Several mechanisms are responsible for brain dysfunction after intracerebral haemorrhage (ICH) [1,2]. Mechanical disruption of tissue leads to reduction of cerebral blood flow (CBF) [3,4] and cascade of cellular and biochemical reactions such as inflammation [5,6], breakdown of haemoglobin  and complement activation .
In these processes the microglial cells play important role. Although under physiological conditions the central nervous system contains almost exclusively the resting form of microglial cells, even minor pathology may induce morphological transformation of these cells into amoeboid ones. Those cells are able to migrate  and are capable of proliferating and of phagocytosis .
In cases of ICH, a prominent role in pathology is played by substances released from activated microglia, namely tumour necrosis factor-alpha (TNF-α) and interleukin-1 beta (IL-1β). They are responsible for damage to the blood–brain barrier (BBB) .
Only some investigations concerning the reaction of microglial cells in response to experimentally induced ICH are available [12,13]. The qualitative assessment of changes in morphology of astroglial and microglial reaction in the course of the ICH was presented by us . Evolution of microglial response to haemorrhage was established by analysis of the pattern of morphological changes and sequence of antigen expression, e.g. the complement type-3 receptor (CR3/CD11b) and major histocompatibility complex (MHC) class II receptor.
The main goal of therapeutic management of ICH is maintenance of intracranial homoeostasis. Surgical management has been proposed as the treatment of choice for ICH [15,16].
The influence of anaesthetic agents on intracranial homoeostasis as well as their neuroprotective properties are of fundamental importance [17,18].
Volatile anaesthetic agents inhibit various components of polymorphonuclear neutrophil (PMN) and platelet function [19–21]. Kowalski and colleagues  suggested that halothane, isoflurane and sevoflurane reduce PMN adhesion in the reperfused coronary system. According to Heindl and colleagues  this action of sevoflurane is related to the adhesion receptor CR3/CD11b. The fact, that microglial cells show particular immunophenotypical features and behaviour similar to that of other macrophages entitled us to conclude that sevoflurane has an ability to modulate microglial reaction in the course of various pathophysiological processes in the central nervous system. Thus the attempt at evaluating the possible influence of sevoflurane on the reactivity of microglial cells in the course of ICH was performed.
Material and methods
Forty adult male rats (weight 275–350 g; Table 1) were used in the study. Animal care and treatment guidelines outlined by the European Community Council Directive of 24 November 1986 (86/609/ EEC), as well as the local Ethics Committee were followed. Animals were divided into two groups depending on anaesthesia. In all the groups animals were anaesthetized with fentanyl (Fentanyl; Polfa-Warszawa, Poland) 0.02 mg kg−1, dehydrobenzperidol (Droperidol; G. Richter, Hungary) 0.75 mg kg−1 administered intraperitoneally and midazolam maleate (Dormicum; Roche, Switzerland) 0.3 mg kg−1 given intramuscularly every half an hour. No additional agents were given to animals of the control group. In the sevoflurane group animals received sevoflurane 2.2 vol% end-tidal concentration, just after tracheotomy. Sevoflurane (Sevorane; Abbott, UK) was administrated by means of Sigma Elite Vaporizer (Penlon; UK). Inspiratory and end-tidal sevoflurane concentrations were monitored with Cardiocap 5 (Datex-Ohmeda, Finland/USA). Each group was divided into five subgroups (four animals in each subgroup) depending on the length of survival period: 1, 3, 7, 14, 21 days.
Using infiltration anaesthesia with 1% lidocaine (Lignocainum hydrochloricum; Polfa-Warszawa, Poland) tracheotomy was performed using a 16-G cannula (Abbocath 16-G; Abbott, USA). The rats were mechanically ventilated with the Small Animal Ventilator SAR 830/p (CWE Inc., USA), using a mixture of air and oxygen (FiO2 = 0.5). Tidal volume and ventilatory frequency were adjusted to keep end-tidal concentration of CO2 (etCO2) constant between 5.19 and 5.45 kPa and O2 saturation of the haemoglobin (SaO2) above 95%. etCO2 was monitored with a carbon dioxide analyser Capstar 100 (CWE Inc., USA). Continuous monitoring of the SaO2 was performed with a pulse oximeter (Novametix; USA) with sensor placed across the hind foot.
