Temporal Changes of the Apparent Diffusion Coefficients of Water and Metabolites in Rats With Hemispheric Infarction: Experimental Study of Transhemispheric Diaschisis in the Contralateral Hemisphere at 7 Tesla

Abstract

The purpose of the present study was to clarify the temporal changes of the apparent diffusion coefficients (ADCs) of cerebral metabolites during early focal ischemia using stimulated echo acquisition mode with short echo time at a 7 T magnet to assess the pathophysiology of the reduction in diffusion properties observed both in the ischemic cerebral hemisphere and in the contralateral hemisphere. The ADCs of metabolites in the infarcted hemisphere 1 hour and 3 hours after the onset of ischemia decreased with 25% and 29% for choline-containing compounds (Cho), 16% and 26% for creatine and phosphocreatine (Cre), and 19% and 19% for N-acetylaspartate (NAA), respectively, compared with the ADC values 2 hours later in the contralateral hemisphere. There were decreases in the ADC of Cho, Cre, and NAA with 21%, 7%, and 18% 8 hours later, respectively, in the noninfarcted hemisphere, which suggested transhemispheric diaschisis in rats with focal cerebral ischemia. The present study proposed that the diffusion characteristics of the brain metabolites might offer new insights into circulatory and metabolic alteration in the cerebral intracellular circumstance.

Keywords

Cerebral ischemia Brain metabolites Rat Diffusion Magnetic resonance spectroscopy Diaschisis

Cerebral vascular disorder is one of the most fatal diseases despite current advances in medical science. Acute cerebral infarction causes a long-term, serious sequelae because of the lack of any established therapy to treat this disorder. Unlike cerebral infarction in animal models, a stroke in humans involves much variability in terms of infarcted volume, infarcted location, patterns of collateral flows, complication of disease, and patient's age (Sorensen et al., 1996; Alison and Warach, 1998). In the diagnosis of acute cerebral ischemia, it is important to understand the extent of reversible ischemic penumbra, to distinguish it from unsalvageable infarcted tissues, and to determine appropriate therapies such as interventional therapy including an intra-arterial injection of thrombolytic agents or conservative medical therapies (The National Institute of Neurological Disorders and Stroke rt-PA Stroke Study Group, 1995; Fisher, 1997). However, methods that enable noninvasive observation of the pathophysiologic state of the affected brain tissue have been limited. In current clinical settings, a therapeutic strategy is often determined according to the duration from the onset of stroke. However, in some cases this may not be pertinent because of the variability between pathologic conditions described above.

Diffusion weighted imaging (DWI) may provide useful information with respect to the structures of cells, membrane permeability, transport processes, and temperature (Stejskal and Tanner, 1965; Le Bihan et al., 1986). Clinically, DWI plays an indispensable role in delineating hyperacute cerebral ischemia (Moseley et al., 1995; Sorensen et al., 1996), a demyelinating disease (Christiansen et al., 1993), and diffusional anisotropy of the white matter (Moseley et al., 1990a) and also aids in distinguishing arachnoid cyst from epidermoid cyst (Tsuruda et al., 1990). Especially, DWI is used to detect ischemic injury of the brain within a few minutes after an arrest of blood flow (Moseley et al., 1990b). Diffusion weighted imaging is, however, based on the diffusivity of water protons. Water is a ubiquitous substance in the living tissue, existing both in the intracellular and extracellular spaces. The signal intensity of DWI depends on the apparent diffusion coefficients (ADC) of the two compartments, the relative volume fractions, and the exchange rate of water through the membranes between these compartments. Therefore, water may not be the suitable molecule to elucidate structures and functions of the affected cells. Many investigators have been trying to discriminate between irreversible infarcted tissue and salvageable ischemic penumbra with the ADC of water protons, but in vain (Roussel et al., 1994; Hoehn-Berlage, 1995; Kohno et al., 1995; Moseley et al., 1995; Sorensen et al., 1996).

Ischemic penumbra is still perfused at a level between the threshold of functional impairment and morphological integrity, which still has a capacity to recover if the perfusion is restored to a normal level (Astrup et al., 1981). Most articles have suggested that though DWI cannot visualize ischemic penumbra, it can delineate eventually infarcted tissues due to the deterioration of energy state and ion homeostasis. Moreover, the pathophysiologic etiologies for the ADC reduction as well as the diagnostic implications of the ADC changes of water protons for tissue viability remain unclear.

To gain more specific insights into molecular mobility in the intracellular environment with regard to cytotoxic edema during early focal cerebral ischemia, it would be beneficial to assess the diffusion properties of metabolites, such as N-acetylaspartate (NAA), creatine and phosphocreatine (Cre), and choline containing compounds (Cho), which exist exclusively in the intracellular space under normal conditions (Merboldt et al., 1993; Nicolay et al., 1995; Wick et al., 1995; van der Toorn et al., 1996).

