Comparison between end-tidal CO2 and tidal volume changes calculated from the respiratory motion tracing used for correction of respiratory fluctuations in a functional MRI experiment with normal breathing and hyperventilation

Introduction: Variations in end-tidal carbon dioxide (ETCO2) are known to be correlated to BOLD signal changes as a result of cerebral blood flow (CBF) adjustments driven by the arterial CO2 concentration that is reflected in expired air [1]. This suggests that measuring expired CO2 concentration could be an important addition to the respiratory motion and pulse plethysmography tracings commonly recorded for FMRI physiologic noise correction [2]. However, because CO2 sampling is not included as a component of MR scanner hardware, collecting this data is technically challenging. A simpler alternative measure has been suggested, derived from the respiratory motion tracing, that accounts for changes in respiratory volume over time (RVT), presumably modeling ETCO2 [3]. This FMRI study employs conditions of paced breathing and hyperventilation that uncouple respiration from ETCO2. Under these conditions, the MR signal changes were more widely correlated to the ETCO2 values than to the calculated RVT values.