Functional MRI data has to deal with physiological phenomena related to respiratory and cardiac functions. Correction techniques can minimize their effects with the risk of discarding relevant information. This is particularly significant when the focus is the study of emotions or ANS-CNS link. Here, we describe a study on the central control of breathing during voluntary breath hold (VBH) and CO2 challenge paradigms. Twenty 3mm thick oblique slices comprising brainstem, putamen, thalamus and cortical areas above corpus callosum were acquired. In VBH study, we compared linear and nonlinear models considering the time course of end-tidal CO2 (PETCO2) (1). Specifically, we used a polynomial static nonlinearity (2) applied to the individual PETCO2 time series within a multiple regression framework. Adjusted R squared (R2-adj) measures were compared at single subject level using different delays (till 12s) and polynomial orders, from 1, i.e. linear, up to 5. Data were analyzed both after applying and without applying retrospective cardiac and respiratory correction (RETROICOR) (3). The R2-adj results show an improvement using the fifth order nonlinearity: this finding is not spatially homogenous and is modified by RETROICOR. The R2-adj was higher without RETROICOR correction with respect to the corrected data, irrespectively of the model. These results indicate a complex relationship between the signal changes related to heartbeat and respiration and the model of PETCO2. The multiple regression analysis show that both the linear and nonlinear models highlight activations in the thalamus, putamen and anterior cingulate with a reduction of significant clusters after RETROICOR. Nonetheless, the nonlinear model applied to corrected data highlight significant clusters in the cingulate cortex, thalamus and pons, indicating a more complex relation of the brain activation with the PETCO2 than that captured by RETROICOR correction. This observation was supported by an exploratory analysis of CO2 challenges data (block design administration of 3% and 7% concentration levels). Group level Independent Component Analysis (ICA) was performed. The spatial distributions of task-related ICs obtained from both challenges were highly correlated (r=0.7). A similar IC was found from a free breathing task. Interestingly, a nonlinear analysis of IC amplitudes obtained from corrected data, normalized to reflect percent signal change, revealed a super-linear relationship with the CO2 changes in the putamen, thalamus and pons. Overall these results indicate a complex relationship between PETCO2 and BOLD changes, and the need of further exploring the effect of correction techniques for the study of breath control with fMRI.

An analysis of fMRI signal during voluntary breath hold and carbon dioxide challenge: physiological correction and modeling issues

Vanello, N.;Callara, A. L.;
2018-01-01

Abstract

Functional MRI data has to deal with physiological phenomena related to respiratory and cardiac functions. Correction techniques can minimize their effects with the risk of discarding relevant information. This is particularly significant when the focus is the study of emotions or ANS-CNS link. Here, we describe a study on the central control of breathing during voluntary breath hold (VBH) and CO2 challenge paradigms. Twenty 3mm thick oblique slices comprising brainstem, putamen, thalamus and cortical areas above corpus callosum were acquired. In VBH study, we compared linear and nonlinear models considering the time course of end-tidal CO2 (PETCO2) (1). Specifically, we used a polynomial static nonlinearity (2) applied to the individual PETCO2 time series within a multiple regression framework. Adjusted R squared (R2-adj) measures were compared at single subject level using different delays (till 12s) and polynomial orders, from 1, i.e. linear, up to 5. Data were analyzed both after applying and without applying retrospective cardiac and respiratory correction (RETROICOR) (3). The R2-adj results show an improvement using the fifth order nonlinearity: this finding is not spatially homogenous and is modified by RETROICOR. The R2-adj was higher without RETROICOR correction with respect to the corrected data, irrespectively of the model. These results indicate a complex relationship between the signal changes related to heartbeat and respiration and the model of PETCO2. The multiple regression analysis show that both the linear and nonlinear models highlight activations in the thalamus, putamen and anterior cingulate with a reduction of significant clusters after RETROICOR. Nonetheless, the nonlinear model applied to corrected data highlight significant clusters in the cingulate cortex, thalamus and pons, indicating a more complex relation of the brain activation with the PETCO2 than that captured by RETROICOR correction. This observation was supported by an exploratory analysis of CO2 challenges data (block design administration of 3% and 7% concentration levels). Group level Independent Component Analysis (ICA) was performed. The spatial distributions of task-related ICs obtained from both challenges were highly correlated (r=0.7). A similar IC was found from a free breathing task. Interestingly, a nonlinear analysis of IC amplitudes obtained from corrected data, normalized to reflect percent signal change, revealed a super-linear relationship with the CO2 changes in the putamen, thalamus and pons. Overall these results indicate a complex relationship between PETCO2 and BOLD changes, and the need of further exploring the effect of correction techniques for the study of breath control with fMRI.
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11568/960345
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact