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


     

Published online before print January 21, 2005, 10.1148/radiol.2343040359
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2343040359v1
234/3/869    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 HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Manka, C.
Right arrow Articles by Kuhl, C. K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Manka, C.
Right arrow Articles by Kuhl, C. K.
(Radiology 2005;234:869-877.)
© RSNA, 2005

Neuroradiology

Three-dimensional Dynamic Susceptibility-weighted Perfusion MR Imaging at 3.0 T: Feasibility and Contrast Agent Dose1

Christoph Manka, MD, Frank Träber, PhD, Juergen Gieseke, PhD, Hans H. Schild, MD and Christiane K. Kuhl, MD

1 From the Department of Radiology, University of Bonn, Sigmund-Freud-Strasse 25, D-53105 Bonn, Germany. Received February 24, 2004; revision requested April 30; revision received May 25; accepted June 15. Address correspondence to C.K.K. (e-mail: kuhl@uni-bonn.de).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
PURPOSE: To prospectively investigate if T2*-weighted dynamic susceptibility-weighted first-pass perfusion magnetic resonance (MR) imaging is feasible at 3.0 T and which dose of contrast agent is suitable for high-field-strength imaging.

MATERIALS AND METHODS: Informed consent was obtained from all participants; study protocol was approved by the institutional review board. Study included three volunteers (two men, one woman aged 35, 39, and 52 years) and 26 patients (mean age, 49 years ± 12.8 [standard deviation]; range, 19–76 years). Volunteers underwent 3.0-T perfusion MR imaging with 0.20, 0.10, and 0.05 mmol per kilogram body weight of gadopentetate dimeglumine; patients underwent imaging with 0.10- and 0.05-mmol doses. Perfusion MR imaging was performed with three-dimensional echo-shifted echo-planar imaging (repetition time msec/echo time msec, 14/21; isotropic 4 mm3 voxels; 50 dynamic volumes with 30 sections each, covering entire brain at temporal resolution of 1.5 seconds per MR image). Quality of source echo-planar images and perfusion maps was assessed; perfusion maps obtained at studies with different contrast media doses were compared. Quantitative perfusion values and diagnostic sensitivity of perfusion studies with 0.10-mmol dose were compared with results with 0.05-mmol dose. Image quality scores were compared with marginal homogeneity test for multinomial variables (Mantel-Haenszel statistics for ordered categorized values). Signal-to-noise ratio and baseline signal intensity in perfusion studies were tested (Student t test for paired samples). Mean transit time (MTT), negative integral (NI), and maximum T2* effect from region-of-interest analysis were compared (one-tailed Student t test for paired samples). Quantitative data on number of gamma-fitted pixels were compared (t test for paired samples). Difference with P = .05 (t test for paired samples) was considered significant.

RESULTS: Perfusion image quality was satisfactory even in areas close to skull base (47 of 52 images, minor distortions; remaining images, marked distortions). Perfusion imaging with 0.20-mmol dose caused almost complete signal cancellation during first pass, particularly in cortical gray matter, since mean maximum T2* effect of 98%, 99%, and 98% for gray matter was reached such that the accurate calculation of perfusion maps was impossible. With 0.10-mmol dose, the NI and maximum T2* effect were comparable to published data for 1.5-T perfusion imaging with 0.20- and 0.05-mmol doses; perfusion maps of sufficient diagnostic quality were obtained. For gray matter, mean maximum T2* effect was 25.4% ± 9.8 with 0.10-mmol dose and 17.5% ± 9.0 with 0.05-mmol dose. For white matter, mean maximum T2* effect was 15.2% ± 4.5 with 0.10-mmol dose and 7.7% ± 2.9 with 0.05-mmol dose. Difference in maximum signal intensity decrease was significant (P < .01). For NI, the difference between 0.10- and 0.05-mmol doses was significant: For gray matter, mean NI was 3.1 ± 1.60 for 0.10-mmol dose and 1.56 ± 1.16 for 0.05-mmol dose. For white matter, mean NI was 1.35 ± 0.59 with 0.1-mmol dose and 0.59 ± 0.30 with 0.05-mmol dose.

CONCLUSION: With echo-shifted multishot echo-planar imaging, dynamic susceptibility-weighted perfusion MR imaging at high field strength is feasible without relevant image distortions. Compared with contrast agent dose for 1.5 T imaging, the dose for 3.0 T can be reduced to 0.10 mmol.

© RSNA, 2005


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Dynamic susceptibility-weighted contrast material–enhanced magnetic resonance (MR) imaging has been increasingly used for the assessment of cerebral perfusion in many different clinical settings, such as ischemic stroke (1–5), neurovascular diseases (6–9), brain tumors (10–19), and neurodegenerative disorders (20–25). Unlike MR angiography, which depicts flow within large vessels, perfusion-weighted MR techniques are sensitive to perfusion on a microscopic (ie, capillary) level (26–28).

Currently, most MR systems that are used for clinical cerebral perfusion imaging operate at a field strength of 1.5 T. Although structural contrast-enhanced brain imaging at 1.5 T usually is performed with a dose of 0.10 mmol per kilogram body weight, a higher dose (0.20 mmol of a gadolinium chelate) is most widely used and is considered the optimal dose for T2*-weighted dynamic susceptibility-weighted contrast-enhanced first-pass perfusion MR imaging (1,18,29–34). This holds true even for the higher concentrations of gadolinium-based contrast agents (eg, gadobutrol, Gadovist; Schering, Berlin, Germany) (35,36).

Since the Food and Drug Administration established guidelines for nonsignificant risk of MR imaging up to 8.0 T, MR systems operating at higher field strengths became increasingly available in a clinical setting. In general, imaging at higher magnetic field strengths offers an at least linear increase in signal-to-noise ratio (37). Higher field strengths are, however, also associated with specific difficulties: For example, magnetic susceptibility scales also with field strength, such that image distortions may become critical, particularly if echo-planar imaging pulse sequences are used, as is the case in most imaging protocols for dynamic susceptibility-weighted contrast-enhanced perfusion MR imaging. On the other hand, in regard to T2*-weighted dynamic susceptibility-weighted contrast-enhanced perfusion MR imaging, 3.0-T systems offer some potential advantages. The shorter T2 and T2* relaxation times and an increased signal-to-noise ratio (38–40) should translate into a more effective T2* reduction of a given amount of contrast medium during its capillary passage (41). Accordingly, it is conceivable that the contrast agent dose to be used for dynamic susceptibility-weighted contrast-enhanced perfusion MR imaging needs to be adjusted to the high-field-strength setting.

Thus, the objective of this study was to prospectively investigate whether clinical T2*-weighted dynamic susceptibility-weighted contrast-enhanced first-pass perfusion MR imaging is feasible at 3.0 T and which dose of contrast agent is suitable for high-field-strength perfusion MR imaging.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Informed consent was obtained from all patients and volunteers after the experimental character of the study had been fully explained to them. The study protocol was reviewed and approved by our institutional review board.

Study Setup
In an initial dose-finding study, three volunteers underwent T2*-weighted perfusion MR imaging three times: once with a gadolinium-based contrast agent (gadopentetate dimeglumine, Magnevist; Schering) in a dose of 0.20 mmol (ie, the recommended dose for perfusion MR imaging at 1.5 T [1,18,29–35]), once with 0.10 mmol of the same contrast agent (short form, 0.10 mmol), and once with 0.05 mmol of the same contrast agent (short form, 0.05 mmol). After reviewing the data generated in this preliminary study, the clinical protocol was set up as an intraindividual comparative study, for which the same patients underwent dynamic susceptibility-weighted contrast-enhanced perfusion MR imaging at 3.0 T twice, once with a dose of 0.10 mmol and once with a dose of 0.05 mmol.