Both femoral artery and vein were cannulated with polyethylene catheters (Abbocath 24-G and 22-G, respectively) for continuous MAP monitoring, blood sampling for biochemical analysis as well as fluid infusion. MAP was monitored with direct blood pressure monitor (Stoelting; USA). The calibration zero point for the transducer (Statham; Stoelting, USA) was set at the level of the external auditory meatus. 0.9% NaCl was administered at the mean rate of 2.5 mL kg−1 h−1 to maintain CPP above 70 mmHg. Body temperature was maintained at 37.0°C (Table 1) using a rectal thermometer and a feedback-controlled heating pad (EST; Stoelting, USA).
MicroSensor ICP Transducer (Codman Johnson & Johnson Medical Ltd, USA) was placed epidurally in the parietal region. For continuous measurement of ICP, the transducer was connected to Neuro Monitor Interface Control Unit (Codman, Johnson & Johnson Medical Ltd, USA) and a Monitoring System 8000 (Simonsen & Weell, Denmark). ICP measurement system was checked by ICP rise following exertion of pressure on the animal's abdomen.
Using an operating microscope, a hole was drilled in the frontal bone on the opposite side to the ICP monitor. The Hamilton syringe connected by the tubing system with 26-G needle, filled with the autologous nonheparinized arterial blood, was placed in the infusion pump. After introducing the needle into the striatum (stereotaxic coordinates: B = 1.2 mm; L = 2.5 mm; H = 5.5 mm), 100 μL of blood was injected over 5 min and the needle was left in place for another 4 min, then it was slowly withdrawn and the skull was sealed.
During production of haematoma, plasma level of glucose and haemoglobin (Table 1) were measured (HEMOCUE AB, Sweden). In order to maintain the comparable physiological conditions during the experiments, continuous monitoring of etCO2, SaO2 and CPP were performed. CPP was calculated from the difference between MAP and ICP (Table 1).
After the experiment the rats were allowed to awake under control of etCO2 and SaO2. Efficient breathing, values of etCO2 below 5.32 kPa and SaO2 above 95%, predisposed to extubation.
The animals were deeply anaesthetized with lethal doses of thiopental sodium (Thiopental; Biochemie GmbH, Austria) 50 mg kg−1, then transcardially perfused with a 0.9% solution of NaCl, followed by a 4% paraformaldehyde solution in 0.1 M phosphate buffer (pH 7.4). The brains were post-fixed in 4% paraformaldehyde fixative for 3–4 h, and then were kept in 0.1 M phosphate buffer containing 15% sucrose (overnight at 4°C) and 30% sucrose (until sunk). Coronal 50-μm-thick serial sections of the brain (five from each animal) were cut with a JUNG 1800 cryostat (Leica, Germany). Free floating sections were blocked with 3% NGS (normal goat serum) containing 0.4% Triton X-100 for 1 h and then were incubated with the monoclonal antibody OX42 recognizing the complement type-3 receptor (CR3/CD11b) (mouse, dilution 1: 50, MCA275G, Serotec; Cambridge, UK) and monoclonal antibody OX6 visualizing MHC class II antigen (mouse, dilution 1: 50, MCA46G, Serotec; Cambridge, UK). After multiple rinses in PBS (phosphate-buffered saline), the sections were incubated (2–3 h, room temperature) with the secondary antibody conjugated with Cy3 (dilution: 1: 500, Jackson Immuno Research, USA) or FITC (dilution: 1: 50, Jackson Immuno Research, USA). The chosen set of brain sections of both the experimental as well as control groups underwent negative control with omission of the primary antibody.
Qualitative assessment of glial activation
Structures as well as the localization and morphology of haematoma were assessed on the sections stained with cresyl violet using Labophot (Nikon, Japan) referring to the stereotactic atlas of the rat brain.
Immunohistochemically labelled sections were analysed using fluorescent microscope BX-51 (Olympus, Japan), and confocal laser scanning μ-Radiance head (Bio-Rad, UK) mounted on an Eclipse E600 microscope (Nikon, Japan). FITC and Cy3 were excited by a monochromatic beam of 488 nm and 514 nm wavelength, respectively. 3D stacks of optical confocal scans were obtained by using high quality 20× and 60× oil lenses and optimal iris in LaserSharp 2000 (Bio-Rad, UK). The distribution of immunopositive cells and their morphological types was determined.
