Numerous experimental studies indicate that volatile anesthetics improve outcome from brain ischemia (1–5). However, the literature is confusing regarding which dose of volatile anesthetic yields the largest therapeutic benefit. A dose-dependent effect of isoflurane in attenuating glutamatergic excitotoxicity has been demonstrated. Patel et al. (6) reported that 2.2% isoflurane was more effective than 0.6% isoflurane in reducing intraischemic extracellular glutamate accumulation. In normal rat cortex reduction of necrosis induced by microinjection of N-methyl-d-aspartate (NMDA) was greater with 2.2% versus 1.2% isoflurane (7). In vitro studies using cultured hippocampal neurons (8), mixed neuronal-glial cell cultures (9), and hippocampal brain slices (10) have also reported a dose-dependent benefit of isoflurane against NMDA receptor antagonism. There are in vivo studies reporting that an isoflurane dose sufficient to induce electroencephalographic (EEG) burst suppression failed to improve outcome (11–12). In contrast 0.5 minimum alveolar concentration (MAC) isoflurane failed to have a protective effect against middle cerebral artery occlusion (13) whereas other studies reported ischemic neuroprotection by isoflurane when it was evaluated at a single dose of approximately ∼1 MAC (1.1% ∼ 1.8%) (1–5,14).
These results suggest that outcome is dependent on the dose of isoflurane used during ischemia and that there might be an optimal concentration for neuroprotection. Accordingly, we performed an experiment examining the dose-response relationship between the intraischemic isoflurane dose and histologic/neurologic outcome from a severe forebrain ischemic insult in the rat. To assure uniformity of postischemic arterial blood pressure during early reperfusion, isoflurane was discontinued and replaced with fentanyl/nitrous oxide after ischemia.
The Yamagata University Animal Care and Use Committee approved this study.
Male Sprague-Dawley rats (8–10 wk of age, Kumagai Shigeyasu, Sendai, Japan) were fasted from food for 12–18 h but allowed free access to water before the experiment. Anesthesia was induced with 5% halothane in oxygen. After orotracheal intubation, the inspired halothane concentration was reduced to 1.5%–2.0%, and the lungs were mechanically ventilated (50% O2/balance N2). End-tidal medical gas concentrations were continuously monitored during the experiment (Narcotica, model HC-510A; Fukuda Denshi, Inc., Sendai, Japan). Minute ventilation was adjusted to maintain normocapnia. Bilateral cortical EEG was continuously monitored from active subdermal electrodes positioned over the parietal cortex bilaterally, a reference electrode was placed on the nasion, and a ground lead was positioned in the tail. A 22-gauge needle thermistor (Unique Medical, Tokyo, Japan) was percutaneously placed adjacent to the skull beneath the temporalis muscle. Pericranial temperature was servoregulated at 37.5°C ± 0.1°C by surface heating or cooling from beginning of the surgery until 2 h after recirculation. Surgery was performed with aseptic technique, and all surgical fields were infiltrated with 1% lidocaine. The tail artery was cannulated to monitor arterial blood pressure and sample arterial blood. Normal saline, 4 mL/kg/h, was continuously infused in the tail artery catheter. Via a ventral neck incision, the right jugular vein was cannulated with a silicone catheter for drug infusion and blood withdrawal. The common carotid arteries were encircled with suture. The vagus nerves and cervical sympathetic plexus were left intact. Muscle paralysis was provided by a 1 mg intraarterial bolus of succinylcholine, repeated as necessary to allow control of ventilation. Heparin 50 IU was given IV.
There is accumulating evidence that preischemic exposure to volatile anesthetics modulates ischemic outcome (15–17). To control for this we used halothane for surgical preparation, followed by conversion of the anesthetic to fentanyl (10 μg/kg bolus followed by 25 μg · kg−1 · h−1 IV) and 70% nitrous oxide/30% O2, allowing halothane to be discontinued. After 30 min, nitrous oxide and fentanyl were discontinued and replaced with isoflurane (in 30% O2/balance N2). The isoflurane dose (0.5, 1.0, 1.5, 2.0, or 2.5 MAC) was defined by a randomization schedule with 10–13 animals studied at each isoflurane concentration. The 1 MAC value for isoflurane used in this experiment was 1.4% (1). Stabilization at the assigned isoflurane dose was allowed for 30 min. Pilot studies were performed to confirm that rats did not express escape effort in the absence of succinylcholine during this interval with 0.5 MAC isoflurane. Arterial blood pressure was not pharmacologically manipulated to avoid pressor-related effects on outcome (18). At 10 min preischemia, baseline values for Paco2, Pao2, arterial pH, plasma glucose, hematocrit, and mean arterial blood pressure (MAP) were obtained. Immediately before ischemia onset, the isoflurane concentration was adjusted 0.5 MAC (0.7%). Cerebral blood flow is reduced to near-zero values during severe forebrain ischemia (19). We reasoned, therefore, that the anesthetic delivery/elimination in brain is negligible during ischemia. This allowed us to “clamp” brain parenchymal anesthetic concentrations during the ischemic insult and also provide a uniform hemodynamic recovery profile given similar circulating anesthetic concentrations at the onset of reperfusion.
