Secondary Logo

Journal Logo

Alphaxalone, a Neurosteroid Anesthetic, Inhibits Norepinephrine Transporter Function in Cultured Bovine Adrenal Medullary Cells

Horishita, Takafumi*; Minami, Kouichiro*; Yanagihara, Nobuyuki; Shiraishi, Munehiro*; Okamoto, Takashi*; Shiga, Yousuke*; Ueno, Susumu; Shigematsu, Akio*

doi: 10.1097/00000539-200212000-00034
ANESTHETIC PHARMACOLOGY: Research Report
Free

We studied the effects of alphaxalone, a neurosteroid anesthetic, on norepinephrine transporter (NET) function in cultured bovine adrenal medullary cells and the effect of a bolus injection of alphaxalone on blood pressure and serum norepinephrine (NE) levels in anesthetized rats. Alphaxalone (10–100 μM) inhibited the desipramine-sensitive uptake of [3H]-NE by bovine adrenal medullary cells in a concentration-dependent manner. Eadie-Hofstee analysis of [3H]-NE uptake showed that alphaxalone increased the apparent Michaelis constant without altering the maximal velocity, indicating that inhibition occurred via competition for the NET. Alphaxalone inhibited the specific binding of [3H]-desipramine to plasma membranes isolated from bovine adrenal medulla. Scatchard analysis of [3H]-desipramine binding revealed that alphaxalone increased the apparent dissociation constant for binding without altering maximal binding, indicating competitive inhibition. Bolus IV administration of alphaxalone had little effect on blood pressure but slightly, and significantly, increased the serum NE levels in anesthetized rats. These findings suggest that alphaxalone competitively inhibits NET function by interfering with both desipramine binding and NE recognition on the NET in adrenal medullary cells and probably in sympathetic neurons.

Departments of *Anesthesiology and †Pharmacology, University of Occupational and Environmental Health, School of Medicine, Kitakyushu, Japan

Supported, in part, by Grants-in-Aid (Nos. 11671532, 10770778, 12770851, 11770878, 12671515, 12671516, and 12770849) from the Ministry of Education, Science, and Culture of Japan, by a UOEH Research Grant for the Promotion of Occupational Health, by the Japan Research Foundation for Clinical Pharmacology, by the Uehara Memorial Foundation, and by the Kanehara-Ichiro Memorial Medical Foundation.

August 23, 2002.

Address correspondence and reprint requests to Kouichiro Minami, Department of Anesthesiology, University of Occupational and Environmental Health, School of Medicine, 1–1, Iseigaoka, Yahatanishiku, Kitakyushu 807–8555, Japan. Address e-mail to kminami@med.uoeh-u.ac.jp.

Neurosteroids are synthesized from cholesterol in both the central and peripheral nervous systems (1). Many steroids in the pregnane series have hypnotic and anesthetic activity (2). Based on this knowledge, the anesthetic alphaxalone (3α-hydroxy-5α-pregnane-11,20-dione) was discovered and tested for clinical use (3). The potent, stereoselective potential of the γ-aminobutyric acid A receptor-mediated responses of alphaxalone was first shown in extracellular recordings performed on rat brain slices (4) and was subsequently confirmed in voltage-clamp experiments using cultured rat hippocampal neurons (5) and adrenal medullary cells (6).

Pain perception is modulated by a variety of neurotransmitters (7), including endogenous norepinephrine (NE) (8). Several lines of evidence have shown that the descending inhibitory system consists of noradrenergic neurons (9). Two experiments have revealed that several anesthetics and analgesics inhibit NE transporter (NET) function at clinical concentrations. The IV anesthetics propofol and ketamine inhibit NET function (10,11). More recently, it was reported that long-term treatment with propofol leads to upregulation of the NET, suggesting that propofol modulates the neuronal transmission of NE during anesthesia (12). Therefore, NETs are thought to be one of the targets of anesthetics.

The NETs located in the presynaptic membranes of noradrenergic nerve termini regulate neurotransmission by taking up NE released into the synaptic cleft (13). Pacholczyk et al. (14) first cloned the human NET, and its messenger RNA is expressed at high levels in the brainstem and adrenal medulla. The complimentary DNA of the bovine adrenal medullary NET, which was cloned by Lingen et al. (15), encodes an amino acid sequence that is approximately 94% identical to that of the human NET. Moreover, the pharmacological properties of the bovine adrenal medullary NET are quite similar to those of central noradrenergic neurons (16). Therefore, bovine adrenal medullary cells are a useful model system for studying the effects of anesthetics (10–12) on NETs in noradrenergic neurons.