Assessment of ADCs of these metabolites may provide changes in diffusivity specific to the intracellular environment and can be a noninvasive means for differentiating infarcted tissue from ischemic penumbra. Evaluation of diffusion properties of water protons could not provide these kinds of information.

The deterioration to ischemic cells is exacerbated with duration from the onset of insult. Measuring a temporal profile of ADC values may help clarify underlying pathophysiological changes in neuronal cells during hyperacute ischemia. Changes in signal intensity of DWI and ADC of water at various time points after an ischemic insult have been documented (Jiang et al., 1993; Grohn et al., 1998; Norris et al., 1998). However, a detailed analysis of the temporal evolution in the ADCs of metabolites both in the ischemic and nonischemic hemisphere has never been reported.

Some studies have shown that cerebral blood flow and metabolism in the hemisphere contralateral to the unilateral cerebral infarction may alter because of a loss of remote inhibition or a loss of remote facilitation (transhemispheric diaschisis) (Tamura et al., 1981; Andrews, 1991). Most measurements of the contralateral cerebral blood flow and metabolism show a decrease over the first week, followed by a gradual return toward the baseline. The ADC of water in the contralateral hemisphere during hyperacute unilateral cerebral ischemia does not show any changes. However, the ADCs of metabolites specific to the intracellular environment may show transhemispheric diaschisis.

Almost all ADC measurements on metabolites using proton spectroscopy have been conducted with a relatively long echo time (60 to 136 milliseconds [ms]), although short echo time would increase signal-to-noise ratio. Therefore, in this study we used stimulated echo acquisition mode sequence, which can accomplish both short echo time (TE) and high b value, simultaneously.

MATERIALS AND METHODS

All experiments were performed using a 7.05 T horizontal spectrometer (UNITY plus-SIS 300/183, Varian) equipped with an actively shielded gradient up to 120 mT/m per axis using a 2-cm diameter surface coil. Diffusion-weighted spectroscopy was accomplished with the stimulated echo acquisition mode sequence (TR/TE/TM = 3000/30/40 ms). Unipolar diffusion gradients, whose duration was 6 ms and whose separation was 54 ms, were applied simultaneously in the three axes during both TE/2 intervals. Manual shimming on the selected voxel was preformed with linear gradient shims by maximizing the absorption peak height of the water resonance. The spectral width was 3200 Hz, with a spectral resolution of 4096 points.

Phantom study

The relative concentrations of the metabolites except lactate were adjusted to mimic those in the normal rat brain in vivo: ChoCl 10 mmol/L, creatine 50 mmol/L, and NAA 60 mmol/L (Gyngell et al., 1991; Ryner et al., 1995). The concentration of lactate was 60 mmol/L, and the pH of this aqueous solution was adjusted to 7.4 with phosphate buffer. In the in vitro diffusion experiment (n = 6), the volumes of interest (VOI) were 6 × 6 × 6 mm³ in size. The temperature were maintained at 20°C using a hand-made Plexiglas holder, circulating heated water, and an optical fiber thermometer (FT1110, Takaoka, Japan). Both in water (8 acquisitions) and metabolite (192 acquisitions) diffusion study, using the gradients of 30, 40, 50, 60, 70 mT/m, resulted in the corresponding b values of 361.7, 643.0, 1004.6, 1446.7, 1969.1 sec/mm², where three CHESS water suppression pulses were used.

Control study

Eight male Wistar rats, 250 to 350 g, were orally intubated and ventilated with 1 to 1.5% halothane for anesthesia. Left femoral arteries and veins were catheterized for continuous monitoring of arterial blood pressure and periodic arterial blood gas analysis, and for drug infusion. Rectal temperature was continuously monitored with the optical fiber thermometer and maintained around 37°C by using the hand-made holder circulating heated water. After taking multislice T2 weighted images (TR/TE = 1500/40 ms; slice thickness 2mm), 3.6 × 3.6 × 6.0 mm³ VOIs were selected in both hemispheres in a mirrored fashion. In water diffusion study (8 acquisitions), using the gradients of 30, 40, 50, 60, 70, 80 mT/m, resulted in the corresponding b values of 361.7, 643.0, 1004.6, 1446.7, 1969.1, and 2571.1 sec/mm². In metabolite diffusion study (192 acquisitions), using the gradients of 30, 50, 70, 85, 100 mT/m, resulted in the corresponding b values of 361.7, 1004.6, 1969.1, 2903.4, and 4018.6 sec/mm², where three CHESS water suppression pulses were used.

The data were tested for statistical significance using a paired Student's t-test, where the ADCs of water and metabolites in one hemisphere were compared with those in the other hemisphere.