Ischemia was induced by reduction of MAP to 30 mm Hg by withdrawal of blood from the venous catheter followed by bilateral carotid occlusion using temporary aneurysm clips. Withdrawn blood in a syringe was placed under a heating lamp to maintain normothermia. Ischemia persisted for 10 min. To terminate ischemia, shed blood was reinfused and carotid arteries were deoccluded. NaHCO3 (0.1 mEq IV) was given to minimize systemic acidosis. Rats exhibiting any EEG activity during ischemia were excluded from the experiment.
Isoflurane administration was terminated and fentanyl infusion (12.5 μg · kg−1 · h−1) and 70% nitrous oxide inspiration were restored for 100 min. The animals were then allowed to awaken and breathed 50% O2/50% N2. Mechanical ventilation was continued until rats exhibited escape movements. The animals were allowed to tracheally extubate themselves. The rats were placed in an oxygen-enriched recovery chamber (Fio2 = 0.3–0.4) for several hours, then returned to their home cage with free access to water and food for 5 days. During this interval, the animals were closely observed for evidence of generalized seizures, either defined by direct observation or dispersion of litter outside the cage overnight.
On the fifth postoperative day, three neuromotor tests (screen clinging, horizontal bar, and prehensile traction) were performed by an observer blinded as to group assignment. Details of this neuromotor assessment are described elsewhere (1). From these 3 tests, a neuromotor score (9 = best possible score) was computed.
Rats were then anesthetized with halothane and underwent in situ brain fixation by intracardiac injection of buffered 4% formalin. After overnight stabilization, the brains were removed and stored in 4% formalin. Paraffin-embedded brain sections were serially cut (5 μm thick) and stained with hematoxylin and eosin. An observer blinded as to group assignment using light microscopy evaluated hippocampal CA1 injury at bregma −4.0 mm. Viable and nonviable neurons were manually counted and the percentage of nonviable neurons was calculated (% CA1 dead). At the level where the septal nuclei were widest, damage in the neocortex (Crude Damage Index; CDI) was graded on a 1–4 scale (1 = 0%–25% of neurons damaged; 2 = 25%–50% of neurons damaged; 3 = 50%–75% of neurons damaged, and 4 = 75%–100% of neurons damaged) (19). By convention, values from the hemisphere with the worst damage were used for the final analysis.
Parametric values, including physiologic data and % CA1 dead neurons, were compared with a one-way analysis of variance. When indicated by a significant F ratio, post hoc inter-MAC comparisons were performed with Scheffe's correction for multiple comparisons. Values were reported as mean ± sd.
Nonparametric values (CDI and neuromotor score) were analyzed with the Kruskal-Wallis test. If statistical differences were detected, the Mann-Whitney U-test was used for intergroup comparisons. The χ2 test was used to compare seizure incidence and mortality rates. Values are reported as median ± interquartile range. Statistical significance was assumed when P < 0.05.
Two rats in the 2 MAC group were excluded from statistical analysis because of persistent EEG activity during ischemia. Pericranial temperature was controlled as intended in all animals.
