Potentials of magnesium treatment in subarachnoid haemorrhage

ARTICLE

Auteur(s) :, Walter M van den Bergh^1,*, Rick M Dijkhuizen², Gabriel JE Rinkel¹

¹Departments of Neurology, Room G03.124, University Medical Centre Utrecht, P.O. Box 85500, 3508 GA Utrecht The Netherlands
²Image Sciences Institute^, University Medical Centre Utrecht, P.O. Box 85500, 3508 GA Utrecht The Netherlands

Introduction

After rupture of an intracranial aneurysm, blood is spouting under arterial pressure into the subarachnoid space. If bleeding continuous, the intracranial pressure rises to the level of the arterial pressure. Despite a reactive increase of the arterial pressure, cerebral perfusion pressure (CPP) decreases. This decrease in CPP is accompanied by acute vasoconstriction and the result is a decrease in cerebral blood flow. This can ultimately result in an intracranial circulatory arrest, which promotes haemostasis, but if it endures, leads to unconsciousness or even death [4-9].

The ischaemia in SAH is biphasic. In the early phase acute cerebral ischaemia occurs, while 4 to 12 days later delayed cerebral ischaemia (DCI) can develop [7, 9]. The actual cause of DCI is unknown, but probably vasospasm play an important role, because many patients have spastic and constricted arteries during the period of DCI. However, vasospasm is not a critical factor for the development of DCI, because many patients with vasospasm do not develop DCI. Also, not all patients with DCI have vasospasm. DCI is an important cause of death and dependency after SAH. Reducing the frequency and consequences of DCI will improve the outcome after SAH.

Current strategies to prevent and treat DCI include neuroprotective drugs (nimodipine) and normovolemia, because fluid restriction has been shown to increase the risk of DCI. However, the improvement in clinical outcome has been modest [10]. In several experimental models of cerebral ischaemia a significant neuroprotective effect of magnesium is demonstrated, with reported infarct reduction of 25 to 61 % [11-13]. Nevertheless, these results have not been confirmed by a large clinical trial [14].

Magnesium sulphate treatment is an established therapy against (pre)eclampsia, a disease sharing many similarities with the development of DCI after SAH [15-19]. This gave us the idea of studying the effect of magnesium treatment in SAH. We have shown in a rat model of SAH that pre-treatment with magnesium sulphate reduces acute cerebral lesion volume by more than 60 % [20].

In this article we discuss the neuroprotective potency of magnesium in SAH by describing the pathophysiology of ischaemia after SAH and the many ways in which magnesium may interfere with this (( figure 1 )).

The ischaemic cascade in SAH – pathophysiology and intervention with magnesium

Excitotoxicity and calcium influx

Immediately after SAH there is a release of glutamate, which results in an increase of the extracellular glutamate concentration of 600 % within 40 minutes [9]. This elevation remains present until 60 minutes after the haemorrhage. Glutamate interacts with the N-methyl-D-aspartate (NMDA) receptor complex, a ligand-gated ion channel for Na⁺, K⁺ and Ca²⁺ with receptor sites for glutamate and specific NMDA-agonists such as glycine. For activation of the NMDA receptor complex the simultaneous stimulation by glutamate and the co-agonist glycine is needed, after the voltage-dependent blockade of the ion channel by Mg²⁺ is conquered. Importantly, overactivation of the NMDA receptor, e.g. as a result of excessive glutamate release, results in pathologically high intracellular Ca²⁺ levels, which can eventually lead to neuronal death [21]. The increase in cytosolic Ca²⁺ concentration is not only caused by an influx of extracellular Ca²⁺, but also by the release of calcium ions from intracellular stores such as the endoplasmic reticulum (ER), mitochondria and calciosomes [22, 23].

In addition, a raised intracellular Ca²⁺ and Na⁺ concentration activates the ATP-ases, which results in expenditure of energy. There is further wasting of ATP by useless pendulating of ions.

There are several treatment opportunities to intervene in the excitotoxic cascade. These include administration of Na⁺ channel blockers, and NMDA and Ca²⁺ antagonists. Nimodipine, the only drug that has proved to be clinically effective in SAH, blocks the extracellular N-type calcium channels.

Alternatively, excitotoxicity may be reduced by treatment with magnesium ions. Mg²⁺ is crucial for maintenance of the normal intra- and extracellular Na⁺ and K⁺ gradient through the NMDA receptor complex (see above). Administration of magnesium causes a concentration-dependent and voltage-dependent reversible decrease of Na⁺ currents and in that manner inhibits the Na⁺ influx, subsequent membrane depolarisation and consequent cellular swelling [24].

We have demonstrated in an experimental model that the duration of ischaemic depolarisations after SAH is substantially reduced after pre-treatment with magnesium sulphate [20]. Magnesium also postpones anoxic depolarisation [25]. This maintenance of the membrane potential may at least partly explain the neuroprotective properties of magnesium in SAH.

Magnesium has also been shown to inhibit the ischaemia-induced raise of glutamate [26, 27]. In addition, magnesium improves the recuperation of ATP deficiency during depolarisation; thereby maintaining the function of the Na⁺-dependent glutamate re-uptake, so that magnesium can contribute to the decrease in extracellular glutamate levels [28].

Within the extracellular space, magnesium is competitive with calcium and reduces the Ca²⁺ influx [29] by blocking both voltage-sensitive and NMDA-activated calcium channels [30, 31]. The blockage of the NMDA receptor by Mg²⁺ is voltage-dependent, but electrophysiological extracellular Mg²⁺ behaves as a non-competitive NMDA antagonist, without the side effects reported from other non-competitive NMDA antagonists [32]. The activated NMDA receptor is not only blocked by extracellular Mg²⁺, but also by intracellular Mg²⁺[33].

Magnesium is not only active in the extracellular space by blocking NMDA and voltage-sensitive calcium channels, but also regulates cytosolic Ca²⁺ homeostasis.

Nitric oxide (NO)

The family of nitric oxide synthases comprises neuronal NOS (nNOS), endothelial (eNOS), and inducible NOS (iNOS). The nNOS and eNOS isoforms are constitutive, Ca²⁺-dependent enzymes that modulate many physiological functions, including the regulation of smooth muscle contraction and blood flow. The iNOS isoform is Ca²⁺-independent, and can be stimulated by stress, inflammation, and infection. In general, nNOS and eNOS release NO in the nM range whereas iNOS, following an induction/latency period, can release NO in the μM range for extended periods of time [34].