Study Cohort
The three volunteers who were included in the preliminary dose-finding study were two men aged 35 and 39 years and one woman aged 52 years. For the actual study, a total of 26 patients (eight men and 18 women) were examined. All the patients were included in the study during a period of 4 weeks. The patients were referred to undergo contrast-enhanced MR imaging on a clinical basis; all patients who consented to undergo the two perfusion studies were consecutively included in the study. Mean age of all patients was 49 years ± 12.8 (standard deviation), with a range of 19–76 years; mean age of men was 49.1 years ± 12.7, and mean age of women was 48.9 years ± 13.2. The difference in age distribution in male and female patients was not statistically significant (P = .974, t test for independent variables). Of 26 patients, 12 were referred for the evaluation of cerebral ischemia; seven, for preoperative evaluation of intra- or extraaxial brain tumors; and seven, for the evaluation of psychiatric disorders.

Imaging Protocol
All examinations were conducted with a clinical 3.0-T whole-body system (Intera; Philips Medical Systems, Best, the Netherlands) with high-performance gradients (master gradients), with a maximum slew rate of 150 mT/m/msec and a maximum strength of 30 mT/m. A standard transmit-receive birdcage head coil was used. The contrast agent gadopentetate dimeglumine (Magnevist; Schering, Berlin, Germany) was injected intravenously via an antecubital 16-gauge cannula by using a power injector (Spectris Solara; Medrad, Maastricht, the Netherlands) at a standardized flow rate of 8 mL/sec. This high flow rate is considered optimal for dynamic susceptibility-weighted contrast-enhanced perfusion MR imaging because it helps in obtaining a sharp bolus profile (42).

A transverse T2*-weighted pulse sequence with three-dimensional echo-planar principles of echo shifting with a train of observations (PRESTO) was applied for all perfusion studies (1,43–47). Parameters included repetition time msec/echo time msec, 14/21; echo-planar imaging factor, 17; field of view, 255 mm; and rectangular field of view, set at 80%. A total of 30 sections with a section thickness of 4 mm and a noninterpolated acquisition matrix of 64 x 64 were acquired, resulting in an isotropic voxel size of 3.98 x 4.00 x 4.00 mm. The PRESTO sequence is a multishot echo-planar imaging technique, with three shots for an echo-planar imaging factor of 17, and a signal bandwidth of 1735 Hz, which corresponds to a bandwidth per pixel, or BWpix, of 102 Hz. This results in a profile acquisition time calculated as follows: Tacq = 1/BWpix = 9.8 msec, which represents the length of the echo train and thus determines the T2* sensitivity. Because of the echo shifting, this pulse sequence allows one to keep the echo train very short and still acquire a total of 30 sections to cover the entire brain with high temporal resolution (1.5 seconds per volume). Each perfusion series consisted of 50 dynamic images; the total acquisition time for one perfusion series with 1500 images was 75 seconds. The perfusion studies were performed in a randomized order, once with a dose of 0.10 mmol and once with a dose of 0.05 mmol. The waiting time between the two perfusion studies was 4 minutes.

The perfusion studies were integrated into the standard clinical imaging protocol, which varied depending on the clinical question. All patients, however, underwent at least the following: T1-weighted spin-echo imaging (500/15, 25 sections, 256 x 256 matrix) before and after injection of both contrast agent boluses (ie, a total of 0.15 mmol gadopentetate dimeglumine), T2-weighted turbo spin-echo imaging (3540/100, 31 sections, 512 x 400 matrix), diffusion-weighted imaging (single-shot echo-planar imaging with 4225/74 and 128 x 128 matrix) with additional apparent diffusion coefficient mapping, and fluid-attenuated inversion-recovery imaging (repetition time msec/echo time msec/inversion time msec of 6000/80/2000 and 256 x 256 matrix) in two orientations (coronal and transverse).

Postprocessing
All perfusion studies were assessed for presence of gross head motion prior to further analysis; studies were discarded from analysis if gross head motion was identified during the dynamic acquisition. Assessment of gross motion was performed by a radiologist (C.M.) who had about 4 years of experience with cerebral MR imaging.

All perfusion imaging data were transferred to a workstation (EasyVision, release 4.4; Philips Medical Systems) and analyzed as follows: By using software that is part of the workstation (Perfusion Software Package), the perfusion data were converted into concentration-time curves with the assumption that the change in relaxivity, calculated as {Delta}R2* = {Delta}(1/T2*), is proportional to the time-dependent contrast medium concentration, calculated as {Delta}R2*(t) = k[CA], where t is time, k is a field-strength- and pulse sequence–specific constant, and [CA] represents the concentration of the contrast agent.

The signal intensity changes were converted into concentration-time curves according to the following equation: Cmeas(t) = –ln[SItiss(t)/S0]/[k · TE], where Cmeas is the measured concentration, t is time, S0 is the averaged baseline signal before injection, SItiss(t) is the tissue signal intensity at time t, and TE is echo time. These concentration-time curves were subsequently fitted with the least squares method to a gamma variant function that included a correction for recirculation. From these curves, parameter maps were computed on a voxel-by-voxel basis with methods similar to other two-dimensional perfusion techniques (1,26–28). No arterial input function and deconvolution were calculated for quantification of the perfusion data (48).

The following parametric maps were calculated: the negative integral (NI) map, which refers to the area under the fitted concentration-time curve, and the mean transit time (MTT) map, which refers to the time needed to reach half the area of the NI.

Data Analysis
Image and time–signal intensity curve review of preliminary dose-finding study data in volunteers.—The parametric maps for the studies in volunteers were evaluated in the same way as was done for the studies in patients: Regions of interest (ROIs) were placed in regions in white matter and gray matter to calculate time–signal intensity curves from the data. The shape of the curve for the bolus and the peak signal change in the gray matter were especially evaluated.

Image quality of the T2*-weighted source images.—The source images (ie, the T2*-weighted echo-shifted three-dimensional echo-planar [PRESTO] images of the 52 perfusion series obtained in patients) were reviewed by two radiologists (C.C.K., C.M., who had 8 and 4 years of experience, respectively, in interpretation of brain perfusion images). Readers were asked to rate, in consensus, the image quality, particularly in regard to susceptibility-mediated image distortions at tissue interfaces in the following anatomic areas: posterior fossa, basal temporal and frontal lobes, and brainstem. Image quality was judged on a four-point scale, with a score of 1 for an image without distortions and with excellent delineation of all structures including the posterior fossa and the base of the skull, a score of 2 for images with minor distortions and with delineation of the target structures in the posterior fossa rated as still possible, a score of 3 for images with marked distortions and with delineation of the target structures rated as impaired, and a score of 4 for images with severe distortions and with delineation of the target structures rated as impossible. Distribution of the image quality scores was analyzed.

Analysis of image quality of perfusion maps.—The image quality of the perfusion maps (MTT and NI) for the 52 perfusion studies in 26 patients was assessed visually by two radiologists (C.M., C.C.K., who had 4 and 8 years of experience, respectively, in interpretation of T2* dynamic susceptibility-weighted contrast-enhanced perfusion images). Readers were asked to rate the perfusion maps in regard to the completeness of gamma fitting in gray matter and white matter areas by using a five-point scoring system. A score of 5 (nondiagnostic) was assigned when a large number of voxels were not fitted in gray matter and white matter areas and gamma-fitted voxels were only in gray matter areas. A score of 4 (poor) was assigned when gamma-fitted voxels were only in gray matter areas. A score of 3 (adequate for diagnosis) was assigned when most (>90%) voxels were gamma fitted in gray matter areas and at least 40%–60% of voxels were fitted in white matter areas. A score of 2 (good) was assigned when all voxels were gamma fitted in gray matter areas and 60%–80% of voxels were fitted in white matter areas. A score of 1 (excellent) was assigned when all voxels were gamma fitted in gray matter areas and 90%–100% of voxels were fitted in white matter areas. In addition, and to provide a more objective measure of the quality of the gamma fitting, the mean number of pixels that were accepted for gamma fitting at perfusion analysis was determined, and these mean values were compared among the perfusion images.