Ten to 13 rats were required to achieve 8 to 9 survival animals per group. Rats given 1.5–2.5 MAC frequently developed seizures within 12–24 h after ischemia (number of rats operated/seizure/survived: 0.5 MAC 10/1/8, 1.0 MAC 11/0/9, 1.5 MAC 11/5/8, 2.0 MAC 12/5/8, 2.5 MAC 13/6/8). The frequency of seizures was different among groups (P = 0.0034). Most rats with seizures did not survive (survival from seizures/total number with seizures: 13/17). The mortality rate, however, was not different among groups (P = 0.79). The remaining deaths were from undefined causes. Physiological values for survivors are reported in Table 1. Values, except MAP 10 min preischemia, were similar among MAC groups. Histological damages and neuromotor score are illustrated in Figure 1. No difference was detected regarding % CA1 dead neurons among MAC groups (0.5 MAC = 93% ± 5%, 1 MAC = 92% ± 4%, 1.5 MAC = 93% ± 9%, 2 MAC = 88% ± 14%, 2.5 MAC = 92% ± 3%; P = 0.89). There was a difference in the CDI among MAC groups (0.5 MAC = 2 ± 0, 1 MAC = 2 ± 0.5, 1.5 MAC = 3 ± 0, 2 MAC = 3 ± 1, 2.5 MAC = 3 ± 1; P = 0.01). The CDI in the 1 MAC group was less than that in the 2.5 MAC group (P = 0.015). A difference was present in the neuromotor scores (0.5 MAC = 8 ± 0.5, 1 MAC = 9 ± 0, 1.5 MAC = 7 ± 2, 2 MAC = 7.5 ± 1.5, 2.5 MAC = 4 ± 2; P = 0.036), whereas the 1 MAC score was higher than the 2.5 MAC score (P = 0.003).
Long-lasting neuroprotection from isoflurane has been questioned (14). However, it seems certain that isoflurane protects the brain against ischemic insults for at least several days (1–5,14). The precise mechanism for isoflurane's neuroprotection is unknown but has been attributed to the same mechanisms that produce anesthesia (20). Isoflurane is an antagonist of glutamate at the NMDA receptor (21) and also enhances the gamma-aminobutyric acid (GABA)A receptor (22). It would be expected that isoflurane acts on these receptors in a dose-dependent manner. Dose-dependent antagonism of the NMDA receptor by isoflurane has been reported in studies using patch-clamped neurons (8), mixed neuronal/glial cultures (9), brain slices (10), and in vivo rat brain cortex exposed to microinjections of excitotoxins (7). Similarly, isoflurane's enhancement of GABAergic-mediated chloride flux is also dose-dependent (23). Thus, if inhibition of the NMDA receptor and enhancement of GABAergic neurotransmission are the principal mechanisms of isoflurane neuroprotection, it would be expected that there is also a dose-dependent relationship between intra-ischemic brain isoflurane concentration and outcome from an ischemic insult. Because this has not been investigated, we conducted the current study with the hypothesis that isoflurane affects outcome in a dose-dependent manner.
In the current experiment, isoflurane did not exhibit a dose-dependent effect on hippocampal CA1 injury. This could indicate absence of protection or more likely reflect the severity of insult. This is indicated by the relatively frequent incidence of postischemic seizures observed, especially in the >1.5 MAC isoflurane groups. In the original description of the rat 2-vessel occlusion model, 10% of rats developed seizures 1–2 days after ischemia (24). This is in contrast to the 50% seizure incidence observed in our experiment, which is similar to that reported in rats having hyperglycemia-augmented ischemic brain injury (25). However, preischemic plasma glucose concentrations were within a normoglycemic range in all groups as a result of overnight fasting from food before the ischemic insult. This indicates that the current study examined a more severe ischemic insult than has been typically observed in rats subjected to 10 minutes of bilateral carotid occlusion combined with hemorrhagic hypotension. Because hippocampal CA1 damage was typically more than 90% in all groups, cortical injury and associated motor function are the most informative aspects of this investigation. Cortical damage was dependent on intraischemic isoflurane concentration with worsened outcome observed both histologically and behaviorally when rats were treated with isoflurane doses larger than 1.0 MAC immediately before ischemia.
The 1.0 MAC isoflurane group had better outcome defined by reduced cortical damage and improved neuromotor scores relative to the 2.5 MAC group. This parallels the incidence of postischemic seizures that occurred in no 1.0 MAC rats versus 46% of 2.5 MAC rats. These results suggest that there is a U-shape response curve for isoflurane dose, with the optimal concentration approximating 1 MAC. Larger concentrations were either numerically or statistically associated with a greater magnitude of injury.
The lack of effect of isoflurane's protection against hippocampal CA1 injury should be considered in the context of other work using the same model. In contrast to the current findings, Miura et al. (1) and Mackensen et al. (2) observed substantial reduction of CA1 damage in rats subjected to 10-minute severe forebrain ischemia when anesthetized with 1.4% (1.0 MAC) isoflurane as opposed to fentanyl/nitrous oxide. Preischemic plasma glucose values were similar to those observed in the current experiment. Differences between these experiments may be the result of the respective anesthetic regimens. Both Miura et al. (1) and Mackensen et al. (2) administered isoflurane for an extended interval before, during, and after ischemia. In the current experiment, isoflurane was administered only during the immediate interval preceding the ischemic insult. To our knowledge, there has been no previous effort to isolate anesthetic exposures to the immediate peri-ischemic interval. One study examined postischemic halothane exposure and identified no effect, but this study was performed in a focal ischemia model (26), and there is no information regarding isoflurane.
