|Home | About | Journals | Submit | Contact Us | Français|
Magnesium sulfate is used extensively for prevention of eclamptic seizures. Empirical and clinical evidence supports the effectiveness of magnesium sulfate; however, questions remain as to its safety and mechanism. This review summarizes current evidence supporting the possible mechanisms of action and several controversies for magnesium sulfate treatment.
Several mechanisms are presented, including the effects of magnesium sulfate on peripheral and cerebral vasodilation, blood-brain barrier protection, and as an anticonvulsant.
Though the specific mechanisms of action remain unclear, the effect of magnesium sulfate in the prevention of eclampsia is likely multi-factorial. Magnesium sulfate may act as a vasodilator, with actions in the peripheral vasculature or the cerebrovasculature, to decrease peripheral vascular resistance and/or relieve vasoconstriction. Additionally, magnesium sulfate may also protect the blood-brain barrier and limit cerebral edema formation, or it may act through a central anticonvulsant action.
Magnesium sulfate (MgSO4) has been used throughout the 20th century for prevention of eclamptic seizures1, 2 and continues to be used extensively3-5. Empirical evidence supports the effectiveness of MgSO4 in preventing and treating eclamptic seizures1, 6-8, in addition to recent controlled clinical trials5, 9, 10. For eclamptic seizure prophylaxis in preeclamptic women, MgSO4 is superior to phenytoin11, 12, nimodipine13, diazepam14, and placebo9. In the multinational Collaborative Eclampsia Trial, MgSO4 reduced the risk of recurrent seizures in eclamptic women by 52% when compared to diazepam and by 67% when compared to phenytoin15. The publication of these clinical trials significantly increased the use of magnesium sulfate versus other anticonvulsants in the United Kingdom and Ireland where the reported use in preeclampsia increased from 2% to 40%16. In addition, 60% of providers surveyed indicated they would use magnesium as an anticonvulsant for eclampsia in 1998, up from only 2% of eclamptic women who received magnesium sulfate in 199216, 17.
Although the effectiveness of MgSO4 in treating and preventing eclampsia has been established, questions still exist as to its safety. There are concerns regarding the possibility of hypermagnesemia toxicity in eclampsia treatment. Normal serum concentrations of Mg+2 are 1.5-2.5 mEq/L (1.8-3.0 mg/dL), with one-third to one-half bound to plasma proteins18, 19. Total magnesium serum concentrations advocated for the treatment of eclamptic convulsions are 3.5-7 mEq/L (4.2-8.4 mg/dL)2, 20, 21, which can be obtained by administering it intramuscularly (6 g loading dose followed by 2 g/h), intravenously (2-4 g dose up to 1 g/min) or a combination of both6, 18, 22. Areflexia, particularly loss of the patellar deep tendon reflex, has been observed at 8-10 mEq/L, and respiratory paralysis seen at >13 mEq/L6, 18, 22. Progressively higher serum magnesium levels can ultimately lead to cardiac arrest18, 22, 23. Some suggest that using standard infusion protocols may not lead to therapeutic serum magnesium levels in all patients, with 36.2% of patients found to have total serum magnesium lower than 4 mEq/L at 30 minutes after treatment initiation in one study24, though no eclamptic seizures were reported during MgSO4 treatment. In addition, there are reports that in some patients eclamptic seizures do not cease even with elevated levels of MgSO46, 7, 25, suggesting that MgSO4 is not effective in treating all cases of eclampsia.
As technologic advances allow for ionized magnesium to be more readily measured, questions have arisen as to whether it is more appropriate to monitor total serum magnesium or the ionized, physiologically active, form. Studies have shown little correlation between total and ionized magnesium levels, either at baseline prior to treatment or during MgSO4 treatment for preeclampsia19, 24. In preeclamptic patients treated with a loading dose of 4 g intravenously followed by 2 g per hour infusion, it was found that both total and ionized Mg+2 concentrations increased quickly after infusion, but steady-state concentrations for total magnesium were 4.84 ± 0.24 mg/dL, whereas for ionized magnesium it was 2.04 ± 0.14 mg/dL19. Similar results have been found by other groups using the same infusion protocol24. Interestingly, as MgSO4 infusion caused significant increases in ionized Mg+2 levels, serum ionized calcium (Ca+2) concentrations were unchanged26, suggesting that the effect of MgSO4 is not exerted through modulations of ionized calcium levels.
