This Article Abstract Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Ulug, A. M. Articles by van Zijl, P. C.M. Search for Related Content PubMed PubMed Citation Articles by Ulug, A. M. Articles by van Zijl, P. C.M.

(Stroke. 1997;28:483-490.)
© 1997 American Heart Association, Inc.

Articles

Absolute Quantitation of Diffusion Constants in Human Stroke

Presented in part at the Joint 3rd World Stroke Congress and 5th European Stroke Conference, Munich, Germany, September 1-4, 1996.

Aziz M. Ulu, PhD; Norman Beauchamp, Jr, MD; R. Nick Bryan, MD, PhD; Peter C.M. van Zijl, PhD

From the Department of Radiology, The Johns Hopkins Medical Institutions, Baltimore, Md.

	Abstract

Top Abstract Introduction Subjects and Methods Results Discussion References

 
Background and Purpose Animal studies have shown that MR diffusion imaging can outline acute ischemic regions before irreversible damage (infarction) occurs. To study evolution of ischemic lesions in humans, it is therefore important to quantify absolute diffusion constants (D values), but quantitation has not been reproducible among different clinics. These problems are explained, and a method for reproducible quantitation is suggested.

Methods Diffusion-weighted and absolute diffusion images were acquired, and the absolute apparent diffusion constants in three orthogonal spatial directions (D_xx, D_yy, and D_zz) were measured. These were combined to calculate images of the orientation-independent apparent diffusion parameter D_av=1/3 Trace{}=1/3(D_xx+D_yy+D_zz). Values of the individual diffusion constants and D_av were evaluated in 6 patients and 6 normal volunteers.

Results Patient data show that comparison of diffusion constants between contralateral and ipsilateral hemispheres after ischemia may give results varying by more than 100% depending on orientation. Findings in normal-appearing regions containing a mixture of gray and white matter in patients (n=5) and in normal volunteers (n=6) show that D_av=(0.92±0.11)x10^–3 mm²/s, with a small intersubject variation, whereas D_xx, D_yy, and D_zz vary strongly. Hemispheric ratios (ipsilateral/contralateral [I/C]) in these subjects were (I/C)_Dav=1.00±0.05, (I/C)_Dxx=1.02±0.15, (I/C)_Dyy=1.07±0.24, and (I/C)_Dzz=0.96±0.28. The individual subjects in this group all had an (I/C)_Dav within 10% of unity, while the other three ratios showed intersubject variations as large as 100%.

Conclusions (I/C)_Dav ratios are a reliable means to quantitate changes in absolute diffusion constants for the study of stroke evolution independent of tissue orientation, gradient orientation, and diffusion time. The use of these ratios will enable reproducible intersubject and interclinic quantitation.

Key Words: diffusion • magnetic resonance imaging • stroke, acute • stroke assessment

	Introduction

Top Abstract Introduction Subjects and Methods Results Discussion References

 
It is believed that early intervention is the key to successful therapeutic outcome in stroke. Therefore, the ability to rapidly diagnose the status of brain perfusion and the extent of tissue that is at risk versus that which is already irreversibly damaged is critical. MRI and MR spectroscopy hold the potential for spatially assessing these parameters in a single clinical examination. MR angiography and dynamic contrast imaging can assess tissue perfusion, while gradient-recalled echo (T2*) imaging has the potential to detect hemorrhage.¹ In addition, hyperintensity in T2 imaging has been used as an indicator of irreversible damage. The initial report by Moseley et al² that the apparent water diffusion constant (or ADC, which we refer to as D in the remainder of the article) of brain tissue is extremely sensitive to ischemic changes has opened the possibility of assessing tissue regions that are in danger of infarction but may still be salvaged.

DW MRI of acute stroke patients has confirmed expectations that ischemic regions are identifiable before T2 increases occur.^{3 4 5 6 7 8} However, since the eventual tissue damage depends on the length of the ischemic period, and human studies are generally obtained at a single point in time, it is presently impossible to determine the exact extent of ischemic evolution after onset. Thus, to obtain clues concerning outcome and the need for treatment, it is necessary to evaluate more than one tissue parameter. In addition to the essential estimate of tissue perfusion, clinical MRI studies have therefore used combined T2-weighted and DW imaging to assess the viability of acute ischemic regions.^{3 4 5 7 8} Welch et al⁹ recently suggested that in analogy to animal studies,¹⁰ human stroke signatures can be designed using changes in absolute D values and T2s with respect to normal tissue in the same slice after segmenting out CSF. However, results by Warach et al⁴ in a large patient population indicate that the time course of absolute D values after ischemic insults in humans differs from that in animal models. This is not unexpected given the great variability in the length of occlusion and the extent of collateral flow in humans. The results of Warach et al on ischemic evolution were confirmed by Sorensen et al⁷ and Marks et al⁸ but contradicted by data from the Welch group.⁹ This controversy was the subject of a recent letter to the editor of Stroke by Warach et al.¹¹ In the present article, we will show that the results found by both groups are not mutually exclusive but appear contradictory, primarily on the basis of differences in quantification of the absolute diffusion constants.

The interlaboratory variation in diffusion constants has previously been noticed in animal brain studies¹²; this discrepancy has been explained in terms of the presence of different apparent diffusion constants in different directions with respect to the white matter tracts (diffusional anisotropy). As a consequence, exact description of diffusional properties requires the use of nine diffusion constants instead of a single constant.^{13 14} These constants describing the apparent diffusion in different directions are given in the so-called "diffusion tensor"^{13 14} :

(1)

in which the directional indices x, y, and z relate to the reference frame in which the experiment is performed (namely, the laboratory frame), which is described by the orientation of the gradient axis system in the scanner (magnetic field along z axis). For supine data acquisition in the axial plane, the x direction is right-left, and the y direction corresponds to anterior-posterior. The diffusional attenuation of the MR signal is now given by¹³

in which is the proton gyromagnetic ratio, the gradient length, and t_dif the diffusion time t_dif=(–/3), with the time between the starts of the two gradients in the diffusion experiment. Thus, measured diffusion constants will depend on the orientation of the subject with respect to the gradient axes (scanner), as well as on the orientation of the gradient axes used for the diffusion measurement. For instance, use of a single gradient direction i measures a single diagonal element D_ii, while combined use of all three gradient directions measures the sum of all nine elements. As a consequence, the use of arbitrary gradient directions may lead to large variations in reported absolute diffusion constants between laboratories. This problem may be further magnified by ischemic overlapping of both white and gray matter regions. Animal studies have shown that the directional dependence and most of the diffusional contrast between gray and white matter can be removed by studying the so-called trace (or average) of the diffusion tensor

(3)

and that use of this isotropic quantity facilitates the detection of ischemic regions.^{14 15} One prerequisite to establishing whether the use of absolute diffusion constants (in conjunction with other tissue parameters) is important for diagnosis and prognosis of human stroke is reproducible D values. In the present study, we evaluate the reproducibility of the absolute values and I/C ratios of the individual tensor components D_ii and D_av in regions containing both gray and white matter in normal volunteers and stroke patients.

