Diffusion-weighted imaging studies of cerebral ischemia in gerbils. Potential relevance to energy failure

Diffusion weighting by the trace of the diffusion tensor within a single scan

The anisotropy of the water diffusion tensor inside brain causes contrast in diffusion images, which depends on the relative orientation of the diffusion gradients and the subject. Because the trace of a tensor is invariant upon rotation, measurement of this trace can reduce the orientation effect. A family of imaging pulse sequences is presented in which the signal intensity is weighted by the trace of the diffusion tensor in a single scan. The methods are demonstrated for chicken gizzard in several orientations with respect to the gradient frame of reference, and for ischemic injury in cat brain after middle cerebral artery occlusion. The sensitivity of the techniques to the presence of background gradients is measured and discussed in detail. As a result, pulse sequences are suggested that provide reliable diffusion constants in both homogeneous and inhomogeneous magnetic fields. The efficiency of the techniques for clinical application is also evaluated.

q-Space imaging of the brain

q-Space imaging (Callaghan, J. Magn. Reson. 88, 493 (1990)) has been used to obtain mouse brain water displacement profiles. These profiles take the form of a unidirectional incoherent-displacement probability density distribution. Two groups of mice were studied, a normal group and one in which surgery had been performed to reduce the supply of blood to the forebrain. In the normal group the incoherent displacement of water was reduced postmortem. Four of the surgically treated mice yielded displacement profiles that resembled those obtained postmortem; the remaining two were near normal. This study demonstrates the feasibility of in vivo q-space imaging. The displacement profile changes that occur subsequent to an interruption of the forebrain blood supply are consistent with the hyperintensity changes seen in diffusion-weighted imaging.

Early evolution and recovery from excitotoxic injury in the neonatal rat brain: a study combining magnetic resonance imaging, electrical impedance, and histology

We explored the therapeutic potentials of two N-methyl-D-aspartate (NMDA) receptor antagonists in vivo using different techniques. NMDA injected into the striatum of neonatal rats (20 nmol/0.5 microliters) induced a rapid increase in the diffusion-weighted (DW) image intensity, spreading over a large part of the ipsilateral hemisphere. Subcutaneous injection of the NMDA receptor antagonist MK-801 (1 mg/kg) or D-(E)-4-(3-phosphono-2-prop-enyl)-2-piperazine-carboxylic acid (D-CPPene; 1.5 mg/kg) reversed both the volume and the grading of the NMDA-induced hyperintensity of DW images, the reversal by MK-801 being more rapid than that by D-CPPene. In the cerebral cortex, there was an inverse relationship between changes in DW image intensity and the size of the extracellular space, assessed by electrical impedance measurements. The reduction of the hyperintense volume in DW images 1 or 2 h after MK-801 or D-CPPene treatment of NMDA-injected animals depended on the type of antagonist used and on the interval between intrastriatal NMDA injection and antagonist treatment. The reduction was 95% when MK-801 was given with a delay of 90 min and decreased to 20% when it was given at 360 min. With D-CPPene, the reduction was 80% after a delay of 30 min and virtually absent when it was administered at 360 min. Quantitative analysis showed significant correlations between the residual hyperintense volume 1 or 2 h after MK-801 or D-CPPene treatment and the final lesion volume, assessed from either T2-weighted images (R = 0.89, p < 0.001) or histology (R = 0.80, p < 0.001) 5 days after the insult. This study illustrates the sensitivity of DW magnetic resonance imaging to monitor in vivo early events after an excitotoxic insult and the effect of putative protective drugs that may counteract the resulting damage.

NMR imaging of the apparent diffusion coefficient (ADC) for the evaluation of metabolic suppression and recovery after prolonged cerebral ischemia