Quantitative assessment of microglial activation
Intensity of microglial reaction was quantified by the number of labelled cells per unit area using the C.A.S.T. grid stereological system (Olympus, Denmark). The reference area bordered by the ventricle, corpus callosum, external capsule and level of rhinal sulcus with the exclusion of haematoma region was marked and measured under small magnification (Fig. 1a). Cells were counted in test frames (area 0.2675 mm2) chosen in systematic random fashion. As a minimum, one third of the reference area was examined. On the basis of their morphology cells were classified into types: amoeboid (Fig. 1c,e) and resting (Fig. 1b,d). Data were then transferred to a spreadsheet and number of cells per unit area calculated. Finally the mean value of the numerical density of glial cells as well as the percentage distribution of their types in each case was estimated.
The significance of differences of values of body weight, glucose, haemoglobin, temperature, etCO2, SaO2, CPP as well as MAP and ICP were assessed by U-test. The influence of anaesthetic agent and period of survival on the numerical density and percentage distribution of microglial cells were tested using analysis of variance followed by LSD (least significant difference) test for planned study of differences between groups and subgroups. 95% confidence intervals for the differences between means were calculated.
Statistical analysis was undertaken with Statistica v. 6.0 (Statsoft, USA) and GraphPad Instat (Graph Pad Inc., USA). Results are expressed as mean ± standard deviation (SD) or mean ± standard error of mean (SE). A P-value less than 0.05 was considered statistically significant.
There were no significant differences in glucose, haemoglobin and temperature between the control and sevoflurane groups (Table 1). During production of ICH, no significant differences in etCO2, SaO2, CPP, MAP and ICP between both groups were observed. The values remained stable in the whole observation period in accordance with the assumptions of the study.
Quantitative assessment of microglial cells activation
In the control group significant differences in the density of OX42-ir microglia (Fig. 2) were present between 3rd and 7th survival day and between 14th and 21st survival day (Fig. 2, Table 2).
Significant increase of percent of the amoeboid form of OX42-ir microglia between 3rd and 7th and between 7th and 14th survival day with a subsequent decrease between 14th and 21st observation days were observed (Fig. 3, Table 2). In the sevoflurane group OX42-ir microglia were not visualized in any of the subgroups.
In the control group there was significant increase of the density of OX6-ir microglial cells only between 3rd and 7th observation days, whereas in the sevoflurane group significant increase of the number of OX6-ir microglia was found between 3rd and 7th observation day and also between 7th and 14th observation day (Fig. 4, Table 3). After that the density of OX6-ir microglia in both groups did not change significantly.
On the third observation day a statistically significant difference in the numerical density of OX6-ir microglia between control and sevoflurane groups was observed (Fig. 4, Table 3).
In the control group a significant increase in the ameboid form of OX6-ir microglia were observed between 1st and 3rd observation day. During the last week of observation a significant decrease was noticed (Fig. 5, Table 3). By contrast, in the sevoflurane group a significant increase in the amoeboid form of OX6-ir microglial cells between 3rd and 7th observation day was found. Then until the end of observation the values of this parameter decreased. There were also significant differences between control and sevoflurane groups on 3rd, 7th and 14th survival days (Fig. 5, Table 3).
To our knowledge, sevoflurane has not been previously investigated as a glial cells antagonist. The main finding in the present study was fact that sevoflurane in clinically relevant concentration inhibits the reaction of the microglial cells in the course of experimental ICH in rats. We have affirmed that sevoflurane delays appearance and decreases the level of microglial cell activation.
The few investigations already published report that volatile anaesthetics have direct effects on PMNs that are likely to change their physiologic properties. Kowalski and colleagues  reported that sevoflurane, isoflurane and halothane in concentrations of 1 and 2 minimum alveolar concentration each inhibited ischaemic-induced PMNs adhesion. This phenomenon can be partially explained by reduced expression of the antigen CR3/CD11b in presence of sevoflurane . Heindl and colleagues suggested that interference of volatile anaesthetics with intracellular messenger systems such as cAMP or cGMP may lead to inhibition of receptor expression through change of its conformation . The present study has also demonstrated that sevoflurane completely inhibited CR3/CD11b expression but on the surface of the microglial cells. Our observation is consistent with results of investigations of Kowalski and colleagues  and Mobert and colleagues . Probably, sevoflurane has a gift for modulation of activation cascade for CR3/CD11b of PMNs and microglial cells.