	Subjects and Methods

Top Abstract Introduction Subjects and Methods Results Discussion References

 
Subjects
Six normal volunteers (subjects 1 to 6, 2 women and 4 men) and 6 stroke patients (subjects 7 to 12, 2 women and 4 men) were studied. The stroke cases ranged from acute (<24 hours) to chronic (>1 year). The age range of the volunteers was 30 to 40 years. The age range of the patients was 29 to 75 years. All patients suffered ischemic stroke. All patients or immediate family members of patients gave written consent for the procedure, which was approved by The Johns Hopkins University Joint Committee on Clinical Investigation. Because this is a technical study evaluating quantitation, we will only provide a brief description of the two cases discussed in Figs 1 and 2 (subjects 7 and 8).

View larger version (113K): [in this window] [in a new window]   Figure 1. Different image types for patient 1 (subject 7 in Fig 3, day 3 after stroke). Except for the absolute diffusion images, the complete gray scale was used. The absolute diffusion images are windowed to highlight regions of different apparent diffusion constants within the lesion. DW images obtained using a single diffusion axis (top row) look very similar because of T2 contamination, as confirmed by the T2-weighted image in row 3. The pure diffusion images (2nd row) show that this is misleading, causing contradictory conclusions about stroke evolution (Table 1). The pure Dav image (row 3) gives reproducible values independent of direction, while the Dav-weighted image (row 3) still has T2 contamination.

View larger version (49K): [in this window] [in a new window]   Figure 2. Different image types for patient 2 (subject 8 in Fig 3, 23 hours after onset). The absolute diffusion image is windowed to highlight regions of different apparent diffusion constant. Old and new infarct can be easily separated by comparing the T2-weighted and DW images or by evaluating the absolute Dav values and (I/C)Dav values (Table 2).

Subject 7 (patient 1) was a 29-year-old black woman with a history of idiopathic cardiomyopathy presenting with acute onset of right-sided hemiparesis and expressive aphasia. CT scan performed on the day of admission demonstrated no evidence of infarction, whereas MRI showed increased T2-weighted signal in the region of the left insular cortex and the left temporal lobe (time between onset and initial MRI was approximately 8 hours). Diffusion imaging was not possible on the first day. The follow-up MRI on day 3 showed extension of the area of T2 prolongation with involvement of the left caudate and lentiform nuclei, as well as the region near the left central sulcus. Patient 2 (subject 8) was a 41-year-old woman with a history of hypertension and prior infarcts in the right caudate nucleus and lentiform nucleus. On admission (23 hours after onset), combined T2 and diffusion imaging showed a new lesion. Results of conventional angiography were normal.

Methods
DW images were collected using a 1.5-T GE Signa clinical scanner. A modified combined multislice spin-echo/EPI sequence with a pair of diffusion gradients centered around the 180° pulse was used. The dephase gradients for spatial encoding were placed directly before acquisition to avoid interference with the diffusion measurement. No additional dephase-rephase gradient crushers around the 180° pulse were used, since crushing of unwanted signals is automatic in the first diffusion experiment at low b value. Thus, no interference of other pulsed gradients is present between the diffusion gradient pair. The current hardware allows diffusion gradients of up to 2.2 G/cm, with a rise time of 320 µs. Raw data sets of 128x128 points each were collected at eight interleaves, zero-filled, and Fourier transformed to obtain 256x256 images. One patient was scanned using single-shot EPI (128x128). Total scanning time for the diffusion study was 20 to 25 minutes for the studies using eight interleaves; it was a few minutes for the single-shot EPI. Cardiac gating with repetition time of 3 or 4 cardiac cycles was used to obtain 20 slices per repetition time; echo time was 100 milliseconds, field of view was 24 cm, and slice thickness was 5 mm. A head holder was used to minimize involuntary head motions. The length () of the diffusion gradients was 25 ms, and the diffusion time t_dif was 40.2 milliseconds.

In each experiment, three sets of DW images with diffusion gradients in the x, y, and z directions were taken. Six gradient strengths corresponding to a b-value range between 2 and 527 s/mm² were used to calculate pure diffusion maps for D_xx, D_yy, and D_zz using Equation 2. These were combined to calculate D_av images. The images at the lowest b value correspond to T2-weighted EPI images in the Figures. The images at the highest b value correspond to the DW images. The sum of the DW images in the three directions are the isotropic DW images. For different brain structures, ROIs were chosen and the apparent diffusion constants evaluated. The error numbers reported in Tables 1 and 2 are the maximum deviations from the average, obtained by repeating the area evaluations three times for newly defined boundaries and independently by two different individuals. In normal subjects, ROIs for different brain structures were analyzed and compared with the contralateral region. In stroke patients, D_av was evaluated in several ischemic regions and in the same regions contralaterally, as well as in normal-appearing regions in both hemispheres.

	Results

Top Abstract Introduction Subjects and Methods Results Discussion References

 
Fig 1 shows EPI images for an axial slice through the cortex in patient 1. This case, acquired at day 3 after onset, illustrates several of the important issues in quantitation of both the actual stroke region and the absolute diffusion constants. The images represent the different types of image display used by different research groups. The most common display is the DW image shown on the top row for x, y, and z gradient weighting. The stroke regions in these images seem very similar. However, the pure diffusion images for D_xx, D_yy, and D_zz and the numbers in Table 1 show that the use of diffusion weighting in a single direction can be very misleading when the slice contains different white matter structures with different fiber orientations. ROIs A1 and B1 (see image in Table 1) contain parts of central white matter of the corona radiata and the centrum semiovale, whereas sections A2 and B2 consist mainly of peripheral white matter of the subcortical U fibers. The results in Table 1 show that the individual D_ii values and ipsilateral/contralateral ([I/C]_Du) ratios vary greatly between white matter areas in these regions. In gray matter, the variation is small, but the gray matter area is small in this slice, and the stroke regions generally include some white matter. Calculation of the actual D_av image (3rd row in Fig 1) and of the isotropic DW (D_av-weighted) images shows that the stroke region becomes better described. However, although gray/white matter contrast is almost completely removed in both images, the stroke region in the D_av-weighted image may still be contaminated by contrast due to spin-density, T2, and T1. Specifically, the two stroke regions (A1 and A3) that are hyperintense in the spin-density/T2-weighted EPI image have a different absolute D_av value, while they look similar in the isotropic DW image.

