Lumped and transfer rate constants in ischemic brain tissue must be measured to estimate accurately cerebral glucose utilization by the deoxyglucose method. We studied the bilateral middle cerebral artery occlusion model in 17 cats, 5 with a 1-hour occlusion, 5 with a 4-hour occlusion, and 7 with a sham operation. The time course of cerebral tissue radioactivity (C,*(t)) was monitored by external coincidence counting during a programmed infusion of [ 18 F]-2-deoxy-2-fluro-d-glucose ("F-2-FDG).

more »

... rterial plasma concentration (C p *(t)) of the tracer was kept constant during the first 45 minutes. Rate constants were estimated from C,*(t) and C *(t) by a nonlinear least-squares fitting routine. The lumped constant was estimated from the fit of the ratio of extraction fractions of glucose and "F-2-FDG by non weigh ted, nonlinear least-squares fitting. In the 4-hour-occlusion model, the transfer constant k,* was 23% lower, k 3 * 39% lower, and the lumped constant 78% higher than in the shamoperated animals. In the 1-hour-occlusion model, k 3 * was 26% lower than in the sham-operated animals but the lumped constant was not significantly different. The rate of glucose utilization was significantly different in the 4-hour-occlusion model compared to the sham-operated animals (48% decrease, p<0.05) but was not significantly different in the 1-hour-occlusion model. (Stroke