A number of studies have demonstrated that magnesium salts given after trau
matic brain injury improve subsequent neurologic outcome. However, given th
at these earlier studies have used a number of different salts, dosages, an
d routes of administration, follow-up studies of the neuroprotective proper
ties of magnesium are complicated, with comparisons to the earlier literatu
re virtually impossible. The present study has therefore characterized the
dose-response characteristics of the most commonly used sulfate and chlorid
e salts of magnesium in a severe model of diffuse traumatic axonal injury i
n rats. Both magnesium salts improved neurologic outcome in rats when admin
istered as a bolus at 30 min after injury. The i.v. and i.m. optima of each
salt was 250 mu mol/kg and 750 mu mol/kg, respectively. The identical conc
entrations required for improved neurologic outcome suggest that improvemen
t in outcome was dependent on the magnesium cation and not the associated a
nion. Subsequent magnetic resonance studies demonstrated that the administe
red magnesium penetrated the blood-brain barrier after injury and resulted
in an increased brain intracellular free magnesium concentration and associ
ated bioenergetic state as reflected in the cytosolic phosphorylation poten
tial. Both of these metabolic parameters positively correlated with resulta
nt neurologic outcome measured daily in the same animals immediately before
the magnetic resonance determinations.