Neuroprotective Effect of Delayed Moderate Hypothermia After Focal Cerebral Ischemia
An MRI Study
Background and Purpose— In contrast to early hypothermia, the effects of delayed hypothermia in focal cerebral ischemia have not been widely addressed. We examined the influence of delayed hypothermia on secondary ischemic injury, MRI lesion size, and neurological outcome after transient focal cerebral ischemia in a rat model.
Methods— Rats (n=30) were subjected to transient middle cerebral artery occlusion (MCAO, 120 minutes) by use of the intraluminal filament model. Animals of the treatment group (n=12) were exposed to whole-body hypothermia of 33°C for 5 hours starting 3 hours after MCAO, whereas the control group (n=18) was kept at 37°C throughout the whole experiment. The normothermia- and hypothermia-treated animals were investigated daily by using the Menzies neurological score. Serial MRI was performed 1, 3, and 6 hours after MCAO and on days 1, 2, 3, and 5. After the final MRI scan, the rats were euthanized, and brain slices were stained by 2,3,5-triphenyltetrazolium chloride.
Results— Delayed hypothermia resulted in a significant increase of survival rate and a significant improvement of the Menzies score. Moreover, a significant decrease in the extent of hyperintense volumes in T2-weighted scans and a reduction of cerebral edema as calculated from T2-weighted scans throughout the examination period were obvious. The extent of cerebral infarct volume and cerebral brain edema examined by MRI was consistent with 2,3,5-triphenyltetrazolium chloride staining.
Conclusions— Our results suggest that even delayed postischemic hypothermia can reduce the extent of infarct volume and brain edema after transient focal cerebral ischemia.
Hypothermia has been shown to be neuroprotective in animal models of focal cerebral ischemia,1–10⇓⇓⇓⇓⇓⇓⇓⇓⇓ provided that it is induced soon after the onset of neurological symptoms and maintained for an adequately long time period.2,3,9,10⇓⇓⇓ Nevertheless, the applicability of these results to stroke patients is limited, because these patients cannot receive therapeutic hypothermia directly after the ischemic injury. The time window used in clinical studies is so long11,12⇓ that the initial cerebral ischemia is potentially aggravated by secondary ischemic damage. This has been shown to increase both ischemic damage13 and mortality7 compared with permanent occlusion.13 The underlying pathophysiological mechanisms include hemodynamic disturbances,10 inflammation,14 and breakdown of the blood-brain barrier.10 Although the effects of hypothermia initiated directly with the onset of reperfusion have been previously addressed,1,3,6,7⇓⇓⇓ the effects of hypothermic treatment initiated within the reperfusion period remain uncertain.
MRI, including diffusion-weighted, T2-weighted, and perfusion-weighted sequences, is a powerful tool for the investigation of experimental cerebral ischemia as well as the time course of therapeutical approaches.15–17⇓⇓ The major advantage of MRI over histological examinations is the noninvasiveness, which allows MRI to be used as a monitoring method. For that purpose, MRI has been used to evaluate therapies with neurotrophins16 or N-methyl-d-aspartate antagonists,17 whereas MRI studies investigating the effects of hypothermia are rare.18,19⇓
Hypothermia induced at stroke onset has already been studied, whereas the characteristics of the delayed application of hypothermia have not been evaluated by MRI.
We undertook the present study to investigate the effects of delayed moderate hypothermia on (1) lesion size and brain swelling as evaluated on MRI, (2) functional outcome, and (3) mortality in transient focal cerebral ischemia. Additionally, we examined the persistence of potential neuroprotective effects in the subacute stages of cerebral infarction.
Materials and Methods
Male Wistar rats weighting 280 to 320 g were used in the present study. All animals were allowed access to food and water ad libitum and were randomly assigned to 1 of the following experimental groups (see below) before surgery. The numbers of animals suffering premature death (ie, before day 5) were added to both groups without randomization.
Group 1 (normothermic control group) was subjected to transient middle cerebral artery (MCA) occlusion (MCAO) for 120 minutes. Normothermia of 37°C body core temperature was maintained for 8 hours.
