1 Department of Anesthesiology and Critical Care Medicine and 2 Department of Neuropathology, Medical Research Institute, Tokyo Medical and Dental University, School of Medicine, 1-5-45 Yushima, Bunkyo-ku, Tokyo 113-8519, Japan*Corresponding author
Accepted for publication: May 14, 2002
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Methods. After the right common carotid artery had been ligated, rats were exsanguinated to maintain a mean arterial pressure of 35, 42, or 50 mm Hg (n=6, each group). Apparent diffusion coefficient (ADC) maps were calculated from DWIs and lesion volume (area) was defined based on ADC values (ADC lesion volume (area)).
Results. ADC lesion volume during exsanguination in the 35 mm Hg group (417 (111) mm3, P<0.01) was significantly larger than in the 42 mm Hg group (87 (84) mm3) and 50 mm Hg group (42 (58) mm3). The low relative cerebral blood flow area, calculated from PWI, was significantly larger during exsanguination in the 35 mm Hg group than in the other groups. ADC lesion volume in the six rats that died within 3 days of the MRI study was significantly larger (median 421 mm3, range 205476 mm3, P<0.005) than in the 12 rats that survived for 3 days (median 26 mm3, range 3517 mm3). Rats with an ADC lesion area over 14 mm2 on the coronal slice including the caudate putamen during exsanguination died within 3 days or revealed a more severe histopathological outcome than those that survived for 3 days.
Conclusions. Incomplete cerebral ischaemia created by the combination of common carotid artery occlusion and exsanguination could be detected by DWI and PWI both qualitatively and quantitatively. The size of the lesion on ADC mapping was found to correlate with mortality and outcome.
Br J Anaesth 2002; 89: 60513
Keywords: brain, ischaemia; measurement techniques, diffusion-weighted magnetic resonance image; measurement techniques, perfusion-weighted magnetic resonance image; rats
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Diffusion and perfusion weighted imaging (DWI and PWI) are two relatively new modalities of magnetic resonance imaging (MRI). DWI has been well established as a reliable non-invasive method for the early detection of cerebral ischaemic stroke.5 6 Even before lesions become apparent on conventional images, DWI can detect abnormalities within minutes after the onset of ischaemia in animal models of stroke.79 With DWI, the signal intensity in ischaemic brain regions appears high because the diffusion coefficient of water is reduced, likely because of the formation of cytotoxic oedema.10 11 This can be quantified by the calculation of apparent diffusion coefficient (ADC) images.12 13 PWI uses gradient echo imaging to follow a fast bolus injection of magnetically susceptible contrast agent. Following ischaemia, perfusion deficits can be observed as decreases in contrast bolus transit.
Three hypotheses formed the basis of the present study. First, it was postulated that intracerebral haemodynamic disturbance can be detected by DWI and PWI, not only qualitatively but also quantitatively. To date, DWI and PWI have been used in experimental animal studies using the middle cerebral artery occlusion model.14 There have been few investigations that use DWI and PWI to document the changes that occur with incomplete cerebral ischaemia. Secondly, if these MRI modalities can detect the spatial evolution of cerebral ischaemic lesions, then perhaps a relationship between lesion volume and mortality could be established. As yet, despite the high mortality rate of perioperative stroke,15 there are no reliable indicators which can predict mortality as a result of cerebral ischaemia. Finally, DWI should be able to predict the severity of cerebral ischaemic injury. In experimental studies, the extent of cerebral ischaemic injury is evaluated histopathologically by examination of specific brain slices. Thus, if the cerebral ischaemia detected by DWI reflects the severity of cerebral injury, a relationship between the ischaemic lesion seen on DWI and histopathologic diagnosis may exist.