ROI-based quantitative analysis of perfusion variables.—The perfusion images in volunteers and patients were loaded onto the workstation with the perfusion analysis software package. ROIs were placed manually in the following anatomic locations of each hemisphere: the gray matter areas of the cerebellum, the hippocampus, the thalamus; the white matter areas of the occipital and frontal lobes; the gray matter areas of the motor cortex; and the middle cerebral artery of both hemispheres. ROIs were placed by one of the authors (C.M.). ROI sizes varied between 3 and 30 pixels, according to the anatomic location. Because the geometric parameters of the two perfusion imaging sequences were kept unchanged, ROIs could be copied from one perfusion image to the other, thus ensuring equivalent ROI settings between the perfusion images in the same patient. In regard to ROI placement in patients, special care was taken to exclude areas with visible enhancement on the postcontrast images obtained with T1-weighted spin-echo sequences to minimize false parametric estimations, such as those caused by blood-brain–barrier disruptions (49).

For each anatomic location, an average of both hemispheres was calculated in regard to the following perfusion variables: MTT, NI, and maximum T2* effect, that is, the maximum signal intensity decrease during the passage of the bolus. The perfusion variables of the perfusion images in patients and volunteers were compared.

Clinical MR image reading.—In two reading sessions that were 6 weeks apart, two radiologists (C.M., C.C.K., who had 4 and 8 years of experience, respectively, in interpretation of perfusion images) interpreted the perfusion maps in consensus. During the first reading session, MR images from the dynamic susceptibility-weighted contrast-enhanced studies performed with the contrast agent dose of 0.05 mmol were displayed; in the second reading session, MR images from the studies performed with 0.10 mmol of contrast agent were presented. This procedure was performed to ensure an independent analysis of the images from the two perfusion studies performed in the same patient. During each reading session, the perfusion maps were displayed together with the corresponding structural MR images, which included the diffusion-weighted images and the respective apparent diffusion coefficient maps, T2-weighted images, and fluid-attenuated inversion-recovery images. These images were displayed together because perfusion maps can only be interpreted in conjunction with structural and/or diffusion-weighted images; accordingly, this procedure was followed because it is the way perfusion images are evaluated in the clinical setting as well. Readers were asked to rate the presence or absence of perfusion deficits and to measure the size (in-plane size) of the areas with abnormal perfusion. The in-plane size was measured manually by using an ROI placed freehandedly at visual detection of the borders of the abnormality by one radiologist (C.M.). The findings on the perfusion images were classified according the following categories: 1, definitely normal; 2, probably normal; 3, probably abnormal; and 4, definitely abnormal. Images with findings assigned to categories 1 and 2 were rated as negative for perfusion abnormalities, and those with findings assigned to categories 3 and 4 were rated as positive for these abnormalities. The sensitivity values with which perfusion deficits were diagnosed from images obtained in the studies with contrast agent doses of 0.05 and 0.10 mmol were calculated as follows: number of true-positive diagnoses of abnormal perfusion (diagnostic categories 3 and 4) divided by the number of all patients who received the final diagnosis of a perfusion abnormality.

The reference standard for the final diagnosis was established by combining information about the results of structural brain imaging (fluid-attenuated inversion-recovery and T2-weighted images, pre- and postcontrast T1-weighted spin-echo images, and diffusion-weighted images) and the clinical findings in the patients.

Analysis of recirculation effects.—To find out whether the second perfusion series would be contaminated with recirculation contrast agent effects and/or a reduced baseline signal, we compared signal-to-noise ratio and baseline signal of the first perfusion series with those of the second perfusion series in each patient. To accomplish this comparison, the precontrast images (set at bolus arrival time point minus 3 dynamic frames, ie, 4.5 seconds) that were obtained with 0.05 and 0.10 mmol in the perfusion series were loaded into the perfusion software package. One of the authors (C.M.) placed ROIs with a mean ROI size of 150 mm2 in the white matter areas of the centrum semiovale, in the putamen, in the thalamus, in the gray matter areas of the cerebellum, and in the ghosting-free part of the image background. Signal-to-noise ratios were calculated by dividing the signal intensity of the brain regions by the standard deviation of the background signal.

Statistical Analysis
A statistical software package (SPSS, version 10; SPSS, Chicago, Ill) was used for analysis. The image quality scores of the perfusion echo-planar PRESTO source images and of the corresponding perfusion maps obtained with contrast agent doses of 0.05 and 0.10 mmol were compared by using a marginal homogeneity test for multinomial variables that was based on the Mantel-Haenszel statistic for ordered categorized values. To investigate whether the first injection of contrast material had a significant influence on the results of the second study (ie, to identify recirculation effects), the signal-to-noise ratio and the baseline signal of the two perfusion studies performed in each subject were compared and tested for statistically significant differences with the Student t test for paired samples. The MTT, NI, and maximum T2* effect values from the ROI analysis were compared by using a one-tailed Student t test for paired samples. Quantitative data about the number of gamma-fitted pixels and the derived perfusion data were compared by using the t test for paired samples. A difference with a P value of .05 with the t test for paired samples was considered significant.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
None of the data sets had to be excluded because of gross head motion, so those in all 26 patients, with a total of 52 perfusion studies that were performed, were accepted for further analysis.

Initial Dose-Finding Study in Volunteers
In the three volunteers, dynamic susceptibility-weighted contrast-enhanced perfusion MR imaging with 0.20 mmol of gadopentetate dimeglumine induced an almost complete signal intensity loss, with a mean maximum T2* effect of 98%, 99%, and 98% for gray matter voxels. With contrast agent doses of 0.10 and 0.05 mmol, a mean maximum T2* effect of 25% and 17% for gray matter voxels was observed. We found that no sharp peak (gamma curve–like) signal was detectable in highly perfused gray matter areas and vessels during bolus passage of the full dose of 0.20 mmol of gadopentetate dimeglumine, but a plateau was reached, which was followed by signal recovery. This activity was due to saturation effects. The absence of a gamma-fittable perfusion time course made the calculation of perfusion variables (MTT, NI) impossible. Accordingly, the clinical perfusion study was performed with doses of 0.10 and 0.05 mmol of gadopentetate dimeglumine.

Clinical Perfusion Study
Quality of source images.—Given the fact that we were dealing with T2*-weighted gradient-echo echo-planar images at 3.0 T, image quality was surprisingly stable (Fig 1). On images in all 52 perfusion imaging studies, only minor distortions were found in the cerebral tissue close to the base of the skull, particularly in the posterior fossa. A score of 2 was assigned to images in 47 of 52 studies; in the remaining studies, a score of 3 was assigned to the images. In none of the 52 studies was image quality determined to be unacceptable (score 5).



View larger version (67K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1. Transverse T2*-weighted PRESTO perfusion MR source images (14/21; field of view, 255 mm; rectangular field of view, set at 80%; isotropic voxel size, 3.98 x 4.00 x 4.00 mm; and noninterpolated acquisition matrix, 64 x 64) show three sections close to skull base. Despite high-field-strength situation, image distortions were read as minor and were confined to sections closest to skull base.

 
Quality of the perfusion maps.—The perfusion maps were rated to be of sufficient diagnostic quality in all 52 studies; however, maps generated from the dynamic susceptibility-weighted contrast-enhanced studies with a contrast agent dose of 0.05 mmol received slightly lower scores compared with the scores assigned to the maps from studies performed with a contrast agent dose of 0.10 mmol: Perfusion maps obtained with a dose of 0.10 mmol were judged to be of good or excellent diagnostic quality in 25 of 26 patients (mean score, 1.42 ± 0.58); the perfusion maps obtained with a dose of 0.05 mmol were rated as good or as adequate for diagnosis (mean score, 2.00 ± 0.58) (Fig 2). None of the perfusion maps received a score of 4 (poor) or 5 (nondiagnostic). The marginal homogeneity test for multinomial variables showed different values in three cases and 15 off-diagonals with an asymptotic significance (two-tailed t test) of P < .001. The results of the visual analysis were corroborated by the results of the analysis of the total number of gamma-fitted voxels. The mean number of voxels that were analyzed per brain was 73 058 ± 14 073. The mean number of voxels that were gamma fitted for dynamic susceptibility-weighted contrast-enhanced perfusion MR imaging with the 0.05-mmol dose was 68 890 ± 13 750, with a mean percentage of gamma-fitted pixels of 94%. With the dose of 0.10 mmol, the mean number of pixels accepted for gamma fitting was 77 226 ± 14 245, with a mean percentage of gamma-fitted pixels of 99%. At perfusion imaging, the decrease in fitted voxels between the 0.10-mmol dose and the 0.05-mmol dose (Fig 3) proved to be significant with a P value of less than .001.