Focal ischemia study

The manner of animal preparation was analogous to that described in the control study section. Focal cerebral ischemia (n = 10) was induced by occluding the left MCA with poly-L-lysine-coated 4-0 nylon sutures (Longa et al., 1989; Belayev et al., 1996). After obtaining multislice DWIs (TR/TE = 1500/40 ms, Δ/δ = 19/16 ms; b = 1226.7 sec/mm²; slice thickness 2 mm), a 3.6 × 3.6 × 6.0 mm³ VOI was located to cover the ischemic region of the left hemisphere as much as possible and another VOI was selected in the right hemisphere in a mirrored fashion. The b values used were the same as those in the control study. Diffusion-weighted spectra with and without water suppression were obtained 1 to 2, 3 to 4, 5 to 6, 7 to 8 hours after the onset of ischemia in the left hemisphere and 2 to 3, 4 to 5, 6 to 7, 8 to 9 hours later in the right hemisphere. All brains were sectioned coronally with 2 mm slices about 24 hours after MCA occlusion, immersed, and stained with 2% triphenyltetrazolium chloride solution for 30 minutes to evaluate the histopathological tissue damage and the propriety of VOI selection.

Statistical analyses were performed using a repeated measures analysis of variance and Fisher's protected least significant difference. The ADCs of water and the metabolites in the noninfarcted hemisphere 2 hours after MCA occlusion were compared with those in the infarcted hemisphere at various time points, and the ADCs of water and the metabolites were compared in the same hemisphere at the following time point.

All free induction decays were zero-filled at the 8192 points. The free induction decays acquired in the phantom study were not linebroadened to assess distortions induced from residual eddy currents. However, the free induction decays acquired in the animal experiments were processed with a 2 Hz linebroadening. After manual correction for phase, the peaks of Cho, Cre, NAA, and lactate were fitted by means of a least squares curve fitting method, assuming a Lorentzian lineshape. Semilogarithmic plots of the peak heights acquired versus b factor were fitted to the equation S/S_o = e^−b·ADC to obtain the ADCs of metabolites. If the correlation coefficient of the fitting was less than 0.9, the acquired ADC value was abandoned.

RESULTS

Phantom study

The measured ADC values of the phantoms agreed well with the reported values (Merboldt et al., 1993; Nicolay et al., 1995; van der Toorn et al., 1996); water 21.0 ± 0.4 (× 10⁻⁴ mm²/s), choline 10.5 ± 0.6 (× 10⁻⁴ mm²/s), creatine 8.5 ± 0.3 (× 10⁻⁴ mm²/s), and NAA 6.5 ± 0.2 (× 10⁻⁴ mm²/s) (Table 1). The ADC of lactate was 9.5 ± 0.2 (× 10⁻⁴ mm²/s). We did not observe any eddy currents effects, though we did not perform the line-broadening process (Fig. 1).

TABLE 1.

ADC values (in 10⁻⁴mm²/s) of water and metabolites in the aqueous solution (n = 6) and in the normal rat brain in vivo (n = 8)

Substance	Phantom	Rat brain (right)	Rat brain (left)
Water	21.0 ± 0.4	7.0 ± 0.2	7.0 ± 0.2
Choline	10.5 ± 0.6	1.5 ± 0.1	1.5 ± 0.2
Creatine	8.5 ± 0.3	1.8 ± 0.2	1.9 ± 0.2
NAA	6.5 ± 0.2	1.8 ± 0.2	1.8 ± 0.1
Lactate	9.5 ± 0.2

The phantom data were measured at 20°C. The values are given as mean ± standard deviation. There were no significant differences of the ADCs of water and metabolites in the left and right hemispheres (paired Student's t-test).

FIG. 1.

Representative diffusion-weighted localized proton spectra from an aqueous solution of 10 mmol/L ChoCl, 50 mmol/L creatine, 60 mmol/L N-acetylaspartate, and 60 mmol/L lactate adjusted to pH 7.4. The diffusion-weighting increases from top to bottom (b value: 361.7, 643.0, 1004.6, 1446.7, 1969.1 sec/mm², TR/TE/TM = 3000/30/40 ms, volume of interest = 6 × 6 × 6mm³) and are delineated without linebroadening to identify residual eddy current effects. Assignments: 1, choline; 2, creatine; 3, N-acetylaspartate; and 4, lactate. TE, echo time; TR, repetition time.

Control study

The blood pressure and arterial blood gas data during the experiments were within normal limits as shown in Table 2. The ADCs of water and metabolites in the right and left hemispheres were 7.0 ± 0.2, 7.0 ± 0.2 (× 10⁻⁴ mm²/s) for water, 1.5 ± 0.1, 1.5 ± 0.2 (× 10⁻⁴ mm²/s) for Cho, 1.8 ± 0.2, 1.9 ± 0.2 (× 10⁻⁴ mm²/s) for Cre, 1.8 ± 0.2, 1.8 ± 0.1 (× 10⁻⁴ mm²/s) for NAA, respectively (Table 1). There were no significant differences of the ADCs of water and metabolites in the left and right hemispheres (paired Student's t-test).

TABLE 2.