Though the use of MgSO4 is wide-spread and effective, its mechanism of action remains unclear. Several possible mechanisms of action have been proposed, including acting as a vasodilator, with actions either peripherally or in the cerebral circulation to relieve vasoconstriction, protecting the blood-brain barrier (BBB) to decrease cerebral edema formation, and acting as a central anticonvulsant. Each of these possible mechanisms of action are discussed below.
Magnesium is a unique calcium antagonist as it can act on most types of calcium channels in vascular smooth muscle27 and as such would be expected to decrease intracellular calcium. One major effect of decreased intracellular calcium would be inactivation of calmodulin-dependent myosin light chain kinase activity and decreased contraction27, causing arterial relaxation that may subsequently lower peripheral and cerebral vascular resistance, relieve vasospasm, and decrease arterial blood pressure. The vasodilatory effect of MgSO4 has been investigated in a wide variety of vessels. For example, both in vivo and in vitro animal studies have shown that it is a vasodilator of large conduit arteries such as the aorta28, 29, as well as smaller resistance vessels including mesenteric27, 30-32, skeletal muscle27, uterine33, and cerebral arteries27, 30, 34. However, the importance of magnesium-induced vasodilation in the treatment and prevention of eclampsia is not completely understood.
The theory of cerebrovascular vasospasm as the etiology of eclampsia seemed to be reinforced by transcranial Doppler (TCD) studies which suggested that MgSO4 treatment caused dilation in the cerebral circulation35-37 as well as in animal studies that used large cerebral arteries34. However, a vasodilator such as MgSO4 would seem to be a paradoxical treatment choice for eclamptic encephalopathy. Eclampsia is thought to be a form of posterior reversible encephalopathy syndrome (PRES) and similar to hypertensive encephalopathy, in which acute elevations in blood pressure cause forced dilatation of the myogenic vasoconstriction of cerebral arteries and arterioles, increased BBB permeability and edema formation38-40. Studies from our lab have shown that MgSO4 causes concentration-dependent vasodilatation in both cerebral and mesenteric resistance arteries; however, mesenteric arteries were significantly more sensitive to MgSO4, particularly during pregnancy30. The finding of a modest vasodilatory effect in the cerebral circulation are consistent with other findings that MgSO4 treatment caused no significant change in cerebral blood flow (CBF), large cerebral artery diameter, or mean middle cerebral artery velocity as determined by magnetic resonance imaging (MRI)41 and TCD42, 43. Together, these results suggest that the effects of MgSO4 as an eclamptic seizure prophylaxis may be more closely related to an effect on peripheral vascular resistance and lowering of systemic blood pressure than to a direct effect on CBF (Table 1 and Figure 1).
Vascular Effects of Magnesium Sulfate.
Reports of the effects of MgSO4 treatment on arterial blood pressure have been mixed. Hypotensive effects have been noted in various studies particularly with bolus injections2, 36, 44, though the duration of decreased blood pressure was varied. In pregnant rats treated with the nitric oxide synthase inhibitor L-NAME to induce hypertension, combination treatment with MgSO4 resulted in significantly lower blood pressures at term and better neonatal outcomes versus animals treated with L-NAME alone45. However, it has been cautioned that MgSO4 should not be considered primarily an anti-hypertensive agent, as there are other drugs better suited for that purpose in eclampsia, including hydralazine, labetalol, and nifedipine 20, 22.
Several reports have suggested that gestation may influence vascular reactivity to MgSO4 and that sensitivity varies with vascular bed28-30, 33. Human uterine arteries from pregnant patients are three-fold more reactive to MgSO4 than uterine arteries from non-pregnant patients33. In aorta from pregnant and non-pregnant rats, both increased and decreased sensitivity to MgSO4-induced vasodilation have been shown based on the preconstriction agent used for in vitro studies. These studies also suggest that pregnancy may differentially affect receptor versus voltage-operated calcium channels in aortic smooth muscle28. In another study of rat aortic rings, the effect of MgSO4 was dependent on gestation and nitric oxide production such that vasodilation was less at term than during late pregnancy29. Our studies found that while mesenteric resistance arteries showed no change in sensitivity with gestation, posterior cerebral resistance arteries from late-pregnant and postpartum animals were significantly less sensitive to MgSO4-induced vasodilation versus those from nonpregnant animals30. This may be due to gestation-induced changes in the cerebral endothelial vaodilatory mechanisms that have been demonstrated during pregnancy and the postpartum state46.
