Previous studies have suggested that metabolic inhibition is neuroprotective, but little evidence has been provided to support this proposal. Using the in vitro rabbit retina preparation as an established model of the central nervous system (CNS), we measured the rate of glucose utilization and lactate production, and the light-evoked compound action potentials (CAPs) as indices of neuronal energy metabolism and electrophysiologic function, respectively. We examined the effect of three (3) treatments options: hypothermia (i.e., 33 degrees C and 30 degrees C), a six-member pharmacologic "cocktail" (tetrodotoxin (0.1 microM), 2-amino-4-phosphonobutyric acid (20 microM), 2-amino-5-phosphonovaleric acid (1 mM), amiloride (1 mM), magnesium (10 mM) and lithium (10 mM) and the combination of magnesium (Mg2+ 1 mM) and mexiletine (Mex, 300 microM) on in vitro rabbit retinas, to see if there is a correlation between neuronal energy metabolism during ischemia (simulated by the reduction of oxygen from 95% to 15% and glucose from 6 mM to 1 mM), and the subsequent recovery of function. Hypothermia and the "cocktail" significantly inhibited both the rate of glucose utilization and lactate production, whereas Mg2+ and/or Mex showed only a nonsignificant tendency toward a reduction, compared to control retinas. Recovery of light-evoked CAPs was significantly improved in hypothermia- and cocktail-treated retinas, as well as with retinas exposed to the combination of Mg2+ plus Mex, but not with Mg2+ or Mex alone, relative to control retinas. A linear regression analysis of the % recovery of function versus the % reduction in the rate of glucose utilization during ischemia showed a significant correlation (r2 = 0.80, correlation coefficient = 0.9, p < 0.05) between these two parameters. This and other data discussed provide convincing evidence that there is a correlation between metabolic inhibition, achieved during ischemia, and neuroprotection.