View larger version (107K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2a. Images obtained in 68-year-old woman with symptoms of acute stroke and left-sided hemiparesis. (a) Diffusion-weighted single-shot echo-planar MR images (4225/74; b value, 1000 sec/mm2) show acute ischemia (arrow) in right internal capsule. (b) Corresponding fluid-attenuated inversion-recovery MR image (6000/80/2000) reveals lesion with hyperintense signal. White circles are ROIs that were placed in the infarcted areas in right internal capsule (Ll), ipsilateral lentiform nucleus (L2), and contralateral internal capsule (L3). Respective time-signal intensity curves are presented in e for perfusion study with 0.10-mmol dose and in f for study with 0.05-mmol dose. (c, d) NI parametric maps of perfusion series with 0.10-mmol and 0.05-mmol doses, respectively. Arrow indicates site of infarction. (e) Gamma-fitted time-signal intensity curves from perfusion study with 0.10-mmol dose. Perfusion time course is given for the ischemic lesion (dark gray line), contralateral internal capsule (black line), and ipsilateral caudate nucleus (light gray line). (f) Gamma-fitted time-signal intensity curves from perfusion study with 0.05-mmol dose. Denomination of curves same as in e. Overall image quality of perfusion maps with 0.10- and 0.05-mmol doses is acceptable (scores of 5 and 4, for excellent and good diagnostic quality, respectively, were assigned). Small ischemic lesion is visible on NI parametric maps for perfusion studies with both doses, as in c and d.

 


View larger version (87K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2b. Images obtained in 68-year-old woman with symptoms of acute stroke and left-sided hemiparesis. (a) Diffusion-weighted single-shot echo-planar MR images (4225/74; b value, 1000 sec/mm2) show acute ischemia (arrow) in right internal capsule. (b) Corresponding fluid-attenuated inversion-recovery MR image (6000/80/2000) reveals lesion with hyperintense signal. White circles are ROIs that were placed in the infarcted areas in right internal capsule (Ll), ipsilateral lentiform nucleus (L2), and contralateral internal capsule (L3). Respective time-signal intensity curves are presented in e for perfusion study with 0.10-mmol dose and in f for study with 0.05-mmol dose. (c, d) NI parametric maps of perfusion series with 0.10-mmol and 0.05-mmol doses, respectively. Arrow indicates site of infarction. (e) Gamma-fitted time-signal intensity curves from perfusion study with 0.10-mmol dose. Perfusion time course is given for the ischemic lesion (dark gray line), contralateral internal capsule (black line), and ipsilateral caudate nucleus (light gray line). (f) Gamma-fitted time-signal intensity curves from perfusion study with 0.05-mmol dose. Denomination of curves same as in e. Overall image quality of perfusion maps with 0.10- and 0.05-mmol doses is acceptable (scores of 5 and 4, for excellent and good diagnostic quality, respectively, were assigned). Small ischemic lesion is visible on NI parametric maps for perfusion studies with both doses, as in c and d.

 


View larger version (115K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2c. Images obtained in 68-year-old woman with symptoms of acute stroke and left-sided hemiparesis. (a) Diffusion-weighted single-shot echo-planar MR images (4225/74; b value, 1000 sec/mm2) show acute ischemia (arrow) in right internal capsule. (b) Corresponding fluid-attenuated inversion-recovery MR image (6000/80/2000) reveals lesion with hyperintense signal. White circles are ROIs that were placed in the infarcted areas in right internal capsule (Ll), ipsilateral lentiform nucleus (L2), and contralateral internal capsule (L3). Respective time-signal intensity curves are presented in e for perfusion study with 0.10-mmol dose and in f for study with 0.05-mmol dose. (c, d) NI parametric maps of perfusion series with 0.10-mmol and 0.05-mmol doses, respectively. Arrow indicates site of infarction. (e) Gamma-fitted time-signal intensity curves from perfusion study with 0.10-mmol dose. Perfusion time course is given for the ischemic lesion (dark gray line), contralateral internal capsule (black line), and ipsilateral caudate nucleus (light gray line). (f) Gamma-fitted time-signal intensity curves from perfusion study with 0.05-mmol dose. Denomination of curves same as in e. Overall image quality of perfusion maps with 0.10- and 0.05-mmol doses is acceptable (scores of 5 and 4, for excellent and good diagnostic quality, respectively, were assigned). Small ischemic lesion is visible on NI parametric maps for perfusion studies with both doses, as in c and d.

 


View larger version (118K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2d. Images obtained in 68-year-old woman with symptoms of acute stroke and left-sided hemiparesis. (a) Diffusion-weighted single-shot echo-planar MR images (4225/74; b value, 1000 sec/mm2) show acute ischemia (arrow) in right internal capsule. (b) Corresponding fluid-attenuated inversion-recovery MR image (6000/80/2000) reveals lesion with hyperintense signal. White circles are ROIs that were placed in the infarcted areas in right internal capsule (Ll), ipsilateral lentiform nucleus (L2), and contralateral internal capsule (L3). Respective time-signal intensity curves are presented in e for perfusion study with 0.10-mmol dose and in f for study with 0.05-mmol dose. (c, d) NI parametric maps of perfusion series with 0.10-mmol and 0.05-mmol doses, respectively. Arrow indicates site of infarction. (e) Gamma-fitted time-signal intensity curves from perfusion study with 0.10-mmol dose. Perfusion time course is given for the ischemic lesion (dark gray line), contralateral internal capsule (black line), and ipsilateral caudate nucleus (light gray line). (f) Gamma-fitted time-signal intensity curves from perfusion study with 0.05-mmol dose. Denomination of curves same as in e. Overall image quality of perfusion maps with 0.10- and 0.05-mmol doses is acceptable (scores of 5 and 4, for excellent and good diagnostic quality, respectively, were assigned). Small ischemic lesion is visible on NI parametric maps for perfusion studies with both doses, as in c and d.

 


View larger version (24K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2e. Images obtained in 68-year-old woman with symptoms of acute stroke and left-sided hemiparesis. (a) Diffusion-weighted single-shot echo-planar MR images (4225/74; b value, 1000 sec/mm2) show acute ischemia (arrow) in right internal capsule. (b) Corresponding fluid-attenuated inversion-recovery MR image (6000/80/2000) reveals lesion with hyperintense signal. White circles are ROIs that were placed in the infarcted areas in right internal capsule (Ll), ipsilateral lentiform nucleus (L2), and contralateral internal capsule (L3). Respective time-signal intensity curves are presented in e for perfusion study with 0.10-mmol dose and in f for study with 0.05-mmol dose. (c, d) NI parametric maps of perfusion series with 0.10-mmol and 0.05-mmol doses, respectively. Arrow indicates site of infarction. (e) Gamma-fitted time-signal intensity curves from perfusion study with 0.10-mmol dose. Perfusion time course is given for the ischemic lesion (dark gray line), contralateral internal capsule (black line), and ipsilateral caudate nucleus (light gray line). (f) Gamma-fitted time-signal intensity curves from perfusion study with 0.05-mmol dose. Denomination of curves same as in e. Overall image quality of perfusion maps with 0.10- and 0.05-mmol doses is acceptable (scores of 5 and 4, for excellent and good diagnostic quality, respectively, were assigned). Small ischemic lesion is visible on NI parametric maps for perfusion studies with both doses, as in c and d.