The purpose of this study was to examine the effects of alphaxalone on NET function. For this purpose, we investigated whether alphaxalone inhibits NET function by assessing the effects of alphaxalone on [3H]-NE uptake and [3H]-desipramine binding in bovine adrenal medullary cells. We also examined serum NE levels and blood pressure using rats to study the effects of alphaxalone on sympathetic nervous system activity.

Back to Top | Article Outline

Methods

This study conformed to the Guide for the Care and Use of Laboratory Animals adopted and promulgated by the Japanese National Institute of Health, and the experiment was approved by the Animal Research Committee of the University of Occupational and Environmental Health. Adrenal medullary cells were isolated by collagenase digestion of slices of bovine adrenal medulla, as described previously (17). Cells were plated on 35-mm Falcon dishes (4 × 106 cells/dish) in Eagle’s minimum essential medium containing 10% calf serum, 60 μg/mL of aminobenzylpenicillin, 100 μg/mL of streptomycin, 0.3 μg/mL of amphotericin B, and 3.0 μM of cytosine arabinoside (17). The cells were cultured under 5% CO2/95% air in an incubator at 37°C and were used for experiments at 2–4 days of culture. In the trypan blue test, more than 99% of the cells were viable.

[3H]-NE uptake by adrenal medullary cells was performed as follows. Cultured cells were incubated at 37°C for 15 min in Krebs-Ringer-HEPES (KRH; pH value of 7.4) buffer containing 100 μM of pargyline, 100 μM of ascorbic acid, and 500 nM of [3H]-NE in the presence or absence of alphaxalone (0.1–100 μM). KRH buffer was composed of 10 mM of HEPES-NaOH (pH value of 7.4), 154 mM of NaCl, 5.6 mM of KCl, 1.1 mM of MgSO4, 2.2 mM of CaCl2, and 10 mM of glucose. To determine the saturation kinetics of [3H]-NE uptake, various concentrations (1–30 μM) of [3H]-NE were added in the presence or absence of 100 μM of alphaxalone or 10 μM of desipramine, a selective inhibitor of NETs. After incubation, the cells were rapidly washed four times with 1 mL of ice-cold KRH buffer and solubilized in 1 mL of 10% Triton X-100. The radioactivity in the solubilized cells was counted with a liquid scintillation counter (LSC-3500E, Aloka, Tokyo, Japan). The specific uptake of [3H]-NE was determined by subtracting the nonspecific uptake (determined in the presence of desipramine) from the total uptake.

[3H]-desipramine was bound to plasma membranes using the following protocol. Plasma membranes isolated from bovine adrenal medulla were prepared as described previously (11). Membranes (20 μg of protein) were resuspended in 10 mM of Tris-HCl (pH value of 7.4), 135 mM of NaCl, 5 mM of KCl, and 1 mM of MgSO4 (buffer B). The membrane suspension was incubated for 30 min at 25°C with [3H]-desipramine (1–24 nM) in the presence or absence of 100 μM of alphaxalone or 10 μM of nisoxetine, a selective NET inhibitor, in a final reaction volume of 250 μL. After incubation, binding was terminated by the addition of 2 mL of ice-cold buffer B followed by rapid vacuum filtration of the membrane suspension through Whatman GF/C glass fiber filters (Whatman, Maidstone, UK). The filters were rapidly washed twice with 2 mL of ice-cold buffer B, and the radioactivity retained on the filters was determined by liquid scintillation counting. Specific binding of [3H]-desipramine was defined as the binding inhibited by nisoxetine.