Physiologic parameters in healthy rats and in rats with unilateral cerebral infarct at various time points during experiments

	pH	PaO₂ (mm Hg)	PaCO₂ (mm Hg)	MABP (mm Hg)
Normal rat	7.36 ± 0.03	187 ± 32	34.2 ± 9.2	108 ± 9
Before ischemia	7.38 ± 0.03	165 ± 30	39.1 ± 6.2	118 ± 9
2 hours after	7.36 ± 0.04	159 ± 34	41.4 ± 9.9	112 ± 15
5 hours after	7.35 ± 0.04	145 ± 38	42.0 ± 10.1	115 ± 18

The values are given as mean ± standard deviation. MABP, mean arterial blood pressure.

Focal ischemia study

Diffusion-weighted imaging and triphenyltetrazolium chloride histological staining specimens confirmed that all rats had hemispheric infarct. The blood press and arterial blood gas data during the experiment were within the normal range as shown in Table 2. Figure 2 shows a representative DWI of a rat with hemispheric infarct, measured about 30 minutes after MCA occlusion. A representative series of diffusion-weighted spectra is shown in Fig. 3.

FIG. 2.

Representative diffusion-weighted images (TR/TE = 1500/40 ms, Δ/δ = 19/16 ms, b = 1226.7 sec/mm², slice thickness 2 mm) of a rat brain about 30 minutes after MCA occlusion. The selected voxels (3.6 × 3.6 × 6 mm³) in the infarcted hemisphere is outlined in the images. The volume of interest in the contralateral hemisphere is placed in a mirrored fashion. TE, echo time; TR, repetition time.

FIG. 3.

Representative diffusion-weighted localized proton spectra from the volume of interest (VOI) in the infarcted hemisphere about one hour after MCA occlusion. The diffusion-weighting increases from top to bottom (b value: 361.7, 1004.6, 1969.1, 2903.4, and 4018.6 sec/mm², TR/TE/TM = 3000/30/40 ms, VOI = 3.6 × 3.6 × 6mm³) and are delineated with 2 Hz linebroadening. Assignments: 1, choline containing compounds; 2, creatine and phosphocreatine; 3, N-acetylaspartate; and 4, lactate. TE, echo time; TR, repetition time.

Apparent diffusion coefficient of water

The ADC of water in the infarcted hemisphere significantly decreased 1, 3, 5, and 7 hours after occlusion, compared with that of the noninfarcted hemisphere 2 hours later (P < 0.001, Fisher's protected least significant difference) (Fig. 4A). In the infarcted hemisphere, there was no significant change from 1 to 3 hours, 3 to 5 hours and 5 to 7 hours. But from 1 to 5 hours (P < 0.05) and 1 to 7 hours (P < 0.01), the ADC of water significantly reduced, suggesting it has gradually decreased in the infarcted hemisphere during the acute phase. In the non-infarcted hemisphere, there was no significant difference between the ADCs of water throughout the experiment.

FIG. 4.

Temporal change of the apparent diffusion coefficients (ADCs) of water and metabolites. Individual data are plotted from 10 rats with focal cerebral ischemia. Mean values are given by lines. (A) ADC of water. (B) ADC of choline containing compounds. (C) ADC of creatine/phosphocreatine. (D) ADC of N-acetylaspartate. (E) ADC of lactate. Statistical significance is analyzed with repeated measures analysis of variance and Fisher's protected least significant difference. (Figure continues on next page)

Apparent diffusion coefficient of choline-containing Compounds

The ADC of Cho in the infarcted hemisphere significantly decreased 1, 3, 5, and 7 hours after occlusion, compared with that of the noninfarcted hemisphere 2 hours later (P < 0.001) (Fig. 4B). In the infarcted hemisphere, the ADC significantly reduced from 1 to 7 hours (P < 0.05). In the noninfarcted hemisphere, there was no significant decrease from 2 hours to 4 hours. However, there was significant reduction from 4 to 6 hours (P < 0.01) and 6 to 8 hours (P < 0.01).

Apparent diffusion coefficient of creatine/phosphocreatine

The ADC of Cre in the infarcted hemisphere significantly decreased 1, 3, 5, and 7 hours after occlusion, compared with that of the noninfarcted hemisphere 2 hours later (P < 0.001) (Fig. 4C). Though the ADC showed significant reduction from 1 to 3 hours (P < 0.01) and from 5 hours to 7 hours (P < 0.05) in the infarcted hemisphere, there was no significant change from 3 to 5 hours. In the noninfarcted hemisphere, there was no significant decrease from 2 to 6 hours. However, there was a significant reduction from 6 to 8 hours (P < 0.05).

Apparent diffusion coefficient of N-acetylaspartate

The ADC of NAA in the infarcted hemisphere was significantly decreased 1, 3, 5, and 7 hours after occlusion, compared with that of the noninfarcted hemisphere 2 hours later (P < 0.001) (Fig. 4D). In the infarcted hemisphere, there were no significant changes at any time. In the noninfarcted hemisphere, there was no significant decrease from 2 hours to 6 hours. However, there was significant decrease from 6 to 8 hours (P < 0.001).