Group 2 (hypothermic group) was subjected to transient MCAO for 120 minutes and normothermia of 37°C body core temperature for 120 minutes of the MCAO and a further 60 minutes of the initial reperfusion period, followed by hypothermia of 33°C body core temperature for 5 hours.
Animals from both groups were further evaluated for a total of 5 days after MCAO.
Anesthesia was induced with an intraperitoneal injection of 100 mg/kg ketamine hydrochloride (Wirtschaftsgenossenschaft deutscher Tierärzte) and, if necessary, maintained with a further 25 mg/kg. PE-50 polyethylene tubing was inserted into the right femoral artery for the monitoring of mean arterial blood pressure and arterial blood gas analysis. Catheter placement in the right femoral vein was performed with PE-50 tubing for injection of contrast medium for perfusion-weighted imaging (PWI). Monitoring of physiological parameters (arterial blood gases, plasma glucose, and body temperature) was done at day 1 for a total of 9 hours. This period included surgery, induction of hypothermia, rewarming, and MRI scans at 1, 3, and 6 hours. For pain relief, the surgical wounds were anesthetized by using 2% lidocaine (0.1 mL). For MRI scans from days 2 to 5, anesthesia was induced by a single intraperitoneal injection of 50 mg/kg ketamine hydrochloride. Body temperature was measured during anesthesia, and MRI scans were performed without additional surgery.
The intraluminal suture model was used for induction of focal cerebral ischemia, as formerly described by Longa et al.20 Briefly, the right common carotid, internal carotid, and external carotid arteries were exposed through a midline incision of the neck. Further dissection identified the origin of the pterygopalatine artery, which was ligated by a 6-0 silk suture. After careful separation from the adjacent vagal nerve, the proximal parts of the right common and external carotid arteries were ligated with 5-0 surgical sutures. A 4-0 monofilament silicone-coated nylon suture was inserted into the distal right common carotid artery and gently advanced 18 to 19 mm from the external carotid artery into the internal carotid artery until the tip occluded the origin of the MCA. For permanent cerebral ischemia, the silicone-coated suture remained in the MCA, whereas cerebral reperfusion was achieved by removal of the occluder filament 2 hours later.
Temperature Control and Adjustment
Body core temperature was continuously monitored 30 minutes before vessel occlusion and during the following 9 hours of the experiment with a thermostatically controlled heating pad (Föhr Medical Instruments). For normothermic animals, temperature was adjusted to 37°C. Hypothermia was induced with ice and by spraying alcohol on the animal’s surface. The target temperature of 33°C was reached after 10 minutes and was maintained by readjusting the heating pad to 33°C. Experimental data indicate that the body core temperature is correlated with pericranial temperature during normothermia and hypothermia.21 For MRI at days 2, 3, and 5, animals were anesthetized repeatedly. After the MRI scan, body temperature was controlled until the animals awoke.
The animals were examined in a 2.35-T scanner (Biospec 24/40, Bruker Medizintechnik). An actively shielded gradient coil with a 120-cm inner diameter was used. This coil was driven by the standard 150-V/100-A gradient power supply. In this configuration, 180 mT/m could be reached in 180 ms. As a radiofrequency coil, a home-built birdcage resonator with a 40-mm inner diameter was used.
Magnetic resonance examination was performed 60 minutes, 3 hours, 6 hours, 1 day, 2 days, and 5 days after MCAO. The animals were kept in the same position throughout imaging. The MRI protocol entailed diffusion-weighted imaging (DWI), which was achieved by using a spin-echo echo-planar imaging sequence (repetition time [TR] 3 seconds, echo time [TE] 67.7 ms, number of averages [NA] 3, 8 different b values from 0 to 1260 s/mm2, diffusion time 50 ms, duration of diffusion gradient 5 ms, field of view [FOV] 4 cm×4 cm, matrix 128×64, 6 slices, and slice thickness 2 mm), T2-weighted imaging, which was achieved by using a RARE sequence (TR 3 seconds, TE 87 ms, NA 4, FOV 4 cm×4 cm, matrix 256×256, 6 slices, and slice thickness 2 mm), and PWI, which was achieved by using a gradient-echo echo-planar imaging sequence (TR 1 second, TE 12 ms, NA 1, FOV 4 cm×4 cm, matrix 128×64, 4 slices, slice thickness 2 mm, number of repetitions 20, and time resolution 1 second). For PWI, a bolus of 0.5 mmol/kg body wt Gd-DTPA (Magnevist, Schering AG) was injected before acquisition of the fifth image data set.
