Previous investigations have demonstrated that the cerebral effects of thiopental and propofol are similar in many respects: both anesthetics cause a dose-dependent suppression of electroencephalographic activity and cerebral metabolic rate (CMR) (1), both reduce intracranial pressure (ICP), and both have free-radical scavenging activity (2–4). Regarding their effects on the ischemic brain, there is a consensus that thiopental is beneficial for certain anoxic and ischemic insults, whereas propofol may not be (5).
Early studies indicated that the mechanism of thiopental action might be via the diminution of CMR, which slows adenosine triphosphate depletion during ischemia. However, more recent studies argued that the protective effect of thiopental was not caused solely by the diminution of CMR because the degree of protection provided by this anesthetic does not correlate with the magnitude of CMR depression (6). Regarding the effect of propofol on ischemic brain, results are conflicting. Although an early study showed that propofol could improve neurologic outcome in a rat forebrain ischemia model (7), other studies have not shown a beneficial effect (5,8).
Cell swelling is an early sequel to cerebral ischemic insults, developing within seconds or minutes after onset of ischemia (9). Although the mechanisms involved in ischemic brain cell swelling are not well understood, it is thought that cell swelling contributes to cell death and an increase of ICP. The aim of this study was to compare the effects of thiopental and propofol on ischemic neuronal cell swelling by use of an in vitro model of oxygen/glucose deprivation (OGD).
This study was approved by the committee for the guidelines on animal experimentation of Niigata University. The method for hippocampal slice preparation was similar to that described in detail elsewhere (10). Adult male Wistar rats weighing 90–120 g (28–35 days old) were preoxygenated in a Plexiglas chamber for 3 min and anesthetized with 4% isoflurane in oxygen. The rats were decapitated, and the brains were removed rapidly and immersed in ice-cold oxygenated artificial cerebrospinal fluid (ACSF), which had the following composition (in mM): NaCl 117, KCl 3.6, NaH2PO4 1.2, CaCl2 2.5, MgCl2 1.2, NaHCO3 25, and glucose 11.5. Cooled brains were hemisected through the midline and trimmed, leaving a rectangular block of tissue that included the dorsal hippocampus. The tissue block was glued on the stage of a vibrating tissue slicer (DTK-1000; Dosaka, Kyoto, Japan) and cut coronally into 400-μm-thick sections at 4°C. Slices were incubated in ACSF at room temperature (22°C–23°C) for a minimum of 1.5 h before being transferred to a recording chamber.
Each hippocampal slice was transferred into a recording chamber (effective volume 0.3 mL), where it was weighted at the edges with silver wire and superfused continuously with 95% oxygen and 5% CO2-equilibrated ACSF. The flow rate was 6–7 mL/min, and the temperature in the recording chamber was kept within 36.8°C–37.2°C by use of a heating bath. The slices were viewed under an inverted microscope (Diaphot; Nikon Co., Tokyo, Japan) with only a coverslip between the slice and the objective. The levels of CA1 pyramidal cell swelling were continuously monitored by light transmittance (LT) changes by use of a microscopic fluorospectrometer (JASCO CAM-500; Japan Spectroscopic Co., Tokyo, Japan). Excitation light (500 nm) was obtained from a xenon lamp (150 W) equipped with a rotating wheel filter. The intensity of transmitted light through a 500 × 125-μm area of the CA1 pyramidal cell layer detected with a photomultiplier at 1 Hz was initiated at 500 (arbitrary unit). Light intensity changes were recorded simultaneously with a chart writer.
The hypoosmotic solutions were prepared by reducing NaCl from the standard ACSF. For preparation of hyperosmotic solutions, proportional amounts of mannitol were added to the standard ACSF. Slices were exposed to each osmolality for 5 min consecutively before being returned to normoosmotic ACSF. For N-methyl-d-aspartate (NMDA) application, the slices were exposed to NMDA (100 μM) for 30 s. The changes of LT were recorded through the hippocampal CA1 pyramidal layer also. OGD was produced by superfusing slices with ACSF equilibrated with a mixture of 95% nitrogen/5% CO2 and replacing glucose with 11.5 mM sucrose to maintain osmolality. The duration of OGD for the LT experiments was 10 min.