![]() |
Methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Animal preparation
Eighteen adult male SpragueDawley rats weighing between 330 and 400 g were used. Anaesthesia was induced with 4% isoflurane in a Plexiglas chamber. The concentration of isoflurane was then reduced to 1.5%, and 1% lidocaine was injected locally for analgesia. Under an operating microscope, the bifurcation of the right common carotid artery was exposed through a midline neck incision and was ligated in duplicate with 3-0 silk surgical thread. A polyethylene catheter (PE-50) was introduced into the right femoral vein for drug infusion. Physiologic saline solution at 4 ml kg1 h1 was administered i.v. throughout the study. Two PE-50 catheters were introduced into femoral arteries on both sides. One arterial catheter was used for continuous arterial pressure monitoring and the other was used for blood withdrawal and arterial blood sampling. After tracheal intubation, each rat was ventilated with 40% oxygen and the ventilatory frequency was controlled to maintain a PaCO2 of approximately 40 mm Hg except during exsanguination. The ventilatory frequency was not changed during exsanguination, and a relative hypocapnoea was allowed to develop because frequent arterial blood sampling, which is necessary to maintain normocapnea, would have made it difficult to maintain the pre-set mean arterial pressure.
Experimental procedure
After placing the animals supine in the probe, anaesthesia was maintained with 1.5% isoflurane in 40% oxygen. The rats body temperature was monitored with a rectal probe, and maintained at 37.5 (0.5)°C using a water-circulating heating pad during the MRI study. Details of the MRI study and the data analysis are given in the Appendix. After animal positioning and shimming, a set of baseline DWIs was acquired.
The animals were randomly allocated to three groups according to mean arterial pressure (MAP) (35, 42, and 50 mm Hg groups, n=6 each) maintained for 30 min by blood withdrawal. Arterial blood was withdrawn through one of the femoral arterial catheters to the pre-set MAP to induce intracerebral haemodynamic disturbance. Fluctuation of ±1 mm Hg from the target arterial pressure was allowed. DWIs were commenced at 10 min after the MAP reached the pre-set value. After 30 min of the ischaemic period, a paramagnetic nuclear magnetic resonance contrast agent (0.5 mmol kg1 gadolinium diethylenetriamine pentaacetic acid; Magnevist, Berlex, USA) was rapidly injected into the right femoral vein. Acquisition of PWIs began approximately 20 s before contrast injection and continued for a total of 80 images. After the completion of PWIs, the withdrawn blood was re-administered i.v. for approximately 10 min. Post-ischaemic DWIs were acquired at 0 and 30 min after completion of the administration of the withdrawn blood.
After extubation, the animals were kept in an oxygen-rich chamber. Their rectal temperature was maintained at 37.5 (0.5)°C using a water-circulating pad for 3 h. They were then released into their home cage.
Arterial blood samples were collected at baseline and 10 min after the completion of the re-infusion of the withdrawn blood. Other arterial blood samples were collected as needed. Samples were analysed on a laboratory blood gas analyzer (STAT profile 5TM gas analyzer, NOVA Biomedical, USA). Metabolic acidosis was treated by the i.v. administration of sodium bicarbonate.
Definition of lesion area (volume) and low perfusion area
The ischaemic lesion area and the low perfusion area were defined arbitrarily as the pixels in which the ADC, relative cerebral blood volume (rCBV), and relative cerebral blood flow (rCBF) value decreased to less than a predetermined threshold on each map. Calculation of ADC, rCBV, and rCBF are given in the Appendix. The threshold was set at 500x106 mm2 s1 on ADC map,16 17 50% of the mean rCBV and rCBF of the left (contralateral) hemisphere on rCBV map and rCBF map, respectively. The lesion area on the ADC map and the low perfusion area on the rCBV and the rCBF map (ADC lesion area, low rCBV area, and low rCBF area, respectively) were calculated by multiplying the number of these pixels by the area of one pixel. Lesion volume of ischaemic injury was determined by integration of the ADC lesion areas for the nine slices (ADC lesion volume).
Histopathology
Neurohistopathology was rated in those rats that survived for 3 days following the MRI study. After each rat was anaesthetized with isoflurane, the chest was opened, and the brain was perfused by transcardial perfusion with 20 ml of isotonic saline followed by 20 ml of 10% buffered formalin. The brain was then removed and stored in 10% formalin for 1 week. The brain was cut into coronal blocks and embedded in paraffin, and 6-µm sections were sliced and mounted on slides. The slides were stained with hematoxylin and eosin and examined using light microscopy. The examiner was blinded to the results of the MRI study. Neuronal histopathology was evaluated in the coronal section at the level of the caudate nucleus (8.69.7 mm anterior to the interaural line18). The ischaemic damage of this section had been evaluated by ADC map, rCBV map, and rCBF map. Neuronal damage was graded on a five-point scale with the following markers:19 0=no observable neuronal death; 1=scattered neuronal death; 2=small focal damage in caudate and cortical areas; 3=large infarcts involving 50% of ischaemic hemisphere; and 4=total hemisphere infarct.