 


View larger version (20K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2f. Images obtained in 68-year-old woman with symptoms of acute stroke and left-sided hemiparesis. (a) Diffusion-weighted single-shot echo-planar MR images (4225/74; b value, 1000 sec/mm2) show acute ischemia (arrow) in right internal capsule. (b) Corresponding fluid-attenuated inversion-recovery MR image (6000/80/2000) reveals lesion with hyperintense signal. White circles are ROIs that were placed in the infarcted areas in right internal capsule (Ll), ipsilateral lentiform nucleus (L2), and contralateral internal capsule (L3). Respective time-signal intensity curves are presented in e for perfusion study with 0.10-mmol dose and in f for study with 0.05-mmol dose. (c, d) NI parametric maps of perfusion series with 0.10-mmol and 0.05-mmol doses, respectively. Arrow indicates site of infarction. (e) Gamma-fitted time-signal intensity curves from perfusion study with 0.10-mmol dose. Perfusion time course is given for the ischemic lesion (dark gray line), contralateral internal capsule (black line), and ipsilateral caudate nucleus (light gray line). (f) Gamma-fitted time-signal intensity curves from perfusion study with 0.05-mmol dose. Denomination of curves same as in e. Overall image quality of perfusion maps with 0.10- and 0.05-mmol doses is acceptable (scores of 5 and 4, for excellent and good diagnostic quality, respectively, were assigned). Small ischemic lesion is visible on NI parametric maps for perfusion studies with both doses, as in c and d.

 


View larger version (31K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3. Box plot shows number of voxels analyzed for studies with 0.05- and 0.10-mmol doses. Number of gamma-fitted voxels differs by only 10% between the two doses.

 
Quantitative analysis of perfusion parameters.—The maximum T2* effect during the first pass was analyzed. The mean peak signal decrease (maximum T2* effect) for gray matter pixels was 25.4% ± 9.8 for the perfusion series with the 0.10-mmol dose, compared with 17.5% ± 9.0 for the perfusion series with the 0.05-mmol dose. The respective mean maximum T2* effect in white matter pixels was 15.2% ± 4.5 and 7.7% ± 2.9, respectively, for the two doses. The difference regarding maximum signal intensity decrease was statistically significant, with a P value of less than .01.

In regard to MTT, this value did not change when we compared studies performed with doses of 0.10 mmol with those performed with doses of 0.05 mmol. Neither the white matter voxels nor the gray matter voxels revealed differences in regard to MTT with different doses of contrast agents (P > .5). Mean MTT values for gray matter voxels in perfusion studies were 6.7 seconds ± 1.2 for the 0.10-mmol dose and 6.3 seconds ± 1.7 for the 0.05-mmol dose. Mean MTT values for white matter voxels was 7.8 seconds ± 1.3 and 6.9 seconds ± 1.9, respectively, for the two doses (Fig 4).



View larger version (39K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4. Box plot shows mean MTTs for gray matter (GM) and white matter (WM) with 0.05- and 0.10-mmol doses. MTT did not vary significantly (P > .05) between the doses.

 
With respect to NI, there was a significant difference in this value between the 0.10-mmol and the 0.05-mmol doses (Fig 5). Mean NI for gray matter voxels was 3.10 ± 1.60 for the 0.10-mmol dose and 1.56 ± 1.16 for the 0.05-mmol dose. Mean NI for white matter voxels was 1.35 ± 0.59 with the 0.10-mmol dose and 0.59 ± 0.30 with the 0.05-mmol dose.



View larger version (29K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5. Box plot shows mean NI, which represents the relative regional cerebral blood volume (rrCBV) for gray matter (GM) and white matter (WM) with 0.05- and 0.10-mmol doses. NI is reduced by half with half the contrast agent dose (0.05 mmol, compared with 0.10 mmol).

 
Clinical reading of perfusion maps.—Seven of 26 patients received the final diagnosis of perfusion abnormalities. Of seven, six prospectively received a diagnosis at dynamic susceptibility-weighted contrast-enhanced perfusion MR imaging in both the study with the 0.05-mmol dose and in that with the 0.10-mmol dose. This translates into a sensitivity of 86% (six of seven patients) for both protocols. One subacute ischemic lesion in a 62-year-old patient was missed at both perfusion studies; it was a small ischemic lesion in the dorsofrontal cortex, which was identified with the diffusion-weighted pulse sequence. The sizes of the detected perfusion abnormalities ranged from 64 to 345 mm2 (in plane); the maximum diameter of the perfusion deficit did not vary between the dynamic susceptibility-weighted contrast-enhanced perfusion study performed with the 0.05-mmol dose and that performed with the 0.10-mmol dose (mean variation, ±3%).

Analysis of possible recirculation effects.—In the dynamic susceptibility-weighted contrast-enhanced perfusion MR studies with the 0.10- and 0.05-mmol doses, the mean signal-to-noise ratio was 122.8 ± 36.9 and 117.5 ± 36.8, respectively. The difference between these values was not statistically significant (P = .13).

A mean signal-to-noise ratio of 124.4 ± 39.5 was obtained in the first perfusion study and that of 121.7 ± 36.8 was obtained in the second study in the respective patients when we compared the influence of already applied contrast medium on baseline signal, which did not reveal a statistically significant difference (P = .4). These results indicate that no recirculation effects or baseline signal intensity shift had to be considered for the data analysis of the two perfusion studies in a given patient.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
At higher magnetic field strengths, the shorter T2 and T2* relaxation times and higher sensitivity for susceptibility effects may influence T2*-mediated first-pass perfusion imaging in several ways. The stronger susceptibility effects at 3.0 T may impair image quality secondary to image distortions and blurring around tissue interfaces and, therefore, may interfere with an accurate assessment of signal intensity changes mediated by the bolus passage of a paramagnetic contrast agent. On the other hand, the shorter T2 and T2* relaxation times and the higher signal-to-noise ratio that are brought about by higher field strengths may translate into a higher effectiveness of a given amount of contrast agent for perfusion imaging.

Thus, this study was undertaken to investigate whether whole-brain perfusion imaging by using T2*-weighted gradient-echo echo-planar imaging pulse sequences is feasible at 3.0 T and to find out which dose of contrast agent is sufficient to generate perfusion maps of adequate diagnostic quality.

In regard to the first objective, our results suggest that image quality of the perfusion source images was not impaired. Despite that we were using susceptibility-weighted T2*-weighted echo-planar imaging pulse sequences at 3.0 T, image distortions and blurring were rated as minor in the majority of images obtained in the studies; this was true also for problematic anatomic areas, such as the posterior fossa and the regions close to the base of the skull (eg, hippocampus and brainstem) (1,43–46). This was probably caused by the use of a multishot echo-shifted three-dimensional gradient-echo echo-planar imaging (PRESTO) sequence. The short echo train length in this pulse sequence seems to compensate for the field-dependent increase in susceptibility effects, while an adequate sensitivity for gadolinium chelate–mediated perfusion effects is maintained. Overall image quality of the source images seemed comparable to that in studies that have been published about the same dynamic susceptibility-weighted contrast-enhanced perfusion technique performed at 1.5 T (1).

In regard to the quality of the resulting parametric maps, the perfusion maps obtained with the very low dose of contrast agent (0.05 mmol) were assigned scores significantly lower than were those obtained with a dose of 0.10 mmol. This was also reflected by the number of pixels that were accepted for gamma fitting (99% vs 94% for the 0.10- and 0.05-mmol dose perfusion studies, respectively). Yet, even with the very low dose of 0.05 mmol, perfusion maps at 3.0 T seemed to offer an at least equivalent quality compared with that in previously published studies about the reference technique performed with the full dose of 0.20 mmol at 1.5 T. With that technique, the mean number of gamma-fitted pixels was reported to be 90% (1), which was calculated by using the same type of software and procedure for perfusion analysis as we used in this study.

