In 27 cats treated to vary arterial serum glucose concentrations, we measured cerebral highenergy phosphate metabolite concentration and intracellular pH using in vivo phosphorus-31 nuclear magnetic resonance spectroscopy during transient global cerebral ischemia and reperfusion. Hypoglycemia was induced with 4 units/kg i.v. insulin in six cats before ischemia; hyperglycemia was induced with 1.5 g/kg i.v. glucose in six cats before and in six cats during ischemia. Nine untreated cats subjected to ischemia without manipulation of blood glucose concentration served as controls. During ischemia, intracellular pH fell to similar levels in the control and both hyperglycemic groups. During reperfusion, the hyperglycemic before ischemia group initially exhibited a severe further decline in intracellular pH (p<0.003); this further decline was not observed in the control or the hyperglycemic during ischemia groups. Intracellular acidosis was attenuated both during ischemia and early after reperfusion in the hypoglycemic before ischemia group. In all groups, cerebral high-energy phosphate metabolite concentrations were depleted during ischemia and then recovered to the same degree during reperfusion. Our data suggest that brain glucose stores before ischemia determine the severity and time course of intracellular acidosis during ischemia and reperfusion. (Stroke 1988;19:1383-1387) T he influence of systemic blood glucose concentration on the neurologic outcome of stroke has been repeatedly documented in experimental animals and patients.1 -8 Experiments have suggested that this effect is mediated by brain acidosis. 8 -12 Efforts to control blood glucose concentration could have promise in the clinical management of stroke, but the precise conditions under which this might be accomplished remain uncertain. For example, if preexisting high blood and tissue glucose concentrations dictate an unfavorable outcome, control of hyperglycemia after stroke onset will be ineffective. However, if control of blood glucose concentration is worthwhile, should levels be controlled to the normal range and will subnormal (even hypoglycemic) levels produce added benefit? We have begun to address these questions by observing energy metabolism and pH in an animal model of global ischemia using noninvasive in vivo phosphorus-31 nuclear magnetic resonance spectroscopy ( 31 P NMR), a technique that has potential for the dynamic monitoring of stroke patients who might be candidates for manipulation of systemic glucose concentrations.
Materials and MethodsWe studied 27 female cats weighing 2.2-3.4 kg. The cats were fasted for 24 hours, with water given ad libitum. Anesthesia induced with 4% halothane in O 2 was followed by tracheotomy, intravenous injection of 0.02 mg atropine and 0.08 mg/kg pancuronium bromide, and mechanical ventilation with <1% halothane/33% O 2 /66% N 2 O + CO 2 . Ventilation gases were adjusted to maintain blood gases within the physiologic range. Rectal temperature was monitored and maintained at 38° C. Both femoral arteri...