Is magnesium neuroprotective following global and focal cerebral ischaemia? A review of published studies

ARTICLE

Auteur(s) : Bruno P Meloni, Hongdong Zhu, Neville W Knuckey

Centre for Neuromuscular and Neurological Disorders, University of Western Australia, Australian Neuromuscular Research Institute, Department of Neurosurgery, Sir Charles Gairdner Hospital, Nedlands, WA, 6009, Australia

Efficacy of magnesium in global (forebrain) cerebral ischaemia animal models

In the first study by Blair et al. [18], magnesium chloride administered intravenously immediately before ischaemia was not neuroprotective. In fact, CA1 neuronal injury was higher in magnesium treated rats than in rats treated with saline. In this case, the dose of magnesium given (5 mmol/kg/476 mg/kg) would be considered high compared to subsequent studies, and additionally was believed to be responsible for raising serum glucose levels from 150 mg/dL to 220 mg/dL immediately after administration. The authors attributed the increased level of CA1 injury to the high serum glucose levels observed in these rats. When glucose levels were controlled by the simultaneous administration of magnesium and insulin the level of CA1 injury was no different to saline treated controls. One potential confounding factor in this study was the high magnesium dose used, which raised serum magnesium levels 10 fold when measured 5 minutes after infusion.

Tsuda et al. [19] observed a neuroprotective effect with magnesium chloride when applied directly to the CA1 sector (1 μL; 50 mM solution) 10 minutes before ischaemia or 0, 2, 12 and 24 hours after ischaemia, but not at 48 hours post-ischaemia in rats. Interestingly, a lower dose of magnesium chloride (1 μL; 10 mM) administered at the 24 hour time point showed a neuroprotective trend, but was not significant at the p < 0.01 level. In this study a potential confounding factor was the possibility of animals becoming hypothermic after ischaemia, as animal body temperature was not monitored during recovery. The potential for hypothermia may have been further compounded as the animals were re-anaesthetised for post-ischaemia magnesium administration. The implications of post-ischaemia hypothermia confounding the findings of magnesium studies will be discussed later.

Next, Okawa [20] evaluated the effects of magnesium sulphate administered as an intravenous loading dose (0.664 mmol/kg) followed by an infusion (0.332 mmol/kg/h for 3 hours followed by 0.083 mmol/kg/h for 45 hours) immediately after ischaemia in dogs. Magnesium treated dogs showed improvement of post-ischaemic (cerebral) hypoperfusion at 48 hours and improvement in neurological outcome at 7 days. The loading dose and initial infusion dose of magnesium used in this study are relatively high compared with other reports, while the 45 hour infusion dose was low. However, Okawa [20] showed a significant increase in magnesium levels in the CSF in magnesium treated ischaemic dogs after 48 hours. Like the previous study, animal body temperature was not monitored post-ischaemia and histological examination of CA1 neurons was not performed.

Sirin et al. [21] used a subcutaneous dose of magnesium sulphate (600 mg/kg or 5 mmol/kg) administered 48 hours before ischaemia and examined the neurological outcome and CA1 and CA3 neuronal injury 4 days post-ischaemia in rats. Magnesium treated rats showed a slightly higher neurological score at 3 and 4 days post-ischaemia, while CA1 and CA3 neuronal damage was decreased from 47% to 36% and 23% to 14% respectively after 4 days. Although Sirin et al. [21] showed positive outcomes in the magnesium treated rats, they were relatively modest and no measures were taken to ensure normal animal body temperature was maintained post-ischaemia. The authors refer to a previous study reporting that a 600 mg/kg subcutaneous dose of magnesium sulphate can significantly increase brain magnesium levels after 48 hours.

In a study by Milani et al. [22] different subcutaneous doses of magnesium chloride (2.5 mmol/kg, 5.0 mmol/kg, 7.5 mmol/kg) were administered to rats at multiple time points post-ischaemia (1, 2, 24 and 48 hours). The 5 mmol/kg dose was evaluated as a single dose administered 2 hours post-ischaemia. In addition, the single and multiple 5 mmol/kg magnesium doses were combined with the CNS suppressant diazepam. In an attempt to minimise post-ischaemia hypothermia, animal body temperatures were maintained between 37-38°C during the first 3 hours after ischaemia by placing rats in a warming box at 30°C when necessary. Seven days after ischaemia, magnesium treatments alone or in combination with diazepam did not significantly decrease neuronal loss in the subiculum and CA1 regions. Interestingly, body temperature in magnesium treated rats remained relatively normal while rats treated with magnesium and diazepam recorded a drop in body temperature of between 1-2.5°C during the 3 hours post-ischaemia monitoring period. Although the authors did not measure magnesium levels in serum or brain they referred to several studies using similar dosage regimens that indicated that their magnesium doses would have increased levels in the brain, especially after ischaemia.

The next report by Miles et al. [23] originated from our own laboratory and evaluated the neuroprotective efficacy of magnesium sulphate when administered intravenously following global cerebral ischaemia in rats. In this study, magnesium treatment initially involved a loading dose at 0.36 mmol/kg, commencing immediately before ischaemia either alone or followed by an infusion of magnesium at 0.06, 0.12, 0.24 or 0.48 mmol/kg/h over 48 hours. We observed that the loading dose alone increased CA1 neuronal survival from 5% to 33%, while rats receiving the magnesium loading dose followed by the infusions at 0.06, 0.12, 0.24 or 0.48 mmol/kg/h demonstrated 30%, 80%, 16% and 5% CA1 neuronal survival respectively. Based on the dose response data for magnesium on CA1 neuronal survival, the loading dose and 0.12 mmol/kg/h infusion dose was selected for assessment of efficacy in post-ischaemic treatment protocols. In these experiments, the magnesium loading dose followed by the infusion was commenced 4, 8, 12 or 24 hours after cerebral ischaemia and resulted in 82%, 71%, 53% and 33% CA1 neuronal survival respectively.

We also assessed the efficacy of the magnesium 0.36 mmol/kg loading dose and 0.12 mmol/kg/h infusion dose in the form of magnesium chloride when administered immediately before ischaemia or at 8 hours after ischaemia. Magnesium chloride administration before ischaemia resulted in 50% CA1 neuronal survival, while 8 hour post-ischaemia administration resulted in 5% CA1 survival.

Our findings confirmed a neuroprotective effect for magnesium that appeared to follow a dose response pattern. We also confirmed that all doses were capable of increasing serum magnesium levels. Hence, based on these results we hypothesised that the failure of previous magnesium studies may have been due to an inappropriate magnesium dose. There are, however, two issues that need to be addressed with respect to our findings. The first is the unexpected ineffectiveness of magnesium chloride when administered 8 hours post-ischaemia. Secondly, we did not monitor animal body temperature post-ischaemia, allowing the possibility that the combination of magnesium treatment and post-ischaemic hypothermia was responsible for the observed neuroprotective effects. The lack of neuroprotective effect with magnesium chloride when administered 8 hours post-ischaemia (5% CA1 survival) was surprising, since magnesium sulphate had been highly effective (71% CA1 survival). Animals treated with the magnesium chloride loading and infusion dose do not experience hyperglycaemia (unpublished observation), therefore the negative finding could not be attributable to elevated blood glucose levels. One possible explanation is that the animals treated with magnesium sulphate, but not those treated with magnesium chloride, experienced a significant period of post-ischaemic hypothermia. In subsequent experiments [24] we have obtained evidence to support the likelihood that post-ischaemic hypothermia was a confounding factor in our study, and this is discussed below.

Zhou et al. [25] administered magnesium sulphate (16.6 mmol/kg) intraperitoneally 30 minutes before global ischaemia in gerbils. At 12, 24 and 48 hours post-ischaemia neuronal apoptosis was assessed in the hippocampus using TUNEL staining. Immunohistochemistry was also used to assess the levels of Bax, Bcl2 and caspase-3. TUNEL staining and levels of Bax and caspase-3 were significantly reduced in magnesium treated animals. Bcl2 levels remained unchanged in treated and untreated animals. While this study has provided evidence that magnesium treatment reduced the level of pro-apoptotic markers, the neuronal outcome after an extended post-ischaemic period (7 days) is unknown and the possibility of post-ischaemic hypothermia has not been adequately accounted for.

The next two studies [26, 27] that assessed magnesium treatment following global ischaemia are from our laboratory. The first evaluated the efficacy of an intravenous magnesium sulphate loading dose at 0.36 mmol/kg administered immediately before ischaemia followed by an infusion of magnesium at 0.06, 0.12 or 0.24 mmol/kg/h over 48 hours. In these experiments we carefully monitored animal body temperature post-ischaemia to ensure animals did not experience any significant hypothermia. At 7 days post-ischaemia we assessed CA1 neuronal survival and observed that none of the magnesium treatments increased CA1 neuronal survival.

We next assessed the 0.36 mmol/kg loading dose followed by an infusion of magnesium at either 0.12 or 0.24 mmol/kg/h over 48 hours. In these experiments we monitored animal body temperature, but made no attempt to keep animals normothermic. In control and magnesium treated animals that self regulated their body temperature a drop in rectal temperature of 0.5-1.5°C during the immediate 4 hour post-ischaemic period was recorded. Moreover, we observed a significantly increased level of CA1 neuronal survival (34%) in animals treated with the loading dose and the 0.12 mmol/kg/h infusion. Animals treated with the loading dose and the 0.24 mmol/kg/h infusion also had increased neuronal survival (20%), but it did not reach statistical significance, while CA1 survival in saline treated control animals was 5%. Taken together, the results from this study indicate that magnesium is only neuroprotective following global ischaemia when combined with post-ischaemic hypothermia.

To address this issue further, our subsequent study [27] first assessed the efficacy of an intravenous magnesium sulphate loading dose at 0.36 mmol/kg administered immediately before ischaemia followed by an infusion of magnesium at 0.12 mmol/kg/h over 48 hours in rats. Immediately post-ischaemia one group of rats had their body temperature lowered to 35°C for 6 hours, while a second group had their body temperature maintained at 37°C. Animals receiving a 6 hour period of hypothermia demonstrated 9.4% CA1 neuronal survival, whereas animals treated with magnesium alone or magnesium and 6 hours of hypothermia demonstrated 5.1% and 37.9% CA1 neuronal survival respectively. These results are in line with our previous study [24] showing that for magnesium to be effective following global ischaemia it must be associated with post-ischaemic hypothermia.

We next assessed if the same magnesium/hypothermia treatment protocol would still be effective if administration was commenced 2 hours after global cerebral ischaemia. In these experiments rats made hypothermic showed 6.1% CA1 neuronal survival and rats receiving magnesium and hypothermia showed 8.1% CA1 survival. Based on this finding we decided to extend the duration of hypothermia and evaluated the efficacy of magnesium treatment combined with a 12 or 24 hour duration of mild hypothermia (35°C) commencing 2 hours post-ischaemia. Rats receiving 12 or 24 hours of hypothermia alone showed 5% and 43% CA1 neuronal survival respectively. Rats receiving the combination of magnesium and 12 or 24 hours of hypothermia showed 9% and 76% CA1 survival, respectively. In summary, these results show that for magnesium to be effective post-ischaemia it must be combined with a prolonged duration of hypothermia. It should also be noted that while prolonged hypothermia alone (24 hour) was neuroprotective the combination of magnesium and hypothermia was significantly more effective.
Table 1 Summary of studies using magnesium following global cerebral ischaemia.

Efficacy of magnesium in focal cerebral ischaemia animal models

The first study to evaluate magnesium in a focal cerebral ischaemia model was by Izumi et al. [28]. A rat permanent middle cerebral artery occlusion (MCAO) model was utilized and magnesium chloride was administered intraperitoneally (1 mmol/kg) immediately before and 1 hour after MCAO. In a separate group of animals the initial magnesium dose was also administered with insulin. Total infarct volumes in magnesium (122 mm³) and magnesium/insulin (92 mm³) treated rats were significantly reduced compared with saline treated controls (165 mm³). The reduction in infarct volume in rats treated with magnesium and insulin (44%) compared with magnesium alone (26%) is most likely attributable to the neuroprotective effect of insulin. Importantly, the authors measured rectal temperature at 1.5, 4, 24 and 48 hours after MCAO and measurements ranged from 36.7 - 37.7°C during this period.

A second study, published in the form of a conference abstract, Roffe et al. [29] compared magnesium chloride treatment in mice when administered intraperitoneally (1 mmol/kg) immediately after permanent MCAO and again at 1 hour. A second group of mice received magnesium and insulin and controls received saline. After 24 hours infarct volumes did not significantly differ, although there was a trend for larger infarcts in magnesium only treated mice. In addition, magnesium treatment increased oedema in infarcted hemispheres, which was not evident in animals treated with magnesium and insulin.

Marinov et al. [30] compared two doses of magnesium sulphate (0.75 mmol/kg and 0.25 mmol/kg) administered intra-arterially (carotid artery) before MCAO in rats. Two periods of reversible focal ischaemia were used (1.5 or 2 hours) and infarct assessment was at 24 hours. Following 1.5 hours of MCAO the 0.75 and 0.25 mmol/kg magnesium doses reduced infarct volume from 24% to 9% and 17%, respectively. The 0.75 mmol/kg dose also significantly reduced brain edema. Following 2 hours of MCAO, only treatment with the 0.25 mmol/kg magnesium dose was assessed, with cortical infarct volume being reduced from 24% to 19%. Animal body temperature was not monitored during the 24 hour post-ischaemia period.

The next study by Schmid-Elsaesser et al. [31] compared the efficacy of magnesium alone and in combination with the anti-oxidant tirilazad following 1.5 hours of transient MCAO in rats. Magnesium chloride was administered intravenously (1 mmol/kg) before and immediately after ischaemia, and animals were allowed to survive for 7 days. Magnesium treatment reduced total infarct volume by 25%, but this was not statistically significant. However this treatment did significantly reduce cortical infarct volume by 37%. Tirilazad reduced total infarct volume by 48% and the combination of magnesium and tirilazad reduced total infarct volume by 59%. Generally, improved neurological functional outcomes reflected reductions in infarct volume for the different treatments. Animal body temperature was monitored for 1 hour after reperfusion and animals were kept in a warm environment for the first eight post-operative hours (Robert Schmid-Elsaesser personal communication).

Lee et al. [32] compared the efficacy of magnesium and the sodium channel blocker mexiletine, alone and in combination, in a permanent MCAO rat model. Intra-arterial administration of magnesium sulphate (0.75 mmol/kg) 10 minutes prior to MCAO significantly reduced infarct volume. Pre- and early (0.5 hour) post-MCAO intraperitoneal treatment with mexiletine also significantly reduced infarct volume, however combined pre-treatment with magnesium and mexiletine did not. The additative adverse affects on cardio-pulmonary function of combined magnesium/mexiletine treatment was suggested as a reason for a lack of neuroprotection. Animal body temperature was not monitored over the 22-24 hour post-ischaemic period. Interestingly, the authors of this study commented that in preliminary experiments the same magnesium treatment protocol administered intravenously did not reduce infarct volume.