View this table: [in this window] [in a new window]   Table 1. Absolute Diffusion Constants (10-3 mm2/s) and Ipsilateral/Contralateral (I/C) Ratios for Patient 1 (tdif=40.2 ms)

Fig 2 shows images for patient 2 (23 hours after onset), who had both an old and a relatively new lesion. Both infarcts are bright on the T2-weighted image, but the newer infarct is brighter on the D_av-weighted image, which also contains T2 contrast. The pure D_av image shows the newer infarct dark and the older infarct bright. With use of diffusion images in individual directions, contradictions again arise about the status of the infarct, as illustrated by the absolute diffusion constants in Table 2. First of all, in the new lesion, the D_zz value is close to normal, whereas D_xx and D_yy are significantly lower than in the contralateral region. For the older infarct, the diffusion values are higher than normal, in agreement with the progression to vasogenic edema or encephalomalacia. In the completely malacic stage, a diffusion constant of free water is expected, namely, 3.0x 10^–3 mm²/s, which should be independent of gradient orientation. In agreement with this, values of D_av=D_zz=D_xx=D_yy=2.6 to 3.0x10^–3 mm²/s were found in two older stroke patients, one of whom had an infarct of more than 1 year old. When patients have old infarcts, the absolute diffusion constants are already directionally independent, but confusion about infarct evolution would occur on comparison of the diffusion constant of the old lesion with that of the normal contralateral region, where the white matter structure is still intact (eg, Table 2, region 2). This illustrates the incorrectness of using a single diffusion direction for stroke evaluation.

View this table: [in this window] [in a new window]   Table 2. Absolute Diffusion Constants (10-3 mm2/s) and Ipsilateral/Contralateral (I/C) Ratios for Patient 2 (tdif=40.2 ms)

Fig 3 shows absolute D_ii values, D_av values, and hemispheric ratios (I/C) for ROIs containing both white and gray matter in normal volunteers (subjects 1 to 6) and in normal-appearing regions in stroke patients (subjects 7 to 12). The D_av value is the mean of two similar regions (one right, one left), and the I/C ratio is the ratio of the diffusion constants measured in these regions. In normal subjects, the right/left ratio was used for I/C, and in stroke patients the I/C ratio was used. The D_ii values scatter widely, and so do the I/C ratios for D_ii. The D_av values, on the other hand, are between 0.79x10^–3 mm²/s and 0.99x10^–3 mm²/s, which is the range covered by the D_av for white and gray matter, which we found to be (0.83±0.06)x10^–3 mm²/s and (0.92±0.15)x10^–3 mm²/s (average±SD), respectively. One exception to this range was subject 11, for whom we had major motion artifacts leading to an artificially high diffusion constant. An interesting aspect of this case was that the I/C ratio for D_av was again within 10% of unity. The average±SD for the 12 subjects was D_av=(0.92±0.10)x10^–3 mm²/s. We would like to point out that the measurement of a D_av value of gray matter that is higher than that of white matter may be due to difficulties in selecting clean gray matter areas in the proximity of CSF, which has a high diffusion constant. The images show negligible contrast between white and gray matter, indicating that they are indistinguishable within the signal-to-noise ratio attainable for the D_av image.

View larger version (34K): [in this window] [in a new window]   Figure 3. Absolute (apparent) diffusion constants and hemispheric ratios for normal-appearing ROIs containing both white and gray matter in normal volunteers (subjects 1 to 6) and stroke patients (subjects 7 to 12). The Dav values are averages between right and left. In normal subjects, the right/left ratio was used for I/C; in stroke patients the I/C ratio was used. Notice the wide scatter for the individual diffusion constants Dii and (I/C)Dii, while Dav and (I/C)Dav are very constant. The exception, subject 9, that falls outside 10% of unity for (I/C)Dav is due to a technical error (see "Results"); the increased Dav for subject 11 was due to motion (see "Results").

Of the I/C ratios for D_av in normal-appearing ROIs, only 1 patient (subject 9) was outside the 10% deviation from unity. On investigation, we found that this person had once moved permanently between acquisition of two of the different tensor component images, leading to interslice contamination. The result for this person seems normal because it is an average of the ipsilateral and contralateral ROIs. Thus, the D_av value for this person should be regarded as a technical error and does not contribute to the true meaning of the results. We have left it to illustrate that care should be taken in analyzing the data. These displacements between acquisition of the images of the individual tensor components can be avoided when using single-scan D_av measurements.^{15 16} Without this erroneous subject, the average±SD values for the hemispheric ratios were (I/C)_Dav=1.00±0.05, (I/C)_Dxx=1.02±0.15, (I/C)_Dyy=1.07±0.24, and (I/C)_Dzz=0.96±0.28. Thus, the SDs for the individual tensor elements are much larger than for D_av, and a large range (10%) of possible I/C values is found for the individual subjects (Fig 3).

	Discussion

Top Abstract Introduction Subjects and Methods Results Discussion References

 
Absolute D Values
The stroke case in Fig 1 (Table 1) clearly demonstrates all potential problems in quantitation or even qualitative assessment when the proper approach in analyzing diffusion images is not taken. When evaluating the individual stroke regions A1 to A3 (Table 1) by simply comparing them with contralateral areas, the interpreter may come to several different conclusions if using only a single gradient axis. For instance, in region A3, users of the x gradient will measure a reduced D value, while z-gradient users will find an increase in D. For region A1, x- and z-gradient users will find a reduced D value, while y-gradient users will see no change in the pure diffusion images. In the study of stroke evolution, mistakes such as these could be detrimental. We would like to point out here that Welch et al⁹ have consistently used the z-gradient axes, which may explain the apparently faster stroke evolution measured by them when compared with the other groups,^{5 7 8} who have predominantly used the x gradient (which gives consistently low diffusion constants) or different combinations of gradients, including measurement of the tensor trace. For instance, the increased T2 in region A3 combined with the larger-than-normal D indicates faster evolution when the z axis is used. When the anisotropic influence of the white matter is removed by study of the tensor trace (D_av), the effects are better defined: a small reduction in A1, a large reduction in A2, and approximately no reduction in A3. Thus, parallel to animal studies,^{14 15 17} the D_av images provide a reliable picture of the ischemic evolution and the stroke area. It should also be noted that although isotropic DW images are better than those obtained with individual gradients, they are still contaminated by spin-density, T2, and T1 influences. This becomes clear when studying the T2-weighted EPI image, which shows a contrast similar to that found in the DW images. Thus, it is essential to evaluate the absolute D_av images for a correct assessment of diffusion changes.

In addition to the incorrect interpretations that are possible when assessing I/C differences in stroke using single gradient axes, it is also impossible to give a general reference D value for white or gray matter in this situation. This is clearly illustrated when comparing the normal white matter regions B1, B2, B4, and A4 in Table 1, which show variation between 0.57x10^–3 mm²/s and 0.81x10^–3 mm²/s for D_xx, between 0.60x10^–3 mm²/s and 1.38x10^–3 mm²/s for D_yy, and between 0.94x10^–3 mm²/s and 1.51x10^–3 mm²/s for D_zz, obviously due to the differences in fiber orientations between the white matter structures. These problems are confirmed by the results for the other patients and volunteers in Figs 2 and 3. In addition to the variation of the individual tensor components for different white matter structures, stroke diagnosis will be complicated by the fact that most stroke regions encompass multiple white and gray matter areas. The results for arbitrary regions containing both white and gray matter chosen in normal volunteers and in normal-appearing regions in stroke patients (Fig 3) show that D_av is indeed very reproducible for most brain structures, whereas the results for the individual tensor components vary greatly. Because the average values for white and gray matter are comparable within error in the individual absolute D_av images, the contrast between white and gray is generally negligible. This results in a picture for the D_av images in which basically only two components are present: brain tissue and CSF, allowing clean evaluation of stroke regions.¹⁴ Because CSF consists mainly of water, its diffusion constant is supposed to be that of pure water at 37°C, which is 3.02x10^–3 mm²/s.¹⁸ The CSF diffusion values can therefore actually be used as a check for the presence of artifacts. The results in Tables 1 and 2 indeed confirm the correct CSF value, indicating good accuracy of the measurements.