Apparent diffusion coefficients of lactate

The peak of the lactate in the noninfarcted hemisphere could not be detected (Fig. 4E). In the ADC of lactate in the infarcted hemisphere, there were no statistically significant changes throughout the experiment.

DISCUSSION

There are three important findings in the present study: (1) ADC reduction of water and metabolites in the ischemic hemisphere at each time point, compared with the ADCs in the contralateral hemisphere 2 hours after stroke, (2) sequential reduction of the ADCs of water, Cho, and Cre in the infarcted hemisphere, and (3) sequential reduction of the ADCs of metabolites except lactate in the contralateral hemisphere.

The ADC of water showed about 15% reduction in the hemispheric infarction (5.8 × 10⁻⁴mm²/s). It was lower than the previously reported values (Knight et al., 1991; Benveniste et al., 1992; Davis et al., 1994; van der Toorn et al., 1994; van Gelderen et al., 1994; Hoehn-Berlage, 1995; van der Toorn et al., 1996). The reasons were believed to be that the size of the voxel measured in this study was larger than the histologically confirmed infarcted area and that the ischemic penumbra and the nonischemic areas in the infarcted hemisphere were included in the measured VOI. Therefore, the ADC of the metabolites in the infarcted hemisphere was thought to partly include penumbra and noninfarcted areas. It is very difficult to have a rat model with complete infarction throughout the hemisphere while having no ischemia in the contralateral hemisphere. As well, the VOI could not be lessened in order to keep enough signal-to-noise ratio of signal intensity originated from metabolites with very low natural abundance, compared with water.

The ADCs of metabolites in the infarcted hemisphere 1 hour and 3 hours after MCA occlusion were 25% and 29% for Cho, 16% and 26% for Cre, and 19% and 19% for NAA, respectively. All these values were significantly lower than the ADCs in the noninfarcted hemisphere 2 hours later. The changes in the ADCs of metabolites were larger than that of water, and it was assumed that the diffusion properties of metabolites were more sensitive for detecting ischemic neuronal damage than that of water. Even in the noninfarcted hemisphere, ischemia-related decrease of ADC was observed (transhemispheric diaschisis), and it may be well that the decrease in diffusivity is present not only in the penumbra, but also in nonischemic areas in the infarcted hemisphere (intrahemispheric diaschisis) (Andrews, 1991). At present, individual measurements of ADC in the above-mentioned three sites (noninfarcted areas in the infarcted hemisphere, penumbra, and the infarcted area) are difficult because the voxel measured was too large, and intrahemispheric diaschisis could not be confirmed. In the future, improvement of spatial and time resolution in the spectroscopic measurement may enable to measure the ADCs in a smaller voxel. By measuring the ADCs of metabolites, we will be able to predict the outcome of neuronal cellular damage in the early stage of cerebral ischemia, resulting in the noninvasive delineation of penumbra along with the infarcted area, which will provide a great impact on the strategy of treatment for patients with cerebral infarction.

Van der Toorn and coworkers (1996) reported that the ADCs of Cre and NAA in the infarcted hemisphere 3 hours after MCA occlusion was significantly lowered (29% and 19% for Cre and NAA, respectively) compared with the noninfarcted hemisphere, although the ADC of Cho did not show significant decrease. These results regarding the ADCs of Cre and NAA agreed well with our data. They did not give explanation for absence of significant reduction in the ADC of Cho. It may not be appropriate to compare our results with their results because of the different experimental models and procedures. In their series, MCA occlusion was accomplished with Fisher rats using Brint's model. Some data concerning the ADC of Cho were excluded in their series because a good fitting could not be obtained. A further investigation would be necessary to elucidate the ADC of Cho in rats with hemispheric infarct.

Wick and coworkers (1995) reported the ADC changes of water and cerebral metabolites using rat models with transient global cerebral ischemia. They reported with the assumption that signal intensity of each metabolite did not change during and after transient ischemia, the decrease in the ADCs during ischemia and about 40 minutes after reperfusion were 28% and 13% for Cho, 35% and 21% for Cre, 12% and 13% for NAA, and 46% and 23% for myo-inositol, respectively, showing statistically significant differences. Their results were similar to our results with respect to the higher rate of changes in the ADCs of Cho and Cre than that of NAA, and that Cho and Cre showed higher sensitivity than NAA as a barometer of ischemic damage. Although the present study does not provide clear evidence for higher sensitivity of Cho and Cre, it may be related to the function of these metabolites in the intracellular space.

It is impossible to compare the ADCs of lactate in the infarcted and noninfarcted hemisphere because the amount of lactate in the noninfarcted cerebral tissue is very minute, and thus cannot be detected by 1H-MRS. However, the results of van der Toorn et al. (1996) and those of our study are in agreement in that the ADC of lactate did not show sequential changes and was significantly lower than the ADCs of other metabolites.