Next, Yang et al. [33] assessed the neuroprotective efficacy of intravenously administered magnesium (0.75 mmol/kg) 2, 6 or 8 hours post MCA embolization in rats. Rat survival significantly increased only in animals treated 2 hours after ischaemia, but improvement in neurological outcome was observed in all magnesium treated groups. Magnesium treatment administered 2 and 6 hours, but not 8 hours post-ischaemia significantly reduced infarct volume. Animal body temperature was not monitored during the 72 hour post-ischaemic period.

Kinoshita et al. [34] administered an intravenous infusion of magnesium sulphate (0.21 mmol/kg) continuously for 2 hours between the onset of MCAO and reperfusion. Twenty-four hours after ischaemia magnesium treated rats exhibited significantly reduced infarct volume. Molecular markers associated with neuronal injury were also generally reduced in the brains of rats treated with magnesium. Animal body temperature was not monitored during the post-ischaemic period.

A gerbil transient focal cerebral ischaemia model was used to assess the effect of magnesium sulphate (0.36 mmol/kg) administered intraperitoneally 30 minutes prior to 60 minutes of ischaemia [35]. Infarct volume was reduced by 38% when measured 24 hours after cerebral ischaemia. Magnesium treated animals also showed better preservation of brain glucose, lactate, pyruvate and glutamate levels compared with untreated controls. In this study post-ischaemic body temperature was not monitored.

Westermaier et al. [36] compared the efficacy of intra-arterial and intravenous magnesium sulphate in a rat model of transient focal ischaemia. Rats were subjected to 90 minutes of MCAO and magnesium sulphate (0.75 mmol/kg) was infused immediately before ischaemia. Both intra-arterial and intravenous treatment improved neurological recovery and although total infarct volume was reduced by ≈25% at 7 days post-ischaemia it was not statistically significant. Animal management post-ischaemia was the same as in the Schmid-Elsaesser et al. study [31].

In a study by Chung et al. [37] the efficacy of magnesium sulphate (0.75 mmol/kg) alone and in combination with FK506 was assessed when administered intraperitoneally 30 minutes prior to permanent focal cerebral ischaemia in the gerbil. Magnesium reduced infarct volume by 25%, FK506 by 41% and the combination of magnesium and FK506 by 50% when measured 24 hours post-ischaemia. Animal body temperature was not monitored in the post-ischaemic period.

In 2004, we were involved in two studies that independently re-evaluated the efficacy of magnesium sulphate when administered immediately prior to transient focal cerebral ischaemia in rats [26]. In the Perth study, MCAO was for 45 minutes and body temperature was controlled during and after ischaemia. In the Canberra study, MCAO was for 2 hours and body temperature was only controlled during ischaemia. Three different doses (0.18, 0.36 or 0.72 mmol/kg) of magnesium were administered intravenously in the Perth study and 2 different doses (0.37 or 0.74 mmol/kg) of magnesium were administered intra-arterially in the Canberra study. No significant differences in total, cortical and striatal infarct volumes between saline and magnesium treated animals were observed when measured 24 or 72 hours after ischaemia in either study. In a subsequent experiment from the Perth laboratory rats were administered with the 0.36 mmol/kg magnesium dose prior to MCAO and allowed to self-regulate their body temperature. Twenty-four hours post-ischaemia infarct volume was reduced by 25%, though it was not statistically significant (unpublished observation).

In a recent study, Westermaier et al. [38] compared the efficacy of different doses of intravenously administered magnesium sulphate following transient focal ischaemia in the rat. Infarct volume was assessed 7 days after ischaemia. Treatments consisted of a single dose (0.75 mmol/kg) immediately before ischaemia, a dual dose before ischaemia (1 mmol/kg) and before reperfusion (1 mmol/kg) and a dose before ischaemia (1 mmol/kg) followed immediately by an infusion (0.5 mmol/kg/h) for 150 minutes. In line with their earlier studies [31, 36], the single magnesium dose, which reduced infarct volume by 31%, was not statistically significant. The dual dose and combined loading and infusion doses of magnesium significantly reduced infarct volume by 32% and 41%, respectively. Animal management post-ischaemia was the same as in the Schmid-Elsaesser et al. study [31].
Table 2 Summary of studies using magnesium following focal cerebral ischaemia^a.

^bAuthors reported that same dose administered IV was ineffective.

Magnesium and stroke clinical trials

Muir and Lees [40] examined the safety and tolerability of magnesium in 60 patients given magnesium sulphate (8 mmol over 15 minutes followed by 65 mmol over 24 hours or 2.7 mmol/h) intravenously within 12 hours of clinically diagnosed acute stroke. Serum magnesium level rose from 0.76 mmol/L to 1.42 mmol/L over 24 hours and remained significantly higher than in the saline placebo group at 48 hours. No differences in blood pressure or adverse events between the magnesium- and placebo-treated patients were observed. They concluded that magnesium is a safe and feasible potential therapy in acute stroke.

