Increased oxygen consumption following activation of brain: theoretical footnotes using spectroscopic data from...

Optical imaging spectroscopy (OIS) and laser Doppler flowmetry (LDF) data sequences from anesthetized rats were used to determine the relationship between changes in oxy-and deoxygenated hemoglobin concentration and changes in blood volume and flow in the presence and absence of stimulation. The data from Jones et al. (accompanying paper) were used to explore the differences between two theoretical models of flow activation coupling. The essential difference between the two models is the extension of the model of Buxton and Frank by Hyder et al. (1998, J. Appl. Physiol. 85: 554--564) to incorporate change in capillary diffusivity coupled to flow. In both models activation-increased flow changes increase oxygen transport from the capillary; however, in Hyder et al.'s model the diffusivity of the capillary itself is increased. Hyder et al. proposed a parameter (Omega), a scaling "constant" linking increased blood flow and oxygen "diffusivity" in the capillary bed. Thus, in Buxton and Frank's theory, Omega = 0; i.e., there are no changes in diffusivity. In Hyder et al.'s theory, 0 < Omega < 1, and changes in diffusivity are assumed to be linearly related to flow changes. We elaborate the theoretical position of both models to show that, in principle, the different predictions from the two theories can be evaluated using optical imaging spectroscopy data. We find that both theoretical positions have limitations when applied to data from brief stimulation and when applied to data from mild hypercapnia. In summary, the analysis showed that although Hyder et al.'s proposal that diffusivity increased during activation did occur; it was shown to arise from an implementation of Buxton and Frank's theory under episodes of brief stimulation. The results also showed that the scaling parameter Omega is not a constant as the Hyder et al. model entails but in fact varies over the time course of the flow changes. Data from experiments in which mild hypercapnia was administered also indicated changes in the diffusivity of the capillary bed, but in this case the changes were negative; i.e., oxygen transport from the capillary decreased relative to baseline under hypercapnia. Neither of the models could account for the differences between the hypercapnia and activation data when matched for equivalent flow changes. A modification to the models to allow non-null tissue oxygen concentrations that can be moderated by changes due to increased metabolic demand following increased neural activity is proposed. This modification would allow modulation of oxygen transport from the capillary bed (e.g., changes in diffusivity) by tissue oxygen tension and would allow a degree of decoupling of flow and oxygen delivery, which can encompass both the data from stimulation and from hypercapnia.