Microglial cells are the principal immune cells. Activated microglia are regarded as cytotoxic effector cells, which release several potentially cytotoxic substances such as: TNF-α and IL-1β . High level of these mediators in the early acute phase of ICH plays a key role in subsequent enlargement of the haematoma . Moreover, TNF-α and IL-1β may be responsible for marked disruption of BBB, that leads to the development of perihaematomal oedema [6,11]. Sevoflurane inhibits the release of TNF-α and IL-1β by human peripheral mononuclear cells . Capacity of this inhibition may limit the grade of disruption of BBB. Nature and location of the lesion as well as extent of disruption of BBB have influence on response of glial cells . Direct action of volatile anaesthetic agent on microvascular membranes and on the size and shape of intracellular tight junctions limit transfer across BBB [26,27]. It may in part explain the outcome of our investigation.
Mechanical injury of brain tissue caused by ICH leads to impairment of CBF and to ischaemia [3,4]. Sevoflurane has been shown to have many diverse effects during rodent cerebral ischaemia including: altered CBF , a global reduction in metabolic rate [18,28] and changes in neurotransmitter release  as well as reduction in apoptosis . We did not define the main mechanism of action of sevoflurane that was responsible for inhibition of microglial reaction in the course of ICH. We can only speculate that some of these mechanisms may also account for the changes in microglia that we found.
Microglia play a key role in the processes of damage and repair of the central nervous system . It is difficult to evaluate the significance of muted reaction of microglia evoked by sevoflurane on restoration of damaged tissue. Our animal model represents a simulation of what might occur in clinical practice but caution must be exercised in extrapolating these data.
In conclusion, this investigation is the first, which examined the influence of sevoflurane on microglial reaction in the course of ICH. Modulation of inflammatory reaction evoked by sevoflurane may find application not only in neuroanaesthesia but also in the treatment of many diseases. However we should take into account that there are several so far unknown factors that can contribute to the sevoflurane-induced inhibition of microglial cells reactivity.
This research was supported by fund from the Polish State Committee of Scientific Research, Grant No. 4 PO5A 007 18.
1. Altumbabic M, Peeling J, Del Bigio MR. Intracerebral hemorrhage in the rat
: effects of hematoma aspiration. Stroke
2. Jellinger K. Pathology and aetiology of supratentorial haemorrhage. In: Pia HW, Langmaid C, Zierski J, eds. Spontaneous Intracerebral Haematomas. Advances in Diagnosis and Therapy
. Berlin, Heidelberg, New York: Springer-Verlag, 1980: 131–135.
3. Nath FP, Kelly PT, Jenkins A, Mendelov AD, Graham DI, Teasdale GM. Effects of experimental intracerebral hemorrhage on blood flow, capillary permeability, and histochemistry. J Neurosurg
4. Nehls DG, Mendelow AD, Graham DI, Sinar EJ, Teasdale GM. Experimental intracerebral hemorrhage: progression of hemodynamic changes after production of a spontaneous mass lesion. Neurosurgery
5. Hickenbottom SL, Grotta JC, Strong R, Denner LA, Aronowski J. Nuclear factor-kappa B and cell death after experimental intracerebral hemorrhage in rats. Stroke
6. Megyeri P, Abraham CS, Temesvari P. Recombinant human tumor necrosis factor a constrics pial arterioles and increases blood–brain barrier permeability in newborn piglets. Neurosci Lett
7. Xi G, Hua Y, Bhasin R, Emis SR, Keep RF, Hoff JT. Mechanisms of edema formation after intracerebral hemorrhage. Effects of extravasatad red blood cells on blood flow and blood–brain barrier integrity. Stroke
8. Xi G, Hua Y, Keep RF, Younger JG, Hoff JT. Systemic complement depletion diminishes perihematomal brain edema in rats. Stroke
