| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Developments |
1 From the Department of Radiology (H.U., M.M., Y. Ishimori, Y. Koshimoto, H.Y., H.K., Y. Kawamura, N.H., Y. Ishii) and the Biomedical Imaging Research Center (N.S., Y.Y.), Fukui Medical University, 23 Shimoaizuki, Matsuoka-cho, Yoshida-gun, Fukui, 910-1193 Japan; GE-Yokogawa Medical Systems, Tokyo, Japan (Tsuyoshi M.); the Department of Medical Informatics, Kyoto University Hospital, Kyoto, Japan (Tetsuya M.); and the Department of Cerebral Research, National Institute for Physiological Sciences, Okazaki, Japan (N.S.). Received July 31, 1998; revision requested September 11; final revision received June 28, 1999; accepted August 12. Supported in part by research grant JSPS-RFTF97L00203 from the Japan Society for the Promotion of Science and by an international exchange grant from the Japan Society of Magnetic Resonance in Medicine. Address reprint requests to H.U. (e-mail: uematsu@fmsrsa.fukui-med.ac.jp).
 |
Abstract |
|---|
 TOP Abstract IntroductionMaterials and MethodsResultsDiscussionAppendixReferences |
|---|
Index terms: Brain, MR, 13.121411, 13.121412 • Brain, neoplasms, 13.3641, 13.366 • Magnetic resonance (MR), diffusion study, 13.12144 • Magnetic resonance (MR), perfusion study, 13.12144 • Magnetic resonance (MR), tissue characterization, 13.121411, 13.121412 • Meninges, MR, 139.121411, 139.121412 • Meninges, neoplasms, 139.366
|
Introduction |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionAppendixReferences |
|---|
The purposes of the present study were to propose a method for evaluating both vascularity and permeability of tissues simultaneously, with use of gadolinium-enhanced and dynamic MR imaging acquisitions with a double-echo technique (14), and to realize a better differentiation between meningioma and neurinoma, both of which show intense contrast enhancement (1).
|
Materials and Methods |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionAppendixReferences |
|---|
R2*) in permeable tissue that is contaminated by T1-shortening effect due to a leakage of the contrast material (12,13). The T1-shortening effect can be corrected by using double-echo technique (14). As this effect is near linearly related to the concentration of the leaked contrast material, it can be used as an index for contrast material leakage:
R2*T1U is the T1-uncorrected R2* from the single-echo data, and R2*T1C is the T1-corrected R2* from the double-echo data. Note that the leakage value should be constant after the first pass ends (t > t0), and the leakage value reflects the total amount of the leaked contrast agent, not the permeability or extraction fraction itself. On the other hand, the vascularity value (relative blood volume) (VV) can be estimated on the basis of the T2*-shortening effect of the intravascular contrast material by using the following equation, fitted with gamma function (R2*T1Cf) to eliminate the second-pass effect (16):
is time.
The leakage value can be normalized by the maximum height of the curve of R2*T1Cf to derive the leakage index (LI):
Subjects
Human studies were performed under the guidelines of the committee on clinical investigations at Fukui Medical University. Written informed consent was obtained from all patients. Eleven consecutive patients (three men and eight women; age range, 26–76 years; mean age, 49.5 years) with acoustic neurinomas and eleven consecutive patients (two men and nine women; age range, 38–73 years; mean age, 56.5 years) with meningiomas were included in the study population. The histologic subtype of the meningiomas was meningotheliomatous in seven patients, transitional in two, fibrous in one, and clear cell in one. All neurinomas originated in the acoustic nerve, and in each case the histologic diagnosis was available. The maximum diameters of meningiomas and neurinomas ranged from 2.0 to 6.0 cm (mean, 3.32 cm) and from 1.5 to 4.0 cm (mean, 2.28 cm), respectively. The diameter of the neurinomas in the vicinity of the internal auditory canal was 1.5 cm or larger; therefore, detection was not hindered by susceptibility artifacts from the mastoid bone.
MR Imaging Techniques
Transverse T1-weighted spin-echo images (repetition time msec/echo time msec = 333/10 with three signals acquired) and T2-weighted fast spin-echo images (3,500/88 [effective] with two signals acquired) were obtained (rectangular field of view, 22 x 16 cm; matrix, 256 x 224; section thickness, 5 mm) with use of a 1.5-T MR system (Horizon; GE Medical Systems, Milwaukee, Wis) before the dynamic study.