For morphological studies, OGD was induced for 6 min. The temperature in the recording chamber was maintained at the same value used in the LT experiments. To provide normal controls, slices (n = 5) were perfused with ACSF for 20 min. In OGD-exposed slices, slices (n = 5) were perfused with ACSF for 14 min before being subjected to 6 min of OGD. To determine the effects of thiopental and propofol on morphological changes induced by the 6-min OGD, the anesthetics were present 10 min before and during the period of OGD. Before anesthetic application, slices were perfused with ACSF for 4 min. At the end of the 20-min perfusion, slices were immediately fixed in freshly prepared 4% paraformaldehyde (in 0.1 M phosphate-buffered saline, pH 7.4). Four hours later, slices were dehydrated in increasing concentrations of ethanol (50%, 70%, 80%, and 90% for 20 min each; 95% for 20 min twice; and 100% for 20 min three times) for histological processing. Paraffin-embedded slices were serially cut into 7-μm sections and stained with hematoxylin-eosin. The center portions of the slices, considered the most viable part, were examined under a microscope and photographed. For quantitative analysis, films were scanned (LS-2000; Nikon) with Adobe Photoshop™ (version 5.5). A 0.5-mm length of the pyramidal cell layer located in the CA1b area was measured for each slice. Both the longest and shortest diameters were measured, and the two values were averaged for individual neurons. The diameter used to represent each slice was the value averaged from all neurons within the 0.5-mm CA1b area.
For all treatments, drugs were prepared as stock solutions and then dissolved in ACSF solution to achieve final concentrations. Propofol (Aldrich Chemical Co., Milwaukee, WI) was dissolved in dimethyl sulfoxide before being diluted in ACSF. The concentration of dimethyl sulfoxide was 0.1% in the controls and in 40 and 160 μM treated slices. NMDA was purchased from Sigma Chemical Co. (St. Louis, MO). Sodium thiopental was a gift of Tanabe Pharmaceutical Co. (Osaka, Japan). All ACSF solutions had an osmolality of 295 ± 1 mOsm as measured with a vapor point osmometer at a pH of 7.3–7.4.
Data are expressed as mean ± sd and were analyzed with Student’s t-tests or one-way analysis of variance, as appropriate. Analysis of variance was followed by Dunnett’s post hoc test for multiple comparisons. Significant differences were assumed at a P value <0.05.
Forty animals were used, and three to five slices prepared from each animal were used for different treatments. The first series of experiments examined whether cell volume changes induced by osmotic changes or NMDA application were associated with alterations in LT through the hippocampal CA1 pyramidal layer. As shown in Figure 1A, the intensity of LT through the slice increased proportionally after hypoosmotic stimuli (−10, −20, −30, and −40 mOsm); upon reexposure to normoosmotic ACSF, LT returned to basal levels in all five slices examined. In contrast to the application of hypoosmotic solution, LT through hippocampal slices decreased after hyperosmotic stimuli (+10, +20, +30, and +40 mOsm); the recovery after returning slices to normoosmotic ACSF was often incomplete (Fig. 1B). However, exposure of slices to NMDA, a neurotoxic substance known to cause rapid neuronal swelling (11), resulted in a sharp LT increase in the CA1 pyramidal cell layer (Fig. 1C).
These results indicate that cell volume changes induced by the changes of osmolality or NMDA application were associated with LT changes through the hippocampal CA1 layer, implying that continuous LT monitoring is useful for evaluating cell volume changes.
There were no significant changes in LT within 30 min in the slices perfused with ACSF at 36.8°C–37.2°C. After exposure of OGD, LT increased slightly during the first several minutes. A sharp increase in LT occurred around 4.5 min after the onset of OGD. After the peak levels of LT increase occurred, LT showed a stable plateau in most slices.
Pretreatment with thiopental dose-dependently inhibited the magnitude of LT increase (Table 1). The LT increases were significantly reduced at 5–7 min after the onset of OGD by thiopental (400 μM). At this dose, thiopental also prolonged the latency to peak LT increase after the onset of OGD (Fig. 2B). However, the inhibitory effect of thiopental on OGD-induced LT increases was statistically insignificant beyond 7 min after the onset of OGD. Propofol altered neither the magnitude of LT increase nor the latency to the sharp increase after the onset of OGD (Table 2 and Fig. 2C).
In normal control slices, CA1 neuronal perikarya, nuclei, apical dendrites, and the gap between cells had a nearly normal appearance (Fig. 3A). Upon OGD treatment, grossly swollen CA1 neuronal cell bodies and nuclei were observed with some disintegrated somata, a decrease in the gap between cells, and the appearance of a nearly empty cytosolic and nuclear space (Fig. 3B). The histological appearance (Fig. 3C) induced by 6 min of OGD in the presence of thiopental (400 μM) was clearly better than that of slices treated with only OGD. The morphological appearance (Fig. 3D) induced by the 6-min OGD in the presence of propofol (160 μM) was similar to the appearance of OGD-treated slices. The diameters of hippocampal CA1 cells in normal control slices and OGD-treated slices with or without anesthetic pretreatments are shown in Table 3. These results indicate that thiopental, but not propofol, attenuated hippocampal CA1 pyramidal cell swelling induced by in vitro ischemia.