In regard to the second issue, the following statements can be made. The contrast agent dose that is currently considered optimal for T2*-weighted dynamic susceptibility-weighted contrast-enhanced perfusion imaging at 1.5 T is 0.20 mmol (29,30,35). There also are reports about the use of only a 0.10-mmol dose at 1.5 T, yet there is agreement that this dose is associated with some substantial disadvantages: With a contrast agent dose of 0.10 mmol, T2*-weighted perfusion MR imaging at 1.5 T needs to be performed with single-shot echo-planar imaging pulse sequences with a very long echo train and a low bandwidth, because only these are sensitive enough to the small susceptibility effects that are associated with small contrast agent doses. The long echo train and the low bandwidth of these pulse sequences, however, lead to increased image distortions that have been observed at 1.5 T, which probably precludes their use with systems that have an even higher field strength. Also, compared with PRESTO, these pulse sequences would be accompanied by a reduced temporal resolution and/or spatial resolution. This, in turn, limits the anatomic coverage and the accuracy with which perfusion-related values are calculated. Accordingly, most of the recently published studies (32,33,50–52) about perfusion imaging at 1.5 T were performed with a contrast agent dose of 0.20 mmol.

When this dose was used for dynamic susceptibility-weighted contrast-enhanced perfusion MR imaging at 3.0 T, the stronger susceptibility effects caused a virtually complete signal void during the first pass of gadolinium-based contrast agents, particularly in gray matter. This complete signal loss interferes with the accurate calculation of perfusion parameters and would reduce the sensitivity for perfusion deficits or variations, particularly in regard to the assessment of hypervascularized lesions, of the technique. Accordingly, the dose of 0.20 mmol probably is not useful or, to put it in positive terms, it is not required for first-pass perfusion imaging at 3.0 T.

With a dose of 0.10 mmol, perfusion imaging is feasible at 3.0 T, with excellent quality of the perfusion maps. On the basis of results of both subjective analysis of the quality of the perfusion maps and quantitative assessment of the perfusion variables, data for perfusion imaging at 3.0 T compared favorably with the data published for dynamic susceptibility-weighted contrast-enhanced perfusion MR imaging with a contrast agent dose of 0.20 mmol at 1.5 T, even if the same pulse sequence was used (1,33,36).

To further investigate how the contrast agent dose influences dynamic susceptibility-weighted contrast-enhanced perfusion MR imaging at 3.0 T, we also performed perfusion imaging with only 0.05 mmol. Compared with the results of the perfusion studies in the same patients with the 0.10-mmol dose, there was a significant decrease in the quality of the perfusion maps, with respect to both visual inspection and in regard to quantitative analysis (number of gamma-fitted pixels). Yet, even with the very small dose of contrast agent, the image quality of the perfusion maps was rated to be good or at least adequate for diagnosis in all 26 patients. Even more important, there was no difference in regard to the sensitivity with which perfusion abnormalities were prospectively diagnosed.

When we compared the maximum T2* effect of perfusion imaging with a dose of 0.10 mmol and that of imaging with a dose of 0.05 mmol, we noted a nonlinear dose dependency of the maximum T2* effect in the gray matter voxels, whereas the reduction was linear in white matter voxels. The maximum T2* effect of gray matter was reduced from 25.4% to 17.5% with doses of 0.10 mmol and of 0.05 mmol, respectively, whereas the maximum T2* effect in white matter voxels decreased from 15.2% to 7.7% with each of these doses, respectively. This can be explained by saturation effects that are secondary to a high local contrast medium concentration in highly perfused gray matter voxels (29). We speculate, therefore, that even with a dose of 0.10 mmol at 3.0 T, saturation effects are already present that may interfere with the calculation of cerebral perfusion parameters. Reduction of the dose could, therefore, translate into a linear relationship between signal intensity loss and actual perfusion and, as such, into a possibly even more accurate delineation of subtle perfusion changes.

As one would anticipate, the NI (formerly called regional cerebral blood volume, or rCBV) was reduced with lower contrast agent doses; this effect is independent of field strength and is in accord with data in previous studies (26). More important, the reduced NI at a dose of 0.05 mmol did not translate into a clinically perceivable reduction in the quality of the parametric maps.

There were several limitations to our study. The most important limitation is that we cannot offer an intraindividual comparison with the imaging technique that is considered the reference standard, which is perfusion imaging at 1.5 T with a contrast agent dose of 0.20 mmol. This comparison would, of course, be desirable to directly correlate the T2* efficiency of the standard technique to that of the high-field-strength technique. Although we can offer a comparison with the published results of another study (1) in which a group of patients were included who underwent perfusion imaging at 1.5 T by using the same pulse sequence (PRESTO) and a contrast agent dose of 0.20 mmol, our conclusions in regard to equivalence with imaging at 3.0 T with a contrast agent dose of 0.10 mmol are preliminary and need to be confirmed in a larger group of patients because of the interindividual setting in this study. For the same reason, the small number of patients with perfusion deficits is another limitation of our study. Our conclusions in regard to the sensitivity for the detection of perfusion abnormalities, especially with respect to very-low-dose perfusion imaging with a contrast agent dose of 0.05 mmol have to be considered as preliminary until confirmed in a larger group of patients with perfusion deficits.

Some technical limitations may apply, as well. One potential limitation is that we performed the studies with two doses (0.10 and 0.05 mmol) within a short time (ie, 4 minutes) of each other. Recirculation effects and different baseline signal intensity levels may, in principle, interfere with the results. Every attempt, however, was made to account for this interference by doing the following: First, the order of the dose studies was randomized such that a systematic bias would be minimized. Second, we compared baseline signal intensity of the images obtained in the first perfusion study with that of those obtained in the second perfusion study in each patient and did not find a significant difference, which strongly suggests that recirculation effects should not be important. This is also in accordance with results in previously published work by Runge and co-workers (53), who demonstrated that a repeat contrast agent injection for first-pass perfusion MR imaging does not affect the perfusion variables, such as the maximum T2* effect (peak signal intensity decreases).

Further, the three-dimensional multishot echo-shifted gradient-echo echo-planar imaging (PRESTO) pulse sequence used in our study belongs, just as does the BURST pulse sequence, to the most advanced family of fast T2*-weighted perfusion imaging pulse sequences that is currently available (1,43–47); PRESTO offers whole-brain coverage at a temporal resolution of 1.5 seconds per dynamic image. The use of this pulse sequence can be considered a study limitation because there are currently only a limited number of institutions that have this type of pulse sequence available. Because of the shorter echo train length of PRESTO compared with that of standard single-shot echo-planar imaging sequences, these pulse sequences are not as affected by susceptibility artifacts as are other sequences; as such, they seem particularly suitable for use at higher magnetic field strengths. If this type of pulse sequence is not available with a high-field-strength system, it is probable that our results in regard to image quality with dynamic susceptibility-weighted contrast-enhanced perfusion MR imaging may not be reproducible. In regard to our conclusions about feasibility of 3.0 T perfusion imaging with a reduced contrast agent dose, however, we are confident that our results are reproducible with other high-field-strength systems, even if PRESTO pulse sequences are not available: For any given amount of contrast agent, the susceptibility-induced signal intensity loss will be smaller for imaging with PRESTO compared with standard single-shot echo-planar imaging. Thus, if a PRESTO pulse sequence works with a reduced dose of contrast agent, then we have even more reason to assume that the more sensitive standard single-shot pulse sequences should work with the reduced dose too.

In conclusion, our results suggest that dynamic susceptibility-weighted contrast-enhanced perfusion MR imaging is feasible at 3.0 T. At 3.0 T, the contrast agent dose for dynamic susceptibility-weighted contrast-enhanced perfusion MR imaging can be reduced to account for the stronger susceptibility effects at higher magnetic field strengths. At 3.0 T, with a dose of 0.10 mmol of contrast agent, T2* effects are achieved that seem equivalent to the effects reported in published data for imaging at 1.5 T with a contrast agent dose of 0.20 mmol; probably, dynamic susceptibility-weighted contrast-enhanced perfusion MR imaging will even be feasible with a dose of 0.05 mmol. Further studies are needed and are under way in our department to corroborate these findings in a larger group of patients with perfusion deficits.


   FOOTNOTES
 
Abbreviations: MTT = mean transit time, NI = negative integral, PRESTO = principles of echo shifting with a train of observations, ROI = region of interest

Authors stated no financial relationship to disclose.

Author contributions: Guarantors of integrity of entire study, C.M., H.H.S., C.K.K.; study concepts and design, C.M., H.H.S., F.T., C.K.K.; literature research, C.M.; clinical and experimental studies, C.M.; data acquisition, C.M., F.T., J.G.; data analysis/interpretation, C.M., F.T., J.G., C.K.K.; statistical analysis, C.M., F.T.; manuscript preparation, C.M., C.K.K.; manuscript definition of intellectual content, C.M., C.K.K., H.H.S.; manuscript editing, C.M.; manuscript revision/review and final version approval, all authors


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Flacke S, Urbach H, Folkers PJ, et al. Ultra-fast three-dimensional MR perfusion imaging of the entire brain in acute stroke assessment. J Magn Reson Imaging 2000; 11:250-259.[CrossRef][Medline]
  2. Fiehler J, von Bezold M, Kucinski T, et al. Cerebral blood flow predicts lesion growth in acute stroke patients. Stroke 2002; 33:2421-2425.[Abstract/Free Full Text]
  3. Liu Y, Karonen JO, Vanninen RL, et al. Cerebral hemodynamics in human acute ischemic stroke: a study with diffusion- and perfusion-weighted magnetic resonance imaging and SPECT. J Cereb Blood Flow Metab 2000; 20:910-920.[Medline]
  4. Parsons MW, Yang Q, Barber PA, et al. Perfusion magnetic resonance imaging maps in hyperacute stroke: relative cerebral blood flow most accurately identifies tissue destined to infarct. Stroke 2001; 32:1581-1587.[Abstract/Free Full Text]
  5. Latchaw RE, Yonas H, Hunter GJ, et al. Guidelines and recommendations for perfusion imaging in cerebral ischemia: a scientific statement for healthcare professionals by the writing group on perfusion imaging, from the Council on Cardiovascular Radiology of the American Heart Association. Stroke 2003; 34:1084-1104.[Free Full Text]
  6. Doerfler A, Eckstein HH, Eichbaum M, et al. Perfusion-weighted magnetic resonance imaging in patients with carotid artery disease before and after carotid endarterectomy. J Vasc Surg 2001; 34:587-593.[CrossRef][Medline]
  7. Calamante F, Ganesan V, Kirkham FJ, et al. MR perfusion imaging in Moyamoya syndrome: potential implications for clinical evaluation of occlusive cerebrovascular disease. Stroke 2001; 32:2810-2816.[Abstract/Free Full Text]
  8. Maeda M, Yuh WT, Ueda T, et al. Severe occlusive carotid artery disease: hemodynamic assessment by MR perfusion imaging in symptomatic patients. AJNR Am J Neuroradiol 1999; 20:43-51.[Abstract/Free Full Text]
  9. Michel E, Liu H, Remley KB, et al. Perfusion MR neuroimaging in patients undergoing balloon test occlusion of the internal carotid artery. AJNR Am J Neuroradiol 2001; 22:1590-1596.[Abstract/Free Full Text]
  10. Hartmann M, Heiland S, Sartor K. Functional MRI procedures in the diagnosis of brain tumors: perfusion- and diffusion-weighted imaging. Rofo Fortschr Geb Rontgenstr Neuen Bildgeb Verfahr 2002; 174:955-964. [German].[Medline]
  11. Kassner A, Annesley DJ, Zhu XP, et al. Abnormalities of the contrast re-circulation phase in cerebral tumors demonstrated using dynamic susceptibility contrast-enhanced imaging: a possible marker of vascular tortuosity. J Magn Reson Imaging 2000; 11:103-113.[CrossRef][Medline]
  12. Le Bihan D, Douek P, Argyropoulou M, Turner R, Patronas N, Fulham M. Diffusion and perfusion magnetic resonance imaging in brain tumors. Top Magn Reson Imaging 1993; 5:25-31.[Medline]
  13. Provenzale JM, Wang GR, Brenner T, Petrella JR, Sorensen AG. Comparison of permeability in high-grade and low-grade brain tumors using dynamic susceptibility contrast MR imaging. AJR Am J Roentgenol 2002; 178:711-716.[Abstract/Free Full Text]
  14. Sartor K. MR imaging of the brain: tumors. Eur Radiol 1999; 9:1047-1054.[CrossRef][Medline]
  15. Alexander E, III. Optimizing brain tumor resection: midfield interventional MR imaging. Neuroimaging Clin N Am 2001; 11:659-672.[Medline]
  16. Lee SJ, Kim JH, Kim YM, et al. Perfusion MR imaging in gliomas: comparison with histologic tumor grade. Korean J Radiol 2001; 2:1-7.[Medline]
  17. Maeda M, Itoh S, Kimura H, et al. Tumor vascularity in the brain: evaluation with dynamic susceptibility-contrast MR imaging. Radiology 1993; 189:233-238.[Abstract/Free Full Text]
  18. Vonken EP, Van Osch MJ, Willems PW, et al. Repeated quantitative perfusion and contrast permeability measurement in the MRI examination of a CNS tumor. Eur Radiol 2000; 10:1447-1451.[CrossRef][Medline]
  19. Hartmann M, Heiland S, Harting I, et al. Distinguishing of primary cerebral lymphoma from high-grade glioma with perfusion-weighted magnetic resonance imaging. Neurosci Lett 2003; 338:119-122.[CrossRef][Medline]
  20. Aichner F, Wagner M, Kremser C, Felber S. MR-imaging of non-Alzheimer’s dementia. J Neural Transm Suppl 1996; 47:143-153.[Medline]
  21. Bozzao A, Floris R, Baviera ME, Apruzzese A, Simonetti G. Diffusion and perfusion MR imaging in cases of Alzheimer’s disease: correlations with cortical atrophy and lesion load. AJNR Am J Neuroradiol 2001; 22:1030-1036.[Abstract/Free Full Text]
  22. Firbank MJ, Colloby SJ, Burn DJ, McKeith IG, O’Brien JT. Regional cerebral blood flow in Parkinson’s disease with and without dementia. Neuroimage 2003; 20:1309-1319.[CrossRef][Medline]
  23. Chabriat H, Pappata S, Ostergaard L, et al. Cerebral hemodynamics in CADASIL before and after acetazolamide challenge assessed with MRI bolus tracking. Stroke 2000; 31:1904-1912.[Abstract/Free Full Text]
  24. Brusa L, Bassi A, Pierantozzi M, Gaudiello SF, Floris R, Stanzione P. Perfusion-weighted dynamic susceptibility (DSC) MRI: basal ganglia hemodynamic changes after apomorphine in Parkinson’s disease. Neurol Sci 2002; 23(suppl 2):S61-S62.
  25. Maas LC, Harris GJ, Satlin A, English CD, Lewis RF, Renshaw PF. Regional cerebral blood volume measured by dynamic susceptibility contrast MR imaging in Alzheimer’s disease: a principal components analysis. J Magn Reson Imaging 1997; 7:215-219.[Medline]
  26. Edelman RR, Mattle HP, Atkinson DJ, et al. Cerebral blood flow: assessment with dynamic contrast-enhanced T2*-weighted MR imaging at 1.5 T. Radiology 1990; 176:211-220.[Abstract/Free Full Text]
  27. Fisel CR, Ackerman JL, Buxton RB, et al. MR contrast due to microscopically heterogeneous magnetic susceptibility: numerical simulations and applications to cerebral physiology. Magn Reson Med 1991; 17:336-347.[Medline]
  28. Rosen BR, Belliveau JW, Buchbinder BR, et al. Contrast agents and cerebral hemodynamics. Magn Reson Med 1991; 19:285-292.[Medline]
  29. Erb G, Benner T, Heiland S, Reith W, Sartor K, Forsting M. Studies on contrast medium dosage in perfusion-weighted MR imaging. Rofo Fortschr Geb Rontgenstr Neuen Bildgeb Verfahr 1997; 167:599-604. [German].[Medline]
  30. Keston P, Murray AD, Jackson A. Cerebral perfusion imaging using contrast-enhanced MRI. Clin Radiol 2003; 58:505-513.[CrossRef][Medline]
  31. Tombach B, Benner T, Reimer P, et al. Do highly concentrated gadolinium chelates improve MR brain perfusion imaging? intraindividually controlled randomized crossover concentration comparison study of 0.5 versus 1.0 mol/L gadobutrol. Radiology 2003; 226:880-888.[Abstract/Free Full Text]
  32. Warmuth C, Gunther M, Zimmer C. Quantification of blood flow in brain tumors: comparison of arterial spin labeling and dynamic susceptibility-weighted contrast-enhanced MR imaging. Radiology 2003; 228:523-532.[Abstract/Free Full Text]
  33. Flacke S, Urbach H, Block W, et al. Perfusion and molecular diffusion-weighted MR imaging of the brain: in vivo assessment of tissue alteration in cerebral ischemia. Amino Acids 2002; 23:309-316.[CrossRef][Medline]
  34. Baird AE, Lovblad KO, Dashe JF, et al. Clinical correlations of diffusion and perfusion lesion volumes in acute ischemic stroke. Cerebrovasc Dis 2000; 10:441-448.[CrossRef][Medline]
  35. Benner T, Reimer P, Erb G, et al. Cerebral MR perfusion imaging: first clinical application of a 1 M gadolinium chelate (Gadovist 1.0) in a double-blinded randomized dose-finding study. J Magn Reson Imaging 2000; 12:371-380.[CrossRef][Medline]
  36. Essig M, Lodemann KP, LeHuu M, Schonberg SO, Hubener M, Van Kaick G. Comparison of MultiHance and Gadovist for cerebral MR perfusion imaging in healthy volunteers. Radiologe 2002; 42:909-915. [German].[CrossRef][Medline]
  37. Trattnig S. 3 Tesla magnetic resonance tomography: clinical applications. Wien Med Wochenschr Suppl 2002; 113:22-27. [German].
  38. Gati JS, Menon RS, Ugurbil K, Rutt BK. Experimental determination of the BOLD field strength dependence in vessels and tissue. Magn Reson Med 1997; 38:296-302.[Medline]
  39. Wang J, Alsop DC, Li L, et al. Comparison of quantitative perfusion imaging using arterial spin labeling at 1.5 and 4.0 Tesla. Magn Reson Med 2002; 48:242-254.[CrossRef][Medline]
  40. Kruger G, Kastrup A, Glover GH. Neuroimaging at 1.5 T and 3.0 T: comparison of oxygenation-sensitive magnetic resonance imaging. Magn Reson Med 2001; 45:595-604.[CrossRef][Medline]
  41. Heiland S, Kreibich W, Reith W, et al. Comparison of echo-planar sequences for perfusion-weighted MRI: which is best? Neuroradiology 1998; 40:216-221.[CrossRef][Medline]
  42. Van Osch MJ, Vonken EJ, Wu O, Viergever MA, van der Grond J, Bakker CJ. Model of the human vasculature for studying the influence of contrast injection speed on cerebral perfusion MRI. Magn Reson Med 2003; 50:614-622.[CrossRef][Medline]
  43. Liu G, Sobering G, Duyn J, Moonen CT. A functional MRI technique combining principles of echo-shifting with a train of observations (PRESTO). Magn Reson Med 1993; 30:764-768.[Medline]
  44. Mattay VS, Frank JA, Duyn JH, et al. Three-dimensional "BURST" functional magnetic resonance imaging: initial clinical applications. Acad Radiol 1996; 3(suppl 2):S379-S383.
  45. van Gelderen P, Grandin C, Petrella JR, Moonen CT. Rapid three-dimensional MR imaging method for tracking a bolus of contrast agent through the brain. Radiology 2000; 216:603-608.[Abstract/Free Full Text]
  46. Grandin CB. Assessment of brain perfusion with MRI: methodology and application to acute stroke. Neuroradiology 2003; 45:755-766.[CrossRef][Medline]
  47. Duyn JH, van Gelderen P, Barker P, Frank JA, Mattay VS, Moonen CT. 3D bolus tracking with frequency-shifted BURST MRI. J Comput Assist Tomogr 1994; 18:680-687.[Medline]
  48. Van Osch MJ, Vonken EJ, Bakker CJ, Viergever MA. Correcting partial volume artifacts of the arterial input function in quantitative cerebral perfusion MRI. Magn Reson Med 2001; 45:477-485.[CrossRef][Medline]
  49. Heiland S, Hartmann M, Sartor K. Is perfusion MRI feasible in lesions with disrupted blood-brain barrier? pitfalls and possible solutions. Rofo Fortschr Geb Rontgenstr Neuen Bildgeb Verfahr 2000; 172:812-816. [German].[Medline]
  50. Fiehler J, Knudsen K, Kucinski T, et al. Predictors of apparent diffusion coefficient normalization in stroke patients. Stroke 2004; 35:514-519.[Abstract/Free Full Text]
  51. Hillis AE, Wityk RJ, Beauchamp NJ, Ulatowski JA, Jacobs MA, Barker PB. Perfusion-weighted MRI as a marker of response to treatment in acute and subacute stroke. Neuroradiology 2004; 46:31-39.[CrossRef][Medline]
  52. Wuerfel J, Bellmann-Strobl J, Brunecker P, et al. Changes in cerebral perfusion precede plaque formation in multiple sclerosis: a longitudinal perfusion MRI study. Brain 2004; 127:111-119.[Abstract/Free Full Text]
  53. Runge VM, Kirsch JE, Wells JW, Dunworth JN, Hilaire L, Woolfolk CE. Repeat cerebral blood volume assessment with first-pass MR imaging. J Magn Reson Imaging 1994; 4:457-461.[Medline]