It was also noted that a larger concentration of isoflurane was associated with worse outcome. Considering that preischemic MAP reduction was greater with larger MAC values, it is possible that low preischemic MAP had some influence on outcome. However, it has been shown that severe systemic hypotension itself, MAP = 25 mm Hg for 20 minutes by exsanguination, does not cause histological injury to the brain (27). No rats in our experiment developed MAP = 25 mm Hg before ischemic insult. Therefore, we do not believe that preischemic relative hypotension contributed to outcome deterioration. It is also reported that administration of large concentrations of isoflurane (>3%) produce mild dose-related cerebral lactic acidosis (28). We speculate that possible existence of cerebral lactic acidosis in the high MAC groups might have adversely contributed to outcome after ischemia.
The goal of our study was to “clamp” the targeted isoflurane concentrations during the ischemic insult and minimize potential confounding effects of ischemic preconditioning caused by preischemic exposure to the volatile anesthetics. The phenomenon of volatile anesthetic preconditioning against ischemic brain injury remains in the initial stages of investigation. However, halothane has been shown to cause acute preconditioning with an exposure interval of 1 hour immediately before focal ischemia onset (17). Similarly, exposure to 1% isoflurane for 3 hours immediately before a focal ischemia insult also caused substantial reduction of cerebral infarct size resulting from focal ischemia (15). The extent to which acute volatile anesthetic exposure increases tolerance to a homogenous forebrain ischemic insult has not been evaluated. However, results from the current experiment provide a rational basis for investigating this possibility. The anesthesia protocols used by Miura et al. (1) and Mackensen et al. (2) provided isoflurane during surgical preparation and physiologic stabilization, which might have preconditioned the brain. Considering their experiments and our results, we speculate that sustained preischemic isoflurane exposure (>30 minutes) with ∼1 MAC is required to induce substantial neuroprotection against a rat model of forebrain ischemia.
In conclusion, we sought to determine if a dose-dependent effect of isoflurane on outcome could be observed in rats subjected to a severe forebrain ischemic insult. Preischemic stabilization of anesthetic concentrations for 30 minutes provided a differential effect on outcome. Postischemic seizures were more frequent in rats given >1.0 MAC isoflurane. MAC 0.5 and 1.0 isoflurane produced numerically similar effects on cortical neuronal necrosis and motor function testing. These variables were worse in rats given 2.5 MAC isoflurane. The data indicate a dose-dependent effect of isoflurane on in vivo ischemic pathogenesis, with moderate dose isoflurane (e.g., 1.0 MAC) providing a better outcome than large-dose (e.g., 2.5%) isoflurane.
The authors thank Mr. Tadayoshi Karube for his expert technical assistance.
1. Miura Y, Grocott HP, Bart RD, et al. Differential effects of anesthetic agents on outcome from near-complete but not incomplete global ischemia in the rat. Anesthesiology 1998;89:391–400.
2. Mackensen GB, Nellgard B, Miura Y, et al. Sympathetic ganglionic blockade masks beneficial effect of isoflurane on histologic outcome from near-complete forebrain ischemia in the rat. Anesthesiology 1999;90:873–81.
3. Soonthon-Brant V, Patel PM, Drummond JC, et al. Fentanyl does not increase brain injury after focal cerebral ischemia in rats. Anesth Analg 1999;88:49–55.
4. Homi HM, Mixco MJ, Sheng H, et al. Severe hypotension is not essential for isoflurane neuroprotection against forebrain ischemia in mice. Anesthesiology 2003;99:1145–51.
5. Blanck TJ, Haile M, Xu F, et al. Isoflurane pretreatment ameliorates postischemic neurologic dysfunction and preserves hippocampal Ca2+/calmodulin-dependent protein kinase in a canine cardiac arrest model. Anesthesiology 2000;93:1285–93.
6. Patel PM, Drummond JC, Cole DJ, Goskowicz RL. Isoflurane reduces ischemia-induced glutamate release in rats subjected to forebrain ischemia. Anesthesiology 1995;82:996–1003.