Statistical analysis
MAP, arterial blood gas data, blood glucose concentration, rectal temperature, and ADC lesion volumes were analysed using two-way repeated measures analysis of variance followed by the Tukey multiple comparison test. Baseline, 10 min after MAP reached the pre-set value, and 10 min after the completion of blood reinfusion, MAP and rectal temperatures were compared. Low rCBV area and low rCBF area were compared among the groups using one-way analysis of variance followed by the Tukey multiple comparison test. The relationships between ADC lesion area and low rCBV area and between ADC lesion area and low rCBF area on the slice including the caudate putamen were analysed using linear fitting. ADC lesion volume was compared between surviving and dead animals using the MannWhitney U test because DAgostino-Pearson test revealed that the spread and skew of data made parametric statistics inappropriate. Statistical significance was established at the P<0.05 level.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
MAP in the 35 mm Hg group was significantly lower than in the 42 mm Hg group (P<0.05) and in the 50 mm Hg group (P<0.01) at 10 min after the completion of the blood reinfusion (Table 1). Both arterial pH and base excess (BE) were also significantly lower in the 35 mm Hg group than in the other two groups during the reinfusion period (P<0.01). Blood glucose concentration in the 35 mm Hg group was lower than in the 50 mm Hg group (P<0.01) at re-infusion, but no significant difference was found between the 35 mm Hg and 42 mm Hg groups. Twelve rats (6, 5, and 1 rats in the 35, 42, and 50 mm Hg groups, respectively) needed i.v. sodium bicarbonate. Their arterial pH and BE were confirmed to be within normal limits at the end of the MRI study (pH 7.40 (0.05), BE 1.4 (1.7)).
|
|
|
|
|
|
|
|
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
By combining DWI and PWI in this study, we were able to show that there were significant linear relationships between the ADC lesion area and low rCBF area, and between ADC lesion area and low rCBV area on the slice including the caudate putamen. The first relationship suggests that these ischaemic lesions caused by low cerebral perfusion enlarged as the exsanguination evolved. Furthermore, our findings suggest that ADC mapping could be used not only to detect intracerebral haemodynamic disturbances qualitatively, but also to quantify them. In fact, with the ADC lesion volume (area) defined using a predetermined threshold,16 17 the spatial evolution of the cerebral ischaemic injury was clearly demonstrated as the exsanguination evolved. The linear relationship between the ADC lesion area and low rCBV area suggests that cerebral vasodilation had become maximal and that the vessel had already begun collapsing when the ischaemic lesion was detected on the ADC map. The different behaviour against cerebral perfusion pressure (CPP) between rCBV and rCBF2 20 may partially explain smaller low rCBV area than low rCBF area in the representative rat (Fig. 1).
In our study, we varied target arterial pressure (35, 42, or 50 mm Hg) to create various degrees of cerebral ischaemic injury. As expected, lesion volume based on the ADC map showed wide variation among the animals. When we compared the results of those animals that died within 3 days with those that survived, there was a significant difference in ADC lesion volume, with surviving animals having smaller lesions. It would appear that ADC lesion volume reflects the spatial evolution of cerebral ischaemia and thus correlates with the severity of the cerebral injury. Large lesions may have been the crucial factor that resulted in the death of the animals.