Adult normothermic cats were submitted to 1-h complete cerebrocirculatory arrest by intrathoracic occlusion of the internal mammary, the innominate, and the subclavian arteries in combination with pharmacologically induced hypotension. After ischemia, recirculation was initiated at different blood pressure levels to manipulate the postischemia resuscitation conditions. The resulting spectrum of postischemic recovery was studied by combining nuclear magnetic resonance imaging of the apparent diffusion coefficient (ADC) with pictorial assays of brain tissue pH, ATP, glucose, and lactate. Before ischemia, the mean ADC (average of seven coronal slices of five cats) was 713 +/- 40 x 10(-6) mm2/s. After 10-min ischemia, ADC declined to 68% of control and after 50 min slightly further to 63% of control. During recirculation after 1-h ischemia, recovery of ADC varied depending on the initial reperfusion pressure and other systemic variables. In two animals ADC only transiently increased followed by a secondary decline below the postischemic level. In three other animals ADC returned to near control within 1 h of recirculation. The comparison of ADC changes with previously reported changes in extracellular volume revealed a close relationship, supporting the notion that ADC is a function of the intra/extracellular water compartmentation. Recovery of ADC correlated closely with tissue pH and metabolic recovery, studied 3 h after the initiation of recirculation. Animals without recovery of ADC exhibited global depletion of ATP and glucose and severe lactacidosis, whereas animals with recovery of ADC showed replenishment of ATP and glucose to near control and a substantial reversal of lactacidosis.(ABSTRACT TRUNCATED AT 250 WORDS)

The effect of hypothermia on transient focal ischemia in rat brain evaluated by diffusion- and perfusion-weighted NMR imaging

The effect of moderate whole-body hypothermia (30 degrees C) on transient middle cerebral artery (MCA) occlusion in the rat was evaluated using diffusion- and perfusion-weighted magnetic resonance imaging. Two hours of transient MCA occlusion was induced by intracarotid insertion of a nylon filament under normothermic (n = 14) and hypothermic (n = 7) conditions. Diffusion- and perfusion-weighted imaging were performed before, during, and after focal ischemia from 30 min up to 7 days. In hypothermic animals, scattered neuronal necrosis was localized to select areas of the caudate putamen and the parietal and insular cortex. In contrast, the normothermic ischemic animals exhibited pan-necrosis and infarct encompassing the damaged area. The diffusion and perfusion data measured from caudate putamen indicate that hypothermia causes a significant reduction in the apparent diffusion coefficient of water (ADCw) and CBF values from normothermic control values (p < 0.01). In both normothermic and hypothermic animals after onset of MCA occlusion, ADCw and CBF values in the core of the ischemic region (striatum) significantly declined from the preischemic and homologous contralateral control ADCw and CBF values (p < 0.05). However, ADCw and CBF in the hypothermic group returned toward control more rapidly than in the normothermic group. These results suggest that the protective effect of hypothermia on ischemic cell damage is reflected in the early return of ADCw during reperfusion and the reduction of ischemic cell damage by hypothermia may be mediated by the improved CBF during acute reperfusion.

Early detection of regional myocardial ischemia in ex vivo piglet hearts: MR imaging with magnetization transfer

The effects of regional myocardial ischemia and reperfusion on magnetization transfer (MT) contrast were investigated in an ex vivo perfused piglet heart model. The extent of the ischemic area was defined with perfusion magnetic resonance (MR) studies performed with use of extracellular contrast agents. Relative MT contrast was calculated for a total of 106 regions of interest in nine hearts. In the areas defined as being severely ischemic in the perfusion studies, a small but significant increase in the MT contrast of 18% +/- 9 (standard deviation) (n = 35) was found as early as 10 minutes after the start of ischemia. This contrast difference was reduced to 11% +/- 10 after 2 hours of total occlusion. The contrast between normal and ischemic tissue can be explained in part by the effect of inflowing blood, which leads to changes in both equilibrium magnetization and the apparent T1 of the perfused tissue. However, theoretical estimation suggests that these flow-related changes would produce a maximal relative change in MT contrast of approximately 4%. The most likely explanation for the rest of the observed changes is alteration in the distribution of cellular water related to the so-called intracellular edema that is known to be associated with the acute phase of myocardial ischemia.

Self-diffusion in CNS tissue by volume-selective proton NMR

We have used a volume-selective NMR pulse sequence to measure the diffusion coefficient (D) inside a selected cuboid volume. The sequence is a variant of the stimulated echo method for volume-selective spectroscopy, and allows sequential measurement of D in any of three orthogonal directions and for a range of diffusion times. The echo attenuation from which D is calculated is free from the cross terms that can arise because of interactions between volume selection and diffusion-sensitizing gradients. Anisotropic and diffusion-time-dependent behavior is observed in brain structures known to have axon fibers aligned primarily in one direction.

Rapid monitoring of changes in water diffusion coefficients during reversible ischemia in cat and rat brain

Changes in the diffusion constant of water during reversible brain ischemia and cardiac arrest were monitored with a 10-s time resolution. Results (five cats, three rats) indicate that these changes are reversible and that the bulk of the changes are not caused by temperature or motion related to brain pulsations and blood flow. The rapid time course of the changes corresponds to the known time course for changes in energy state, signal transduction, and ionic homeostasis.