In terms of water diffusion, anisotropy has been established to correspond with neural fiber orientation (Chenevert et al., 1990; Moseley et al., 1990a; Sakuma et al., 1991). Similarly, anisotropic diffusion in metabolites has been reported in the human cerebral white matter (Henriksen et al., 1995). If anisotropic diffusion exists in the brain of rats, the ADC acquired would be the projected component of diffusion tensor to a certain axis. Therefore, the ADC value may depend on the directional difference between the axes of the magnet and the brain. However, the voxel size of 3.6 × 3.6 × 6.0 mm³ used in this study includes the white matter with various fiber orientation, and therefore, anisotropic diffusion was thought to be averaged and cancelled.

The parameters applied to obtain the ADCs of the metabolites in our study required 48 minutes in one session. Since we examined hyperacute cerebral infarction, the concentration and T2 value of metabolites may be altered during measurement. If this was true, changes in signal intensity in accordance with alteration of concentration and T2 might result in erroneous calculation of ADC. But it was reported that signal intensities and relaxation properties of Cho, Cre, and NAA were not altered between 3 to 6 hours after the onset of ischemia but did decrease 24 hours after stroke (van der Toorn et al., 1995). Therefore, it was unnecessary to consider the alteration in concentration and T2 of metabolites in our study since the measurement was finished within 9 hours after the onset of ischemia.

A major factor contributing to the reduction in the ADC of water in the cerebral ischemia so far has been suggested to be structural distortion of extracellular space. The reduction in volume of the extracellular fluid in the cerebral tissues occurs in accordance with cytotoxic edema (van der Toorn et al., 1996; Niendorf et al., 1996). However, in this study, we observed the decrease in the ADC of metabolites, which could not be attributed to the increased tortuosity in the extracellular space because these metabolites were found mainly inside the cell under normal conditions. The membrane may become leaky under pathological condition such as stroke, and then part of metabolites may move to the extracellular space. This would probably cause the ADC values to be higher, because it is expected that the ADC of metabolites in the extracellular space is higher than that in the intracellular space. The decrease in the ADC of water during cerebral ischemia was ascribed to the increase in tortuosity in the extracellular space, and we could estimate that the decrease in the ADC of metabolites as well as water was originated from the increase of the intracellular tortuosity, which could be caused by the fragmentation of intracellular molecular structures and dissociation of microtubles.

A drawback of the present report was that the duration of anesthesia was quite long for rats. There has been no report that describes the anesthetic effect on diffusion properties, to our knowledge, and thus further investigation on this point is necessary. However, the anesthetic agent halothane is widely used to study cerebral ischemia and has a minor effect on cerebral metabolism. Although the ADC of water did not show any change in the non-infarcted hemisphere, the ADC of metabolites decreased sequentially in our study. This result suggested that the diffusion properties of metabolites might have unique information about intracellular circumstance, which that of water molecule cannot provide.

The result worthy of special mention in the present study was the decrease in the ADC of metabolites in the noninfarcted contralateral hemisphere of rats with cerebral hemispheric infarct. Many reports have shown that the ADC of water in the nonischemic hemisphere contralateral to the ischemic hemisphere does not have any changes, as in this study. However, significant decreases in the ADC was observed in Cho from 4 to 8 hours, in Cre from 6 to 8 hours, and in NAA from 6 to 8 hours, respectively. Compared with the ADC of the noninfarcted hemisphere 2 hours after occlusion, there were decreases with 21% for Cho, 7% for Cre, and 18% for NAA 8 hours later. The decrease in the ADCs of Cho and Cre in the noninfarcted hemisphere was smaller than those in the infarcted hemisphere but the ADC of NAA in the noninfarcted hemisphere was decreased almost equally to that in the ischemic hemisphere 8 hours later. Van der Toorn and coworkers (1996) reported that there were no significant changes in the ADC of Cho at 24 hours, compared with the ADC 3 hours after occlusion; however, the ADC of Cre and NAA did show significant decreases of 18% and 19%, respectively. In our study, the ADC of Cre, a sensitive marker of ischemia in the infarcted hemisphere, reduced less than the ADCs of Cho and NAA in the noninfarcted hemisphere 8 hours after ischemia. This result indicated that the cause of the decrease in the ADC of metabolites was different in the infarcted and noninfarcted hemisphere in rats with focal cerebral ischemia. The decrease in the ADC of Cre may appear to be prominent later than the other two substances although we cannot determine this since we did not measure the ADC of metabolites 24 hours or more after ischemic insult.

