock M, Schoenberg SO, Floemer F, Schad LR. Separation of arteries and veins in 3D MR angiography using correlation analysis. Magn Reson Med 2000; 43:481-487.[CrossRef][Medline]
14. Chiti A, Schreiner FA, Crippa F, Pauwels EK, Bombardieri E. Nuclear medicine procedures in lung cancer. Eur J Nucl Med 1999; 26:533-555.[CrossRef][Medline]
15. Gudbjartsson H, Patz S. The Rician distribution of noisy MRI data. Magn Reson Med 1995; 34:910-914.[Medline]
16. Svanholm H, Starklint H, Gundersen HJG, Fabricius J, Barlebo H, Olsen S. Reproducibility of histomorphologic diagnoses with special reference to the kappa statistic. APMIS 1989; 97:689-698.[Medline]
17. Fishman AP. Dynamic of the pulmonary circulation. In: Hamilton WF, eds. Handbook of physiology, section 2: circulation. Washington, DC: American Physiological Society, 1963; 1668-1743.
18. Musch G, Layfield JD, Harris RS, et al. Topographical distribution of pulmonary perfusion and ventilation, assessed by PET in supine and prone humans. J Appl Physiol 2002; 93:1841-1851.[Abstract/Free Full Text]
19. Nyren S, Mure M, Jacobsson H, Larsson SA, Lindahl SG. Pulmonary perfusion is more uniform in the prone than in the supine position: scintigraphy in healthy humans. J Appl Physiol 1999; 86:1135-1141.[Abstract/Free Full Text]
20. Volk M, Strotzer M, Lenhart M, et al. Renal time-resolved MR angiography: quantitative comparison of gadobenate dimeglumine and gadopentetate dimeglumine with different doses. Radiology 2001; 220:484-488.[Abstract/Free Full Text]
21. Knopp MV, Schoenberg SO, Rehm C, et al. Assessment of gadobenate dimeglumine for magnetic resonance angiography: phase I studies. Invest Radiol 2002; 37:706-715.[CrossRef][Medline]
22. Herborn CU, Lauenstein TC, Ruehm SG, Bosk S, Debatin JF, Goyen M. Intraindividual comparison of gadopentetate dimeglumine, gadobenate dimeglumine, and gadobutrol for pelvic 3D magnetic resonance angiography. Invest Radiol 2003; 38:27-33.[CrossRef][Medline]
23. Kauczor HU. Radiology of pulmonary ventilation by means of magnetic resonance tomography. Rofo Fortschr Geb Rontgenstr Neuen Bildgeb Verfahr 2000; 172:403-406. [German].[Medline]
24. Kauczor HU, Hanke A, Van Beek EJ. Assessment of lung ventilation by MR imaging: current status and future perspectives. Eur Radiol 2002; 12:1962-1970.[Medline]
25. Salerno M, de Lange EE, Altes TA, Truwit JD, Brookeman JR, Mugler JP, 3rd. Emphysema: hyperpolarized helium 3 diffusion MR imaging of the lungs compared with spirometric indexes—initial experience. Radiology 2002; 222:252-260.[Abstract/Free Full Text]
26. Kauczor HU, Hofmann D, Kreitner KF, et al. Normal and abnormal pulmonary ventilation: visualization at hyperpolarized He-3 MR imaging. Radiology 1996; 201:564-568.[Abstract/Free Full Text]

This article has been cited by other articles:

				Y. Ohno, H. Koyama, M. Nogami, D. Takenaka, S. Matsumoto, M. Yoshimura, Y. Kotani, and K. Sugimura Postoperative Lung Function in Lung Cancer Patients: Comparative Analysis of Predictive Capability of MRI, CT, and SPECT Am. J. Roentgenol., August 1, 2007; 189(2): 400 - 408. [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]

This Article Abstract Figures Only Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Fink, C. Articles by Kauczor, H.-U. Search for Related Content PubMed PubMed Citation Articles by Fink, C. Articles by Kauczor, H.-U.