MgSO4 may have other effects within the vasculature that could also explain its effectiveness in eclampsia (included in Figure 1). Magnesium may act by stimulating production of prostacyclin by endothelial cells causing vasodilation47, or by inhibiting platelet aggregation47, 48. In patients with pregnancy-induced hypertension, MgSO4 treatment significantly decreased circulating levels of angiotensin-converting enzyme49. These actions may attenuate the endothelial dysfunction associated with (pre)eclampsia50-52.
The cerebral endothelium that forms the BBB has unique features compared to the peripheral endothelium including a lack of capillary fenestrations53, a low basal rate of pinocytosis54, 55, and the presence of high electrical resistance tight junctions between adjacent endothelial cells54, 56. Disruption of the BBB can result in vasogenic edema formation, an important component in the clinical picture of eclampsia57, 58. Decreased BBB permeability with MgSO4 treatment has been reported in a variety of animal models of BBB disruption including traumatic brain injury59, septic encephalopathy60, hypoglycemia61, and mannitol injection62. We recently reported MgSO4 treatment decreased BBB permeability in response to acute hypertension in late-pregnant rats63. In addition, several studies have shown that MgSO4 decreases cerebral edema formation after brain injury59, 62, 64-67. Together, these studies importantly suggest that one mechanism by which MgSO4 is effective in eclampsia treatment may be through protection of the BBB and decreased cerebral edema formation.
Several mechanisms of action have been proposed to explain the neuroprotective effects of MgSO4 (Table 2 and Figure 2). Magnesium is a calcium antagonist that acts both intracellularly and extracellularly68, and may act directly on cerebral endothelial cells. It is possible that by acting as a calcium antagonist at the level of the endothelial cell actin cytoskeleton, MgSO4 opposes paracellular movement of solutes through the tight junctions (Figure 2). This hypothesis is supported by several studies which demonstrated that inhibition of myosin light chain (MLC) phosphorylation decreases agonist-induced permeability by inhibiting actin stress fiber contraction69-71. Alternatively, pinocytosis is induced by acute hypertension and may contribute to increased BBB permeability during elevated intravascular pressure.72 MgSO4 treatment may therefore decrease pinocytosis caused by acute hypertension and restrict the movement of water and solutes into the brain by transcellular transport, thereby limiting edema formation and improving clinical outcomes in eclampsia.
Although widely used, there is controversy regarding the use of MgSO4 treatment for neurological conditions, such as eclamptic seizures. Concerns have been raised that MgSO4 treatment may mask the outward signs of convulsions through its action at the neuromuscular junction without treating the cause of the seizure in the central nervous system18, 73. Dose-related depression of neuromuscular transmission has been shown in preeclamptic women receiving traditional MgSO4 therapy74. Studies have also shown that there is little to no change in electroencephalograms obtained during MgSO4 treatment, and minimal signs of central nervous system depression in both normal75 and eclamptic patients25, and in animals76. However, clinical trials have demonstrated the efficacy of MgSO4 in the treatment and prevention of eclamptic seizures versus more traditional anticonvulsant drugs, including phenytoin and diazepam 12, 14, 15.
The possible anticonvulsant activity of magnesium may be related to its role as an N-methyl-d-aspartate (NMDA) receptor antagonist77-79, shown in Table 3 and Figure 3. Seizures are thought to be mediated at least in part by stimulation of glutamate receptors, such as the NMDA receptor79, 80. In rats, systemic magnesium treatment results in a resistance to both electrically stimulated81 and NMDA-induced hippocampal seizures82. In addition, systemic treatment with MgSO4 causes a significant reduction in the NMDA receptor binding capacity in the brain78. Animal studies have also shown that MgSO4 reduces epileptic seizure activity83, though these findings have been challenged due to inadequate controls76.
