Retention of 99mTc-bicisate in the human brain after intracarotid injection.

99mTc-bicisate (ECD) was injected as a bolus into the internal carotid artery, and cerebral uptake and retention were recorded with fast-rotating single photon emission computed tomography (SPECT) equipment in four patients suffering from temporal lobe epilepsy. Quantitative regional cerebral blood flow (rCBF) was measured tomographically with the 133Xe inhalation technique. We applied a three-compartment kinetic model and algorithms modified from a previous analysis of 99mTc d,l-hexamethylpropyleneamine oxide (HM-PAO) kinetics. The bicisate brain uptake and retention curve was very similar to that of HM-PAO, and it can be described by a triexponential function including an initial steep component representing the vascular transmitted spike, a second less steep component representing back-diffusion from brain tissue to blood, and a third, very slow component, representing the very slow loss due to incomplete retention of the deesterified hydrophilic metabolites. Computerized curve-fitting on data from three patients gave average kinetic values for the first-passage (unilateral) extraction of E = 0.60 (range, 0.59-0.61); the overall retained fraction of the tracer supplied was R = 0.44 (0.43-0.45), and the conversion/clearance ratio was alpha = k3/k2 = 2.59 (2.38-2.77). This alpha is higher than that for HM-PAO, and therefore bicisate uptake as a function of blood flow is more linear than in HM-PAO. Less correction for backdiffusion is therefore needed. From 1 to 24 h there was an average loss of hydrophilic tracer of 3.5%/h, but the late distribution images were essentially unchanged over time, pointing to practically the same rate of loss in all regions.