For dynamic studies of all patients, we used two echoes with echo times of 7 and 23 msec spoiled gradient-recalled acquisition in the steady state, or SPGR, sequence (33.3/7, 23; flip angle, 10°; 0.75 signal acquired; matrix, 256 x 128; section thickness, 7 mm; rectangular field of view, 24 x 16 cm). The dynamic images were obtained at the level of a single section that corresponded to the levels of the most appropriate abnormalities seen on nonenhanced MR images obtained in selected patients with brain tumors. After five images were acquired, gadopentetate dimeglumine (0.15 mmol per kilogram of body weight) was rapidly injected intravenously at a rate of 4 mL/sec with an MR-compatible power injector (MRS-50; Nemoto, Tokyo, Japan), followed by a 20-mL saline solution flush. After administration of the bolus of gadopentetate dimeglumine, a dynamic series of 50 sets of double-echo images were obtained at 2.4-second intervals. For this sequence, the total acquisition time was approximately 2 minutes. After the dynamic studies, we also obtained contrast-enhanced T1-weighted spin-echo images.
Data Analysis
Region-of-interest analysis.—Regions of interest were placed by one radiologist (H.U.) in both the normal white matter and the solid portion of tumors. Time series of signal intensity were converted to changes in R2*T1C and R2*T1U (Appendix). Gamma fitting (16) of R2*T1C data was performed to generate R2*T1Cf. The vascularity values were then obtained by means of Equation (2), and the vascularity index of the tumor was calculated with the vascularity value of the tumor normalized by that of the white matter. To obtain the leakage value, five points of (R2*T1C - R2*T1U) values immediately after the first pass of time-to-R2*T1Cf curve were averaged.
Parametric maps.—Parametric images of the vascularity and leakage indexes were generated for each subject by using in-house software. MR images were transferred to a workstation (Ultra1 Creator 3D; Sun Microsystems, Mountain View, Calif). On a pixel-by-pixel basis, the signal intensity was converted to changes in R2*T1C and R2*T1U. To obtain the vascularity value, we used a simple numeric integration of time-to-R2*T1C curve across a user-specified time interval instead of gamma fitting, because gamma fitting on a pixel-by-pixel basis is difficult due to the relatively low signal-to-noise ratio on the dynamic images. The leakage value was obtained by using the same procedure as in the region-of-interest analysis, on a pixel-by-pixel basis.
Statistical Analysis
For both meningiomas and neurinomas, a statistical review of the vascularity and leakage indexes was performed with the Mann-Whitney U test. A P value of less than .05 was considered statistically significant.
|
Results |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionAppendixReferences |
|---|
R2*T1Cf curve and low leakage values relative to the gamma-fitted R2* curve of the normal white matter (Fig 1a), which resulted in a high vascularity index and a relatively low leakage index (Fig 1b). Note that the time-to-leakage value was constant after the first pass ended. Conversely, a typical neurinoma was characterized by a large leakage value, and the height of the R2*T1Cf curve was similar to the height of the gamma-fitted R2* of the white matter (Fig 2a). This resulted in a vascularity index similar to that of the normal white matter and a leakage index higher than that of meningioma (Fig 2b). These characteristics are summarized in Figure 3.
|
|
|
|
|
The leakage values for all patients in the study ranged from 3.15 to 11.40 (mean, 6.99 ± 2.12). The mean leakage index of the neurinomas was significantly higher than that of the meningiomas (mean, 0.27 ± 0.08 [meningiomas] vs 0.88 ± 0.32 [neurinomas], P = .001, Mann-Whitney U test). The leakage index separated the two types of neoplasms except in one case.
|
Discussion |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionAppendixReferences |
|---|
Findings in the simulation study (Appendix) showed a nearly linear relationship between the leakage value and T1 change (R1) with long T10 (>1.0 second) and within a certain range of leakage values (from 0 to 12). Note that T10 and T1 are the T1 values before and after the arrival of the contrast agent, respectively. When T10 of tumors, as measured with a 1.5-T system, is greater than 1.0 (17), the leakage value in most cases observed in this study was less than 12, with the leakage value reflecting the concentration of the contrast agent in the extravascular space. The time-to-leakage value remained constant after the first pass ended, which confirmed our assumption that there is no back diffusion of the contrast agent, at least in the case of meningiomas and neurinomas.
Vascularity has been used to characterize the neoplastic process. Maeda et al (10) suggested that vascularity evaluated on the basis of the time-to-R2*T1U curve was significantly higher in meningiomas than in neurinomas, even with contamination of the T1-shortening effect caused by the contrast agent leaking into the extravascular space. Findings in the present study confirmed that meningiomas were characterized by higher vascularity than were neurinomas, although there was substantial overlap even with correction for the T1-shortening effect.
In characterization of the neoplastic process, it is essential to include leakage of the contrast agent into the extravascular space, but, to our knowledge, few studies have been performed to attempt to quantify it (18,19). In these studies, pharmacokinetic two-compartment models are used to quantify the leakage of gadopentetate dimeglumine, assuming a linear relationship between the change in R1 and the tissue concentration of gadopentetate dimeglumine and a multiexponential decay of plasma concentration of gadopentetate dimeglumine. Low temporal resolution of tissue T1 measurements, arterial blood sampling, and complicated curve analysis are cumbersome in a clinical setting. In the present study, we quantified the amount of the leaked contrast material during the first pass by measuring its counter effect on the T2* shortening due to the intravascular fraction of the contrast agent (the leakage value). Note that the leakage value reflects the total amount of the leaked contrast material and not the permeability or extraction fraction itself. Hence, we normalized the leakage value with the maximum height of R2*T1Cf to obtain the leakage index. As R2*T1Cfmax is known to be proportional to blood flow, the leakage index may in part reflect tissue permeability.
In the present study, we attempted to differentiate meningiomas and neurinomas, because they frequently originate in similar locations such as the cerebellopontine angle and the skull base, with similar patterns of contrast enhancement. Differential diagnosis is usually possible when morphologic characteristics are found such as extension along the course of cranial nerves in neurinomas or dural enhancement adjacent to the tumor in meningiomas (20,21). Previous attempts at tissue characterization on the basis of signal intensity differences or relaxation time measurements resulted in considerable overlap (21,22).
Our data showed that the leakage indexes of neurinomas were much larger than those of meningiomas except in one case, which suggests that neurinomas are more permeable to contrast media in tumor vessels than are meningiomas. This is consistent with findings in the study by Watabe and Azuma (23), who reported that an injection of gadopentetate dimeglumine caused a greater T1 increment in neurinomas than in meningiomas. This indicates that the contrast material achieved greater access into extravascular spaces in neurinomas.
According to findings in previous studies (23,24), endothelial fenestration and open-gap junctions, which are commonly found in the capillaries of neurinomas and meningiomas, function as routes into the extravascular space, where the T1 effect of gadopentetate dimeglumine can be achieved. The capillary structures in neurinomas are simple, and the gap junctions are usually short, straight, and patent (24), freely communicating with extravascular space. In meningiomas, on the other hand, the gap junctions are often tortuous, elongated, and sinusoidal, and they are frequently lined with abnormal endothelial cells similar to tumor cells (24). The difference in gap junctions between the two types of tumors may explain why, during the first transit, the contrast agent achieves greater and faster access into the extravascular space in neurinomas, which causes a larger leakage index in neurinomas than in meningiomas.
Although the proposed indexes provide acceptable results for differentiating neurinomas and meningiomas, further studies with more patients are necessary to confirm the usefulness of our theory.
In conclusion, dynamic contrast-enhanced MR imaging with double-echo technique can allow separate evaluation of both the permeability and vascularity of tumors. A differential diagnosis of meningioma and neurinoma may be possible with use of this technique.
|
|
Appendix |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionAppendixReferences |
|---|
> t0), F represents the regional blood flow, and CA() represents the input function. The relationship between the concentration of the contrast agent in the tissue vessels at the time [CV()] and CA() is given with the following:
). Hence, E is given with
Bolus injection of a paramagnetic contrast agent such as gadopentetate dimeglumine causes MR signal changes due to a change in relaxation times (T1, T2, T2*). The relationship between the MR signal (S) of a gradient echo and the relaxation time can be expressed as follows (26):
is the proton density, 
is the flip angle, TR is the repetition time, TE is the echo time, and T1 is the tissue T1 value. When images are obtained before and after the injection of a contrast agent, one can combine these equations and then eliminate and sin , so that the equations become
R2* is the T2* rate change [(1/T2*)]; S0 and S are the MR signals before and after the arrival of the contrast agent, respectively; and T10 and T1 are the T1 values before and after the arrival of the contrast agent, respectively.