Magnesium ions must cross the BBB in order to elicit a central anticonvulsant effect. It has been demonstrated in animals that MgSO4 can cross the intact BBB and enter the central nervous system in correlation with the level of serum hypermagnesemia81. Interestingly, seizure activity increases the movement of magnesium into the brain81. Human studies have also shown small but significant increases in cerebrospinal fluid concentrations of MgSO4 after systemic administration2, 84. Conversely, other work has suggested that the BBB prevents changes in brain and cerebrospinal fluid magnesium concentrations85. However, this same group later suggested that even a small amount of magnesium in the central nervous system may suppress cortical neuronal activity86. The possibility remains that acute hypertension that leads to convulsions and BBB disruption may permit MgSO4 to enter the brain parenchyma and act as an anticonvulsant during eclampsia.
A better understanding of the mechanisms of action of MgSO4 could allow for more directed use in the treatment of eclampsia and other brain injury disorders. An interesting area for future studies is the relationship between MgSO4 and cerebral edema formation, as it has been proposed that MgSO4 may limit cerebral edema formation through an effect on aquaporin (AQP) expression. Aquaporin-4 (AQP4) is a water channel protein that has been localized to astrocytic endfeet87, 88 and has also been reported to have a perivascular domain89. Cerebral edema in response to brain injury is associated with an upregulation of AQP4 in the brain90, 91, and it has been suggested that MgSO4 treatment attenuates cerebral edema formation via downregulation of AQP4 expression in astrocytes65, though the mechanism of action has not been delineated. This idea is particularly interesting with respect to eclampsia as cerebral AQP4 expression is significantly increased during pregnancy92.
One of the difficulties in studying preeclampsia and eclampsia is the lack of appropriate animal models, particularly as (pre)eclampsia is a disease specific to bipedal species93. In our lab, we have used a rat model of hypertensive encephalopathy during pregnancy to study the neurologic outcomes of eclampsia, specifically how acute elevations in blood pressure lead to forced dilatation of myogenic vasoconstriction, causing increased blood-brain barrier permeability and subsequent edema formation63, 94. Other animal models of preeclampsia and eclampsia exist, including reduced uterine placental perfusion (RUPP), Dahl Salt-Sensitive rats, nitric oxide synthase inhibition, and exogenous soluble fms-like tyrosine receptor kinase-1 (sFlt-1). These models focus on different aspects of the disease including the impact of placental perfusion, preexisting hypertension, and the significance of endothelial dysfunction, oxidative stress and circulating anti-angiogenic factors. The pros and cons of the different models have been reviewed elsewhere93, 95, all of which provide opportunities to further study the specific actions of MgSO4 for seizure prophylaxis.
MgSO4 has been shown to be an effective treatment option for the prevention of eclampsia. Its mechanism of action is likely multi-factorial, encompassing both vascular and neurological mechanisms. Being a calcium antagonist, its effect on vascular smooth muscle to promote relaxation and vasodilation may have a role in lowering total peripheral vascular resistance. In addition, MgSO4 may have an effect on the cerebral endothelium to limit vasogenic edema by decreasing stress fiber contraction and paracellular permeability via calcium-dependent second messenger systems such as MLC kinase. Lastly, MgSO4 may also act centrally to inhibit NMDA receptors, providing anticonvulsant activity by increasing the seizure threshold. A more complete understanding of the effects of MgSO4 will likely promote safer and more effective treatments of eclampsia.
We gratefully acknowledge the support of the American Heart Association Established Investigator Award (0540081N to M.J.C.), the American Heart Association Northeast Affiliate Research Committee Predoctoral Fellowship (000019871 to A.G.E.), the National Institute of Neurological Disorders and Stroke (R01 NS045940 to M.J.C.), the Totman Medical Research Trust, and the University of Vermont College of Medicine MD/PhD Program.
Anna G. Euser: Research Grant: AHA Northeast Affiliate Predoctoral Fellowship, Amount: >= $10,000
Marilyn J. Cipolla: Research Grant: NIH NS045849, Amount: >= $10,000
AHA EI 0540083, Amount: >= $10,000
Conflicts of Interest and Disclosures: None.