Image data were subsequently transferred to a SUN Sparcstation 10 (SUN Microsystems). Infarct volume was calculated from the T2-weighted images,15 and the apparent diffusion coefficient was calculated from the diffusion-weighted images, as described in Heiland and Sartor.22 A side-by-side difference of apparent diffusion coefficient value from homologous pixels (ie, the ischemic and normal hemispheres that best define the ischemic lesion volume in vivo) of 29%, which is highly correlated with postmortem infarct volume, was used to define abnormal ischemic pixels.22 Relative regional cerebral blood volume (rrCBV) and the relative mean transit time (rMTT) were calculated from the perfusion-weighted MRI data, as described by Heiland et al.23 These parameters were investigated for 5 different regions of interest (ROIs, 1.5-mm diameter) in each hemisphere, including the thalamus, basal ganglia, temporal cortex, sensory cortex, and cingulate gyrus.
MRI protocol was used as follows: on day 1, PWI, DWI, and T2-weighted imaging 1, 3, and 6 hours after MCAO; on days 2 to 5, DWI and T2-weighted imaging.
MRI scans from animals that died before day 5 were excluded from analysis.
The animals were daily examined for body weight and body temperature. The neurological score was assessed according to the description of Menzies et al24 (the Menzies score is a rating scale that ranges from 0, indicating no deficit, to 5, indicating death).24 This score has been shown to be correlated with infarct volume in different studies.16,24⇓ On day 5, the brains were removed, inspected for subarachnoid hemorrhage, and then coronally sectioned into five 2-mm coronal slices. These slices were incubated for 30 minutes in a 2% solution of 2,3,5-triphenyltetrazolium chloride (TTC) at 37°C and fixed by immersion in a 10% buffered formalin solution. TTC-stained brain sections were photographed by using a charge-coupled device camera (EDC-1000HR Computer Camera, Electrim Corp; slices 1 to 5=bregma coordinates 2.4, 0.4, −1.6, −3.6, and −5.4, respectively).16,25⇓ Infarct volumes were calculated with the observers blinded to the treatment given, and the infarct size was quantified by using an image-processing software package (Bio Scan OPTIMAS) in each of the 5 slices. Infarct areas on each slice were then summed and multiplied by slice thickness to calculate direct infarct volumes.16 To compensate for the effect of brain edema, the corrected infarct volume was calculated as previously described.16 In brief, corrected infarct area was assessed as left hemisphere area minus right hemisphere area minus infarct area.16
Brain edema was assessed by dividing the ipsilateral through contralateral hemisphere.16 Animals that died before day 5 were excluded from the assessment of neurological score, body weight, and TTC staining but were investigated for the probable cause of death (eg, large hemispheric infarction or subarachnoid hemorrhage).
Normally distributed values are expressed as mean±SD and were compared by using the paired t test. ANOVA was used for multiple group comparison. A χ2 test was used for comparison of frequencies. After all the data were acquired, the randomization code was broken. ANOVA and subsequent post hoc Fisher protected least significant difference tests were used to determine the statistical significance of differences in continuous variables, such as physiological parameters and diffusion-, perfusion-, and T2-derived infarct volumes. The t test was used for comparison of postmortem infarct volumes. The Mann-Whitney U test was used for comparison of nonparametric data. Significance was declared at a value of P<0.05.
Physiological parameters including body weight, arterial blood gases, and blood pressure did not differ statistically before the surgical procedure or throughout the complete experiment (see Table).