The presence of constitutive and inducible forms of NOS suggests that they may have distinct functions, hence, constitutive NOS is responsible for a basal or ‘tonal’ level of NO and this keeps particular types of cells in a state of inhibition – and activation of these cells occurs through disinhibition. It thus seems that a modest amount of NO, produced by the endothelium, works in a neuroprotective manner, while increased concentrations, formed by nNOS and iNOS, are neurotoxic [35-37].

In contrary to focal cerebral ischaemia in which there is an increase in NO [38], 10 minutes after experimental SAH there is a significant decrease of nitric oxide metabolites in the brain with a recuperation in 60-180 minutes. This is not caused by NOS inhibition, but by scavenging of NO, because NO reacts with oxyhaemoglobin in the extravasated blood to generate nitrate and ferric heme [39, 40]. Human studies also show a continuous decline in extracellular NO concentrations after SAH [41]. This initial decline of NO may contribute to cerebral damage in SAH, as there is a tendency for lower NO concentration in patients developing DCI.

A shortage of NO leads, via a reduced cGMP production and consequent reduced endothelial-mediated vasodilatation, to vasoconstriction [5, 9, 42, 43]. Treatment with NO has been shown to attenuate acute vasospasm and improve CBF in experimental SAH [43].

Hypomagnesemia, which frequently occurs after SAH, results in a reduced endothelial NO release. In this way hypomagnesemia can induce vasoconstriction [44, 45]. Magnesium inhibits the elevated nNOS activity of cortical neurones in several experimental ischaemia models [46-48], probably by blocking the NMDA-receptor induced Ca²⁺ influx .

Phospholipids and free radicals

ROS play a detrimental role in the pathophysiological development of brain injuries. In SAH, they cause endothelial damage and intima proliferation, which amplify vasospasms [7].

ROS cause lipid peroxidation of cell membranes resulting in a loss of membrane integrity and a disturbance of ion gradients and increased microvascular permeability. The CNS is particularly susceptible to lipid peroxidation because its membrane lipids are rich in polyunsaturated fatty acids that possess reactive hydrogen. In SAH, lipid peroxidation is catalysed by free iron derived from haemoglobin released from extravasated red blood cells. Disruption of neuronal, glial and vascular membranes inhibits Na⁺/K⁺ ATPase and Ca²⁺ATPase. This increases the cellular influx of Ca²⁺, which activates phospholipase A₂ resulting in a release of arachidonic acid. The production of metabolites such as platelet-activating factor, prostaglandin E₂ and leukotriene B₄ enhances inflammation.

Lipid peroxidation of vascular endothelium also causes the increased permeability of the blood–brain barrier associated with ischaemic injury, and may produce prolonged cerebral vasospasm after SAH.

Magnesium reduces free radical production in the mitochondria, probably more by inhibition of NADP oxidation than by improving the mitochondrial capacity of scavenging free radicals [49]. In addition, magnesium protects endothelial cells against cytotoxicity of radicals [50].

Magnesium ions can bind to phospholipids in the cell membrane and decrease the mobility of the phospholipids [51]. It has been suggested that magnesium can alter membrane permeability and receptor function in this way [52].

Mitochondria

Magnesium is necessary in several enzyme reactions of the glycolysis and Krebs-cyclus. Reception and release of Mg²⁺ in the mitochondrion is a respiratory-dependent and uncoupling-sensitive process. Potassium, ATP, ADP and respiration inhibit the admission of Mg²⁺. The mitochondrial Mg²⁺ concentration seems to be equal to the cytosolic concentration. The transfer of a phosphoryl group to an acceptor, by which a kinase is needed, releases the energy delivered by ATP. Almost all kinases need Mg²⁺ for their activity. ATP is only functional when it is complexed with Mg²⁺.

Magnesium plays an important role in the intra- and extramitochondrial ion homeostasis and due to its calcium-antagonistic effect protects the mitochondrion from calcium overload. Magnesium inhibits the swelling and uncoupling of mitochondria which admitted too much Ca²⁺[53].

Calcium has the potential to overactivate destructive enzymes. Reperfusion can result in restoration of Ca²⁺ homeostasis and complete recovery. However, ischaemic changes of mitochondrial oxidative phosphorylation and production of free radicals could lead to chronic or secondary metabolic failure, oxidative stress and changed cellular Ca²⁺ concentrations, so that vulnerable cells are determined to apoptosis. Apoptosis does occur in SAH and might play an important role in the pathogenesis of vasospasm [54-56].

Under certain circumstances, a mitochondrial permeability transition (MPT) pore in the inner membrane is opened (“assembled”). The extent of opening of the pore is increased by Ca²⁺ and diminished by Mg²⁺. One result of the MPT pore opening is that any mitochondrial calcium load will be rapidly discharged. An MPT pore assembly can lead to mitochondrial dysfunction and cell death [57-63]. Blocking of an MPT has neuroprotective effects. In ischaemic conditions this pore is opened and diminishing of this pore by Mg²⁺ protects mitochondria during ischaemia [64].

Mitochondria malfunction after SAH [65-68], probably due to ischaemia or reperfusion, induces low intracellular pH, high intracellular Ca²⁺ and oxidative stress.

Magnesium has been shown to preserve mitochondrial function under several experimental conditions [26, 64, 69]. Magnesium reduces calcium storage by inhibiting the Ca²⁺ influx and as such preserves the mitochondrial membrane potential without influencing the cytosolic Ca²⁺ and Mg²⁺ concentration during reoxygenation. Magnesium is also capable of preserving the oxidative phosphorylation and in that manner provides enough ATP to maintain the intra- and extramitochondrial balance by calcium pumps. This preservation of mitochondria could be an important reason why magnesium protects postischaemic tissue [70].

Magnesium could also have an important effect on repair and regeneration of neurones and maintenance of membrane structures because Mg²⁺ is an essential ion for synthesis of DNA, protein and energy rich material. Magnesium deficit amplifies apoptosis [71], while magnesium therapy most probably has a preventive role in neuronal apoptosis in neonatal asphyxia [72].

Delayed cerebral ischaemia

Vasospasm

The precise mechanism of vasospasm remains unclarified, but it seems that endothelial mechanisms provide the most prominent contribution to this process and there is growing evidence that the constituents of a subarachnoid blood clot, especially oxyhaemoglobin, seem to be the principle initiating factor. After SAH, large numbers of haemoglobin-containing red blood cells are released into the brain’s parenchyma and subarachnoid space. In the sub-acute phase following SAH, a carefully controlled process without rapid haemolysis removes red blood cells and their contents. Red cell lysis occurs slowly with the release of oxyhaemoglobin [77]. Oxyhaemoglobin produces contraction of cerebral arteries either by a direct contractile effect on smooth muscle cells or by scavenging endothelial nitric oxide [78]. Furthermore, auto-oxidation of oxyhaemoglobin releases free radicals (superoxide, hydrogen peroxide and finally hydroxyl), and degradation of oxyhaemoglobin increases free iron and haeme that, in turn, may cause oxidative injury.