One argument against the need for using D_av values to assess stroke evolution is that hemispheric ratios for the individual D_ii values may provide accurate insight into the changes in diffusion in stroke patients. To assess the reliability of this approach, we evaluated hemispheric ratios for D_av and the individual tensor elements in the 12 subjects in this study. The results in Fig 3 again show that only the hemispheric ratios values for D_av ([I/C]_Dav) are reliable, and except for 1 patient (subject 9) who was eliminated because of a technical error related to permanent head displacement between acquisition of the different tensor components (see "Results"), all (I/C)_Dav are restricted to within 10% of unity. This is well reflected in the small SD of 5% in (I/C)_Dav for the 11 subjects, whereas the SDs in the hemispheric ratios of the individual tensor components are 15%, 24%, and 27% for x, y, and z directions, respectively. Thus, although the average hemispheric ratios for the 11 subjects are still within 10% of unity, the intersubject variation is much larger than this range (see Fig 3), and only (I/C)_Dav can be used reliably for quantitation.

Evaluation of the Exact Area of Stroke Regions
When the DW images in the top row of Fig 1 are studied, the stroke regions look similar for all individual gradient axes, which should be attributed to influences from spin-density and T2/T1 contrast that interfere with the diffusion contrast. Similarly, the individual absolute diffusion images (2nd row in Fig 1) cannot provide a correct picture because of the interference of white matter diffusional anisotropy. Thus, these human results confirm the results of animal studies,^{14 15 17} in that a D_av image is needed (3rd row) to accurately outline ischemic regions. This is especially clear in Fig 1 (Table 1), where three regions of different diffusion constant (A1 to A3) can be separated inside the lesion area. Of these, region A1 has a reduced D_av, but a one-time evaluation of only D_av cannot distinguish between an initial lowering in D_av and a D_av that is already increasing because of edema. The increased intensity in the T2-weighted image shows that the latter is the case. Similar reasoning applies to region A3. Region A2, on the contrary, still has a low D_av and no changes in the T2-weighted image, indicating potential salvageability.

For the separation of old and new strokes, DW images also provide clear information (Fig 2), but again the sole use of pure DW images is not conclusive (due to potential T2 changes); additional absolute diffusion and/or T2-weighted information is necessary. For instance, the three images in Fig 2 clearly indicate the presence of an old and a new stroke from comparison of the T2-weighted and the DW images, but the combined information from the pure diffusion image shows that although the new stroke seems already infarcted (hyperintense on T2 weighting¹⁹ ), it has not yet evolved to complete vasogenic edema, as demonstrated by the low D_av. The viability thresholds for absolute values of D_av, spin-density, T1, and T2 are not yet determined in humans, but the animal studies^{10 17 20 21} could provide some guidelines. In any event, use of truly quantifiable parameters such as D_av and changes in spin-density, T1, and T2 should allow correct evaluation of the status of different regions within an ischemic area.

The most important conclusion about stroke evolution, illustrated in Tables 1 and 2, is that variability in individual axes used by different laboratories can lead to contradictory conclusions about stroke evolution. This is true for older strokes (age indicated by high T2) when the contralateral site is a white matter region (eg, Table 2, region 2 where the y direction result indicates much faster evolution than x or z results) and for newer strokes in white/gray matter mixed regions (eg, Table 1, region 3). In the latter example, the z direction gives an apparent diffusion constant indicating fast evolution from ischemia to infarction, with a diffusion value and T2 higher than normal on day 3, similar to the results for the z direction reported by Welch et al.⁹ The diffusion results in the other directions are equal to or lower than normal, indicating slower evolution, similar to results by other groups. It should be noted that the use of individual directions need not always lead to discrepancies in stroke evolution because the effect depends on the constitution of the ROI (percent white or gray matter). To avoid confusion, the trace (D_av) should be used. We also would like to point out that the anisotropic effects measured here could potentially be of interest for understanding stroke evolution and that measurement of the complete tensor would be extremely important if stroke evolution is directionally dependent. Future experiments on large groups of patients will have to define whether this is necessary.

Influence of the Diffusion Time
Although D_av images are independent of the orientation of the subject, it is extremely important to realize that they still contain information about diffusional restriction. This is due to the fact that the signal intensity variation as a function of gradient strength also depends on the diffusion time t_dif. This variation in D_av due to changes in t_dif has been demonstrated in the cat brain^{12 14 17} where, on reducing t_dif, an increase in D_av was found. Contrary to D_av, the hemispheric ratio should be independent of the diffusion time, since this contribution is divided out. Therefore, we recommend use of the (I/C)_Dav ratio to allow comparison of stroke evolution independent of diffusion time.

Data Reproducibility
Our results indicate that it is possible to obtain very reproducible diffusion constants in normal volunteers and normal-appearing regions in stroke patients but that the stroke regions may vary dramatically from case to case. One origin of this may be the lesion heterogeneity and the errors in evaluation resulting from the choice of an ROI for analysis. However, we do believe that this should not lead to major discrepancies between laboratories in conclusions regarding stroke evolution in large patient groups. The consistent normal data also indicate that the methodology used is robust. If laboratories find discrepancies among normal subjects, there may be many underlying causes, mainly related to image quality (eddy currents, b values too low, low signal-to-noise ratio, patient restriction with interleaved EPI), that have to be evaluated carefully.

Meaning of Absolute Diffusion Constants
Finally, we would like to caution the reader about the use of absolute diffusion constants in determining the evolution of stroke. The philosophical basis of a stroke signature is based partially on a series of detailed animal studies, which have shown close spatial relationships between the areas of initially reduced D values, regions of abnormal pH, ATP, and glucose,²² and ultimate areas of histopathological damage.^{10 21 23 24 25 26 27 28} Focal ischemia studies also show that the absolute value of the water diffusion constant evolves during prolonged periods of ischemia and, in individual animal models, correlates with outcome.^{17 26 29 30} In addition, transient focal ischemia studies show that the absolute diffusion constant correlates with cerebral blood flow during occlusion, signifying that the absolute value of D is a good indicator of risk of injury at a certain time point.¹⁷ However, it should be stressed that even the largest reductions in D value are reversible to normal in both global^{31 32} and focal¹⁷ transient ischemia models, therefore indicating that both the absolute D value and the period of reduction in D are the major determinants for the resulting injury. Unfortunately, clinical diffusion MRI can only base prognostic determination on a single examination, which prevents determination of the period of reduction in the D value. Further studies are needed to establish whether the use of multiple MR parameters, including perfusion and T2, can provide data that can substitute for this missing link and thus provide accurate stroke diagnosis and prognosis.