Andrews (1991) described the changes in various parameters, such as electrical activity, blood flow, and metabolism, in the hemisphere contralateral to the unilateral hemispheric infarction. He indicated that there were many reports stating that animal models within 24 hours of the onset of infarction showed decrease of cerebral blood flow and local oxygen metabolites and glucose metabolites. Though there have been no reports that show the relationship between these parameters and the ADC of metabolites, the decreased ADC of metabolites is expected to be one of the transhemispheric diaschisis. One of the important factors contributing to the transhemispheric diaschisis may be the decrease in cerebral blood flow. Compared with the ADC decrease in the infarcted and noninfarcted hemispheres, the decrease in cerebral blood flow of the noninfarcted hemisphere contralateral to the ischemic hemisphere was relatively mild (80% to 90% of the normal level) (Kogure et al., 1980; Tamura et al., 1981). Therefore, we cannot conclude that only the blood flow decrease is responsible for reducing the ADC of metabolites in the noninfarcted hemisphere. There should not have been increases in intracellular and extracellular tortuosity in the noninfarcted hemisphere, as seen in the ADC changes in the infarcted hemisphere. These suggest factors other than intracellular viscosity and intracellular and extracellular tortuosity-such as remote effects through transsynaptic neural pathways-contribute to the decrease in the ADC of metabolites in rats with focal cerebral ischemia.

Footnotes

Abbreviations used

References

Alison

Warach

(1998) Magnetic resonance imaging of acute stroke. J Cereb Blood Flow Metab 18:583–609

Andrews

(1991) Transhemispheric diaschisis. A review and comment. Stroke 22:943–949

Astrup

Siesjo

Symon

(1981) Thresholds in cerebral ischemia—the ischemic penumbra Stroke 12:723–725

Belayev

Alonso

Busto

Zhao

Ginsberg

(1996) Middle cerebral artery occlusion in the rat by intraluminal suture. Neurological and pathological evaluation of an improved model. Stroke 27:1616–1623

Benveniste

Hedlund

Johnson

(1992) Mechanism of detection of acute cerebral ischemia in rats by diffusion-weighted magnetic resonance microscopy. Stroke 23:746–754

Chenevert

Brunberg

Pipe

(1990) Anisotropic diffusion in human white matter. Demonstration with MR techniques in vivo. Radiology 177:401–405

Christiansen

Gideon

Thomsen

Stubgaard

Henriksen

Larsson

(1993) Increased water self-diffusion in chronic plaques and in apparently normal white matter in patients with Mult Scler. Acta Neurol Scand 87:195–199

Davis

Ulatowski

Eleff

Izuta

Mori

Shungu

van Zijl

(1994) Rapid monitoring of changes in water diffusion coefficients during reversible ischemia in cat and rat brain. Magn Reson Med 31:454–460

Fisher

(1997) Characterizing the target of acute stroke therapy. Stroke 28:866–872

10.

Grohn

OHJ

Lukkarinen

Oja

JME

van Zijl

PCM

Ulatowski

Traystman

Kauppinen

(1998) Noninvasive detection of cerebral hypoperfusion and reversible ischemia from reductions in the magnetic resonance imaging relaxation time, T2. J Cereb Blood Flow Metab 18:911–920

11.

Gyngell

Ellermann

Michaelis

Hanicke

Merboldt

Bruhn

Frahm

(1991) Non-invasive 1H NMR spectroscopy of the rat brain in vivo using a short echo time STEAM localization sequence. NMR Biomed 4:150–156

12.

Henriksen

Karnick

Rosenbaum

(1995) Apparent diffusion of metabolites in the human brain. A proton-MRS study, in Proc., SMR, 3rd Annual Meeting, Nice, France, p. 1855

13.

Hoehn-Berlage

(1995) Diffusion-weighted NMR imaging: application to experimental focal cerebral ischemia. NMR Biomed 8:345–358

14.

Jiang

Zhang

Chopp

Helpern

Ordidge

Garcia

Marchese

Qing

Knight

(1993) Temporal evolution and spatial distribution of the diffusion constant of water in rat brain after transient middle cerebral artery occlusion. J Neurol Sci 120:123–130

15.

Knight

Ordidge

Helpern

Chopp

Rodolosi

Peck

(1991) Temporal evolution of ischemic damage in rat brain measured by proton nuclear magnetic resonance imaging. Stroke 22:802–808

16.

Kogure

Busto

Schwartzman

Scheinberg

(1980) The dissociation of cerebral blood flow, metabolism, and function in the early stages of developing cerebral infarction. Ann Neurol 8:278–290

17.

Kohno

Hoehn-Berlage

Mies

Back

Hossmann

(1995) Relationship between diffusion-weighted MR images, cerebral blood flow, and energy state in experimental brain infarction. Magn Reson Imaging 13:73–80

18.

Le Bihan

Breton

Lallemand

Grenier

Cabanis

Laval-Jeantet

(1986) MR Imaging of intravoxel incoherent motions: Application to diffusion and perfusion in neurologic disorders. Radiology 161:401–407

19.

Longa

Weinstein

Carlson

Cummins

(1989) Reversible middle cerebral artery occlusion without craniectomy in rats. Stroke 20:84–91

20.

Merboldt

Horstermann

Hanicke

Bruhn

Frahm

(1993) Molecular self-diffusion of intracellular metabolites in rat brain in vivo investigated by localized proton NMR diffusion spectroscopy. Magn Reson Med 29:125–129

21.