In a subsequent trial [41], to optimise the dosing regimen for a multi-center trial, intravenous magnesium as a loading dose of 8, 12 or 16 mmol, followed by a 65 mmol infusion over 24 hours was administered to 25 patients within 24 hours of the onset of clinically diagnosed stroke. There were no obvious effects of magnesium on heart rate, blood pressure or blood glucose. The 16 mmol loading infusion achieved target serum concentrations (1.49 mmol/L) most rapidly. Survival curve analysis found a trend in favour of magnesium, though no significant differences in outcome measures were observed in magnesium and placebo-treated groups.

Lampl et al. [42] performed a placebo-controlled, double-blind study using a magnesium sulphate intravenous loading dose (16mmol) and infusion (6 mmol/h for 5 days), administered within 24 hours of stroke onset. Twenty-one patients received magnesium and 20 patients received placebo (saline); patients were followed for 1 month. Several outcome scales (Orgogozo, Mathew, Rankin) indicated that magnesium treatment had a significant positive effect on patient outcome. Based on their positive results the authors indicated that a larger trial was needed to confirm their findings.

The two earlier clinical studies [40, 41] demonstrated that intravenous administration of magnesium was tolerated well in stroke patients. Due to the sample sizes however, these studies were not powerful enough to detect any favourable clinical outcome. As a result two large multi-centre clinical trials (IMAGES and FAST-MAG) were recommended: the IMAGES trial [16], which is now complete, and the phase III FAST-MAG trial [17], which commenced in 2005.

The IMAGES stroke trial [16] consisted of a randomised trial comprising 2589 patients who each received 16 mmol magnesium intravenously over 15 minute followed by 65 mmol over 24 hours, or matching placebo, within 12 hours of acute stroke. No significant differences in mortality and disability were found between patients treated with magnesium and placebo. Furthermore, magnesium was found to be largely ineffective with benefit only observed in patients suffering lacunar strokes. The FAST-MAG pilot trial [17, 43] showed that paramedic initiation of magnesium treatment can be achieved within 2 hours in 75% of patients, without serious adverse effects. Clinical outcomes were encouraging, but the full potential of early magnesium administration as an acute stroke treatment will need to wait until completion of the phase III FAST-MAG trial.

Why have studies using magnesium following cerebral events produced conflicting results?

Confounding effects of post-ischaemic hypothermia

Besides our own studies [24, 26], only four focal and one global cerebral ischaemia study have paid consideration to post-ischaemic animal body temperature. In a focal ischaemia study by Izumi et al. [28], while no measures were implemented to maintain animal body temperature, recordings were made at 1.5, 4, 24 and 48 hours after ischaemia. Magnesium treated animals experienced a 0.5°C drop in rectal temperature at the 1.5 and 4 hour time measurements and a 26% reduction in total infarct volume after 48 hours. This level of infarct volume reduction is similar to what was obtained in our laboratory when animal body temperature was not actively controlled (unpublished observation and see below). In focal ischaemia experiments originating from Schmid-Elsaesser’s laboratory [31, 36, 38], animals were monitored and maintained normothermic for 1 hour after reperfusion and housed in a warm environment for eight post-operative hours. Continuous monitoring of post-operative body temperature was not performed. Findings from Schmid-Elsaesser’s laboratory have revealed that, depending on the treatment schedule, magnesium administration can result in significant and non-significant reductions in infarct volume. In the global study [22], in order to maintain rat body temperature between 37 and 38°C in the first few hours after ischaemia, rats were housed, when necessary at 30°C. In this study, no reduction in CA1 and subiculum injury was observed.

The possibility of post-ischaemic hypothermia confounding experimental studies with magnesium is further supported by findings obtained in our laboratory [24, 25]. In an initial study evaluating the neuroprotective efficacy of magnesium following global ischaemia we did not monitor or control post-ischaemic body temperature and we observed a neuroprotective effect with magnesium when administered in pre- and post-ischaemia treatment protocols [23]. However, in subsequent studies we demonstrated that magnesium treatment is not neuroprotective following both global and focal cerebral ischaemia under controlled post-ischaemic normothermic conditions. Furthermore, following global ischaemia we observed that if body temperature was not controlled animals became mildly hypothermic during surgical recovery and magnesium treatment significantly reduced CA1 neuronal death. Similarly, we have observed that if body temperature is not controlled following focal ischaemia, infarct volume is reduced by 25%. While not statistically significant, this finding indicates that the presence of hypothermia does have an effect on magnesium’s ability to reduce brain injury.