Haemoglobin, oxyhaemoglobin and deoxyhaemoglobin induce severe vasospasm in in vivo animal models and the increase in their concentrations in the human perivascular space and CSF is parallel to the occurrence of DCI [79, 80].

Vasodilatation by magnesium

Vasodilatation by magnesium is probably caused by reducing Ca²⁺ influx and the competitive inhibition of Mg²⁺ to Ca²⁺ at binding sites of the myosin light chain kinase (MLCK) regulated protein calmodulin. When Mg²⁺ is bound to calmodulin it is unable to stimulate MLCK. This results in lower MLCK activity. Magnesium also regulates nuclear and perinuclear Ca²⁺ in cerebrovascular smooth muscle cells, probably by means of nuclear, ER-Golgi and cytoplasmic L-type voltage membrane regulated calcium channels [82].

Magnesium has a vasorelaxing effect in oxyhaemoglobin-induced vasospasm and it ameliorates vasospasm in experimental SAH [83-85]. Magnesium induces a dose-dependent vasodilatation, reduces cerebrovascular tone, increases CBF and protects the metabolism [86-89].

Inflammation

Within 36 hours after SAH there is an increase in the permeability of the blood-brain barrier (BBB), with a peak after 48 hours and normalisation at day 3 [98]. This may eventually lead to vasogenic oedema and secondary brain damage.

From day two on, there is cell migration from the pia mater to the blood clot by subarachnoid macrophages. Macrophage activity is related to free radical production, which causes endothelial damage en stenosis (intima proliferation) [7]. On day 3 blood cells and fibrin are largely cleared and after 5 days the macrophages disappear; they are re-transformed to pia-arachnoid cells.

An important consequence of the inflammatory response and endothelial activation is the production of endothelin by mononuclear leukocytes within 5 days after the haemorrhage [99-101]. In vitro studies show that this can be induced by haemolysis. Endothelin is the most prominent endothelium-derived constricting factor and a very potent vasoconstrictive agent, mediated by endothelial mechanisms. Increased endothelin concentrations after SAH have been associated with endothelial damage, DCI and poor outcome [102].

Magnesium reduces leukocyte activity in vitro and might thus have an inhibiting effect on the immune response, whereas hypomagnesemia stimulates the immune response by increasing macrophage synthesis of the cytokines IL-1β and TNFα, probably through a calcium-mediated mechanism [103]. Probably of great importance in the attenuation of the inflammatory response is the suggestion that magnesium reduces BBB permeability [104].

Magnesium reduces the production of endothelin and completely attenuates the vasoconstrictive effect of endothelin, possibly by voltage-dependent blocking of calcium channels [105, 106].

Platelet aggregation on eicosanoids

During ischaemia, NMDA-receptor stimulation leads to Ca²⁺-dependent phospholipase A₂ activation [107, 108]. Activation of phospholipase A₂ causes the release of arachidonic acid from phospholipids in the cell membrane. Arachidonic acid is rapidly converted to a family of active eicosanoids among which are prostaglandins and thromboxanes. Both prostaglandins and thromboxanes have a complex impact on vascular reactivity, stimulation of the inflammatory response, BBB permeability, and platelet aggregation.

Thromboxane A₂ synthesised from platelets, is a potent vasoconstrictor and stimulates platelet aggregation. Both platelet aggregation and the associated release of thromboxane B₂, the stabile metabolite of activated platelet produced TXA₂, are increased from day 3 after SAH, especially in those patients with symptoms of DCI [109-111].

Magnesium has membrane stabilising and protecting properties due to its electrical effect and inhibition of phospholipase A₂[112]. Magnesium ions can bind to phospholipids in the cell membrane and decrease their mobility. It has been suggested that magnesium can alter membrane permeability and receptor function in this way [113].

Magnesium inhibits many agonists of platelet aggregation and adhesion, like thromboxane A₂ and beta-thromboglobin, most probably due to inhibition of intracellular Ca²⁺ mobilisation [45, 114-117]. As a consequence, magnesium inhibits platelet aggregation and the platelet-dependent thrombus formation, and this effect is independent of, and additive to, that of aspirin [118-121].

Magnesium stimulates synthesis and release of the potent vasodilator prostacyclin and has a vasorelaxing effect in prostaglandin-induced vasospasm [85, 122-125].

Hypomagnesemia in subarachnoid haemorrhage

The cause of hypomagnesemia after SAH is still unclear, but it is most likely caused by an intracellular shift of magnesium ions.

In normal conditions the membrane gradient of Mg²⁺ is modest, but this can change as a result of cellular activities. The most important factors are the concentration of nucleotides and the activity of transport systems in cell and mitochondrial membranes. The slightly higher extracellular Mg²⁺ concentration directs magnesium into cells. Efflux of Mg²⁺, which is against the electrochemical gradient and energy dependent, is accomplished by means of the 2Na⁺/Mg²⁺ pump. An increase of the Mg²⁺ influx and/or a decrease in energy consuming efflux is responsible for 40 % of the decrease in extracellular Mg²⁺ concentration during experimental ischaemia [127].

The elevated levels of catecholamines after SAH may play a mediating role in the increased intracellular shift of Mg²⁺[128, 129]. Catecholamines stimulate lipolysis through the β2-receptor with liberation of free fatty acids. This leads to an intracellular deposit of Mg²⁺ as an insoluble soap [130]. The consequent decrease of intracellular Mg²⁺ may lead to Mg²⁺ influx. It has been demonstrated that adrenaline causes a rapid fall in the plasma magnesium concentrations [131].

Another possible cause for the increased intracellular shift is the glutamate-stimulated Mg²⁺ influx by NMDA activated ion channels, which takes place in the absence of extracellular Na⁺ and Ca²⁺[132].

Intracellular Mg²⁺ levels are indeed increased in SAH. However, 90 % of the intracellular Mg²⁺ is complexed with ATP, and the increase of intracellular Mg²⁺ during ischaemia may also be the result of the release of Mg²⁺ from this complex. ATP binds with Mg²⁺ with an associate constant of 4, while binding affinity with ADP is about 2 times smaller. The cytosolic and mitochondrial Mg²⁺ concentrations will increase in cells with a poor energy state and less ATP [53, 133]. The increase of intracellular Mg²⁺ is even less than might be expected from ATP utilisation, probably because of a disappearance of Mg²⁺ by binding to other cell components.

Within red cells there is an increase in Mg²⁺ concentration after deoxygenation on the basis of the greater affinity of deoxy-Hb than oxy-Hb for the cell Mg²⁺ buffers ATP, ADP and BPG [134]. There is, however, also a direct binding of Mg²⁺ to Hb. Although the interaction is of low affinity, it becomes prominent at high concentrations of Mg²⁺ and Hb. This Mg²⁺ buffer capacity of haemoglobin might be an additional reason why serum magnesium is decreased in SAH [133]. It may be an additional reason for the vasoconstrictive potency of oxyhaemoglobin as hypomagnesemia causes vasoconstriction.

The diminished availability, and subsequent decreased extracellular Mg²⁺ after SAH, results in significantly increased intracellular free Ca²⁺ in cerebral vascular muscle cells, and type-2 astrocytes. This may cause cerebral microvascular constriction, followed by a proinflammatory response, inducing vascular smooth muscle, endothelial and neuronal cell damage [135].

Thus, there might be a causal relation between the decreased availability of magnesium in SAH and the deterioration of the acute cerebral damage and the development of vasospasm and DCI.

Decreased serum magnesium levels after SAH might also be the missing link between SAH and ECG abnormalities. In patients with SAH, lower serum magnesium levels are related to less pronounced increase in the QTc interval [136].

Side effects and toxicity of magnesium therapy

Magnesium therapy in subarachnoid haemorrhage

A pilot study showed that 7 out of 10 patients without magnesium therapy developed angiographic vasospasm compared to 2 out of 13 patients receiving magnesium supplementation to serum levels of 1.0-1.5 mmol/L [140]. Yet another study in 14 patients showed that this had no effect on the total blood volume in the middle cerebral artery [141].

Conclusion

We are currently running a prospective randomised, placebo-controlled, multicentre trial to determine whether intravenous administration of magnesium sulphate reduces the frequency of DCI in patients admitted within 4 days after aneurysmal SAH; MASH: Magnesium and Acetylsalicylic acid in Subarachnoid Haemorrhage (p.m.: acetylsalicylic acid is also tested in this study) [142]. Recruitment started in November 2000 and 283 patients have been included to January 10th, 2004. Preliminary data shows that only 2 % of patients have signs of clinical hypermagnesemia, while no severe side effects occurred. It is expected that the MASH trial will report in 2005.

References

2 Hippocrates, Leonicenus N, Laurentianus L. (trans.) Aphorismi, cum Galeni Commentariis; Praedictiones, cum Galeni Commentariis. Paris: Simon Sylvius; 1527.

3 Linn FH, Rinkel GJ, Algra A, Van Gijn J. Incidence of subarachnoid hemorrhage: role of region, year, and rate of computed tomography: a meta-analysis. Stroke 1996; 27: 625-9.

4 Alkan T, Tureyen K, Ulutas M, Kahveci N, Goren B, Korfali E, Ozluk K. Acute and delayed vasoconstriction after subarachnoid hemorrhage: local cerebral blood flow, histopathology, and morphology in the rat basilar artery. Arch Physiol Biochem 2001; 109: 145-53.

5 Bederson JB, Germano IM, Guarino L. Cortical blood flow and cerebral perfusion pressure in a new noncraniotomy model of subarachnoid hemorrhage in the rat. Stroke 1995; 26: 1086-91.

6 Grote E, Hassler W. The critical first minutes after subarachnoid hemorrhage. Neurosurgery 1988; 22: 654-61.

7 Jackowski A, Crockard A, Burnstock G, Russell RR, Kristek F. The time course of intracranial pathophysiological changes following experimental subarachnoid haemorrhage in the rat. J Cereb Blood Flow Metab 1990; 10: 835-49.

8 Prunell GF, Mathiesen T, Diemer NH, Svendgaard NA. Experimental subarachnoid hemorrhage: subarachnoid blood volume, mortality rate, neuronal death, cerebral blood flow, and perfusion pressure in three different rat models. Neurosurgery 2003; 52: 165-75.

9 Bederson JB, Levy AL, Ding WH, Kahn R, DiPerna CA, Jenkins AL, Vallabhajosyula P. Acute vasoconstriction after subarachnoid hemorrhage. Neurosurgery 1998; 42: 352-60.

10 Hop JW, Rinkel GJ, Algra A, Van Gijn J. Case-fatality rates and functional outcome after subarachnoid hemorrhage: a systematic review. Stroke 1997; 28: 660-4.

11 Izumi Y, Roussel S, Pinard E, Seylaz J. Reduction of infarct volume by magnesium after middle cerebral artery occlusion in rats. J Cereb Blood Flow Metab 1991; 11: 1025-30.

12 Marinov MB, Harbaugh KS, Hoopes PJ, Pikus HJ, Harbaugh RE. Neuroprotective effects of preischemia intraarterial magnesium sulfate in reversible focal cerebral ischemia. J Neurosurg 1996; 85: 117-24.

13 Muir KW. New experimental and clinical data on the efficacy of pharmacological magnesium infusions in cerebral infarcts. Magnes Res 1998; 11: 43-56.

14 Muir KW, Lees KR, Ford I, Davis S. Magnesium for acute stroke (Intravenous Magnesium Efficacy in Stroke trial): randomised controlled trial. Lancet 2004; 363: 439-45.

15 Belfort MA, Anthony J, Saade GR, Allen Jr. JC. A comparison of magnesium sulfate and nimodipine for the prevention of eclampsia. N Engl J Med 2003; 348: 304-11.

16 Which anticonvulsant for women with eclampsia? Evidence from the Collaborative Eclampsia Trial. Lancet 1995; 345: 1455-63.

17 Do women with pre-eclampsia, and their babies, benefit from magnesium sulphate? The Magpie Trial: a randomised placebo-controlled trial. Lancet 2002; 359: 1877-90.

18 Greene MF. Magnesium sulfate for preeclampsia. N Engl J Med 2003; 348: 275-6.

19 Lucas MJ, Leveno KJ, Cunningham FG. A comparison of magnesium sulfate with phenytoin for the prevention of eclampsia. N Engl J Med 1995; 333: 201-5.

20 Van den Bergh WM, Zuur JK, Kamerling NA, Van Asseldonk JT, Rinkel GJ, Tulleken CA, Nicolay K. Role of magnesium in the reduction of ischemic depolarization and lesion volume after experimental subarachnoid hemorrhage. J Neurosurg 2002; 97: 416-22.

21 Lee JM, Zipfel GJ, Choi DW. The changing landscape of ischaemic brain injury mechanisms. Nature 1999; 399: A7-14.

22 Paschen W, Doutheil J. Disturbances of the functioning of endoplasmic reticulum: a key mechanism underlying neuronal cell injury? J Cereb Blood Flow Metab 1999; 19: 1-18.

23 Choi DW. Calcium: still center-stage in hypoxic-ischemic neuronal death. Trends Neurosci 1995; 18: 58-60.

24 Sang N, Meng Z. Blockade by magnesium of sodium currents in acutely isolated hippocampal CA1 neurons of rat. Brain Res 2002; 952: 218-21.

25 Van der Hel WS, Van den Bergh WM, Nicolay K, Tulleken KA, Dijkhuizen RM. Suppression of cortical spreading depressions after magnesium treatment in the rat. Neuroreport 1998; 9: 2179-82.

26 Lin JY, Chung SY, Lin MC, Cheng FC. Effects of magnesium sulfate on energy metabolites and glutamate in the cortex during focal cerebral ischemia and reperfusion in the gerbil monitored by a dual-probe microdialysis technique. Life Sci 2002; 71: 803-11.

27 Rothman S. Synaptic release of excitatory amino acid neurotransmitter mediates anoxic neuronal death. J Neurosci 1984; 4: 1884-91.

28 Schanne FA, Gupta RK, Stanton PK. 31P-NMR study of transient ischemia in rat hippocampal slices in vitro. Biochim Biophys Acta 1993; 1158: 257-63.

29 Altura BM, Altura BT. Magnesium ions and contraction of vascular smooth muscles: relationship to some vascular diseases. Fed Proc 1981; 40: 2672-9.

30 Mayer ML, Westbrook GL, Guthrie PB. Voltage-dependent block by Mg2+ of NMDA responses in spinal cord neurones. Nature 1984; 309: 261-3.

31 Nowak L, Bregestovski P, Ascher P, Herbet A, Prochiantz A. Magnesium gates glutamate-activated channels in mouse central neurones. Nature 1984; 307: 462-5.

32 Muir KW, Lees KR. Clinical experience with excitatory amino acid antagonist drugs. Stroke 1995; 26: 503-13.

33 Johnson JW, Ascher P. Voltage-dependent block by intracellular Mg2+ of N-methyl-D-aspartate-activated channels. Biophys J 1990; 57: 1085-90.

34 Katusic ZS, Shepherd JT. Endothelium-derived vasoactive factors: II. Endothelium-dependent contraction. Hypertension 1991; 18: III86-III92.

35 Samdani AF, Dawson TM, Dawson VL. Nitric oxide synthase in models of focal ischemia. Stroke 1997; 28: 1283-8.

36 Gahm C, Holmin S, Mathiesen T. Nitric oxide synthase expression after human brain contusion. Neurosurgery 2002; 50: 1319-26.

37 Gahm C, Holmin S, Mathiesen T. Temporal profiles and cellular sources of three nitric oxide synthase isoforms in the brain after experimental contusion. Neurosurgery 2000; 46: 169-77.

38 Kumura E, Kosaka H, Shiga T, Yoshimine T, Hayakawa T. Elevation of plasma nitric oxide end products during focal cerebral ischemia and reperfusion in the rat. J Cereb Blood Flow Metab 1994; 14: 487-91.

39 Sehba FA, Schwartz AY, Chereshnev I, Bederson JB. Acute decrease in cerebral nitric oxide levels after subarachnoid hemorrhage. J Cereb Blood Flow Metab 2000; 20: 604-11.

40 Cooper CE. Nitric oxide and iron proteins. Biochim Biophys Acta 1999; 1411: 290-309.

41 Sakowitz OW, Wolfrum S, Sarrafzadeh AS, Stover JF, Dreier JP, Dendorfer A, Benndorf G, Lanksch WR, Unterberg AW. Relation of cerebral energy metabolism and extracellular nitrite and nitrate concentrations in patients after aneurysmal subarachnoid hemorrhage. J Cereb Blood Flow Metab 2001; 21: 1067-76.

42 Kim P, Schini VB, Sundt Jr. TM, Vanhoutte PM. Reduced production of cGMP underlies the loss of endothelium-dependent relaxations in the canine basilar artery after subarachnoid hemorrhage. Circ Res 1992; 70: 248-56.

43 Sehba FA, Ding WH, Chereshnev I, Bederson JB. Effects of S-nitrosoglutathione on acute vasoconstriction and glutamate release after subarachnoid hemorrhage. Stroke 1999; 30: 1955-61.

44 Pearson PJ, Evora PR, Seccombe JF, Schaff HV. Hypomagnesemia inhibits nitric oxide release from coronary endothelium: protective role of magnesium infusion after cardiac operations. Ann Thorac Surg 1998; 65: 967-72.

45 Shechter M. The role of magnesium as antithrombotic therapy. Wien Med Wochenschr 2000; 150: 343-7.

46 Sun X, Mei Y, Tong E. Effect of magnesium on nitric oxide synthase of neurons in cortex during early period of cerebral ischemia. J Tongji Med Univ 2000; 20: 13-5; 42.

47 Garnier Y, Middelanis J, Jensen A, Berger R. Neuroprotective effects of magnesium on metabolic disturbances in fetal hippocampal slices after oxygen-glucose deprivation: mediation by nitric oxide system. J Soc Gynecol Investig 2002; 9: 86-92.

48 Honda Y. Cellular and molecular biology of ischemic retina. Nippon Ganka Gakkai Zasshi 1996; 100: 937-55.

49 Garcia LA, Dejong SC, Martin SM, Smith RS, Buettner GR, Kerber RE. Magnesium reduces free radicals in an in vivo coronary occlusion-reperfusion model. J Am Coll Cardiol 1998; 32: 536-9.

50 Dickens BF, Weglicki WB, Li YS, Mak IT. Magnesium deficiency in vitro enhances free radical-induced intracellular oxidation and cytotoxicity in endothelial cells. FEBS Lett 1992; 311: 187-91.

51 Ebel H, Gunther T. Magnesium metabolism: a review. J Clin Chem Clin Biochem 1980; 18: 257-70.

52 Trauble H, Eibl H. Electrostatic effects on lipid phase transitions: membrane structure and ionic environment. Proc Natl Acad Sci USA 1974; 71: 214-9.

53 Saris NE, Mervaala E, Karppanen H, Khawaja JA, Lewenstam A. Magnesium. An update on physiological, clinical and analytical aspects. Clin Chim Acta 2000; 294: 1-26.

54 Aoki K, Zubkov AY, Ross IB, Zhang JH. Therapeutic effect of caspase inhibitors in the prevention of apoptosis and reversal of chronic cerebral vasospasm. J Clin Neurosci 2002; 9: 672-7.