The procedures used to measure arterial blood pressure and to sample blood have been reported previously (18). Briefly, adult male Wistar rats (body weight, 300 g) were anesthetized by intraperitoneal injection of pentobarbital sodium (50 mg/kg). The rat was placed in a restraining cage, and a polyvinyl chloride catheter was inserted into its tail vein to deliver IV solutions. Another catheter was placed in the femoral artery to measure mean arterial blood pressure (MAP). Throughout the experiments, MAP was measured using a calibrated pressure transducer (Baxter Healthcare Corporation, Deerfield, IL) positioned one third of the distance from the brisket to the top of the back and recorded using a polygraph (Datascope 870 monitor, Datascope Corporation, Paramus, NJ) and the MacLab/2 data acquisition system (AD Instruments Pty. Ltd, Castle Hill, NSW, Australia). The rats were allowed at least 60 min to equilibrate after surgical preparation to eliminate the effects of pentobarbital and surgical stimulation on MAP and renal blood flow. During the experiments, the rats breathed spontaneously, and oxygen was supplied. No changes in pH values, blood gas indices, electrolytes, or hematocrit occurred throughout the experiment (data not shown). The body temperature was maintained at 37°C with a heating pad. In our preliminary experiments, the MAP was not affected by intraperitoneal injection of pentobarbital sodium itself and remained stable during the experiment (for approximately 5 h) (data not shown).

After surgery and a 60-min equilibration period after pentobarbital sodium treatment of rats, MAP was measured before (control) and after administering the test compounds. According to our previous reports (18), injection of lactated Ringer’s solution without test compounds does not change the MAP. The maximal change in MAP was observed within 5 min of the administration of each test compound. The MAP is given in mm Hg. The serum NE concentration was also measured before and after the administration of an alphaxalone infusion (0.1, 0.3, and 0.6 μg/kg) to the anesthetized rats. The serum NE concentration was measured using the method reported by Minami et al. (10). We collected arterial blood (600 μL) for NE measurement and simultaneously infused the same volume of lactated Ringer’s solution containing 6% hydroxyethyl starch into the tail vein to reduce any hemodynamic effects caused by the blood withdrawal. This procedure had no effect on the arterial blood pressure (10). After a 15-min rest, alphaxalone (0.1–0.6 μg/kg) was injected into the tail vein continuously for 30 s by an infusion pump (Terumo model STC-253, Terumo, Tokyo, Japan). After treatment with alphaxalone for 15 min, another 600 μL of blood was collected. Blood samples were centrifuged at 1500 g for 5 min at 4°C, and the sera were kept at −80°C until the NE assay. Blood sampling had little effect on NE concentration or blood pressure (10). NE in the sera was absorbed to activated Al2O3 and separated by high-performance liquid chromatography (Shimadzu LC-5A, Kyoto, Japan) using a reverse-phase C18 column (2.6 × 250 mm; Cosmosil, Nacalai Tesque, Inc, Kyoto, Japan). The separated NE was assayed electrochemically using an amperometer (Hitachi L-7480, Tokyo, Japan). NE recovery during this measurement was approximately 95%. In this assay system, inter- and intraassay coefficients of variation were 1.3% and 2.3%, respectively. The limit of detection was 1.8 pg/mL.

Eagle’s minimum essential medium was from Nissui Pharmaceuticals (Tokyo, Japan). Calf serum, (-)-NE, pargyline hydrochloride, (-)-ascorbic acid, atropine sulfate, and HEPES were from Nacalai Tesque, Inc. Collagenase was from Nitta Zerachin (Osaka, Japan). Desipramine hydrochloride and alphaxalone were from Sigma (St Louis, MO). Nisoxetine hydrochloride was from Research Biochemicals International (Natick, MA). (-)-[7, 8-[3H]-NE (34.0 Ci/mmol) was from Amersham Pharmacia Biotech (Buckinghamshire, UK). [Benzene ring, 10, 11-3H]-desmethylimipramine hydrochloride ([3H]-desipramine) (73.0 Ci/mmol) was from New England Nuclear (Boston, MA). L-NE was dissolved in 0.1 M of HCl, adjusted to a neutral pH value with 0.1 M of NaOH, and diluted with distilled water for use.

Kinetic variables for [3H]-desipramine binding, dissociation constant (KD) and maximal binding value (Bmax), were estimated by Scatchard analysis. All values are expressed as the mean ± sd. Statistical evaluation was accomplished by one-way analysis of variance. When a significant P value was found by analysis of variance, Dunnett test for multiple comparisons was performed to identify differences among the groups. When P < 0.05, the differences were considered statistically significant. Curve fitting for the concentration-response curves were performed using Prism version 3.0.2 (GraphPad Software, San Diego, CA).