ADC lesion volume after blood re-infusion was also compared between those animals that died within 3 days and those that survived. Regions with low ADC values after reperfusion are likely to infarct, because they reflect tissues with irreversible ischaemic damage. This has been documented in the transient middle cerebral artery occlusion model.79 In 10 of the 12 animals that survived, the histopathological analyses demonstrated neuronal damage but not pannecrosis. The fact that these animals also showed very small ADC lesions area (less than 0.6 mm2) on the slice including the caudate putamen at 30 min after blood re-infusion supports the assumption that the cerebral ischaemia was not severe in such animals. The cerebral injury was mostly reversible, although damaged neurones were seen predominantly in the border zone area. When ADC lesion areas on the specific coronal slice were compared to the neuronal damage scores, there seemed to be a threshold of ADC lesion area (14 mm2) over which a more severe outcome (death or neuronal damage score 2) was likely. The relationship between the ADC lesion area and the neuronal damage score observed in the present study suggests that ADC mapping may be useful in predicting histopathological outcome in animals that survive intracerebral haemodynamic disturbance.
It was also observed that a significantly lower MAP in the 35 mm Hg group occurred after re-infusion of the withdrawn blood (Table 1). This may in part be explained by the occurrence of a more severe metabolic acidosis in this group. However, as the acidbase abnormality returned to near normal at the end of the MRI study, the effect of transient metabolic acidosis on outcome may be small. It was also noted that a difference in blood glucose concentration occurred among the groups following the blood re-infusion (Table 1). This may be explained by the fact that significantly more blood was withdrawn in the 35 mm Hg group. Blood glucose may have been partly glycolysed by erythrocytes21 and the re-infused blood may have diluted the existing blood glucose of the rats in the 35 mm Hg group. However, as the outcome of the animals in the 35 mm Hg group was the worst despite the lower blood glucose concentration, the impact on outcome of the difference in blood glucose concentration may not be substantial.
Presently, MRI systems are being developed both for clinical use and for intraoperative use during neurosurgical procedures performed under general anaesthesia.2224 Already, DWI can be performed to determine whether cerebral ischaemia or infarction has occurred following completion of the procedure before the patient leaves the operating room.25 The results of the present study suggest that intraoperative DWI and PWI could potentially be used to detect early abnormalities of intracerebral haemodynamics and thus alert the surgeon to the occurrence of cerebral ischaemia as a result of compromised CBF. This would facilitate the introduction of brain protective measures, such as hypothermia. In addition, intraoperative MRI could potentially become a decision-making tool for the neurosurgeon. For example, the decision as to whether or not to insert a shunt during carotid endarterectomy could depend on the results of intraoperative MRI.
However, intraoperative DWI and PWI techniques as they stand are not yet ready for clinical detection of intraoperative cerebral ischaemia or prediction of mortality and outcome in patients. The development process towards clinical application will require several problems to be overcome. First, the cost of the complete system of intraoperative MRI must be markedly reduced before widespread use becomes possible. MRI-compatible surgical instruments including titanium forceps, titanium scissors, and titanium retractors are required, and anaesthesia machines and monitors should also be MRI compatible. Secondly, the size of the MRI scanner must be reduced to allow surgeons to obtain optimal access to patients. However, this should not be achieved at the expense of homogeneity of the magnetic field or function of the MRI system. Lastly, the abilities of DWI and PWI to detect incomplete cerebral ischaemia and to predict mortality and outcome must be validated in humans.
In conclusion, DWI and PWI both qualitatively and quantitatively delineated intracerebral haemodynamic disturbances. They clearly demonstrate the spatial evolution of the cerebral ischaemic injury as the exsanguination evolved. During incomplete cerebral ischaemia, the measurement of lesion volume as detected by ADC mapping predicted mortality. Larger ischaemic lesions detected by ADC map on the coronal section, which included the caudate putamen were associated with a more severe outcome (death or small focal damage). In the future, intraoperative DWI and PWI have the potential to facilitate the early detection of intracerebral haemodynamic disturbances, thus allowing a timely and effective response.
![]() |
Acknowledgement |
---|
![]() |
Appendix |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
DWI
Spin-echo DWIs (TR ms/TE ms=2000/64, 40-mm field of view, 128x128 matrix, 2-mm section thickness) consisted of nine coronal slices covering most of the rat brain. Diffusion weighting was applied along all three axes (b=0 and 1100 s mm2). A single DWI required an acquisition time of approximately 18 min. DWI with a b value of 0 s mm2 corresponds to a T2-weighted image.
