Ischemic contracture may be avoided by the provision of glucose under low flow conditions (Owen et al., 1990). However, accumulation of harmful metabolic end products may inhibit glycolytic flux and lessen the benefit of glucose. We assessed whether during increasingly severe flow restriction, provision of glucose might be harmful rather than beneficial, using the Langendorff perfused rat heart. Ischemic contracture (resting tension expressed as percent of preischemic developed pressure) was measured via a left ventricular balloon. Reductions in flow to 0, 0.015, 0.03, 0.06, 0.1, 0.2 or 0.4 ml/min/g wet wt over 60 min were tested. At zero flow, peak contracture was 61.4±3.5% (±s.e.) but fell to 15.6±6.3% with 0.4 ml/min/g wet wt (P<0.05) in the presence of 11 mmol/l glucose. Time-to-onset of contracture was significantly delayed by the higher coronary flows. At coronary flows down to zero, the effect of glucose was inconstant or absent, but not harmful. With the residual flow at 0.2 ml/min/g wet wt, a dose response to glucose in ischemia was elicited, using concentrations of 0, 2.5, 5.5, 11 or 22 mmol/l. Maximum protection against ischemic contracture was found with 11 mmol/l glucose. However, once contracture occurred, functional recovery was severely impaired in all cases. Reducing glycogen prior to low flow ischemia (0.2 ml/min/g wet wt) with 11 mmol/l glucose increased peak contracture, and reduced the time-to-onset of contracture. Increased preischemic glycogen had little effect on contracture.Glycolytic flux fell in proportion to the coronary flow. However, there was an increased glucose extraction at lower flows of 0.1 and 0.2 ml/min/g wet wt, suggesting that it is the rate of delivery (i.e. coronary flow) which is the rate limiting step rather than enzyme inhibition by accumulated metabolites. If flow were further reduced, metabolite accumulation would become more important, such that with no flow, this would be the determinant of glycolytic flux rate.In our model, the two requirements for optimal protection from ischemia were (i) provision of glucose (11 mmol/l was optimal) and (ii) an adequate coronary flow to deliver the glucose and remove end product inhibition (greater than 0.06 ml/min/g wet wt).