This study demonstrated that thiopental, but not propofol, reduced OGD-induced hippocampal CA1 pyramidal cell swelling in vitro. The hippocampal slice system used in this study provided the capability of directly examining the effects of anesthetics on ischemic neuronal damage. The slice system differs from the in vivo ischemia models in four major respects. First, perfusion by blood flow is absent, and hence interaction with blood-borne elements cannot occur. Second, the bathing medium continues to flow over the slice, so that substances released by the slice (e.g., K+, glutamate, or lactate) are removed from the slice surfaces. Third, the pH buffering ability of the bathing medium is maintained because HCO3− and CO2 contents are kept constant. Fourth, there are no mechanical constraints on the slice, and the slices can freely expand.
In the slice preparation, the increase in LT has been suggested to measure cell swelling (12). This is supported by a study in which the LT changes correlated with another measure of cell volume changes (13). Our results also demonstrated that a rapid increase of LT could be induced by hypoosmotic stimulation and NMDA exposure, indicating that monitoring LT changes is useful for evaluating cell swelling. The limitation of this approach is that if the cellular membrane is already damaged, cells cannot swell further even with a prolonged OGD, and there would not be an increase in LT. This is also why we did not attempt to compare LT changes after reoxygenation.
In this study, we demonstrated that thiopental and propofol had different effects on OGD-induced cell swelling. This is in accordance with a functional study (14) in which thiopental, but not propofol, was found to protect against OGD-induced neurotransmission damage.
Although CMR suppression has long been proposed as a mechanism by which anesthetics protect neurons against ischemia, it seems that reduction in CMR alone was not the cause of attenuating ischemic neuronal swelling. First, a study has shown that thiopental protected neurons against ischemic damage without a reduction of adenosine triphosphate consumption in the slice model (15). Second, although propofol reduces CMR as effectively as thiopental (16), an attenuation of ischemic cell swelling by propofol was not obtained in this study. Therefore, mechanisms other than CMR suppression seem necessary to explain the attenuation of ischemic cell swelling by thiopental.
Ischemic neuronal swelling is associated with rapid or acute cell death that occurs within seconds and minutes after onset of cerebral ischemia. Although the exact mechanisms underlying neuronal swelling remain unclear, evidence has emerged that excitotoxicity is involved (9).
Several lines of evidence indicate that thiopental and propofol modulate glutamate receptors differently, especially the NMDA subtype. Both electrophysiological and intracellular ion-imaging studies have shown that thiopental potently inhibits NMDA receptors (10,17), whereas studies of the effect of propofol on NMDA receptors have produced conflicting results (14,18,19). Despite an earlier investigation demonstrating that propofol slightly inhibited NMDA receptor channels (18), recent functional studies have shown that propofol augmented NMDA receptor activity (14,19). Given the importance of NMDA receptor activation in the development of ischemic neuronal swelling, the differential effects of thiopental and propofol on NMDA receptors may play an important role in the difference between thiopental and propofol on ischemic neuronal swelling.
Beyond the different effects of thiopental and propofol on NMDA receptors, thiopental and propofol may differentially affect ischemia-induced Na+ influx through voltage-dependent sodium channels. Within a few minutes after the onset of ischemia in vitro, there is a massive increase of intracellular sodium (20) that can be partially inhibited by lidocaine, a voltage-dependent sodium channel blocker (21). Results obtained from slice ischemia models have shown that thiopental, but not propofol, was able to reduce intracellular sodium increases in normothermic conditions (15,22,23).
Thiopental has long been used in a variety of clinical conditions to control increased ICP. Similar to thiopental, propofol has been reported to reduce ICP in certain pathological conditions (24,25). On the basis of our results, we propose that different mechanisms may contribute to the decrease of ICP by thiopental and propofol. However, given that the time advantage for thiopental is relatively small, a beneficial effect may be obtained only in subgroups of patients after transient cerebral ischemia during anesthesia.
We conclude that thiopental, but not propofol, offers inhibitory effects on ischemic hippocampal CA1 pyramidal cell swelling in vitro. The differential effects of thiopental and propofol on neuronal cell swelling may underlie their different effects on ischemic neuronal damage.
1. Schwab S, Spranger M, Schwarz S, Hacke W. Barbiturate coma in severe hemispheric stroke: useful or obsolete? Neurology 1997; 48: 1608–13.
2. Smith DS, Rehncrona S, Siesjo BK. Inhibitory effects of different barbiturates on lipid peroxidation in brain tissue in vitro: comparison with the effects of promethazine and chlorpromazine. Anesthesiology 1980; 53: 186–94.
3. Murphy PG, Davies MJ, Columb MO, Stratford N. Effect of propofol and thiopentone on free radical mediated oxidative stress of the erythrocyte. Br J Anaesth 1996; 76: 536–43.
4. Kahraman S, Demiryurek AT. Propofol is a peroxynitrite scavenger. Anesth Analg 1997; 84: 1127–9.
5. Ridenour TR, Warner DS, Todd MM, Gionet TX. Comparative effects of propofol and halothane on outcome from temporary middle cerebral artery occlusion in the rat. Anesthesiology 1992; 76: 807–12.