Back to Top | Article Outline

Results

Alphaxalone (10–100 μM) significantly inhibited [3H]-NE uptake by adrenal medullary cells in a concentration-dependent manner (Fig. 1). [3H]-NE uptake was reduced by 29% ± 13%, 34% ± 17%, and 46% ± 5% at 10, 30, and 100 μM of alphaxalone, respectively. Incubation of the cells with increasing concentrations of [3H]-NE (1–30 μM) showed that the [3H]-NE uptake was saturable (Fig. 2 A). Eadie-Hofstee analysis yielded a maximal velocity (Vmax) of 133 ± 10 pmol/4 × 106 cells/15 min and an apparent Michaelis constant (Km) of 2.8 ± 0.4 μM in control cells (Fig. 2 B). Alphaxalone (100 μM) produced an increase in Km (4 ± 0.8 μM;P < 0.05) with no significant change in Vmax (125 ± 11 pmol/4 × 106 cells/15 min) (Fig. 2 B).

Figure 1

Figure 1

Figure 2

Figure 2

To determine the site of action of alphaxalone on the NET, we next examined the effects of alphaxalone on the binding of [3H]-desipramine to plasma membranes isolated from bovine adrenal medulla. Specific binding of [3H]-desipramine was saturable (Fig. 3 A). Scatchard analysis indicated that there was a single population of binding sites. The apparent KD was 5.1 ± 0.9 pmol/mg, and the Bmax was 1.4 ± 0.4 pmol/mg of protein in control cells (Fig. 3 B). Alphaxalone (100 μM) significantly increased the KD for [3H]-desipramine binding (7.4 ± 1.3 nM) without any change in Bmax (1.5 ± 0.4 pmol/mg of protein).

Figure 3

Figure 3

We next examined the effects of various concentrations of alphaxalone (0.1–100 μM) on [3H]-desipramine binding to the cell membranes. Alphaxalone inhibited [3H]-desipramine binding in a concentration-dependent manner (Fig. 4). [3H]-Desipramine binding was reduced to 82% ± 5%, 68% ± 5%, and 56% ± 7% of the control value at 10, 30, and 100 μM of alphaxalone, respectively.

Figure 4

Figure 4

In this study, alphaxalone significantly inhibited the [3H]-NE uptake (Fig. 1) raising the question of whether alphaxalone affects the NE uptake in sympathetic nerve terminals. To clarify this, we next examined the effects of a bolus injection of alphaxalone (0.1–0.6 mg/kg;n = 10) on serum NE levels in anesthetized rats. As shown in Table 1, the bolus injection of alphaxalone had a slight effect on blood pressure, although the 0.3 mg/kg and 0.6 mg/kg bolus caused a slight, but significant, increase in the serum NE level (Table 1).

Table 1

Table 1

Back to Top | Article Outline

Discussion

In this study, we demonstrated that alphaxalone inhibits NET function in bovine adrenal medullary cells. This is the first report demonstrating that alphaxalone inhibits NET function, suggesting that the NET is one of the sites of neurosteroid action. Sear and Prys-Roberts (19) described the effects of a number of infusion regimens and their associated total drug concentrations. The infusion rates varied between 13.5 and 67.8 μg · kg−1 · min−1, with total drug plasma concentrations of 1.9–3.9 μg/mL. In our study, alphaxalone significantly inhibited NE uptake to 71% of the control value at 10 μM. It has been reported that alphaxalone binds plasma protein, and the quoted plasma protein binding for alphaxalone is approximately 40% in humans (20). From these results, a free drug concentration at clinical concentration would be smaller than 10 μM. However, as much as 80% of NE released from presynaptic terminals is believed to be reuptaken by the neuron (21). Therefore, our present findings suggest that clinically relevant concentrations of alphaxalone suppress NET function.

