Logo Logo
Help
Contact
Switch language to German
Neuroenergetics and function of the default mode network. quantitative measurement of human brain metabolism
Neuroenergetics and function of the default mode network. quantitative measurement of human brain metabolism
Functional brain imaging relies on the fact that neuronal activity and brain metabolism are closely coupled. Changes in neuronal activity evoke vascular responses, more precisely changes in cerebral blood flow (CBF) and blood volume (CBV), as well as metabolic responses, that is oxygen and glucose consumption. Yet, in neuroimaging research, most studies are not measuring these different responses separately. Instead, most researchers acquire the blood-oxygen-level-dependent (BOLD) signal, which is relying on a mismatch between oxygen delivery and the cerebral rate of oxygen (CMRO2). Increases in neuronal activity usually trigger larger CBF than CMRO2 increases. Only if more oxygen is delivered than is consumed, deoxyhemoglobin (dHb) levels in venous blood drop, which gives rise to a positive BOLD response. Thus, metabolic and vascular contributions cannot be disentangled by a simple BOLD experiment. Further, impaired metabolism in different disease states or after drug administration have been shown to influence the amplitude of BOLD signals. Due to its relative nature, its amplitude is not comparable in terms of neuronal activity level across different brain regions within one subject and even less across subjects or patient populations. In contrast to fMRI BOLD, CMRO2 is a physiological signal that directly measures oxidative brain metabolism. Apart from oxygen, the brain needs glucose to fuel its brain activity. The cerebral rate of glucose (CMRglc) quantifies glucose consumption and, via comparing CMRO2 to CMRglc, provides information about levels of oxidative versus nonoxidative metabolism. Changes in both glucose and oxygen metabolism are supposedly more localized at the exact site of neuronal activation than CBF and BOLD changes. In this doctoral thesis, I show the applicability of simultaneous CMRO2 and CMRglc measurements during a visual stimulation experiment within the first study. I then extend these findings to a cognitive design within the second study. I show that multiparametric, quantitative BOLD (mqBOLD) measurements in healthy human subjects are sensitive enough to measure absolute task-induced changes in CMRO2. Further, by comparing fMRI BOLD with CMRO2, I provide evidence that specifically negative BOLD responses are not necessarily indicative of reduced oxygen metabolism and thereby cannot be interpreted as reduced excitatory brain activity. This showed in a dissociation of the sign of BOLD and CMRO2 changes, which was more pronounced for negative than for positive BOLD signal changes and especially apparent in regions of the default mode network (DMN). While parts of the DMN with significant BOLD decreases also showed concomitant task-induced decreases in oxygen metabolism, around half to two-thirds of DMN voxels actually showed increased oxygen consumption, despite consistent negative BOLD signal changes. These results fundamentally question our commonly accepted interpretation of specifically negative BOLD responses as indicators for decreased neuronal activity. Overall, measuring brain metabolism has several advantages over fMRI BOLD. The most important ones are its better localization at the site of neuronal activity and its potential for measuring absolute energy consumption. While there is currently no way of measuring CMRglc without the administration of radiotracers, the non-invasive measurement of CMRO2 developed into a promising area of research. Further steps into improving the applicability of measuring cerebral oxygen consumption are important to consolidate its path into clinical routines as well as within neuroimaging research, as an extension or alternative to fMRI BOLD measurements.
Multiparametric quantitative MRI (mqBOLD), default mode network (DMN), negative BOLD response (NBR), fMRI BOLD, cerebral metabolic rate of oxygen (CMRO2), cerebral metabolic rate of glucose (CMRglc), cerebral blood flow (CBF),, oxygen extraction fraction (OEF)
Epp, Samira Maria
2023
English
Universitätsbibliothek der Ludwig-Maximilians-Universität München
Epp, Samira Maria (2023): Neuroenergetics and function of the default mode network: quantitative measurement of human brain metabolism. Dissertation, LMU München: Graduate School of Systemic Neurosciences (GSN)
[thumbnail of Epp_Samira_Maria.pdf]
Preview
PDF
Epp_Samira_Maria.pdf

5MB

Abstract

Functional brain imaging relies on the fact that neuronal activity and brain metabolism are closely coupled. Changes in neuronal activity evoke vascular responses, more precisely changes in cerebral blood flow (CBF) and blood volume (CBV), as well as metabolic responses, that is oxygen and glucose consumption. Yet, in neuroimaging research, most studies are not measuring these different responses separately. Instead, most researchers acquire the blood-oxygen-level-dependent (BOLD) signal, which is relying on a mismatch between oxygen delivery and the cerebral rate of oxygen (CMRO2). Increases in neuronal activity usually trigger larger CBF than CMRO2 increases. Only if more oxygen is delivered than is consumed, deoxyhemoglobin (dHb) levels in venous blood drop, which gives rise to a positive BOLD response. Thus, metabolic and vascular contributions cannot be disentangled by a simple BOLD experiment. Further, impaired metabolism in different disease states or after drug administration have been shown to influence the amplitude of BOLD signals. Due to its relative nature, its amplitude is not comparable in terms of neuronal activity level across different brain regions within one subject and even less across subjects or patient populations. In contrast to fMRI BOLD, CMRO2 is a physiological signal that directly measures oxidative brain metabolism. Apart from oxygen, the brain needs glucose to fuel its brain activity. The cerebral rate of glucose (CMRglc) quantifies glucose consumption and, via comparing CMRO2 to CMRglc, provides information about levels of oxidative versus nonoxidative metabolism. Changes in both glucose and oxygen metabolism are supposedly more localized at the exact site of neuronal activation than CBF and BOLD changes. In this doctoral thesis, I show the applicability of simultaneous CMRO2 and CMRglc measurements during a visual stimulation experiment within the first study. I then extend these findings to a cognitive design within the second study. I show that multiparametric, quantitative BOLD (mqBOLD) measurements in healthy human subjects are sensitive enough to measure absolute task-induced changes in CMRO2. Further, by comparing fMRI BOLD with CMRO2, I provide evidence that specifically negative BOLD responses are not necessarily indicative of reduced oxygen metabolism and thereby cannot be interpreted as reduced excitatory brain activity. This showed in a dissociation of the sign of BOLD and CMRO2 changes, which was more pronounced for negative than for positive BOLD signal changes and especially apparent in regions of the default mode network (DMN). While parts of the DMN with significant BOLD decreases also showed concomitant task-induced decreases in oxygen metabolism, around half to two-thirds of DMN voxels actually showed increased oxygen consumption, despite consistent negative BOLD signal changes. These results fundamentally question our commonly accepted interpretation of specifically negative BOLD responses as indicators for decreased neuronal activity. Overall, measuring brain metabolism has several advantages over fMRI BOLD. The most important ones are its better localization at the site of neuronal activity and its potential for measuring absolute energy consumption. While there is currently no way of measuring CMRglc without the administration of radiotracers, the non-invasive measurement of CMRO2 developed into a promising area of research. Further steps into improving the applicability of measuring cerebral oxygen consumption are important to consolidate its path into clinical routines as well as within neuroimaging research, as an extension or alternative to fMRI BOLD measurements.