Additional support for hypothermia confounding magnesium’s neuroprotective effect comes from recent data from our laboratory [27]. When we compared magnesium efficacy in normothermic animals and in animals rendered mildly hypothermic (35°C) for 6 hours immediately post-ischaemia we observed that treatment with magnesium and mild hypothermia together increased CA1 neuronal survival from 5% to 38%. Importantly no neuroprotection was observed in normothermic animals receiving magnesium or in animals rendered hypothermic for 6 hours post-ischaemia. These findings illustrate two important points with respect to hypothermia confounding experimental data. One is that ischaemic control animals, which experience some level of hypothermia, may not show any sign of neuroprotection. The second is that the combination of magnesium and hypothermia has unmasked a neuroprotective effect. To this end, as in other studies [58, 59], we have shown that mild hypothermia (34-35°C) must be prolonged (24 hours) when commenced post-ischaemia to produce a protective effect and that post-ischaemic treatment with magnesium and mild hypothermia is more effective than either treatment alone [27]. Taken together these findings further highlight the need to maintain post-ischaemic animal body temperature in drug evaluation studies and, since in the majority of magnesium/cerebral ischaemia studies this was not done, questions must be raised regarding the validity of the data generated.

Dosage, route and time of magnesium administration

The time of administration is also likely to influence any neuroprotective effects afforded by magnesium. Most studies have administered magnesium before or immediately after cerebral ischaemia, hence in these studies it is likely that increased magnesium levels were present in the brain at the time of ischaemia. Post-ischaemia treatments with magnesium following focal and global cerebral ischaemia have produced positive outcomes, but as mentioned above, the contribution of hypothermia in these studies was not determined. However, even if hypothermia has confounded treatment outcomes, it is encouraging to note that delayed treatment with magnesium can be effective.

Why was IMAGES trial unsuccessful?

The phase III FAST-MAG trial [17], which is currently underway will assess whether field administration of magnesium within 2 hours of stroke onset improves clinical outcomes. Importantly, the FAST-MAG trial will address whether the delayed magnesium treatment that occurred in the IMAGES trial was a reason for its ineffectiveness. However, based on our own assessment of magnesium under normothermic conditions we predict that if FAST-MAG patients are maintained normothermic this trial will also show little benefit. We believe that for magnesium to produce a positive outcome after stroke it needs to be combined with mild hypothermia. Hypothermia induction may only require a reduction in body temperature of 1-2°C involving basic cooling measures in conscious patients. Experiments in our laboratory are currently assessing the neuroprotective effect of combined magnesium/mild hypothermia (35°C) treatments when applied several hours after both global and focal cerebral ischaemia. It is anticipated that our experimental findings will enable better design of future clinical trials to test the efficacy of combined magnesium/mild hypothermia to improve patient outcome following cerebral ischaemia/stroke.

Conclusion

References

2 Altura BT, Altura BM. The role of magnesium in etiology of strokes and cerebrovasospasm. Magnesium 1982; 1: 277-91.

3 Lampl Y, Geva D, Gilad R, Eshel Y, Ronen L, Sarova-Pinhas I. Cerebrospinal fluid magnesium level as a prognostic factor in ischaemic stroke. J Neurol 1998; 245: 584-8.

4 McIntosh TK, Faden AI, Yamakami I, Vink R. Magnesium deficiency exacerbates and pretreatment improves outcome following traumatic brain injury in rats: ³¹P magnetic resonance spectroscopy and behavioural studies. J Neurotrauma 1988; 5: 17-30.

5 Vink R, McIntosh TK, Demediuk P, Weiner MW, Faden AI. Decline in intracellular free Mg²⁺ is associated with irreversible tissue injury after brain trauma. J Biol Chem 1988; 263: 757-61.

6 Helpern JA, Vande Linde AMQ, Welch KMA, Levine SR, Schultz LR, Ordidge RJ, et al. Acute elevation and recovery of intracellular [Mg²⁺] following human focal cerebral ischemia. Neurology 1993; 43: 1577-81.

7 Vande Linde AMQ, Chopp M. Chronic changes in brain Mg²⁺ concentration after forebrain ischemia in the rat. Metab Brain Dis 1991; 6: 199-206.

8 Vink R, Heath DL, McIntosh TK. Acute and prolonged alterations in brain free magnesium following fluid percussion-induced brain trauma in rats. J Neurochem 1996; 66: 2477-83.

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

10 Cotton DB, Hallak M, Janusz C, Irtenkauf SM, Berman RF. Central anticonvulsant effects of magnesium sulphate on N-methyl-D-aspartate-induced seizures. Am J Obstet Gynecol 1993; 168: 974-8.

11 Muir JK, Raghupathi R, Emery DL, Bareyre FM, McIntosh TK. Post-injury magnesium treatment attenuates traumatic brain injury-induced cortical induction of p53 mRNA in rats. Exp Neurol 1999; 159: 584-93.

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

13 Muir KW. Magnesium for neuroprotection in ischaemic stroke. Rationale for use and evidence of effectiveness. CNS Drugs 2001; 15: 921-30.

14 Van den Bergh WM, on behalf of the MASH study group. Magnesium sulfate in aneurysmal subarachnoid hemorrhage. Stroke 2005; 36: 1011-5.

