2711 Reproducibility of M , CMRO 2 and OEF measurements using QUO 2 MRI and dual-echo pCASL

Target audience: Researchers interested in calibrated MRI and in imaging of cerebral metabolic rate of O2 consumption (CMRO2) by MRI. Purpose: Recently, several groups have introduced a method, which allows the noninvasive imaging of resting cerebral metabolic rate of O2 consumption (CMRO2) by MRI. The approach described by our team, dubbed QUO2, is based on a hybrid calibration of the BOLD signal that includes the use of both hypercapnia (HC) and hyperoxia (HO) as iso-metabolic manipulations. During the latter conditions, BOLD and CBF responses are measured simultaneously using an implementation of pCASL that includes a dual-echo readout. The resultant end-tidal O2 (ETO2), fractional BOLD change, and fractional CBF change are put into the Generalized Calibration Model (GCM), yielding a system of two equations with two unknowns: the BOLD calibration parameter M (extrapolated maximum BOLD signal increase when venous saturation approaches 100%) and Oxygen Extraction Fraction (OEF; the fraction of delivered oxygen that is consumed). CMRO2 can then be determined by multiplying O2 delivery, itself the product of CBF and arterial O2 content, with the OEF. In a parallel study, our group has demonstrated the reproducibility of the ETO2 and ETCO2 traces during the respiratory manipulation, as well as that of the resting CBF, hypercapnia-induced change in CBF, and percent change in BOLD during hypercapnia and hyperoxia. The goal of the present study was to assess the test-retest reliability of the QUO2 outputs M, OEF0 and CMRO20, on a group of healthy control subjects, to aid in calculation of statistical power in future applications.