Monitoring the initial expansion of focal ischaemic changes by diffusion-weighted MRI using a remote controlled method of occlusion

The purpose of the present study was to quantify the early changes which occur on diffusion-weighted imaging following focal cerebral ischaemia. We have developed a method for the remote occlusion of the middle cerebral artery in the rat, allowing early changes to be monitored and images to be acquired before and after ischaemia under identical conditions. Sequential diffusion-weighted images were acquired before and for up to 6 h following the occlusion. The data show that a hyperintense area appears soon after occlusion which expands during the first 2 h. It has previously been shown that the hyperintensity observed after ischaemia is linked to events associated with energy failure. Our findings are in good agreement with previous pathophysiological studies of focal ischaemia, which suggest a growing of the energy depleted area. In conclusion, our data extend the usefulness of diffusion-weighted imaging to the study of the early spatial evolution of the most compromised area after focal ischaemia. Diffusion-weighted imaging may constitute a very powerful tool to investigate pharmacological therapies which may interact with this evolution, both in experimental and clinical studies.

Comparison of diffusion-weighted MRI with changes in cell volume in a rat model of brain injury

The precise mechanisms that underlie acute changes in tissue water diffusion following cerebral ischemia or related insults such as glutamate exposure remain unexplained, but it has been suggested that these may be caused by cell swelling due to water uptake. This study was undertaken to compare the changes observed in diffusion-weighted MR images with changes in the cellular volume measured by electrical impedance in a model of N-methyl-D-aspartate-induced brain injury in perinatal rats. The results show that the temporal course of the intensity changes in the diffusion-weighted images parallelled the progressive shrinkage of the extracellular space measured from the electrical impedance. After administration of the N-methyl-D-aspartate antagonist MK-801 the signal enhancement in the images was reversed, which paralleled the normalization of the extracellular space observed by the impedance measurements. It was estimated that the extracellular space decreased from 24 to 12% while the apparent diffusion coefficient of water decreased from 0.89 x 10(-9) in normal tissue to 0.42 x 10(-9) m2/s in tissue exposed to N-methyl-D-aspartate. These data indicate that changes in tissue water diffusion are related to changes in cell volume.

Water diffusion and acute stroke

The occlusion of the middle cerebral artery was used as an experimental acute stroke model in 30 cats. The diffusion of water was followed by diffusion-sensitized MRI between 1 and 15 h after induction of stroke. It is demonstrated that images representing the trace of the diffusion tensor provide a much more accurate delineation of affected area than images representing the diffusion in one direction only. The reason is that the strong contrast caused by the anisotropy and orientation of myelin fibers is completely removed in the trace of the diffusion tensor. The trace images show a small contrast between white and gray matter. The diffusion coefficient of white matter is decreased in acute stroke to approximately the same extent as gray matter. It is further shown that the average lifetime of water in extra and intracellular space is shorter than 20 ms both for healthy and ischemic tissue indicating that myelin fibers are permeable to water. The anisotropy contrast did not change before or after induction of stroke, nor after sacrifice. Together, these observations are consistent with the view that the changes in water diffusion during acute stroke are directly related to cytotoxic oedema, i.e., to the change in relative volume of intra- and extracellular spaces. Changes in membrane permeability do not appear to contribute significantly to the changes in diffusion.

Early changes in cerebral sodium distribution following ischaemia monitored by 23Na magnetic resonance imaging

23Na magnetic resonance imaging has been used in this preliminary study to investigate early changes in brain sodium signal intensity during and after cerebral ischaemia in a gerbil model. The total sodium signal in selected brain regions decreased between 15 and 30% within 4 min of the onset of ischaemia, and then remained constant throughout the ischaemic period. The same pattern was observed in the eyes. On reperfusion, there was no significant change in the sodium signal over the first 4 min, but by 8 min the signal intensity had returned to or passed through control levels in all regions measured, with the exception of the eyes. These observations are consistent with the loss and resynthesis of ATP as seen in this model, and may be reflecting the redistribution of tissue sodium resulting from energy failure and recovery.

