Table 2.

Clinical data regarding the use of magnesium therapy in CVS.

Author/ date	Study type	Mg2+ administration type	N treatment/ placebo	Major results
Mees et al., 2012 [40]	RCT	64 mMol/day (IV) of MgSO4 for 20 days	606/597	Outcomes assessed by the modified Rankin Scale were similar in the treatment and placebo groups.
Wong et al., 2010 [41]	RCT	20 mMol bolus followed by 80 mMol/day (IV) for 14 days	169/158	(i) Outcome measured by GOS was the same at six months after treatment (ii) Incidence of CVS was similar in both groups
Westermaieret al., 2010 [42]	RCT	16 mMol bolus of MgSO4   followed by  8 mMol/hour continuous infusion for 10 days	55/55	(i) Lower incidence of CVS in Mg2+ group (ii) Lower incidence of delayed ischemic infarction in Mg2+ group (iii) DIND nonsignificantly reduced in Mg2+ group, though fewer patients with DIND expressed delayed ischemic infarction
Shah et al., 2009 [43]	Retrospective case series	8–32 mMol (IA) via super-select catheterization	14	(i) No long-term outcomes reported (ii) No cerebral infarction in 12/14 patients
Mori et al., 2009 [44]	Prospective case series	15 mMol/L MgSO4 at 20 mL/hour (intracisternal) for 20 days	10	(i) Five patients had good recovery (ii) One patient exhibited moderate disability (iii) One patient exhibited severe disability (iv) Two patients progressed to a vegetative state (v) One patient died
Muroi et al., 2008 [45]	RCT	16 mMol bolus in 150 mL  followed by 64 mMol continuous infusion for 12 days	31/27	(i) Magnesium treatment had to be discontinued in 52% of patients due to adverse side effects (ii) Trend towards improved outcome observed
Dorhout Mees et al., 2007 [21]	Retrospective case series	64 mMol/day for 20 days	155/194	(i) Risk of DCI was lower in patients with higher serum magnesium concentrations when compared to the lowest quartile (ii) No effect on incidence of poor outcome
Schmid-Elsaesser et al., 2006 [46]	RCT	10 mg/kg bolus  followed by 30 mg/kg (IV) infusion of MgSO4 for 7 days	53/51	(i) No difference in outcome measured by GOS after 12 months (ii) Similar incidence of CVS in both groups (iii) Rate of cerebral infarction similar in both groups.
Wong et al., (2006) [47]	RCT	20 mMol bolus  followed by  80 mMol/day (IV) for 14 days	30/30	(i) CVS incidence decreased, but not statistically significant (ii) Vasospasm detected via TCD shorter in duration (iii) No difference in outcome measured by GOS at 6 months
Prevedello et al., 2006 [48]	RCT	20 mMol bolus of MgSO4   followed by  120–150 mMol/day	48 treated with nimodipine, Triple-H therapy, bed rest/ 24 treated with MgSO4 adjunct	(i) Incidence of vasospasm was reported to be equal in both groups. (ii) Vasospasm occurring in the nimodipine-only group was correlated with longer hospital stays, when compared to the MgSO4 adjunct group
Stippler et al., 2006 [49]	Retrospective historically controlled.	100 mMol MgSO4/day continuous infusion	38/38	(i) Incidence of vasospasm in the Mg2+ adjunct group decreased by 18% (ii) Outcome not changed in Mg2+ group (P > 0.05)
Yahia et al., (2005) [50]	Prospective pilot study	100 mMol/hour MgSO4 for 10 days	19	(i) No adverse effects from continuous magnesium infusion (ii) Lower incidence of both angiographic and clinical CVS observed than the literature values
van den Bergh, 2005 [51]	RCT	64 mMol (IV) MgSO4 for 14 days	139/144	Suggested benefit for reduction of DCI, though results were inconclusive
Venya et al., 2002 [52]	RCT	6 g bolus followed by  2 g/hour MgSO4 for 10 days	20/20	(i) No significant reduction in incidence of CVS (ii) Trend toward improved neurological outcome measured by GOS at 3 months
Chia et al., 2002 [53]	Retrospective case series	24–52 mMol/day continuous infusion MgSO4	13/10	(i) Significant reduction in the incidence of CVS (ii) Neurologic outcome was similar in both groups
Wong et al., 2011 [54]	Meta-analysis	—	441	(i) Lowered odds ratio for incidence of CVS and DCI in the magnesium treatment groups (ii) Increased odds ratio for favorable outcomes
Wong et al., 2010 [41]	Meta-analysis	—	875	(i) No benefit from magnesium infusion on the incidence of cerebral infarction (ii) Nonsignificant increase for odds ratio of favorable outcome at 3 and 6 months.
Chen and Carter, 2011 [55]	Meta-analysis	—	936	(i) Decreased risk of poor outcome at 3–6 months in the magnesium treatment groups (ii) Risk of mortality after SAH was unaffected
Ma et al., 2010 [22]	Meta-analysis	—	699	(i) Magnesium infusion reduced the risk for DCI and poor outcome after SAH (ii) Serum levels need to be monitored closely to prevent adverse side effects