55 Nau R, Haase S, Bunkowski S, Bruck W. Neuronal apoptosis in the dentate gyrus in humans with subarachnoid hemorrhage and cerebral hypoxia. Brain Pathol 2002; 12: 329-36.

56 Zubkov AY, Tibbs RE, Clower B, Ogihara K, Aoki K, Zhang JH. Morphological changes of cerebral arteries in a canine double hemorrhage model. Neurosci Lett 2002; 326: 137-41.

57 Zoratti M, Szabo I. The mitochondrial permeability transition. Biochim Biophys Acta 1995; 1241: 139-76.

58 Schinder AF, Olson EC, Spitzer NC, Montal M. Mitochondrial dysfunction is a primary event in glutamate neurotoxicity. J Neurosci 1996; 16: 6125-33.

59 Zamzami N, Hirsch T, Dallaporta B, Petit PX, Kroemer G. Mitochondrial implication in accidental and programmed cell death: apoptosis and necrosis. J Bioenerg Biomembr 1997; 29: 185-93.

60 Zamzami N, Susin SA, Marchetti P, Hirsch T, Gomez-Monterrey I, Castedo M, Kroemer G. Mitochondrial control of nuclear apoptosis. J Exp Med 1996; 183: 1533-44.

61 Green DR, Reed JC. Mitochondria and apoptosis. Science 1998; 281: 1309-12.

62 Kroemer G, Dallaporta B, Resche-Rigon M. The mitochondrial death/life regulator in apoptosis and necrosis. Annu Rev Physiol 1998; 60: 619-42.

63 Mignotte B, Vayssiere JL. Mitochondria and apoptosis. Eur J Biochem 1998; 252: 1-15.

64 Dubinsky JM, Brustovetsky N, Pinelis V, Kristal BS, Herman C, Li X. The mitochondrial permeability transition: the brain’s point of view. Biochem Soc Symp 1999; 66: 75-84.

65 Marzatico F, Gaetani P, Silvani V, Lombardi D, Sinforiani E, Baena R. Experimental isobaric subarachnoid hemorrhage: regional mitochondrial function during the acute and late phase. Surg Neurol 1990; 34: 294-300.

66 Marzatico F, Gaetani P, Baena R, Silvani V, Paoletti P, Benzi G. Bioenergetics of different brain areas after experimental subarachnoid hemorrhage in rats. Stroke 1988; 19: 378-84.

67 Baena R, Gaetani P, Silvani V, Spanu G, Marzatico F. Effect of nimodipine on mitochondrial respiration in different rat brain areas after subarachnoid haemorrhage. Acta Neurochir Suppl (Wien) 1988; 43: 177-81.

68 D’avella D, Cicciarello R, Zuccarello M, Albiero F, Romano A, Angileri FF, Salpietro FM, Tomasello F. Brain energy metabolism in the acute stage of experimental subarachnoid haemorrhage: local changes in cerebral glucose utilization. Acta Neurochir (Wien) 1996; 138: 737-43.

69 Xu M, Dai W, Deng X. Effects of magnesium sulfate on brain mitochondrial respiratory function in rats after experimental traumatic brain injury. Chin J Traumatol 2002; 5: 361-4.

70 Sharikabad MN, Ostbye KM, Brors O. Increased [Mg2+]o reduces Ca2+ influx and disruption of mitochondrial membrane potential during reoxygenation. Am J Physiol Heart Circ Physiol 2001; 281: H2113-H2123.

71 Martin H, Richert L, Berthelot A. Magnesium deficiency induces apoptosis in primary cultures of rat hepatocytes. J Nutr 2003; 133: 2505-11.

72 Turkyilmaz C, Turkyilmaz Z, Atalay Y, Soylemezoglu F, Celasun B. Magnesium pre-treatment reduces neuronal apoptosis in newborn rats in hypoxia-ischemia. Brain Res 2002; 955: 133-7.

73 Kagstrom E, Greitz T, Hanson J, Galera R. Changes in cerebral blood flow after subarachnoid hemorrhage. Excerpta Med Int Congr Series 1966: 629-33.

74 Bergvall U, Steiner L, Forster DM. Early pattern of cerebral circulatory disturbances following subarachnoid haemorrhage. Neuroradiology 1973; 5: 24-32.

75 Bergvall U, Galera R. Time relationship between subarachnoid haemorrhage, arterial spasm, changes in cerebral circulation and posthaemorrhagic hydrocephalus. Acta Radiol Diagn (Stockh) 1969; 9: 229-37.

76 Weir B, Grace M, Hansen J, Rothberg C. Time course of vasospasm in man. J Neurosurg 1978; 48: 173-8.

77 Macdonald RL, Weir BK. A review of hemoglobin and the pathogenesis of cerebral vasospasm. Stroke 1991; 22: 971-82.

78 Cook DA, Vollrath B. Free radicals and intracellular events associated with cerebrovascular spasm. Cardiovasc Res 1995; 30: 493-500.

79 Clark JF, Reilly M, Sharp FR. Oxidation of bilirubin produces compounds that cause prolonged vasospasm of rat cerebral vessels: a contributor to subarachnoid hemorrhage-induced vasospasm. J Cereb Blood Flow Metab 2002; 22: 472-8.

80 Pluta RM, Afshar JK, Boock RJ, Oldfield EH. Temporal changes in perivascular concentrations of oxyhemoglobin, deoxyhemoglobin, and methemoglobin after subarachnoid hemorrhage. J Neurosurg 1998; 88: 557-61.

81 Dietrich HH, Dacey Jr. RG. Molecular keys to the problems of cerebral vasospasm. Neurosurgery 2000; 46: 517-30.

82 Altura BM, Zhang A, Cheng TP, Altura BT. Extracellular magnesium regulates nuclear and perinuclear free ionized calcium in cerebral vascular smooth muscle cells: possible relation to alcohol and central nervous system injury. Alcohol 2001; 23: 83-90.

83 Ram Z, Sadeh M, Shacked I, Sahar A, Hadani M. Magnesium sulfate reverses experimental delayed cerebral vasospasm after subarachnoid hemorrhage in rats. Stroke 1991; 22: 922-7.

84 Pyne GJ, Cadoux-Hudson TA, Clark JF. Magnesium protection against in vitro cerebral vasospasm after subarachnoid haemorrhage. Br J Neurosurg 2001; 15: 409-15.

85 Miura K. Changes in Mg++ concentration of CSF after subarachnoid hemorrhage and Mg++--effects on the contractions of bovine cerebral artery. No Shinkei Geka 1988; 16: 1251-9.

86 Altura BT, Altura BM. Interactions of Mg and K on cerebral vessels--aspects in view of stroke. Review of present status and new findings. Magnesium 1984; 3: 195-211.

