Objectives. This study was designed to determine the effects of glucose-insulin infusion on ischemia-induced changes in extracellular potassium ([K+]o) accumulation and the associated electrophysiologic abnormalities in the canine heart. Background. Although glucose-insulin-potassium infusion has been shown to limit myocardial injury in acute ischemia, its effect on ischemia-induced electrophysiologic alterations has not been investigated. Methods. Recordings of [K+]o and local electrograms from the normal, border and ischemic zones were obtained during serial (10-min) left anterior descending coronary artery occlusions in the control state and after infusion of glucose-insulin (eight dogs), glucose alone (six dogs) or insulin alone (eight dogs). Results. Glucose-insulin infusion caused significant reduction in the rise of [K+]o during the entire period of ischemia in both ischemic and border zones associated with significant improvement in the degree of intramyocardial conduction delay. At 10 min of ischemia, [K+]o was reduced from a mean control level of 15.9 +/- 3.7 to 10.1 +/- 4.3 mmol/liter (p < 0.005) in the ischemic zone and from 6.8 +/- 1.9 to 5.5 +/- 1.1 mmol/liter (p < 0.05) in the border zone. The electrogram duration was shortened from a mean control value of 102 +/- 13 to 78 +/- 12 ms in the ischemic zone and from 79.2 +/- 7.8 to 58.1 +/- 6.6 ms in the border zone (p < 0.005). Glucose alone caused significant reduction in [K+]o during the initial 6 min of ischemia, only in the ischemic zone. Conversely, insulin caused no changes in [K+]o accumulation during ischemia. Neither glucose nor insulin alone had any effect on ischemia-induced intramyocardial conduction delay. Conclusions. The present study demonstrated that the combination of glucose and insulin is essential for the salutary effect of reducing [K+]o accumulation during ischemia and improving the associated intramyocardial conduction delay. It could be postulated that glucose in the presence of insulin increases the glycolytic flux, thereby providing adequate adenosine triphosphate for suppressing the cardiac adenosine triphosphate-sensitive potassium ion channels. The latter are, at least partially, responsible for the [K+]o rise in the early phase of ischemia. This study highlights the antiarrhythmic potential of interventions that modulate the metabolic consequences of ischemia.