PWI
PWIs were acquired using a gradient echo imaging sequence (TR ms/TE ms=14/8, 10° flip angle, 40-mm field of view, 64x64 matrix, 2-mm section thickness, 1 image s1) to follow a fast bolus injection of magnetic susceptibility contrast agent. The single PWI slice was chosen to coincide with one of the DWI slices and the histopathology examination.
MRI data analysis
Pairs of DWIs with two b values were used to generate maps of the ADC. ADC maps were calculated using the standard equation:26
ADC=ln (S0/S1)/(b1b0)
where S0 and S1 are the signals of the two DWI scans and b0 and b1 are 0 and 1100 s mm2.
Signal intensity of PWIs was converted to the change in the T2 rate, R2*(t), using the standard equation:27 28
R2*(t)=ln [S(t)/S(0)]/TE
where S(t) is the signal intensity at time t during passage of the contrast agent, S(0) is the baseline value of the pre-contrast signal intensity, and TE is echo time. As R2*(t) is proportional to the contrast agent concentrationtime curves, it can be used to estimate perfusion parameters.
Assessment of rCBV can be calculated by integrating the R2*(t) time curve with respect to time. The behaviour of the physiological time series data as a contrast agent passes through the vascular bed is closely represented by the gamma-variate curve.29 To generate rCBV data without the effects of contrast recirculation, the area under the fitted gamma-variate curve was integrated.
An estimate of the vascular transit time (VTT) was obtained by subtracting the time of arrival of the contrast agent in the cerebral tissue from the first moment of the fitted gamma-variate curve. The estimates of VTT and rCBV were used to calculate the rCBF by using the equation of the central volume principle:30
rCBF=rCBV/VTT
where rCBF was determined for each pixel.
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
2 Powers WJ. Cerebral hemodynamics in ischemic cerebrovascular disease. Ann Neurol 1991; 29: 231240[ISI][Medline]
3 Derdeyn CP, Grubb RL Jr, Powers WJ. Cerebral hemodynamic impairment: methods of measurement and association with stroke risk. Neurology 1999; 53: 251259
4 Stamou SC, Hill PC, Dangas G, et al. Stroke after coronary artery bypass: incidence, predictors, and clinical outcome. Stroke 2001; 32: 15081513
5 Rordorf G, Koroshetz WJ, Copen WA, et al. Regional ischemia and ischemic injury in patients with acute middle cerebral artery stroke as defined by early diffusion-weighted and perfusion-weighted MRI. Stroke 1998; 29: 939943
6 van Everdingen KJ, van der Grond J, Kappelle LJ, Ramos LMP, Mali WPTM. Diffusion-weighted magnetic resonance imaging in acute stroke. Stroke 1998; 29: 17831790
7 Minematsu K, Fisher M, Li L, et al. Effects of a novel NMDA antagonist on experimental stroke rapidly and quantitatively assessed by diffusion-weighted MRI. Neurology 1993; 43: 397403[Abstract]
8 Minematsu K, Fisher M, Li L, Sotak CH. Diffusion and perfusion magnetic resonance imaging studies to evaluate a noncompetitive N-methyl-D-aspartate antagonist and reperfusion in experimental stroke in rats. Stroke 1993; 24: 20742081[Abstract]
9 Lo EH, Matsumoto K, Pierce AR, Garrido L, Luttinger D. Pharmacologic reversal of acute changes in diffusion-weighted magnetic resonance imaging in focal cerebral ischemia. J Cereb Blood Flow Metab 1994; 14: 597603[ISI][Medline]
10 Fisher M, Sotak CH, Minematsu K, Li L. New magnetic resonance techniques for evaluating cerebrovascular disease. Ann Neurol 1992; 32: 115122[ISI][Medline]
11 Benveniste H, Hedlund LW, Johnson A. Mechanism of detection of acute cerebral ischemia in rats by diffusion-weighted magnetic resonance microscopy. Stroke 1992; 23: 746754[Abstract]