If the contrast agent remains in the vascular space without leaking into the interstitial space, as would be the case in a normal brain with an intact blood-brain barrier, the T1 of the tissue is not affected by the contrast agent: T1 
T10. Hence, Equation (A6) can be simplified as follows (5–13):
R2* curve is proportional to the relative regional cerebral blood volume (rrCBV), on the basis of the linear relationship between R2* and the local concentration of the paramagnetic contrast agent and the principles of the indicator dilution theory for nondiffusible intravascular tracers (5):
and ß represent arbitrary parameters determined by means of nonlinear least squares fit. The fitted gamma variate curve is used to extrapolate the concentration curve and approximate its appearance while avoiding the confounding recirculation effects of the contrast agent.
Even during the first pass, gadopentetate dimeglumine enters the extravascular space of many tissues, except for the central nervous system with an intact blood-brain barrier. Leakage of gadopentetate dimeglumine has two effects on changes in the signal intensity of tissues. First, the loss of field heterogeneity between the vessels and the surrounding tissues affects [-ln(S/S0)/TE] in Equation (A6)(10); this effect was not considered in this study. Second, when leaked, gadopentetate dimeglumine shortens the T1 of the tissues,
R2* will be underestimated due to a leakage of the contrast agent (R2*T1U = R2* uncorrected for T1 shortening) (12,13).
Miyati et al (14) evaluated tumor vascularity with correction for T1 shortening by using double-echo MR imaging. R2* is
R2* with correction to T1 shortening (R2*T1C) is
R2* corrected for T1 shortening (R2*T1C) (14) fitted with the gamma function to eliminate the second-pass effect (16):
Miyati et al (14) did not point out that the correction factor for the T1-shortening effect may be used as an index for contrast material leakage. Denote as R2*T1U the T1-uncorrected R2* that was calculated from the single-echo MR data (from Eq [A7]). If the concentration of the contrast agent in the extravascular space of the tissue (CT) and R2*T1C(t) - R2*T1U(t)(t > t0) are related linearly, the latter can be used as a parameter of contrast material leakage to extravascular space. An index of the leaked contrast agent (the leakage value LV) is given with the following:
R2*T1Cfmax:
R2*T1Cfmax represents the maximum height of the curve R2*T1C. As R2*T1Cfmax is known to be proportional to blood flow (25), the leakage index may in part reflect tissue permeability.
Simulation of the relationship between the leakage value and the concentration of the contrast agent.—We performed a computer simulation to evaluate the relationship between the leakage value and the concentration of the contrast agent. The simulation was based on the fact that both the leakage value and concentration of the contrast agent in the extravascular space are related to T1 values with and those without contrast material. The leakage value is given with Equation (A13), with the flip angle ( = 10°), repetition time (TR = 33.3 msec), and second echo time (TE = 23 msec) fixed to match those used in our study.
Denote R1 as (1/T1 - 1/T10). Since the linear relationship between R1 and the tissue concentration of the contrast agent is well known (5), R1 was plotted against the leakage value by changing T1. T1 is expressed with T10 as
T1 is the shortened T1 due to the leaked contrast material. Assuming T10 of 1.2 second, the most representative T1 value of neoplasms measured with a 1.5-T system (17), the leakage value was plotted against R1 by changing T1 from 0.1 to 1.2 seconds with a step of 0.1 second. We performed similar simulations with T10 of 1.0, 1.4, and 1.6 seconds.
Results of the simulation.—Figure 4 shows the simulated plot of R1 against the leakage value. With T10 of 1.2 second, in the range of leakage values from 0 to 12, a nearly linear relationship was found (r2 = 0.94), whereas an exponential relationship was demonstrated with a wider range from 0 to 16. The relationship became more linear as T10 grew larger.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
R2* = T2* rate change
R2*T1C = T1-corrected change in R2* from the double-echo data
R2*T1Cf = gamma-fitted R2*T1C data
R2*T1U = T1-uncorrected change in R2* from the single-echo data Author contributions: Guarantors of integrity of entire study, H.U., M.M., N.S.; study concepts and design, H.U.; definition of intellectual content, H.U., M.M., N.S.; literature research, H.U., M.M., N.S.; clinical studies, H.U., M.M., H.Y., H.K., Y. Koshimoto, Y. Kawamura; data acquisition, Tsuyoshi M., Y. Ishimori; statistical analysis, H.U.; manuscript preparation, H.U.; manuscript editing, M.M., N.S.; manuscript review, Tetsuya M., N.H., Y.Y., Y. Ishii.