87 Seelig JM, Wei EP, Kontos HA, Choi SC, Becker DP. Effect of changes in magnesium ion concentration on cat cerebral arterioles. Am J Physiol 1983; 245: H22-H26.

88 Torregrosa G, Perales AJ, Salom JB, Miranda FJ, Barbera MD, Alborch E. Different effects of Mg2+ on endothelin-1- and 5-hydroxytryptamine-elicited responses in goat cerebrovascular bed. J Cardiovasc Pharmacol 1994; 23: 1004-10.

89 Van den Bergh WM, Algra A, Van der Sprenkel JW, Tulleken CA, Rinkel GJ. Hypomagnesemia after aneurysmal subarachnoid hemorrhage. Neurosurgery 2003; 52: 276-82.

90 Frijns CJ, Rinkel GJ, Castigliego D, Van Gijn J, Sixma JJ, Fijnheer R. Endothelial cell activation after subarachnoid hemorrhage. Neurosurgery 2002; 50: 1223-9.

91 Handa Y, Kubota T, Kaneko M, Tsushida A, Kobayashi H, Kawano H. Expression of intercellular adhesion molecule 1 (ICAM-1) on the cerebral artery following subarachnoid haemorrhage in rats. Acta Neurochir (Wien) 1995; 132: 92-7.

92 Dumont AS, Dumont RJ, Chow MM, Lin CL, Calisaneller T, Ley KF, Kassell NF, Lee KS. Cerebral vasospasm after subarachnoid hemorrhage: putative role of inflammation. Neurosurgery 2003; 53: 123-33.

93 Mathiesen T, Lefvert AK. Cerebrospinal fluid and blood lymphocyte subpopulations following subarachnoid haemorrhage. Br J Neurosurg 1996; 10: 89-92.

94 Mathiesen T, Edner G, Ulfarsson E, Andersson B. Cerebrospinal fluid interleukin-1 receptor antagonist and tumor necrosis factor-alpha following subarachnoid hemorrhage. J Neurosurg 1997; 87: 215-20.

95 Mathiesen T, Andersson B, Loftenius A, Von Holst H. Increased interleukin-6 levels in cerebrospinal fluid following subarachnoid hemorrhage. J Neurosurg 1993; 78: 562-7.

96 Fujii Y, Takeuchi S, Sasaki O, Minakawa T, Koike T, Tanaka R. Serial changes of hemostasis in aneurysmal subarachnoid hemorrhage with special reference to delayed ischemic neurological deficits. J Neurosurg 1997; 86: 594-602.

97 Spallone A, Acqui M, Pastore FS, Guidetti B. Relationship between leukocytosis and ischemic complications following aneurysmal subarachnoid hemorrhage. Surg Neurol 1987; 27: 253-8.

98 Germano A, D’Avella D, Imperatore C, Caruso G, Tomasello F. Time-course of blood-brain barrier permeability changes after experimental subarachnoid haemorrhage. Acta Neurochir (Wien) 2000; 142: 575-80.

99 Bertsch T, Kuehl S, Muehlhauser F, Walter S, Hodapp B, Rossol S, Schmeck J, Ragoschke A, Schutt S, Frintzinger M, Horn P, Vajkoczy P, Wellner MW, Fassbender K. Source of endothelin-1 in subarachnoid hemorrhage. Clin Chem Lab Med 2001; 39: 341-5.

100 Fassbender K, Hodapp B, Rossol S, Bertsch T, Schmeck J, Schutt S, Fritzinger M, Horn P, Vajkoczy P, Wendel-Wellner M, Ragoschke A, Kuehl S, Brunner J, Schurer L, Schmiedek P, Hennerici M. Endothelin-1 in subarachnoid hemorrhage: An acute-phase reactant produced by cerebrospinal fluid leukocytes. Stroke 2000; 31: 2971-5.

101 Fassbender K, Hodapp B, Rossol S, Bertsch T, Schmeck J, Schutt S, Fritzinger M, Horn P, Vajkoczy P, Kreisel S, Brunner J, Schmiedek P, Hennerici M. Inflammatory cytokines in subarachnoid haemorrhage: association with abnormal blood flow velocities in basal cerebral arteries. J Neurol Neurosurg Psychiatry 2001; 70: 534-7.

102 Juvela S. Plasma endothelin and big endothelin concentrations and serum endothelin-converting enzyme activity following aneurysmal subarachnoid hemorrhage. J Neurosurg 2002; 97: 1287-93.

103 Shogi T, Oono H, Nakagawa M, Miyamoto A, Ishiguro S, Nishio A. Effects of a low extracellular magnesium concentration and endotoxin on IL-1beta and TNF-alpha release from, and mRNA levels in, isolated rat alveolar macrophages. Magnes Res 2002; 15: 147-52.

104 Kaya M, Kucuk M, Kalayci RB, Cimen V, Gurses C, Elmas I, Arican N. Magnesium sulfate attenuates increased blood-brain barrier permeability during insulin-induced hypoglycemia in rats. Can J Physiol Pharmacol 2001; 79: 793-8.

105 Berthon N, Laurant P, Fellmann D, Berthelot A. Effect of magnesium on mRNA expression and production of endothelin-1 in DOCA-salt hypertensive rats. J Cardiovasc Pharmacol 2003; 42: 24-31.

106 Kemp PA, Gardiner SM, March JE, Rubin PC, Bennett T. Assessment of the effects of endothelin-1 and magnesium sulphate on regional blood flows in conscious rats, by the coloured microsphere reference technique. Br J Pharmacol 1999; 126: 621-6.

107 Dumuis A, Pin JP, Oomagari K, Sebben M, Bockaert J. Arachidonic acid released from striatal neurons by joint stimulation of ionotropic and metabotropic quisqualate receptors. Nature 1990; 347: 182-4.

108 Dumuis A, Sebben M, Haynes L, Pin JP, Bockaert J. NMDA receptors activate the arachidonic acid cascade system in striatal neurons. Nature 1988; 336: 68-70.

109 Juvela S, Ohman J, Servo A, Heiskanen O, Kaste M. Angiographic vasospasm and release of platelet thromboxane after subarachnoid hemorrhage. Stroke 1991; 22: 451-5.

110 Juvela S, Hillbom M, Kaste M. Platelet thromboxane release and delayed cerebral ischemia in patients with subarachnoid hemorrhage. J Neurosurg 1991; 74: 386-92.

111 Juvela S. Cerebral infarction and release of platelet thromboxane after subarachnoid hemorrhage. Neurosurgery 1990; 27: 929-35.