Conclusions
In summary, our data on human stroke patients and normal volunteers show that measurement of apparent diffusion constants in individual directions (D_ii) and the I/C ratios of D_ii may lead to erroneous interpretation of stroke data with respect to stroke evolution, which could be detrimental when used to obtain stroke signatures. In agreement with previous animal studies,^{14 15 17} we found that measurement of D_av is very reproducible in normal volunteers and in normal-appearing regions in stroke patients. Our data also show that the ischemic area indicated by isotropic DW images and absolute D_av images may differ because of interference of T2 contrast. Thus, although a qualitative interpretation of the stroke status can be obtained by comparing T2-weighted and DW images, absolute quantitation can only be achieved using absolute diffusion constants. Finally, we showed that the hemispheric ratio for D_av is very reproducible in normal volunteers and in normal-appearing regions in stroke patients. Based on the fact that hemispheric ratios are independent of the diffusion time, we recommend using these ratios to evaluate diffusional changes in stroke. To assess the influence of motion artifacts on the data, one can use (I/C)_Dav ratios and absolute D_av values in normal regions of tissue and CSF as a reference for data quality. The use of a reliable general approach for quantitation of diffusion constants should facilitate the diagnosis of stroke status and therefore the testing of new therapies.

	Selected Abbreviations and Acronyms

 

D = apparent diffusion constant CSF = cerebrospinal fluid DW = diffusion-weighted EPI = echo-planar imaging I/C = ipsilateral/contralateral ROI = region of interest

	Acknowledgments

 
This research is supported in part by National Institutes of Health (NIH) grant NS31490 and by NIH research training grant CA 09630 (Dr Ulu). Part of this work was done during the tenure of an Established Investigatorship from the American Heart Association (Dr van Zijl). The authors are grateful to Theodore J. Passe, MD, for assistance in patient recruitment.

	Footnotes

 
Reprint requests to Peter C.M. van Zijl, PhD, Department of Radiology, Traylor 217, The Johns Hopkins University Medical School, 720 Rutland Ave, Baltimore, MD 21205.

Received October 25, 1996; revision received December 18, 1996; accepted December 18, 1996.

	References

Top Abstract Introduction Subjects and Methods Results Discussion References

 
1. Thulborn KR, Atlas SW. Intracranial hemorrhage. In: Atlas SW, ed. Magnetic Resonance Imaging of the Brain and Spine. New York, NY: Raven Press Publishers; 1991:223-326.

2. Moseley ME, Cohen Y, Mintorovitch J, Chileuitt L, Shimizu H, Kucharczyk J, Wendland MF, Weinstein PR. Early detection of regional cerebral ischemia in cats: comparison of diffusion- and T2-weighted MRI and spectroscopy. Magn Reson Med.. 1990;14:330-346. [Medline] [Order article via Infotrieve]

3. Warach S, Chien D, Li W, Ronthal MM, Edelman RR. Fast magnetic resonance diffusion-weighted imaging of acute human stroke. Neurology.. 1992;42:1717-1723. [Abstract/Free Full Text]

4. Warach S, Gaa D, Siewert B, Wielopolski P, Edelman RR. Acute human stroke studied by whole brain echo planar diffusion-weighted magnetic resonance imaging. Ann Neurol.. 1995;37:231-241. [Medline] [Order article via Infotrieve]

5. Warach S, Benfield A, Edelman RR. Time course of abnormal apparent diffusion coefficient in human stroke. In: Proceedings of the 3rd meeting of the Society of Magnetic Resonance in Medicine; August 19-25, 1995; Nice, France. Abstract, page 82.

6. Ulu A, Oliverio P, Beauchamp N, Soher B, Bryan RN, Oppenheimer S, van Zijl PCM. In: Proceedings of the 3rd meeting of the Society of Magnetic Resonance in Medicine; August 19-25, 1995; Nice, France. Abstract, page 1384.

7. Sorenson AG, Buananno FS, Gonzalez RG, Schwamm LH, Lev MH, Huang-Hellinger FR, Reese TG, Weisskoff RM, Davis TL, Suwanwela N, Can U, Moreira JA, Copen WA, Look RB, Finklestein SP, Rosen BR, Koroshetz WJ. Hyperacute stroke: evaluation with combined multisection diffusion-weighted and hemodynamically weighted echo-planar imaging. Radiology.. 1996;199:391-401. [Abstract/Free Full Text]

8. Marks MP, de Crespiguy A, Lentz D, Enzmann DR, Albers GW, Moseley ME. Acute and chronic stroke: navigated spin-echo diffusion-weighted MR Imaging. Radiology.. 1996;199:403-408. [Abstract/Free Full Text]

9. Welch KMA, Windham J, Knight RA, Nagesh V, Hugg JW, Jacobs M, Peck D, Booker P, Dereski MO, Levine SR. A model to predict the histopathology of human stroke using diffusion and T2-weighted magnetic resonance imaging. Stroke.. 1995;26:1983-1989. [Abstract/Free Full Text]

10. Knight RA, Dereski MO, Helpern JA, Ordidge RJ, Chopp M. Magnetic resonance imaging assessment of evolving focal cerebral ischemia: comparison with histopathology in cats. Stroke.. 1994;25:1252-1262. [Abstract]

11. Warach S, Moseley M, Sorenson AG, Koroshetz W. Time course of diffusion imaging abnormalities in human stroke. Stroke.. 1996;27:1254-1256. Letter.

12. van Zijl PCM, Davis D, Moonen CTW. Diffusion spectroscopy in living systems. In: Gillies RJ, ed. NMR in Physiology and Biomedicine. 1994:185-198.

13. Basser PJ, Matiello J, Le Bihan D. MR diffusion tensor spectroscopy and imaging. Biophys J.. 1994;66:259-267. [Medline] [Order article via Infotrieve]

14. van Gelderen P, de Vleeschonwer MHM, DesPres D, Pekar J, van Zijl PCM, Moonen CTW. Water diffusion and acute stroke. Magn Reson Med.. 1994;31:154-163. [Medline] [Order article via Infotrieve]

15. Mori S, van Zijl PCM. Single-scan magnetic resonance imaging of the trace of the diffusion tensor. Magn Reson Med.. 1995;33:41-52. [Medline] [Order article via Infotrieve]

16. Wong EC, Cox RW, Song AW. Optimized isotropic diffusion weighting. Magn Reson Med.. 1995;34:139-143. [Medline] [Order article via Infotrieve]

17. Miyabe M, Mori S, van Zijl PCM, Kirsch JR, Eleff SM, Koehler RC, Traystman RJ. Correlation of the average water diffusion constant with cerebral blood flow and ischemic damage after transient middle cerebral artery occlusion in cats. J Cereb Blood Flow Metab.. 1996;16:881-891. [Medline] [Order article via Infotrieve]

18. Harris KR, Woolf LA. Pressure and temperature dependence of the self-diffusion coefficient of water and oxygen-18 water. J Chem Soc Faraday.. 1980;176:377-385.