To examine the site of action of alphaxalone on the NET, we studied the effects of alphaxalone on the kinetic variables for [3H]-NE uptake and [3H]-desipramine binding. Alphaxalone increased the Km for [3H]-NE uptake without altering the Vmax, indicating competitive inhibition. We also examined the effects of alphaxalone on [3H]-desipramine binding to plasma membranes isolated from bovine adrenal medulla. Alphaxalone significantly increased the KD for [3H]-desipramine binding without changing Bmax, suggesting that alphaxalone competitively inhibits [3H]-desipramine binding. Molecular studies of monoamine transporters have produced evidence that distinct, but overlapping, regions within NET molecules determine substrate recognition, translocation, and antagonist affinity (22). In our study, the competitive inhibition of [3H]-NE uptake and [3H]-desipramine binding by alphaxalone suggests that alphaxalone binds to a region that overlaps the sites responsible for NE recognition and antidepressant binding. Recent reports have suggested that two residues located in transmembrane domains (TMD) 6 and 7 of the human NET play an important role in tricyclic antidepressant interaction and that a critical region in TMD 8 is likely involved in the tertiary structure allowing high-affinity binding of tricyclic antidepressants (23). To identify the alphaxalone binding site, it is required to determine whether the mutation of residues in TMD 6, TMD 7, or TMD 8 of the NET abolishes alphaxalone action.

Several lines of evidence have shown that the descending inhibition system consists of noradrenergic neurons (9). The antinociceptive effects of some drugs, such as tricyclic antidepressants, are partially explained by enhanced noradrenergic neurotransmission caused by suppression of the NET in the descending inhibitory system in the brain and spinal cord (24). Moreover, several anesthetics and analgesics, including ketamine and propofol, suppress the NET function (10–12). From our results, the inhibition of NET by alphaxalone could play a partial role in its anesthetic mechanisms. It has previously been reported that alphaxalone produces sedative and anesthetic effects without antinociception in rats (25). In contrast, Gilron and Coderre (26) reported the preemptive analgesic effects of alphaxalone through γ-aminobutyric acid A receptors in the rat formalin test. The antinociception of alphaxalone is still controversial. To clarify the role of the inhibitory effects of alphaxalone on NET in its anesthetic mechanism, it would be interesting to study the analgesic effects of alphaxalone using NET knockout mice.

Alphaxalone had little effect on MAP but slightly increased the serum NE level in rats (Table 1). In contrast to our findings, it has been reported that alphaxalone reduces catecholamine secretion from the dog adrenal medulla (27). This raises the question of how alphaxalone increases serum NE levels. The plasma concentration of NE is the result of a balance between the rate of NE spillover into the circulation and the rate of NE clearance from the circulation; the spillover is determined by the balance between the release and uptake of NE in sympathetic nerve terminals (28). We found that the NE uptake was reduced to 86% at 10 μM of alphaxalone. The neuron is believed to take up approximately 80%–90% of the NE released from presynaptic terminals, terminating neurotransmission (21). Based on this evidence and our findings, we propose that the alphaxalone-induced increase in serum NE levels is predominantly mediated by the inhibition of NE uptake from sympathetic nerve terminals. However, it should be noted that anesthetized rats were used in the present study, and the effects of alphaxalone on NE levels using awake animals have not been studied. It may be required to clarify the effects of alphaxalone on sympathetic nervous activity to study awake rats.

In conclusion, our results suggest that clinically relevant concentrations of alphaxalone inhibit NET function by blocking desipramine-binding and NE recognition sites. These findings add to our understanding of the anesthetic mechanisms of alphaxalone.