112 Bara M, Guiet-Bara A, Durlach J. Analysis of magnesium membraneous effects: binding and screening. Magnes Res 1988; 1: 29-34.

113 Trauble H, Eibl H. Electrostatic effects on lipid phase transitions: membrane structure and ionic environment. Proc Natl Acad Sci USA 1974; 71: 214-9.

114 Ravn HB, Vissinger H, Kristensen SD, Husted SE. Magnesium inhibits platelet activity--an in vitro study. Thromb Haemost 1996; 76: 88-93.

115 Ravn HB, Vissinger H, Kristensen SD, Wennmalm A, Thygesen K, Husted SE. Magnesium inhibits platelet activity--an infusion study in healthy volunteers. Thromb Haemost 1996; 75: 939-44.

116 Steurer G, Yang P, Rao V, Mohl W, Glogar D, Smetana R. Acute myocardial infarction, reperfusion injury, and intravenous magnesium therapy: basic concepts and clinical implications. Am Heart J 1996; 132: 478-82.

117 Serebruany VL, Herzog WR, Schlossberg ML, Gurbel PA. Bolus magnesium infusion in humans is associated with predominantly unfavourable changes in platelet aggregation and certain haemostatic factors. Pharmacol Res 1997; 36: 17-22.

118 Ravn HB, Kristensen SD, Vissinger H, Husted SE. Magnesium inhibits human platelets. Blood Coagul Fibrinolysis 1996; 7: 241-4.

119 Adams JH, Mitchell JR. The effect of agents which modify platelet behaviour and of magnesium ions on thrombus formation in vivo. Thromb Haemost 1979; 42: 603-10.

120 Hughes A, Tonks RS. Magnesium, adenosine diphosphate and blood platelets. Nature 1966; 210: 106-7.

121 Bennett JS, Kolodziej MA. Disorders of platelet function. Dis Mon 1992; 38: 577-631.

122 Alborch E, Salom JB, Perales AJ, Torregrosa G, Miranda FJ, Alabadi JA, Jover T. Comparison of the anticonstrictor action of dihydropyridines (nimodipine and nicardipine) and Mg2+ in isolated human cerebral arteries. Eur J Pharmacol 1992; 229: 83-9.

123 Perales AJ, Torregrosa G, Salom JB, Miranda FJ, Alabadi JA, Monleon J, Alborch E. In vivo and in vitro effects of magnesium sulfate in the cerebrovascular bed of the goat. Am J Obstet Gynecol 1991; 165: 1534-8.

124 Watson KV, Moldow CF, Ogburn PL, Jacob HS. Magnesium sulfate: rationale for its use in preeclampsia. Proc Natl Acad Sci USA 1986; 83: 1075-8.

125 Nadler JL, Goodson S, Rude RK. Evidence that prostacyclin mediates the vascular action of magnesium in humans. Hypertension 1987; 9: 379-83.

126 Warren BB, Muizelaar JP, Choi SC. Magnesium’s role in cerebral vasospasm and outcome in patients with aneurysmal subarachnoid hemorrhage. In: Findlay JM, ed. Cerebral Vasospasm. 1993: 401-4.

127 Lee MS, Wu YS, Yang DY, Lee JB, Cheng FC. Significantly decreased extracellular magnesium in brains of gerbils subjected to cerebral ischemia. Clin Chim Acta 2002; 318: 121-5.

128 Kawahara E, Ikeda S, Miyahara Y, Kohno S. Role of autonomic nervous dysfunction in electrocardio-graphic abnormalities and cardiac injury in patients with acute subarachnoid hemorrhage. Circ J 2003; 67: 753-6.

129 Lambert G, Naredi S, Eden E, Rydenhag B, Friberg P. Monoamine metabolism and sympathetic nervous activation following subarachnoid haemorrhage: influence of gender and hydrocephalus. Brain Res Bull 2002; 58: 77-82.

130 Flink EB, Brick JE, Shane SR. Alterations of long-chain free fatty acid and magnesium concentrations in acute myocardial infarction. Arch Intern Med 1981; 141: 441-3.

131 Whyte KF, Addis GJ, Whitesmith R, Reid JL. Adrenergic control of plasma magnesium in man. Clin Sci (Lond) 1987; 72: 135-8.

132 Stout AK, Li-Smerin Y, Johnson JW, Reynolds IJ. Mechanisms of glutamate-stimulated Mg2+ influx and subsequent Mg2+ efflux in rat forebrain neurones in culture. J Physiol 1996; 492(Pt 3): 641-57.

133 Jung DW, Apel L, Brierley GP. Matrix free Mg2+ changes with metabolic state in isolated heart mitochondria. Biochemistry 1990; 29: 4121-8.

134 Raftos JE, Lew VL, Flatman PW. Refinement and evaluation of a model of Mg2+ buffering in human red cells. Eur J Biochem 1999; 263: 635-45.

135 Barbour RL, Gebrewold A, Altura BT, Altura BM. Optical spectroscopy and prevention of deleterious cerebral vascular effects of ethanol by magnesium ions. Eur J Pharmacol 2002; 447: 79-86.

136 Van den Bergh WM, Algra A, Rinkel GJ. Electrocardiographic abnormalities and serum magnesium in patients with Subarachnoid Hemorrhage. Stroke 2004.

137 Veyna RS, Seyfried D, Burke DG, et al. Magnesium sulfate therapy after aneurysmal subarachnoid hemorrhage. J Neurosurg 2002; 96: 510-4.

138 Kirmani JF, Yahia AM, Qureshi AI, Kim SR, Bendok BR, Levy EI, et al. Pilot trial of safety and effectiveness of continuous intravenous magnesium sulfate for prevention of cerebral vasospasm in patients with aneurysmal subarachnoid hemorrhage. American Association of Neurological Surgeans Annual Meeting 2002.

139 Van den Bergh WM, Albrecht KW, Berkelbach van der Sprenkel JW, Rinkel GJ. Magnesium therapy after aneurysmal subarachnoid haemorrhage a dose-finding study for long term treatment. Acta Neurochir (Wien) 2003; 145: 195-9.

140 Chia RY, Hughes RS, Morgan MK. Magnesium: a useful adjunct in the prevention of cerebral vasospasm following aneurysmal subarachnoid haemorrhage. J Clin Neurosci 2002; 9: 279-81.

141 Brewer RP, Parra A, Lynch J, Chilukuri V, Borel CO. Cerebral blood flow velocity response to magnesium sulfate in patients after subarachnoid hemorrhage. J Neurosurg Anesthesiol 2001; 13: 202-6.

142 Major Ongoing Stroke Trials. Stroke 2004; 35: E46-E57.