Two animals in the normothermic group and 1 in the hypothermic group died within the first 6 hours of subarachnoidal hemorrhage. During the first 48 hours, 8 normothermic animals and 2 hypothermic animals died (P<0.05, χ2 test) and were excluded from analysis by MRI and TTC staining. Removal of the brains identified a total infarction of the MCA region with a large cerebral edema in all cases as cause of death.
Twenty-four hours after MCAO, the neurological deficit measured by the Menzies score was 2.96±0.31 for the hypothermic group and 3.35±0.41 for the normothermic group (P=NS). During the following days, the neurological deficit decreased continuously. At day 5, the neurological score was 2.51±0.21 in the hypothermic animals compared with 3.02±0.13 in the normothermic animals (P<0.05).
Cerebral Infarct Volume and Cerebral Edema on TTC Staining
At day 5, the infarct volume corrected for cerebral edema was 215.4±23.8 mm3 for the normothermic group and was significantly higher than that in the hypothermic group, with an infarct volume of 159.3±21.9 mm3 (P<0.05). Cerebral edema at day 5 was significantly different: 13.0±3.6% in the normothermic group versus 7.9±2.5% in the hypothermic group (P<0.05).
Magnetic Resonance Imaging
No hyperintense areas were observed in either group on T2-weighted imaging 60 minutes after the onset of ischemia. After 180 minutes, the volume of the hyperintense area was 155±37 versus 147±32 mm3 in the normothermic and hypothermic groups, respectively (P=NS). Six hours after occlusion, the hyperintense area was larger in the normothermic compared with the hypothermic group (216±27 versus 179±36 mm3, respectively; P<0.05). Animals treated with hypothermia showed significantly smaller T2 lesions on days 1, 2, and 5 (P<0.05, Figures 1 and 2⇓). On day 5, hyperintense regions were significantly larger in the normothermic group compared with the hypothermic group (221±42 versus 169±38 mm3, respectively; P<0.05). Significantly less cerebral swelling was observed in the hypothermic compared with the normothermic group on the first, second, and fifth days (Figures 2 and 3⇓).
Volumes of hyperintense regions measured by DWI were significantly lower in the hypothermic compared with the normothermic group. This difference was statistically significant throughout the observation period (Figure 4).
Cerebral perfusion as measured by PWI was decreased after the induction of cerebral ischemia in the regions typically supplied by the MCA. Therefore, rrCBV in the ischemic hemisphere was <50% of the corresponding ROIs in the nonischemic hemisphere including thalamus, basal ganglia, temporal cortex, and sensory cortex. rMTT was prolonged in both groups for the named ROIs in the ischemic hemisphere, as expected. rrCBF and rMTT were not significantly different between the hypothermic and normothermic groups. After reperfusion 3 and 6 hours after MCAO, rrCBV was increased in all regions supplied by the MCA, and rMTT was decreased. There were no significant differences between PWI at 3 and 6 hours after MCAO between both groups and within the group itself. However, the values of PWI did not return to normal after reperfusion. Compared with the nonischemic hemisphere, rrCBV was diminished and rMTT was prolonged. There was a nonsignificant trend toward a better perfusion in cortical regions of the ischemic hemisphere in the hypothermic group.
Our results suggest that delayed moderate hypothermia can persistently decrease the secondary ischemic injury after focal cerebral ischemia: the volumes of cerebral infarct and edema, as evaluated by serial MRI and TTC staining, were significantly lower over the complete observation period of 5 days. The initial growth of lesions in DWI and T2 was less pronounced and remained stable below the lesions of normothermic animals throughout 5 days. Moreover, hypothermia significantly increased the survival rate and was associated with better neurological scores.