Incidence of apparent restricted diffusion in three different models of cerebral infarction

High speed magnetic resonance imaging (MRI) and short diffusion times are used to investigate the appearance of restricted diffusion in three different models of cerebral infarction. The models are: the middle cerebral artery occlusion (MCAO) model in the rat, the carotid occlusion model in the gerbil, and the Rose Bengal microvascular occlusion model in the rat. All three were investigated for 16 b-values equally spaced between 10 and 1510 s/mm2 using two distinct experiments. In the ct (constant time) experiment, the diffusion time was held constant at 11.7 ms while the b-value was varied with the gradient strength. In the cg (constant gradient) experiment, the gradient strength was held constant and the b-value increased by varying the diffusion time from 4.4 to 11.7 ms. A monoexponential decay of the signal intensity with b-value in the ct experiment accompanied by nonmonoexponential (NME) decay in the cg experiment is indicative of restricted diffusion. As this phenomenon is detectable only at short diffusion times, it cannot be due to restriction by impermeable membranes, and we have thus termed this apparent restriction. For the MCAO model and the carotid occlusion model, apparent restriction was found both inside the infarct territory and in some regions outside it. No definite evidence for restriction was found for the Rose Bengal model, which was, however, only studied from 24 h post-insult.

Apparent diffusion coefficient mapping of experimental focal cerebral ischemia using diffusion-weighted echo-planar imaging

Diffusion-weighted, echo-planar imaging (EPI) was used to map regional changes in the apparent diffusion coefficient (ADC) during experimental focal ischemia in the rat brain following permanent middle cerebral arterial occlusion (MCAO). Sixteen 64 x 64 diffusion-weighted EPIs were acquired in 32 s with successively increasing amplitudes of the diffusion-sensitive gradient pulses. A linear least-squares regression algorithm was used to fit 15 of the 16 two-dimensional matrices, on a pixel-by-pixel basis, to solve for the slope from which the ADC value was calculated. The correlation coefficient of the fit, R2, was used to filter the final ADC maps, and the ADCs were then scaled appropriately to be displayed in a 256 gray level format. Ranges (bins) of 0.05 x 10(-3) mm2/s were then grouped and color coded to qualify and quantify the evolution of ischemia in the MCA territory. The percentage of area in the ischemic and contralateral hemispheres in seven ADC bins were calculated at 30, 60, and 120 min after MCAO for 10 animals and demonstrated a significant increase in ADC bins below 0.45 x 10(-3) mm2/s and a decrease in bins above 0.50 x 10(-3) mm2/s over time. The postmortem infarct area, as measured by TTC staining, was highly correlated with the portion of the ischemic hemisphere falling below ADC values of 0.55 x 10(-3) mm2/s at 2 h after stroke onset. These studies suggest that focally ischemic brain tissue can be quantitatively subdivided according to ADC values and that ADC values below 0.55 x 10(-3) mm2/s 2 h following ischemia highly predict infarction in a rat permanent occlusion stroke model.

Changes in water diffusion and relaxation properties of rat cerebrum during status epilepticus

Diffusion-weighted (DW) imaging has been used to record changes associated with status epilepticus (SE) in rat brain. It was found that the apparent diffusion coefficient (ADC) of water in brain decreased 14-18% during SE, and it fell a further 20-22% when the animals were sacrificed. The transverse decay time constant T2* showed corresponding reductions, but no significant changes were seen in relaxation times T1 or T2 values. Changes in ADC in status epilepticus are similar to those seen in stroke and ischemia but occur under very different conditions of blood flow and metabolism.

Controllable graded cerebral ischaemia in the gerbil: studies of cerebral blood flow and energy metabolism by hydrogen clearance and 31P NMR spectroscopy

A technique for remotely controlling the degree of carotid artery occlusion in the gerbil model of cerebral ischaemia has been developed. The technique relies on manually adjustable nylon snares around the carotid arteries, in conjunction with a computer-based monitoring system, to control the degree of occlusion. This has allowed us to determine the dependence of energy metabolism (as assessed by 31P NMR spectroscopy) on blood flow in greater detail than was possible in our previous studies. Data obtained show that energy changes first appear at flows of 25-30 mL/100 g/min, while at flows below 20 mL/100 g/min there is a major derangement of energy metabolism. The model was used to determine the sensitivity of cerebral energy metabolism to reduced cerebral blood flow under normothermic conditions and in mild hypothermia (30 degrees C). Hypothermia had a protective effect in that energy metabolites were maintained at flows significantly below the normothermic threshold.