12 Bihan DL, Breton E, Lallemand D, Grenier P, Cabanis E, Laval-Jeantet M. MR imaging of intravoxel incoherent motions: application to diffusion and perfusion in neurologic disorders. Radiology 1986; 161: 401407[Abstract]
13 Bihan DL, Moonen CTW, van Zijl PCM, Pekar J, DesPres D. Measuring random microscopic motion of water in tissues with MR imaging: a cat brain study. J Comput Assist Tomogr 1991; 15: 1925[ISI][Medline]
14 Li F, Liu KF, Silva MD, et al. Transient and permanent resolution of ischemic lesions on diffusion-weighted imaging after brief periods of focal ischemia in rats: correlation with histopathology. Stroke 2000; 31: 946954
15 Parikh S, Cohen JR. Perioperative stroke after general surgical procedures. NY State J Med 1993; 93: 162165
16 Ishikawa S, Yokoyama K, Kuroiwa T, Makita K. Evolution of cerebral ischaemia induced by thromboembolism in rats detected by early sequential MR imaging. Br J Anaesth 2001; 87: 469476
17 Morikawa S, Inubushi T, Ishii H, Nakasu Y. Effects of blood sugar level on rat transient focal brain ischemia consecutively observed by diffusion-weighted EPI and 1H echo planar spectroscopic imaging. Magn Reson Med 1999; 42: 895902[ISI][Medline]
18 Paxinos G, Watson C. The Rat Brain in Stereotaxic Coordinates, 4th Edn. New York, NY: Academic Press Inc., 1982
19 Hoffman WE, Braucher E, Pelligrino DA, et al. Brain lactate and neurologic outcome following incomplete ischemia in fasted, nonfasted, and glucose-loaded rats. Anesthesiology 1990; 72: 10451050[ISI][Medline]
20 Sette G, Baron JC, Mazoyer B, Levasseur M, Pappata S, Crouzel C. Local brain haemodynamics and oxygen metabolism in cerebrovascular disease: positron emission tomography. Brain 1989; 112: 931951[Abstract]
21 Glycolysis and the catabolism of hexoses. In: Lehninger AL, Nelson DL, Cox MM, eds. Principles of Biochemistry, 2nd Edn. New York, NY: Worth Publishers, 1993: 400445
22 Black PMcL, Moriarty T, Alexander E III, et al. Development and implementation of intraoperative magnetic resonance imaging and its neurosurgical applications. Neurosurgery 1997; 41: 831845[ISI][Medline]
23 Hall WA, Martin AJ, Liu H, Nussbaum ES, Maxwell RE, Truwit CL. Brain biopsy using high-field strength interventional magnetic resonance imaging. Neurosurgery 1999; 44: 807814[ISI][Medline]
24 Steinmeier R, Fahlbusch R, Ganslandt O, et al. Intraoperative magnetic resonance imaging with the Magnetom Open scanner: concepts, neurosurgical indications, and proceduresa preliminary report. Neurosurgery 1998; 43: 739748[ISI][Medline]
25 Hall WA, Liu H, Martin AJ, Pozza CH, Maxwell RE, Truwit CL. Safety, efficacy, and functionality of high-field strength interventional magnetic resonance imaging for neurosurgery. Neurosurgery 2000; 46: 632642
26 Le Bihan D, Breton E, Lallemand D, et al. Separation of diffusion and perfusion in intravoxel incoherent motion MR imaging. Radiology 1988; 168: 497505[Abstract]
27 Belliveau JW, Rosen BR, Kantor HL, et al. Functional imaging by susceptibility-contrast NMR. Magn Reson Med 1990; 14: 538546[ISI][Medline]
28 Hamberg LM, Macfarlane R, Tasdemiroglu E, et al. Measurement of cerebrovascular changes in cats after transient ischemia using dynamic magnetic resonance imaging. Stroke 1993; 24: 444451[Abstract]
29 Thompson HK, Starmer CF, Whalen RE, McIntosh HD. Indicator transit time considered as a gamma variate. Circ Res 1964; 14: 502515[ISI]
30 Meier P, Zierler KL. On the theory of the indicator-dilution method for measurement of blood flow and volume. J Appl Physiol 1954; 6: 731744[ISI]