Standardization of Input Function for Quantification of Regional Cerebral Blood Flow in N-isopropyl-p-[ 123 I]iodoamphetamine SPECT

Abstract

Quantification of regional cerebral blood flow (rCBF) using N-isopropyl-p-[123I]iodoamphetamine (IMP) and single photon emission computed tomography (SPECT) requires continuous or frequent arterial blood sampling and accurate measurement of radioactivity of many blood samples. In order to avoid these invasive and troublesome procedures, a simplified method with a single blood sampling and two SPECT scans is proposed based on the two-compartment model We tried to find the optimum time of single blood sampling for calibration of the standardized input function. There were ten subjects; four patients with cerebrovascular disease, three with brain tumor, one with head trauma, and two healthy volunteers. Their ages ranged from 25 to 74 years with the average of 56.4 years. IMP (167 MBq) was infused into the right cubital vein at a constant rate in 15 s. Following the start of IMP infusion, arterial blood was sampled from the left radial artery every 15 s for 2 min, every 30 s for the next 3 min, every 1 min for the following 5 min, and with a longer interval thereafter until 40 min after. The radioactivity of the whole blood was counted and its octanol extraction fraction was measured for each sample using the well counter. The whole blood radioactivities obtained at each time point from individual subject were averaged for all 10 subjects. The standardized input function was obtained from the product of the averaged radioactivity of the whole blood and the octanol extraction fraction at each time point. Early and late SPECT data were acquired at mid scan time of 40 min and 180 min, respectively. The rCBF was quantified by the “table-look-up” method. The difference between the standardized input function calibrated at each time point and the individual actual input function was 5.7% at most any time after 3 min and was minimal, 3.6%, at 20 min after the start of IMP infusion. These data were better than the reported minimal value of 5.3% at 9–10 min which were obtained by 1-min IMP infusion. The rCBF values figured out by the present method are well correlated with those calculated by the individual actual input function. An IMP infusion in 15 s with single arterial blood sampling any time after 3 min for calibrating the standardized input function is a practical method for rCBF quantification using IMP SPECT.