6. Schmid-Elsaesser R, Schroder M, Zausinger S, et al. EEG burst suppression is not necessary for maximum barbiturate protection in transient focal cerebral ischemia in the rat. J Neurol Sci 1999; 162: 14–9.
7. Kochs E, Hoffman WE, Werner C, et al. The effects of propofol on brain electrical activity, neurological outcome, and neuronal damage following incomplete ischemia in rats. Anesthesiology 1992; 76: 245–52.
8. Tsai YC, Huang SJ, Lai YY, et al. Propofol does not reduce infarct volume in rats undergoing permanent middle cerebral artery occlusion. Acta Anaesthesiol Sin 1994; 32: 99–104.
9. Dirnagl U, Iadecola C, Moskowitz MA. Pathobiology of ischemic stroke: an integrated view [review]. Trends Neurosci 1999; 22: 391–7.
10. Zhan R-Z, Fujiwara N, Yamakura T, et al. Differential inhibitory effects of thiopental, thiamylal and phenobarbital on both voltage-gated calcium channels and NMDA receptors in rat hippocampal slices. Br J Anaesth 1998; 81: 932–9.
11. Colwell CS, Levine MS. Glutamate receptor–induced toxicity in neostriatal cells. Brain Res 1996; 724: 205–12.
12. Lipton P. Effects of membrane depolarization on light scattering by cerebral cortical slices. J Physiol 1973; 231: 365–83.
13. Andrew RD, Adams JR, Polischuk TM. Imaging NMDA- and kainate-induced intrinsic optical signals from the hippocampal slice. J Neurophysiol 1996; 76: 2707–17.
14. Zhan R-Z, Qi S, Wu C, et al. Intravenous anesthetics differentially reduce neurotransmission damage caused by oxygen-glucose deprivation in rat hippocampal slices in correlation with NMDA receptor inhibition. Crit Care Med 2001; 29: 808–13.
15. Kass IS, Abramowicz AE, Cottrell JE, Chambers G. The barbiturate thiopental reduces ATP levels during anoxia but improves electrophysiological recovery and ionic homeostasis in the rat hippocampal slice. Neuroscience 1992; 49: 537–43.
16. Akrawi WR, Drummond JC, Kalkman CJ, Patel PM. A comparison of the electrophysiologic characteristics of EEG burst-suppression as produced by isoflurane, thiopental, etomidate, and propofol. J Neurosurg Anesthesiol 1996; 8: 40–6.
17. Zhan R-Z, Fujiwara N, Endoh H, et al. Thiopental inhibits increases in [Ca2+
induced by membrane depolarization, NMDA receptor activation, and ischemia in rat hippocampal and cortical slices. Anesthesiology 1998; 89: 456–66.
18. Orser BA, Bertlik M, Wang LY, MacDonald JF. Inhibition by propofol (2,6 di-isopropylphenol) of the N-methyl-D-aspartate subtype of glutamate receptor in cultured hippocampal neurones. Br J Pharmacol 1995; 116: 1761–8.
19. Zhu H, Cottrell JE, Kass IS. The effect of thiopental and propofol on NMDA- and AMPA-mediated glutamate excitotoxicity. Anesthesiology 1997; 87: 944–51.
20. Taylor CP, Weber ML, Gaughan CL, et al. Oxygen/glucose deprivation in hippocampal slices: altered intraneuronal elemental composition predicts structural and functional damage. J Neurosci 1999; 19: 619–29.
21. Zhang Y, Lipton P. Cytosolic Ca2+
changes during in vitro ischemia in rat hippocampal slices: major roles for glutamate and Na+
release from mitochondria. J Neurosci 1999; 19: 3307–15.
22. Amorim P, Chambers G, Cottrell J, Kass IS. Propofol reduces neuronal transmission damage and attenuates the changes in calcium, potassium, and sodium during hyperthermic anoxia in the rat hippocampal slice. Anesthesiology 1995; 83: 1254–65.
23. Wang T, Raley-Susman KM, Wang J, et al. Thiopental attenuates hypoxic changes of electrophysiology, biochemistry, and morphology in rat hippocampal slice CA1 pyramidal cells. Stroke 1999; 30: 2400–7.
24. Weinstabl C, Mayer N, Hammerle AF, Spiss CK. The effects of propofol bolus administration on the intracranial pressure in craniocerebral trauma. Anaesthesist 1990; 39: 521–4.
© 2002 International Anesthesia Research Society
25. Nimkoff L, Quinn C, Silver P, Sagy M. The effects of intravenous anesthetics on intracranial pressure and cerebral perfusion pressure in two feline models of brain edema. J Crit Care 1997; 12: 132–6.