15 Heath DL, Vink R. Neuroprotective effects of MgSO₄ and MgCl₂ in closed head injury: a comparative phosphorous NMR study. J Neurotrauma 1998; 15: 183-9.

16 Intravenous Magnesium Efficacy in Stroke (IMAGES) Study Investigators. Magnesium for acute stroke (Intravenous Magnesium Efficacy in Stroke trial): randomised controlled trial. Lancet 2004; 363: 439-45.

17 Saver JL, Kidwell C, Eckstein M, Starkmam S. Prehospital neuroprotective therapy for acute stroke: results of the field administration of stroke therapy-magnesium (FAST-MAG) pilot trial. Stroke 2004; 35: e106-e108.

18 Blair JL, Warner DS, Todd MM. Effects of elevated plasma magnesium versus calcium on cerebral ischemic injury in rats. Stroke 1989; 20: 507-12.

19 Tsuda T, Kogure K, Nishioka K, Watanabe T. Mg²⁺ administered up to twenty-four hours following reperfusion prevents ischemic damage of the CA1 neurons in the rat hippocampus. Neuroscience 1991; 44: 335-41.

20 Okawa M. Effects of magnesium sulfate on brain damage by complete global brain ischemia (Japanese). Masui. Jap J Anesthesiol 1992; 41: 341-55.

21 Sirin BH, Coskun E, Yilik L, Ortac R, Sirin H, Tetik C. Neuroprotective effects of preischemia subcutaneous magnesium sulfate in transient cerebral ischemia. Eur J Cardiothorac Surg 1998; 14: 82-8.

22 Milani H, Lepri ER, Giordani F, Favero-Filho LA. Magnesium chloride alone or in combination with diazepam fails to prevent hippocampal damage following transient forebrain ischemia. Braz J Med Biol Res 1999; 32: 1285-93.

23 Miles AN, Majda BT, Meloni BP, Knuckey NW. Post-ischemic intravenous administration of magnesium sulfate inhibits hippocampal CA1 neuronal death after transient global ischemia in rats. Neurosurgery 2001; 49: 1443-51.

24 Zhu H-D, Meloni BP, Moore SR, Majda BT, Knuckey NW. Intravenous administration of magnesium is only neuroprotective following transient global ischemia when present with post-ischemic mild hypothermia. Brain Res 2004; 1014: 53-60.

25 Zhou H, Ma Y, Zhou Y, Liu Z, Wang K, Cheng G. Effects of magnesium sulphate on neuron apoptosis and expression of caspase-3, bax and bcl-2 after cerebral ischemia-reperfusion injury. Chin Med J (Engl) 2003; 116: 1532-4.

26 Zhu H-D, Martin RL, Meloni BP, Oltvolgyi C, Moore S, Majda BT, Knuckey NW. Magnesium sulfate fails to reduce infarct volume following transient focal ischemia in rats. Neurosci Res 2004; 49: 347-53.

27 Zhu H-D, Meloni BP, Bojarski C, Knuckey MW, Knuckey NW. Post-ischemic modest hypothermia (35°C) combined with intravenous magnesium is more effective at reducing CA1 death than either treatment used alone following global cerebral ischemia in rats. Exp Neurol 2005; 193: 361-8.

28 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.

29 Roffe C, Thomas L, Fotheringham A, Davies I. The effect of magnesium on infarct size and oedema after middle cerebral artery occlusion. Cerebrovasc Dis 1996; 6(Supplement 2): 42; (Abstract).

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

31 Schmid-Elsaesser R, Zausinger S, Hungerhuber E, Baethmann A, Reulen HJ. Neuroprotective effects of combination therapy with tirilazad and magnesium in rats subjected to reversible focal cerebral ischemia. Neurosurgery 1999; 44: 163-72.

32 Lee EJ, Ayoub IA, Harris FB, Hassan M, Ogilvy CS, Maynard KI. Mexiletine and magnesium independently, but not combined, protect against permanent focal cerebral ischemia in Wistar rats. J Neurosci Res 1999; 58: 442-8.

33 Yang Y, Li Q, Ahmad F, Shuaib A. Survival and histological evaluation of therapeutic window of post-ischemia treatment with magnesium sulphate in embolic stroke model of rat. Neurosci Lett 2000; 285: 119-22.

34 Kinoshita Y, Ueyama T, Senba E, Terada T, Nakai K, Itakura T. Expression of c-fos, heat shock protein 70, neurotrophins, and cyclooxygenase-2 mRNA in response to focal cerebral ischemia/reperfusion in rats and their modification by magnesium sulfate. J Neurotrauma 2001; 18: 435-45.

35 Lin J-Y, Chung S-Y, Lin MC, Cheng FC. Effects of magnesium sulphate 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.

36 Westermaier T, Hungerhuber E, Zausinger S, Baethmann A, Schmid-Elsaesser R. Neuroprotective efficacy of intra-arterial and intravenous magnesium sulphate in a rat model of transient focal cerebral ischemia. Acta Neurochir (Wien) 2003; 145: 393-9.