Moseley

Cohen

Kucharczyk

Mintorovitch

Asgari

Wendland

Tsuruda

Norman

(1990a) Diffusion-weighted MR imaging of anisotropic water diffusion in cat central nervous system. Radiology 176:439–445

22.

Moseley

Cohen

Mintorovitch

Chileuitt

Shimizu

Kucharczyk

Wendland

Weinstein

(1990b) Early detection of regional cerebral ischemia in cats: comparison of diffusion- and T2-weighted MRI and spectroscopy. Magn Reson Med 14:330–346

23.

Moseley

Butts

Yenari

Marks

de Crespigny

(1995) Clinical aspects of DWI. NMR Biomed 8:387–396

24.

The National Institute of Neurological Disorders and Stroke rt-PA Stroke Study Group (1995) Tissue plasminogen activator for acute ischemic stroke. N Eng J Med 333:1581–1587

25.

Nicolay

van der Toorn

Dijkhuizen

(1995) In vivo diffusion spectroscopy. An overview. NMR Biomed 8:365–374

26.

Niendorf

Dijkhuizen

Norris

van Lookeren Campagne

Nicolay

(1996) Biexponential diffusion attenuation in various states of brain tissue: implications for diffusion-weighted imaging. Magn Reson Med 36:847–857

27.

Norris

Hoehn-Berlage

Dreher

Kohno

Busch

Schmitz

(1998) Characterization of middle cerebral artery occlusion infarct development in the rat using fast nuclear magnetic resonance proton spectroscopic imaging and diffusion-weighted imaging. J Cereb Blood Flow Metab 18:749–757

28.

Roussel

van Bruggen

King

Houseman

Williams

Gadian

(1994) Monitoring the initial expansion of focal ischemic changes by diffusion-weighted MRI using a remote controlled method of occlusion. NMR Biomed 7:21–28

29.

Ryner

Sorenson

Thomas

(1995) Localized 2D J- resolved 1H MR spectroscopy: strong coupling effects in vitro and in vivo. Magn Reson Imaging 13:853–869

30.

Sakuma

Nomura

Takeda

Tagami

Nakagawa

Tamagawa

Ishii

Tsukamoto

(1991) Adult and neonatal human brain: diffusional anisotropy and myelination with diffusion-weighted MR imaging. Radiology 180:229–233

31.

Sorensen

Buonanno

Gonzalez

Schwamm

Lev

Huang-Hellinger

Reese

Weisskoff

Davis

Suwanwela

Can

Moreira

Copen

Look

Finklestein

Rosen

Koroshetz

(1996) Hyperacute stroke: evaluation with combined multisection diffusion-weighted and hemodynamically weighted echo-planar MR imaging. Radiology 199:391–401

32.

Stejskal

Tanner

(1965) Spin diffusion measurements: spinechoes in the presence of a time-dependent field gradient. J Chem Phys 42:288–292

33.

Tamura

Graham

McCulloch

Teasdale

(1981) Focal cerebral ischaemia in the rat: 2. Regional cerebral blood flow determined by [14C]iodoantipyrine autoradiography following middle cerebral artery occlusion. J Cereb Blood Flow Metab 1:61–69

34.

Tsuruda

Chew

Moseley

Norman

(1990) Diffusion-weighted MR imaging of the brain: value of differentiating between extraaxial cysts and epidermoid tumors. Am J Neuroradiol 11:925–931

35.

van der Toorn

Verheul

van der Sprenkel

JWB

Tulleken

Nicolay

(1994) Changes in metabolites and tissue water status after focal ischemia in cat brain assessed with localized proton MR spectroscopy. Magn Reson Med 32:685–691

36.

van der Toorn

Dijkhuizen

Tulleken

CAF

Nicolay

(1995) T1 and T2 relaxation times of the major 1H-containing metabolites in rat brain after focal ischemia. NMR Biomed 8:245–252

37.

van der Toorn

Sykova

Dijkhuizen

Vorisek

Vargova

Skobisova

van Lookeren Campagne

Reese

Nicolay

(1996) Dynamic changes in water ADC, energy metabolism, extracellular space volume, and tortuosity in neonatal rat brain during global ischemia. Magn Reson Med 36:52–60

38.

van der Toorn

Dijkhuizen

Tulleken

CAF

Nicolay

(1996) Diffusion of metabolites in normal and ischemic rat brain measured by localized 1H MRS. Magn Reson Med 36:914–922

39.

van Gelderen

de Vleeschouwer

MHM

DesPres

Pekar

van Zijl

Moonen

(1994) water diffusion and acute stroke. Magn Reson Med 31:154–163

40.

Wick

Nagatomo

Prielmeier

Frahm

(1995) alteration of intracellular metabolite diffusion in rat brain in vivo during ischemia and reperfusion. Stroke 26:1930–1934

41.

Zhang

Silva

Williams

Koretsky

(1995) NMR measurement of perfusion using arterial spin labeling without saturation of macromolecular spins. Magn Reson Med 33:370–376