Experimental data from transient focal cerebral ischemia indicate that delayed reperfusion contributes to ischemic injury and increases the size of the ischemic lesion. This was shown for the Long-Evans rat strain by Aronowski et al,13 who observed a larger volume of cerebral infarct and edema after transient MCAO of 120 minutes compared with permanent ischemia. This finding was further supported in an MRI study in Wistar rats26 that showed a reduction in DWI hyperintensity after transient MCAO of 45 minutes but an increase after MCAO of 120 minutes. Other experiments showed that mortality increased to 45% if reperfusion was initiated 120 minutes after MCAO.7 The pathophysiological mechanism of injury from delayed reperfusion itself is not yet fully understood but may include a breakdown of the blood-brain barrier,10 an increase of postischemic edema,10 a secondary increase of excitatory amino acids,27 an inflammatory reaction,14 and altered microvascular permeability and integrity.10,27–29⇓⇓⇓ From the present study, it can only be postulated that hypothermia can reduce these effects of delayed reperfusion, because only infarct volume, cerebral edema, and MRI parameters were investigated. However, some experimental data indicate that some of the above-mentioned mechanisms are counteracted by hypothermia.2,9,10,29⇓⇓⇓
Only a few studies have examined the influence of hypothermia on delayed reperfusion after ischemic injury.1,3,6,7⇓⇓⇓ Hypothermia applied directly after reperfusion has been reported to be neuroprotective when it is maintained for 3 to 22 hours.1,6,7⇓⇓ Conflicting data exist regarding hypothermia induced later in the reperfusion period: mild hypothermia leads to significant reductions of cerebral cortical infarct volume when it is induced directly after a 120-minute MCAO but not after an additional reperfusion period of 1 hour (Maier et al7). The study of Maier et al is not in accordance with our findings; we were able to demonstrate a neuroprotective effect of moderate hypothermia induced 3 hours after MCAO and 1 hour after reperfusion. This discrepancy is probably due to the longer duration of hypothermia applied in the present study compared with the study conducted by Maier et al, inasmuch as extending the application of cooling has been shown to enhance its neuroprotective effectiveness, at least during intraischemic and postischemic application.1,3,6,7⇓⇓⇓
Serial MRI, including T2- and diffusion-weighted sequences, was used to quantify the effects of delayed hypothermia on the initiation and persistence of possible neuroprotective effects. To our knowledge, only 2 previous studies have used MRI to describe the effect of moderate hypothermia in a rat model of transient cerebral ischemia.18,19⇓ However, in contrast to the present study, Jiang et al18 and Yenari et al19 investigated moderate intraischemic hypothermia in model of transient MCAO. In contrast, we examined the postischemic application of moderate hypothermia, thus using a model incorporating the possible effects of secondary ischemic damage.
Assessment of T2 and DWI revealed smaller areas of hyperintensity in hypothermic compared with normothermic animals, which reached statistical significance 6 hours after MCAO and remained significantly different throughout the observation period. Because the major effect of hypothermia on lesion volume was observed during the first hours, an immediate fashion of neuroprotection can be suggested that is similar to that seen for N-methyl-d-aspartate antagonists.17 In contrast, neurotrophins show a more delayed pattern of neuroprotection, as observed by MRI.16
DWI is very sensitive for the acute phase of cerebral ischemia, whereas it tends to overestimate the ischemic lesion in the chronic phase after ischemia because of changes in the diffusion status of the tissue. In these cases, T2-weighted imaging should be performed.15,16,30⇓⇓ Our data show a close correlation between T2-weighted imaging and DWI in the chronic phase of ischemic injury. Differences in apparent lesion volumes of DWI remained stable between both groups, thus not indicating secondary ischemic damage at later time periods in hypothermia.
PWI can be used to evaluate cerebral perfusion. Although PWI does not precisely reflect regional cerebral blood flow, it can provide information regarding microvascular patency.26,31⇓ Data from PWI in experimental focal cerebral and hypothermia are controversial. In accordance with our results, Yenari et al19 showed no significant differences between normothermic and hypothermic animals during MCAO and reperfusion. Still, PWI delay was severe during occlusion but disappeared in the reperfusion period. In contrast, using the arterial spin-labeling technique, Jiang et al18 reported that regional cerebral blood flow recovered under hypothermia compared with normothermia. Laser-Doppler flowmetry in transient MCAO revealed initial hyperperfusion followed by hypoperfusion, but this effect was blunted by hypothermia.10 Inasmuch as Yenari et al19 found that PWI was not altered by hypothermic treatment after 48 hours up to 7 days, we do not suggest alteration of CBF being a main factor for the neuroprotective effect of hypothermia.