19. Bose B, Jones SC, Lorig R, Friel HT, Weinstein M, Little JR. Evolving focal cerebral ischemia in cats: spatial correlation of NMR imaging, cerebral blood flow, tetrazolium staining, and histopathology. Stroke.. 1988;19:28-37.[Abstract/Free Full Text]

20. Knight RA, Ordidge RJ, Helpern JA, Chopp M, Rodolosi LC, Peck D. Temporal evolution of ischemic damage in rat brain measured by proton nuclear magnetic resonance imaging. Stroke.. 1991;22:802-808. [Abstract/Free Full Text]

21. Hoehn-Berlage M, Eis M, Back T, Kohno K, Yamashita K. Changes of relaxation times (T1, T2) and apparent diffusion coefficient after permanent middle cerebral artery occlusion in the rat: temporal evolution, regional extent and comparison with histology. Magn Reson Med.. 1995;34:824-834. [Medline] [Order article via Infotrieve]

22. Hossmann K-A, Fischer M, Bockhorst K, Hoehn-Berlage MA. NMR imaging of the apparent diffusion constant (ADC) for the evaluation of metabolic suppression and recovery after prolonged ischemia. J Cereb Blood Flow Metab.. 1994;14:723-731. [Medline] [Order article via Infotrieve]

23. Kucharczyk J, Mintorovitch J, Asgari HS, Moseley M. Diffusion/perfusion imaging of acute cerebral ischemia. Magn Reson Med.. 1991;19:311-315. [Medline] [Order article via Infotrieve]

24. Minematsu K, Li L, Sotak CH, Davis MA, Fisher M. Reversible focal ischemic injury demonstrated by diffusion-weighted magnetic resonance imaging in rats. Stroke.. 1992;23:1304-1311. [Abstract/Free Full Text]

25. Helpern JA, Dereski MO, Knight RA, Ordidge RJ, Chopp M, Oing ZX. Histopathological correlations of nuclear magnetic resonance imaging parameters in experimental ischemia. Magn Reson Imaging.. 1993;11:241-246. [Medline] [Order article via Infotrieve]

26. Jiang Q, Zhang ZG, Chopp M, Helpern JA, Ordidge RJ, Garcia JH, Marchese BA, Qing ZX, Knight RA. Temporal evolution and spatial distribution of the diffusion constant of water in rat brain after transient middle cerebral artery occlusion. J Neurol Sci.. 1993;120:123-130. [Medline] [Order article via Infotrieve]

27. Pierpaoli C, Righini A, Linfante I, Tao-Cheng JH, Alger J, Di Chiro G. Histopathologic correlates of abnormal water diffusion in cerebral ischemia: diffusion-weighted MR imaging and light and electron microscopic study. Radiology.. 1993;189:439-448. [Abstract/Free Full Text]

28. Back T, Hoehn-Berlage M, Kohno K, Hossmann K-A. Diffusion nuclear magnetic resonance imaging in experimental stroke. Stroke.. 1994;25:494-500. [Abstract]

29. Dardzinski BJ, Sotak CH, Fisher M, Hasegawa Y, Li L, Minematsu K. Apparent diffusion constant mapping of experimental focal cerebral ischemia using diffusion-weighted echo-planar imaging. Magn Reson Med.. 1993;30:318-325. [Medline] [Order article via Infotrieve]

30. Hasegawa Y, Fisher M, Latour LL, Dardzinski BJ, Sotak CH. MRI diffusion mapping of reversible and irreversible ischemic injury in focal brain ischemia. Neurology.. 1994;44:1484-1490. [Abstract/Free Full Text]

31. Davis D, Ulatowski J, Eleff S, Izuta M, Mori S, Shungu D, van Zijl PCM. Rapid monitoring of changes in water diffusion coefficients during reversible ischemia in cat and rat brain. Magn Reson Med.. 1994;31:454-460. [Medline] [Order article via Infotrieve]

32. Pierpaoli C, Alger J, Righini A, Mattiello J, Dickerson R, DesPres D, Barnett A, Di Chiro G. High temporal resolution diffusion MRI of global cerebral ischemia and reperfusion. J Cereb Blood Flow Metab.. 1996;16:892-905.[Medline] [Order article via Infotrieve]

This article has been cited by other articles:

				H. K. Lim, J. K. Kim, K. A. Kim, and K.-S. Cho Prostate Cancer: Apparent Diffusion Coefficient Map with T2-weighted Images for Detection--A Multireader Study Radiology, January 1, 2009; 250(1): 145 - 151. [Abstract] [Full Text] [PDF]

				A M BLAMIRE The technology of MRI -- the next 10 years? Br. J. Radiol., August 1, 2008; 81(968): 601 - 617. [Abstract] [Full Text] [PDF]

				D. C. Colvin, T. E. Yankeelov, M. D. Does, Z. Yue, C. Quarles, and J. C. Gore New Insights into Tumor Microstructure Using Temporal Diffusion Spectroscopy Cancer Res., July 15, 2008; 68(14): 5941 - 5947. [Abstract] [Full Text] [PDF]

				Y. Liu, V. Soppi, T. Mustonen, M. Kononen, T. Koivisto, A. Koskela, J. Rinne, and R. L. Vanninen Subarachnoid Hemorrhage in the Subacute Stage: Elevated Apparent Diffusion Coefficient in Normal-appearing Brain Tissue after Treatment Radiology, December 19, 2006; (2006) 2422051698. [Abstract] [Full Text]

				S M. Maniega, M E Bastin, P A Armitage, A J Farrall, T K Carpenter, P J Hand, V Cvoro, C S Rivers, and J M Wardlaw Temporal evolution of water diffusion parameters is different in grey and white matter in human ischaemic stroke J. Neurol. Neurosurg. Psychiatry, December 1, 2004; 75(12): 1714 - 1718. [Abstract] [Full Text] [PDF]

				M E Bastin, S Sinha, A J Farrall, J M Wardlaw, and I R Whittle Diffuse brain oedema in idiopathic intracranial hypertension: a quantitative magnetic resonance imaging study J. Neurol. Neurosurg. Psychiatry, December 1, 2003; 74(12): 1693 - 1696. [Abstract] [Full Text] [PDF]

				C. A. Doege, C. M. Kerskens, B. I. Romero, P. Brunecker, J. Junge-Hulsing, W. von Pannwitz, B. Muller, and A. Villringer Assessment of Diffusion and Perfusion Deficits in Patients with Small Subcortical Ischemia AJNR Am. J. Neuroradiol., August 1, 2003; 24(7): 1355 - 1363. [Abstract] [Full Text] [PDF]