9. Stence N, Waite M, Dailey ME. Dynamic of microglial activation a confocal time-lapse analysis in hipocampal slices. Glia
10. Silva Y, Leira R, Tejada J, Lainez JM, Castillo J, Davalos A. Molecular signatures of vascular injury are associated with early growth of intracerebral hemorrhage. Stroke
11. Holmin S, Mathiesen T. Intracerebral administration of interleukin-1beta and induction of inflammation, apoptosis and vasogenic edema. J Neurosurg
12. Gong C, Hoff JT, Keep RF. Acute inflammatory reaction following experimental intracerebral hemorrhage in rat
. Brain Res
13. Koeppen AH, Dickson AC, McEvoy JA. The cellular reactions to experimental intracerebral hemorrhage. J Neurol Sci
(Suppl) 1995; 134
14. Kowiański P, Karwacki Z, Dziewiatkowski J et al
. Evolution of microglial and astroglial response during experimental intracerebral haemorrhage in the rat
. Folia Neuropathol
15. Saver JL, Hankey G, Hon CS. Surgery for primary intracerebral hemorrhage: meta-analysis of CT-era studies. Stroke
16. Auer LM, Deinsberger W, Neiderkorn K et al
. Endoscopic surgery versus medical treatment for spontaneous intracerebral hematoma: a randomized study. J Neurosurg
17. Bundgaard H, von Oettingen G, Larsen KM et al
. Effects of sevoflurane on intracranial pressure, cerebral blood flow and cerebral metabolism. A dose-response study in patients subjected to craniotomy for cerebral tumors. Acta Anaesthesiol Scand
18. Nakajima Y, Moriwaki G, Ikeda K, Fujise Y. The effects of sevoflurane on recovery of brain energy metabolism after cerebral ischemia in the rat
: a comparison with isoflurane and halothane. Anesth Analg
19. Heindl B, Reihle FM, Zhler S, Conzen PF, Becker BF. Sevoflurane and isoflurane protect the reperfused guinea pig heart by reducing postischemic adhesion of polymorphonuclear neutrophils. Anesthesiology
20. Horn NA, de Rossi L, Robitzsch T, Hecker KE, Hutschenreuter G, Rossaint R. Sevoflurane inhibits unstimulated and agonist-induced platelet antigen expression and platelet function in whole blood in vitro
21. Kowalski Ch, Zahler S, Becker BF et al
. Halothane, isoflurane, and sevoflurane reduce postischemic adhesion of neutrophils in the coronary system. Anesthesiology
22. Mobert J, Zahler S, Becker BF, Conzen PF. Inhibition of neutrophil activation by volatile anesthetic decreases adhesion to cultured human endothelial cells. Anesthesiology
23. Otten U, Marz P, Hesse K, Hock C, Kunz D, Rose-John S. Signals regulating neurotrophin expression in glial cells. In: Castellano-Lopez B, Nieto-Sampedro M, eds. Glial Cell Function
. Amsterdam, London, New York, Oxford, Paris, Shanon, Tokyo: Elsevier, 2001: 545–565.
24. Mitsuhata H, Shimizu R, Yokoyama MM. Suppressive effects of volatile anesthetics on cytokine release in human peripheral blood mononuclear cells. Int J Immunopharmacol
25. Gimenez y Ribotta M, Menet V, Privat A. The role of astrocytes in axonal regeneration in the mannalian CNS. In: Castellano-Lopez B, Nieto-Sampedro M, eds. Glial Cell Function
. Amsterdam, London, New York, Oxford, Paris, Shanon, Tokyo: Elsevier, 2001: 588–610.[net]
26. Brugger B, Bauer A, Finsterer U, Bernasconi P, Kreimeier U, Christ F. Microvasular changes during anesthesia: sevoflurane compared with propofol. Acta Anaesthesiol Scand
27. Chi OZ, Anwar M, Sinha AK, Wei H, Klein SL, Weiss HR. Effects of isoflurane on transport across the blood–brain barrier. Anesthesiology
28. Oshima T, Karasawa F, Okazaki Y, Wada H, Sato T. Effects of sevoflurane on cerebral blood flow and cerebral metabolic rate of oxygen in human beings: a comparison with isoflurane. Eur J Anaesthesiol
29. Vinje ML, Moe MC, Valo ET, Berg-Johnsen J. The effect of sevoflurane on glutamate release and uptake in rat
cerebrocortical presynaptic terminals. Acta Anaesthesiol Scand
30. Engelhard K, Werner C, Eberspacher E et al
. Sevoflurane and propofol influence the expression of apoptosis-regulating proteins after cerebral ischaemia and reperfusion in rats. Eur J Anaesthesiol