However, the present study has some limitations. Although compared with normothermia, hypothermia caused a stable reduction of cerebral infarct volume through the observation period, we cannot exclude the possibility that this effect is only transient and vanishes after a longer time period. The data concerning the long-term neuroprotective effect of postischemic hypothermia are inconclusive for global as well as focal cerebral ischemia.7,29,32⇓⇓
In conclusion, the present study showed that even delayed application of moderate hypothermia in the reperfusion period of transient focal cerebral ischemia has neuroprotective effects, as demonstrated by serial MRI and neurological testing. The benefit of moderate hypothermia extends beyond the acute phase and is still evident 5 days after MCAO.
The authors would like to acknowledge the excellent assistance of Frank Malischewsky.
- Received January 31, 2002.
- Revision received March 13, 2002.
- Accepted March 25, 2002.
- ↵Maier CM, Ahern Kv, Cheng ML, Lee JE, Yenari MA, Steinberg GK. Optimal depth and duration of mild hypothermia in a focal model of transient cerebral ischemia: effects on neurologic outcome, infarct size, apoptosis, and inflammation. Stroke. 1998; 29: 2171–2180.
- ↵Yanamoto H, Nagata I, Nakahara I, Tohnai N, Zhang Z, Kikuchi H. Combination of intraischemic and postischemic hypothermia provides potent and persistent neuroprotection against temporary focal ischemia in rats. Stroke. 1999; 30: 2720–2726.
- ↵Karibe H, Zarow GJ, Graham SH, Weinstein PR. Mild intraischemic hypothermia reduces postischemic hyperperfusion, delayed postischemic hypoperfusion, blood-brain barrier disruption, brain edema, and neuronal damage volume after temporary focal cerebral ischemia in rats. J Cereb Blood Flow Metab. 1994; 14: 620–627.
- ↵Schwab S, Schwarz S, Spranger M, Keller E, Bertram M, Hacke W. Moderate hypothermia in the treatment of patients with severe middle cerebral artery infarction. Stroke. 1998; 29: 2461–2466.
- ↵Krieger DW, De Georgia MA, Abou-Chebl A, Andrefsky JC, Sila CA, Katzan IL, Mayberg MR, Furlan AJ. Cooling for acute ischemic brain damage (cool aid): an open pilot study of induced hypothermia in acute ischemic stroke. Stroke. 2001; 32: 1847–1854.
- ↵Fisher M, Albers GW. Applications of diffusion-perfusion magnetic resonance imaging in acute ischemic stroke. Neurology. 1999; 52: 1750–1756.
- ↵Schäbitz WR, Hoffmann TT, Heiland S, Kollmar R, Bardutzky J, Sommer C, Schwab S. Delayed neuroprotective effect of insulin-like growth factor-I after experimental transient focal cerebral ischemia monitored with MRI. Stroke. 2001; 32: 1226–1233.
- ↵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: 2074–2081.
- ↵Longa E, Weinstein PR, Carlson S, Cummins R. Reversible middle cerebral artery occlusion. Stroke. 1989; 20: 84–91.
- ↵Muller TB, Haraldseth O, Jones RA, Sebastiani G, Godtliebsen F, Lindboe CF, Unsgard G. Combined perfusion and diffusion-weighted magnetic resonance imaging in a rat model of reversible middle cerebral artery occlusion. Stroke. 1995; 26: 451–457.
- ↵Neumann-Haefelin T, Kastrup A, de Crespigny A, Yenari MA, Ringer T, Sun GH, Moseley ME. Serial MRI after transient focal cerebral ischemia in rats: dynamics of tissue injury, blood-brain barrier damage, and edema formation. Stroke. 2000; 31: 1965–1972.