				A. K. Kamal, J. P. Dyke, J. M. Katz, B. Liberato, C. G. Filippi, R. D. Zimmerman, and A. M. Ulug Temporal Evolution of Diffusion after Spontaneous Supratentorial Intracranial Hemorrhage AJNR Am. J. Neuroradiol., May 1, 2003; 24(5): 895 - 901. [Abstract] [Full Text] [PDF]

				H. Tamura, S. Takahashi, N. Kurihara, S. Yamada, J. Hatazawa, and T. Okudera Practical Visualization of Internal Structure of White Matter for Image Interpretation: Staining a Spin-Echo T2-Weighted Image with Three Echo-Planar Diffusion-Weighted Images AJNR Am. J. Neuroradiol., March 1, 2003; 24(3): 401 - 409. [Abstract] [Full Text] [PDF]

				J.M. Wardlaw, S.L. Keir, M.E. Bastin, P.A. Armitage, and A.K. Rana Is diffusion imaging appearance an independent predictor of outcome after ischemic stroke? Neurology, November 12, 2002; 59(9): 1381 - 1387. [Abstract] [Full Text] [PDF]

				A. K. Kamal, A. Z. Segal, and A. M. Ulug Quantitative Diffusion-Weighted MR Imaging in Transient Ischemic Attacks AJNR Am. J. Neuroradiol., October 1, 2002; 23(9): 1533 - 1538. [Abstract] [Full Text] [PDF]

				J. R. Carhuapoma, P. B. Barker, D. F. Hanley, P. Wang, and N. J. Beauchamp Human Brain Hemorrhage: Quantification of Perihematoma Edema by Use of Diffusion-Weighted MR Imaging AJNR Am. J. Neuroradiol., September 1, 2002; 23(8): 1322 - 1326. [Abstract] [Full Text] [PDF]

				S. Sinha, M. E. Bastin, I. R. Whittle, and J. M. Wardlaw Diffusion Tensor MR Imaging of High-Grade Cerebral Gliomas AJNR Am. J. Neuroradiol., April 1, 2002; 23(4): 520 - 527. [Abstract] [Full Text] [PDF]

				J. Helenius, L. Soinne, J. Perkio, O. Salonen, A. Kangasmaki, M. Kaste, R. A. D. Carano, H. J Aronen, and T. Tatlisumak Diffusion-Weighted MR Imaging in Normal Human Brains in Various Age Groups AJNR Am. J. Neuroradiol., February 1, 2002; 23(2): 194 - 199. [Abstract] [Full Text] [PDF]

				C. S. Kidwell, J. L. Saver, J. Mattiello, S. Warach, D. S. Liebeskind, S. Starkman, P. M. Vespa, J. P. Villablanca, N. A. Martin, J. Frazee, et al. Diffusion-perfusion MR evaluation of perihematomal injury in hyperacute intracerebral hemorrhage Neurology, November 13, 2001; 57(9): 1611 - 1617. [Abstract] [Full Text] [PDF]

				C. Oppenheim, C. Grandin, Y. Samson, A. Smith, T. Duprez, C. Marsault, and G. Cosnard Is There an Apparent Diffusion Coefficient Threshold in Predicting Tissue Viability in Hyperacute Stroke? Stroke, November 1, 2001; 32(11): 2486 - 2491. [Abstract] [Full Text] [PDF]

				T Murata, H Kimura, H Kado, M Omori, J Onizuka, T Takahashi, H Itoh, and Y Wada Neuronal damage in the interval form of CO poisoning determined by serial diffusion weighted magnetic resonance imaging plus 1H-magnetic resonance spectroscopy J. Neurol. Neurosurg. Psychiatry, August 1, 2001; 71(2): 250 - 253. [Abstract] [Full Text] [PDF]

				D. D. M. Lin, C. G. Filippi, A. B. Steever, and R. D. Zimmerman Detection of Intracranial Hemorrhage: Comparison between Gradient-echo Images and b0 Images Obtained from Diffusion-weighted Echo-planar Sequences AJNR Am. J. Neuroradiol., August 1, 2001; 22(7): 1275 - 1281. [Abstract] [Full Text] [PDF]

				P. Mukherjee and R. C. McKinstry Reversible Posterior Leukoencephalopathy Syndrome: Evaluation with Diffusion-Tensor MR Imaging Radiology, June 1, 2001; 219(3): 756 - 765. [Abstract] [Full Text] [PDF]

				C. G. Filippi, A. M. Ulu, E. Ryan, S. J. Ferrando, and W. van Gorp Diffusion Tensor Imaging of Patients with HIV and Normal-appearing White Matter on MR Images of the Brain AJNR Am. J. Neuroradiol., February 1, 2001; 22(2): 277 - 283. [Abstract] [Full Text] [PDF]

				J. O. Karonen, Y. Liu, R. L. Vanninen, L. Østergaard, P. L. Kaarina Partanen, P. A. Vainio, E. J. Vanninen, J. Nuutinen, R. Roivainen, S. Soimakallio, et al. Combined Perfusion- and Diffusion-weighted MR Imaging in Acute Ischemic Stroke during the 1st Week: A Longitudinal Study Radiology, December 1, 2000; 217(3): 886 - 894. [Abstract] [Full Text]

				T. Back, J. G. Hirsch, K. Szabo, and A. Gass Failure to Demonstrate Peri-Infarct Depolarizations by Repetitive MR Diffusion Imaging in Acute Human Stroke Stroke, December 1, 2000; 31(12): 2901 - 2906. [Abstract] [Full Text] [PDF]

				E. R. Melhem, R. Itoh, L. Jones, and P. B. Barker Diffusion Tensor MR Imaging of the Brain: Effect of Diffusion Weighting on Trace and Anisotropy Measurements AJNR Am. J. Neuroradiol., November 1, 2000; 21(10): 1813 - 1820. [Abstract] [Full Text] [PDF]

				M. C. DeLano, T. G. Cooper, J. E. Siebert, M. J. Potchen, and K. Kuppusamy High-b-value Diffusion-weighted MR Imaging of Adult Brain: Image Contrast and Apparent Diffusion Coefficient Map Features AJNR Am. J. Neuroradiol., November 1, 2000; 21(10): 1830 - 1836. [Abstract] [Full Text] [PDF]

				S. L. Keir and J. M. Wardlaw Systematic Review of Diffusion and Perfusion Imaging in Acute Ischemic Stroke Stroke, November 1, 2000; 31(11): 2723 - 2731. [Abstract] [Full Text] [PDF]

				M. Augustin, R. Bammer, J. Simbrunner, R. Stollberger, H.-P. Hartung, and F. Fazekas Diffusion-weighted Imaging of Patients with Subacute Cerebral Ischemia: Comparison with Conventional and Contrast-enhanced MR Imaging AJNR Am. J. Neuroradiol., October 1, 2000; 21(9): 1596 - 1602. [Abstract] [Full Text]

