Quantitative accuracy of serotonergic neurotransmission imaging with high-resolution 123I SPECT.

Abstract

Serotonin transporter (SERT) imaging can be used to study the role of regional abnormalities of neurotransmitter release in various mental disorders and to study the mechanism of action of therapeutic drugs or drugs' abuse. We examine the quantitative accuracy and reproducibility that can be achieved with high-resolution SPECT of serotonergic neurotransmission. Binding potential (BP) of (123)I labeled tracer specific for midbrain SERT was assessed in 20 healthy persons. The effects of scatter, attenuation, partial volume, misregistration and statistical noise were estimated using phantom and human studies. Without any correction, BP was underestimated by 73%. The partial volume error was the major component in this underestimation whereas the most critical error for the reproducibility was misplacement of region of interest (ROI). The proper ROI registration, the use of the multiple head gamma camera with transmission based scatter correction introduce more relevant results. However, due to the small dimensions of the midbrain SERT structures and poor spatial resolution of SPECT, the improvement without the partial volume correction is not great enough to restore the estimate of BP to that of the true one.