Objectives: Glucose is the primary substrate for the energy requiremen
ts of the nervous system. Nevertheless, administration of glucose to c
ritically ill patients with central nervous system trauma may have adv
erse effects on their neurologic recovery. The purpose of this study w
as to evaluate the effects of other sources of nonprotein calories on
spinal cord lactate accumulation and on electrophysiologic recovery af
ter a period of severe spinal cord ischemia. Design: Two randomized, b
linded studies were performed: one of glycolytic energy substrates (fr
uctose, xylitol, sorbitol, glycerol) and one of ketogenic energy subst
rates (beta-hydroxybutyrate, acetate, butyrate). Setting: College teac
hing hospital laboratory. Subjects: New Zealand albino rabbits (weight
3.5 to 4.5 kg). Interventions: After infusion of the randomly assigne
d treatment, temporary ischemia was produced in the lumbosacral spinal
cord by occluding the abdominal aorta with a balloon catheter. Measur
ements and Main Results: Blood concentrations of glucose, lactate, pyr
uvate, and ketone bodies and spinal cord dialysate concentration of la
ctate were measured before and after infusion of the assigned treatmen
t, and during ischemia and during the first 2 hrs after reperfusion. S
pinal somatosensory evoked potentials were recorded during ischemia to
assure a similar severity of ischemia in all animals and during the f
irst 2 hrs after reperfusion as a measure of electrophysiologic recove
ry. Infusion of the glycolytic nutrients xylitol and fructose increase
d blood glucose and lactate concentrations, and resulted in increased
lactate accumulation in the spinal cord during ischemia and resulted i
n a significantly poorer recovery of the spinal somatosensory evoked p
otential than infusion of saline. Infusion of sorbitol and glycerol di
d not have these adverse effects in the doses administered. None of th
e ketogenic nutrients increased blood glucose concentration or increas
ed lactate accumulation in the spinal cord during ischemia when compar
ed with infusion of saline. Infusion of butyrate and acetate caused ar
terial hypotension and resulted in a poorer recovery of the spinal som
atosensory evoked potential than saline. Infusion of beta-hydroxybutyr
ate did not have an adverse effect on blood pressure or on evoked pote
ntial recovery. Conclusions: Glycerol, sorbitol, and beta-hydroxybutyr
ate deserve further evaluation as potential nonprotein calorie sources
in patients with neurologic injury. Xylitol and fructose are not suit
able since these substrates resulted in hyperglycemia and increased la
ctate accumulation in the central nervous system, and had detrimental
effects on electrophysiologic recovery after ischemia. The short-chain
fatty acids (acetate and butyrate) also had adverse effects on electr
ophysiologic recovery after ischemia, probably because of their hypote
nsive effects when given intravenously, rather than from the effects o
f their metabolism.