				S. T. Engelter, J. M. Provenzale, J. R. Petrella, D. M. DeLong, and J. R. MacFall The Effect of Aging on the Apparent Diffusion Coefficient of Normal-Appearing White Matter Am. J. Roentgenol., August 1, 2000; 175(2): 425 - 430. [Abstract] [Full Text] [PDF]

				S. F. Tanner, L. A. Ramenghi, J. P. Ridgway, E. Berry, M. A. Saysell, D. Martinez, R. J. Arthur, M. A. Smith, and M. I. Levene Quantitative Comparison of Intrabrain Diffusion in Adults and Preterm and Term Neonates and Infants Am. J. Roentgenol., June 1, 2000; 174(6): 1643 - 1649. [Abstract] [Full Text]

				D. G. Gadian, F. Calamante, F. J. Kirkham, M. Bynevelt, C. L. Johnson, D. A. Porter, W. K. Chong, M. Prengler, and A. Connelly Diffusion and Perfusion Magnetic Resonance Imaging in Childhood Stroke J Child Neurol, May 1, 2000; 15(5): 279 - 283. [Abstract] [PDF]

				P. Mukherjee, M. M. Bahn, R. C. McKinstry, J. S. Shimony, T. S. Cull, E. Akbudak, A. Z. Snyder, and T. E. Conturo Differences between Gray Matter and White Matter Water Diffusion in Stroke: Diffusion-Tensor MR Imaging in 12 Patients Radiology, April 1, 2000; 215(1): 211 - 220. [Abstract] [Full Text]

				M. Yoneda, M. Maeda, H. Kimura, A. Fujii, K. Katayama, and M. Kuriyama Vasogenic edema on MELAS: A serial study with diffusion-weighted MR imaging Neurology, December 1, 1999; 53(9): 2182 - 2182. [Abstract] [Full Text] [PDF]

				P. Y-K. Wang, P. B. Barker, R. J. Wityk, A. M. Ulug, P. C. M. van Zijl, and N. J. Beauchamp Diffusion-Negative Stroke: A Report of Two Cases AJNR Am. J. Neuroradiol., November 1, 1999; 20(10): 1876 - 1880. [Abstract] [Full Text]

				Q. Yang, B. M. Tress, P. A. Barber, P. M. Desmond, D. G. Darby, R. P. Gerraty, T. Li, and S. M. Davis Serial Study of Apparent Diffusion Coefficient and Anisotropy in Patients With Acute Stroke Stroke, November 1, 1999; 30(11): 2382 - 2390. [Abstract] [Full Text] [PDF]

				D. G. Darby, P. A. Barber, R. P. Gerraty, P. M. Desmond, Q. Yang, M. Parsons, T. Li, B. M. Tress, and S. M. Davis Pathophysiological Topography of Acute Ischemia by Combined Diffusion-Weighted and Perfusion MRI Stroke, October 1, 1999; 30(10): 2043 - 2052. [Abstract] [Full Text] [PDF]

				A. G. Sorensen, O. Wu, W. A. Copen, T. L. Davis, R. G. Gonzalez, W. J. Koroshetz, T. G. Reese, B. R. Rosen, V. J. Wedeen, and R. M. Weisskoff Human Acute Cerebral Ischemia: Detection of Changes in Water Diffusion Anisotropy by Using MR Imaging Radiology, September 1, 1999; 212(3): 785 - 792. [Abstract] [Full Text]

				N. J. Beauchamp Jr, P. B. Barker, P. Y. Wang, and P. C. M. vanZijl Imaging of Acute Cerebral Ischemia Radiology, August 1, 1999; 212(2): 307 - 324. [Abstract] [Full Text]

				J. L. Sunshine, R. W. Tarr, C. F. Lanzieri, D. M. D. Landis, W. R. Selman, and J. S. Lewin Hyperacute Stroke: Ultrafast MR Imaging to Triage Patients prior to Therapy Radiology, August 1, 1999; 212(2): 325 - 332. [Abstract] [Full Text]

				J. O. Karonen, R. L. Vanninen, Y. Liu, L. Ostergaard, J. T. Kuikka, J. Nuutinen, E. J. Vanninen, P. L. K. Partanen, P. A. Vainio, K. Korhonen, et al. Combined Diffusion and Perfusion MRI With Correlation to Single-Photon Emission CT in Acute Ischemic Stroke : Ischemic Penumbra Predicts Infarct Growth Stroke, August 1, 1999; 30(8): 1583 - 1590. [Abstract] [Full Text] [PDF]

				F. E. Lecouvet, T. P. J. Duprez, J. M. Raymackers, A. Peeters, and G. Cosnard Resolution of early diffusion-weighted and FLAIR MRI abnormalities in a patient with TIA Neurology, March 1, 1999; 52(5): 1085 - 1085. [Abstract] [Full Text]

				J. C. Corvol, C. Oppenheim, R. Manai, M. Logak, D. Dormont, Y. Samson, C. Marsault, and G. Rancurel Diffusion-Weighted Magnetic Resonance Imaging in a Case of Cerebral Venous Thrombosis Stroke, December 1, 1998; 29(12): 2649 - 2652. [Abstract] [Full Text] [PDF]

				V. Nagesh, K. M. A. Welch, J. P. Windham, S. Patel, S. R. Levine, D. Hearshen, D. Peck, K. Robbins, L. D'Olhaberriague, H. Soltanian-Zadeh, et al. Time Course of ADCw Changes in Ischemic Stroke: Beyond the Human Eye! Stroke, September 1, 1998; 29(9): 1778 - 1782. [Abstract] [Full Text] [PDF]

				K.J. van Everdingen, J. van der Grond, L.J. Kappelle, L.M.P. Ramos, and W.P.T.M. Mali Diffusion-Weighted Magnetic Resonance Imaging in Acute Stroke Stroke, September 1, 1998; 29(9): 1783 - 1790. [Abstract] [Full Text] [PDF]

				M. B. Singer, J. Chong, D. Lu, W. J. Schonewille, S. Tuhrim, and S. W. Atlas Diffusion-Weighted MRI in Acute Subcortical Infarction Stroke, January 1, 1998; 29(1): 133 - 136. [Abstract] [Full Text] [PDF]

				P. Mukherjee, J. H. Miller, J. S. Shimony, T. E. Conturo, B. C. P. Lee, C. R. Almli, and R. C. McKinstry Normal Brain Maturation during Childhood: Developmental Trends Characterized with Diffusion-Tensor MR Imaging Radiology, November 1, 2001; 221(2): 349 - 358. [Abstract] [Full Text] [PDF]

This Article Abstract Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Ulug, A. M. Articles by van Zijl, P. C.M. Search for Related Content PubMed PubMed Citation Articles by Ulug, A. M. Articles by van Zijl, P. C.M.