Back to Top | Article Outline

References

1. Compagnone NA, Mellon SH. Neurosteroids: biosynthesis and function of these novel neuromodulators. Front Neuroendocrinol 2000; 21: 1–56.
2. Holzbauer M. Physiological aspects of steroids with anaesthetic properties. Med Biol 1976; 54: 227–42.
3. Child KJ, Currie JP, Davis B, et al. The pharmacological properties in animals of CT1341: a new steroid anaesthetic agent. Br J Anaesth 1971; 43: 2–13.
4. Harrison NL, Simmonds MA. Modulation of the GABA receptor complex by a steroid anaesthetic. Brain Res 1984; 323: 287–92.
5. Harrison NL, Vicini S, Barker JL. A steroid anesthetic prolongs inhibitory postsynaptic currents in cultured rat hippocampal neurons. J Neurosci 1987; 7: 604–9.
6. Cottrell GA, Lambert JJ, Peters JA. Modulation of GABAA receptor activity by alphaxalone. Br J Pharmacol 1987; 90: 491–500.
7. Yaksh TL. CNS mechanisms of pain and analgesia. Cancer Surv 1988; 7: 5–28.
8. Jones SL. Descending noradrenergic influences on pain. Prog Brain Res 1991; 88: 381–94.
9. Fields HL, Basbaum AI. Central nervous system mechanisms of pain modulation. In: Wall PD, Melzack R, eds. Textbook of pain. 4th ed. Edinburgh, Scotland: Churchill Livingstone, 1999: 309–29.
10. Minami K, Yanagihara N, Segawa K, et al. Inhibitory effects of propofol on catecholamine secretion and uptake in cultured bovine adrenal medullary cells. Naunyn-Schmiedebergs Arch Pharmacol 1996; 353: 572–8.
11. Hara K, Yanagihara N, Minami K, et al. Ketamine interacts with the noradrenaline transporter at a site partly overlapping the desipramine binding site. Naunyn-Schmiedebergs Arch Pharmacol 1998; 358: 328–33.
12. Hara K, Yanagihara N, Minami K, et al. Dual effects of intravenous anesthetics on the function of norepinephrine transporters. Anesthesiology 2000; 93: 1329–35.
13. Barker EL, Blakely RD. Norepinephrine and serotonin transporters. In: Bloom FE, Kupfer DJ, eds. Psychopharmacology. 4th ed. New York: Raven Press, 1995: 321–33.
14. Pacholczyk T, Blakely RD, Amara SG. Expression cloning of a cocaine- and antidepressant-sensitive human noradrenaline transporter. Nature 1991; 350: 350–4.
15. Lingen B, Bruss M, Bonisch H. Cloning and expression of the bovine sodium- and chloride-dependent noradrenaline transporter. FEBS Lett 1994; 342: 235–8.
16. Bönisch H, Brüss M. The noradrenaline transporter of the neuronal plasma membrane. Ann N Y Acad Sci 1994; 733: 193–202.
17. Yanagihara N, Minami K, Shirakawa F, et al. Stimulatory effect of IL-1 β on catecholamine secretion from cultured bovine adrenal medullary cells. Biochem Biophys Res Commun 1994; 198: 81–7.
18. Minami K, Segawa K, Uezono Y, et al. Adrenomedullin inhibits the pressor effects and decrease in renal blood flow induced by norepinephrine or angiotensin II in anesthetized rats. Jpn J Pharmacol 2001; 86: 159–64.
19. Sear JW, Prys-Roberts C. Plasma concentrations of alphaxalone during continuous infusion of Althesin. Br J Anaesth 1979; 51: 861–5.
20. Child KJ, Gibson W, Harnby G, Hart JW. Metabolism and excretion of Althesin (CT 1341) in the rat. Postgrad Med J 1972; 48: S37–43.
21. Amara SG, Kuhar MJ. Neurotransmitter transporters: recent progress. Annu Rev Neurosci 1993; 16: 73–93.
22. Buck KJ, Amara SG. Chimeric dopamine-norepinephrine transporters delineate structural domains influencing selectivity for catecholamines and 1-methyl-4-phenylpyridinium. Proc Natl Acad Sci USA 1994; 91: 12584–8.
23. Roubert C, Cox PJ, Bruss M, et al. Determination of residues in the norepinephrine transporter that are critical for tricyclic antidepressant affinity. J Biol Chem 2001; 276: 8254–60.
24. Monks R, Merskey H. Psychotropic drugs. In: Wall PD, Melzack R, eds. Textbook of pain. 4th ed. Edinburgh, Scotland: Churchill Livingstone, 1999: 1155–86.
25. Nadeson R, Goodchild CS. Antinociceptive properties of neurosteroids. II. Experiments with Saffan and its components alphaxalone and alphadolone to reveal separation of anaesthetic and antinociceptive effects and the involvement of spinal cord GABA(A) receptors. Pain 2000; 88: 31–9.
26. Gilron I, Coderre TJ. Preemptive analgesic effects of steroid anesthesia with alphaxalone in the rat formalin test: evidence for differential GABA(A) receptor modulation in persistent nociception. Anesthesiology 1996; 84: 572–9.
27. Sumikawa K, Matsumoto T, Amenomori Y, et al. Selective actions of intravenous anesthetics on nicotinic- and muscarinic-receptor-mediated responses of the dog adrenal medulla. Anesthesiology 1983; 59: 412–6.
28. Deegan R, He HB, Wood AJ, et al. Effects of anesthesia on norepinephrine kinetics: comparison of propofol and halothane anesthesia in dogs. Anesthesiology 1991; 75: 481–8.
© 2002 International Anesthesia Research Society