37 Chung SY, Lin JY, Lin MC, Liu HM, Wang MF, Chung FC. Synergistic efficacy of magnesium sulfate and FK506 on cerebral ischemia-induced infarct volume in gerbil. Med Sic Monit 2004; 10: 105-8.

38 Westermaier T, Zausinger S, Baethmann A, Schmid-Elsaesser R. Dose finding of intravenous magnesium sulphate in focal transient focal cerebral ischemia in rats. Acta Neurochir (Wien) 2005; 147: 525-32.

39 Wester PO, Asplund K, Erkisson S, Hagg E, Lithner F, Strand T. Infusion of magnesium in patients with acute brain infarction. Acta Neurol Scand 1984; 70: 143.

40 Muir KW, Lees KR. A randomised, double blind, placebo-controlled pilot trial of intravenous magnesium sulphate in acute stroke. Stroke 1995; 26: 1183-8.

41 Muir KW, Lees KR. Dose optimization of intravenous magnesium sulphate after acute stroke. Stroke 1998; 29: 918-23.

42 Lampl Y, Gilad R, Geva D, Eshel Y, Sadeh M. Intravenous magnesium sulphate in acute stroke: A randomised double blind study. Clin Neuropharmacol 2001; 24: 11-5.

43 Stroke Trials Directory. www.strokecenter.org/.

44 Busto R, Dietrich WD, Globus M-T, Valdes I, Scheinberg P, Ginsberg MD. Small differences in intra-ischemic brain temperature critically determine the extent of ischemic neuronal injury. J Cereb Blood Flow Metab 1987; 7: 729-38.

45 Coimbra C, Wieloch T. Hypothermia ameliorates neuronal survival when induced 2 hours after ischaemia in the rat. Acta Physiol Scand 1992; 146: 543-4.

46 Coimbra C, Wieloch T. Moderate hypothermia mitigates neuronal damage in the rat when initiated several hours following transient cerebral ischemia. Acta Neuropathol (Berl) 1994; 87: 325-31.

47 Colbourne F, Li H, Buchan AM. Indefatigable CA1 sector neuroprotection with mild hypothermia induced 6 hours after severe forebrain ischemia in rats. J Cereb Blood Flow Metab 1999; 19: 742-9.

48 Corbett D, Nurse S, Colbourne F. Hypothermic neuroprotection: a global ischemia study using 18- to 20-month-old gerbils. Stroke 1997; 28: 2238-43.

49 Dietrich WD, Busto R, Alonso O, Globus M-T, Ginsberg MD. Intra-ischemic but not post ischemic brain hypothermia protects chronically following global forebrain ischemia in rats. J Cereb Blood Flow Metab 1993; 13: 541-9.

50 Welsh FA, Sims RE, Harris VA. Mild hypothermia prevents ischemic injury in gerbil hippocampus. J Cereb Blood Flow Metab 1990; 10: 557-63.

51 Behringer W, Safar P, Kentner R, Wu X, Kagan VE, Radovsky A, Clark RSB, Kochanek PM, Subramanian M, Tyurin VA, Tyurin YY, Tisherman SA. Antioxidant tempol enhances hypothermic cerebral preservation during prolonged cardiac arrest in dogs. J Cereb Blood Flow Metab 2002; 22: 105-17.

52 Buchan A, Pulsinelli WA. Hypothermia but not N-methyl-D-aspartate antagonist, MK-801, attenuates neuronal damage in gerbils subjected to transient global ischemia. J Neurosci 1990; 10: 311-6.

53 Corbett D, Evans S, Thomas C, Wang D, Jonas AR. MK-801 reduced cerebral ischemic injury by inducing hypothermia. Brain Res 1990; 514: 300-4.

54 Dietrich WD, Lin B, Globus M-T, Green EJ, Ginsberg MD, Busto R. Effect of delayed MK-801 (dizocilpine) treatment with or without immediate post-ischemic hypothermia on chronic neuronal survival after global forebrain ischemia in rats. J Cereb Blood Flow Metab 1995; 15: 960-8.

55 Dowden J, Corbett D. Ischemic preconditioning in 18- to 20-month-old gerbils: long term survival with functional outcome measures. Stroke 1999; 30: 1240-6.

56 Kawai K, Okauchi M, Morisaki K, Nagao S. Effects of delayed intraischemic and postischemic hypothermia on a focal model of transient cerebral ischemia in rats. Stroke 2000; 31: 1982-9.

57 Memezawa H, Zhao Q, Smith M-L, Siesjö BK. Hyperthermia nullifies the ameliorating effect of sizocilpine maleate (MK-801) in focal cerebral ischemia. Brain Res 1995; 670: 48-52.

58 Nurse S, Corbett D. Neuroprotection after several days of mild, drug-induced hypothermia. J Cereb Blood Flow Metab 1996; 16: 474-80.

59 Corbett D, Hamilton M, Colbourne F. Persistent neuroprotection with prolonged postischemic hypothermia in adult rats subjected to transient middle cerebral artery occlusion. Exp Neurol 2000; 163: 200-6.