7. Harada H, Kelly PJ, Cole DJ, et al. Isoflurane reduces N
-methyl-d-aspartate toxicity in vivo
in the rat cerebral cortex. Anesth Analg 1999;89:1442–7.
8. Yang J, Zorumski CF. Effects of isoflurane on N
-methyl-d-aspartate gated ion channels in cultured rat hippocampal neurons. Ann N Y Acad Sci 1991;625:287–9.
9. Kudo M, Aono M, Lee Y, et al. Effects of volatile anesthetics on N
-methyl-d-aspartate excitotoxicity in primary rat neuronal-glial cultures. Anesthesiology 2001;95:756–65.
10. Popovic R, Liniger R, Bickler PE. Anesthetics and mild hypothermia similarly prevent hippocampal neuron death in an in vitro
model of cerebral ischemia. Anesthesiology 2000;92:1343–9.
11. Warner DS, Deshpande JK, Wieloch T. The effect of isoflurane on neuronal necrosis following near-complete forebrain ischemia in the rat. Anesthesiology 1986;64:19–23.
12. Drummond JC, Cole DJ, Patel PM, Reynolds LW. Focal cerebral ischemia during anesthesia with etomidate, isoflurane, or thiopental: a comparison of the extent of cerebral injury. Neurosurgery 1995;37:742–8.
13. Sarraf-Yazdi S, Sheng H, Miura Y, et al. Relative neuroprotective effects of dizocilpine and isoflurane during focal cerebral ischemia in the rat. Anesth Analg 1998;87:72–8.
14. Kawaguchi M, Kimbro JR, Drummond JC, et al. Isoflurane delays but does not prevent cerebral infarction in rats subjected to focal ischemia. Anesthesiology 2000;92:1335–42.
15. Kapinya KJ, Prass K, Dirnagl U. Isoflurane induced prolonged protection against cerebral ischemia in mice: a redox sensitive mechanism? Neuroreport 2002;13:1431–5.
16. Zheng S, Zuo Z. Isoflurane preconditioning induces neuroprotection against ischemia via activation of P38 mitogen-activated protein kinases. Mol Pharmacol 2004;65:1172–80.
17. Bhardwaj A, Castro III AF, Alkayed NJ, et al. Anesthetic choice of halothane versus propofol: impact on experimental perioperative stroke. Stroke 2001;32:1920–5.
18. Cole DJ, Drummond JC, Osborne TN, Matsumura J. Hypertension and hemodilution during cerebral ischemia reduce brain injury and edema. Am J Physiol 1990;259:211–7.
19. Mackensen GB, Nellgard B, Kudo M, et al. Periischemic cerebral blood flow (CBF) does not explain beneficial effects of isoflurane on outcome from near-complete forebrain ischemia in rats. Anesthesiology 2000;93:1102–6.
20. Warner DS. Perioperative neuroprotection: are we asking the right questions? Anesth Analg 2004;98:563–5.
21. Aronstam RS, Martin DC, Dennison RL. Volatile anesthetics inhibit NMDA-stimulated 45Ca uptake by rat brain microvesicles. Neurochem Res 1994;19:1515–20.
22. Miao N, Frazer MJ, Lynch CR. Volatile anesthetics depress Ca2+ transients and glutamate release in isolated cerebral synaptosomes. Anesthesiology 1995;83:593–603.
23. Gyulai FE, Mintun MA, Firestone LL. Dose-dependent enhancement of in vivo
GABA(A)-benzodiazepine receptor binding by isoflurane. Anesthesiology 2001;95:585–93.
24. Smith ML, Bendek G, Dahlgren N, et al. Models for studying long-term recovery following forebrain ischemia in the rat. 2. A 2-vessel occlusion model. Acta Neurologica Scandinavica 1984;69:385–401.
25. Warner DS, Smith ML, Siesjo BK. Ischemia in normo- and hyperglycemic rats: effects on brain water and electrolytes. Stroke 1987;18:464–71.
26. Sarraf-Yazdi S, Sheng H, Brinkhous AD, et al. Effects of postischemic halothane administration on outcome from transient focal cerebral ischemia in the rat. J Neurosurg Anesthesiol 1999;11:31–6.
27. Warner DS, Reasoner DK, Todd MM, McAllister A. Secondary hypotensive insults in a rat forebrain ischemia model. Brain Res 1990;536:176–82.
© 2006 International Anesthesia Research Society
28. Newberg LA, Milde JH, Michenfelder JD. The cerebral metabolic effects of isoflurane at and above concentrations that suppress cortical electrical activity. Anesthesiology 1983;59:23–8.