|
References |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionAppendixReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. Okuda, K. Oshio, H. Shinmoto, A. Tanimoto, H. Asada, T. Fujii, Y. Yoshimura, and S. Kuribayashi Semiquantitative Assessment of MR Imaging in Prediction of Efficacy of Gonadotropin-releasing Hormone Agonist for Volume Reduction of Uterine Leiomyoma: Initial Experience Radiology, September 1, 2008; 248(3): 917 - 924. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Cha Update on Brain Tumor Imaging: From Anatomy to Physiology AJNR Am. J. Neuroradiol., March 1, 2006; 27(3): 475 - 487. [Full Text] [PDF]
|
|
|
|
|
|
S. Cha, L. Yang, G. Johnson, A. Lai, M.-H. Chen, T. Tihan, M. Wendland, and W.P. Dillon Comparison of Microvascular Permeability Measurements, Ktrans, Determined with Conventional Steady-State T1-Weighted and First-Pass T2*-Weighted MR Imaging Methods in Gliomas and Meningiomas. AJNR Am. J. Neuroradiol., February 1, 2006; 27(2): 409 - 417. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Jun, J. Garcia, T. Tihan, M.W. McDermott, and S. Cha Perfusion MR Imaging of an Intracranial Collision Tumor Confirmed by Image-Guided Biopsy AJNR Am. J. Neuroradiol., January 1, 2006; 27(1): 94 - 97. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Bulakbasi, M. Kocaoglu, A. Farzaliyev, C. Tayfun, T. Ucoz, and I. Somuncu Assessment of Diagnostic Accuracy of Perfusion MR Imaging in Primary and Metastatic Solitary Malignant Brain Tumors AJNR Am. J. Neuroradiol., October 1, 2005; 26(9): 2187 - 2199. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. H. Dennis Cheong, C. C. Tchoyoson Lim, and T. S. Koh Dynamic Contrast-enhanced CT of Intracranial Meningioma: Comparison of Distributed and Compartmental Tracer Kinetic Models--Initial Results Radiology, September 1, 2004; 232(3): 921 - 930. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Law, K. Kazmi, S. Wetzel, E. Wang, C. Iacob, D. Zagzag, J. G. Golfinos, and G. Johnson Dynamic Susceptibility Contrast-Enhanced Perfusion and Conventional MR Imaging Findings for Adult Patients with Cerebral Primitive Neuroectodermal Tumors AJNR Am. J. Neuroradiol., June 1, 2004; 25(6): 997 - 1005. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Law, S. Yang, J. S. Babb, E. A. Knopp, J. G. Golfinos, D. Zagzag, and G. Johnson Comparison of Cerebral Blood Volume and Vascular Permeability from Dynamic Susceptibility Contrast-Enhanced Perfusion MR Imaging with Glioma Grade AJNR Am. J. Neuroradiol., May 1, 2004; 25(5): 746 - 755. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A Jackson Imaging microvascular structure with contrast enhanced MRI Br. J. Radiol., December 1, 2003; 76(suppl_2): S159 - S173. [Full Text] [PDF]
|
|
|
|
|
|
S. Yang, M. Law, D. Zagzag, H. H. Wu, S. Cha, J. G. Golfinos, E. A. Knopp, and G. Johnson Dynamic Contrast-Enhanced Perfusion MR Imaging Measurements of Endothelial Permeability: Differentiation between Atypical and Typical Meningiomas AJNR Am. J. Neuroradiol., September 1, 2003; 24(8): 1554 - 1559. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Uematsu, M. Maeda, N. Sadato, T. Matsuda, Y. Ishimori, Y. Koshimoto, H. Kimura, H. Yamada, Y. Kawamura, Y. Yonekura, et al. Blood Volume of Gliomas Determined by Double-Echo Dynamic Perfusion-Weighted MR Imaging: A Preliminary Study AJNR Am. J. Neuroradiol., November 1, 2001; 22(10): 1915 - 1919. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Sugahara, Y. Korogi, M. Kochi, Y. Ushio, and M. Takahashi Perfusion-sensitive MR Imaging of Gliomas: Comparison between Gradient-echo and Spin-echo Echo-planar Imaging Techniques AJNR Am. J. Neuroradiol., August 1, 2001; 22(7): 1306 - 1315. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS |
) is constant after the first pass ends, and the leakage value is relatively small compared with the height of 
) with their gamma-fitted curve (fine line) show that the height of 
, n = 11). There is a substantial overlap of the vascularity index between the two types of tumors. The leakage index separated them except in one case.
), 1.2 (•), 1.4 (