This article has been cited by other articles:


Home page
 Am. J. Neuroradiol.Home page
L.S. Hu, L.C. Baxter, K.A. Smith, B.G. Feuerstein, J.P. Karis, J.M. Eschbacher, S.W. Coons, P. Nakaji, R.F. Yeh, J. Debbins, et al.
Relative Cerebral Blood Volume Values to Differentiate High-Grade Glioma Recurrence from Posttreatment Radiation Effect: Direct Correlation between Image-Guided Tissue Histopathology and Localized Dynamic Susceptibility-Weighted Contrast-Enhanced Perfusion MR Imaging Measurements
AJNR Am. J. Neuroradiol., March 1, 2009; 30(3): 552 - 558.
[Abstract] [Full Text] [PDF]

Home page
 RadiologyHome page
C. K. Kuhl, F. Traber, and H. H. Schild
Whole-Body High-Field-Strength (3.0-T) MR Imaging in Clinical Practice * Part I. Technical Considerations and Clinical Applications
Radiology, March 1, 2008; 246(3): 675 - 696.
[Abstract] [Full Text] [PDF]

Home page
 Arch NeurolHome page
M. Inglese, S.-J. Park, G. Johnson, J. S. Babb, L. Miles, H. Jaggi, J. Herbert, and R. I. Grossman
Deep Gray Matter Perfusion in Multiple Sclerosis: Dynamic Susceptibility Contrast Perfusion Magnetic Resonance Imaging at 3 T
Arch Neurol, February 1, 2007; 64(2): 196 - 202.
[Abstract] [Full Text] [PDF]

Home page
 RadiologyHome page
C. K. Kuhl, P. Jost, N. Morakkabati, O. Zivanovic, H. H. Schild, and J. Gieseke
Contrast-enhanced MR Imaging of the Breast at 3.0 and 1.5 T in the Same Patients: Initial Experience
Radiology, June 1, 2006; 239(3): 666 - 676.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2343040359v1
234/3/869    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 HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Manka, C.
Right arrow Articles by Kuhl, C. K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Manka, C.
Right arrow Articles by Kuhl, C. K.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
RADIOLOGY RADIOGRAPHICS RSNA JOURNALS ONLINE