Increased blood flow detected by fMRI scans?

Andrew Gyles syzygium at
Sun Oct 21 20:33:32 EST 2001

"Richard Norman" <rsnorman at> wrote in message
news:lmd3ttgv1s43jeda5eqjforg38t8e4ilh4 at
> On Sat, 20 Oct 2001 18:24:14 +0200, "Brian" <zhil at> wrote:
> >"Richard Norman" <rsnorman at> skrev i melding
> >news:qfh1ttofdui4h57qd8ed6uq2dmt1r5u9lo at
> >> On Fri, 19 Oct 2001 15:16:56 +1000, "Andrew Gyles"
> >> <syzygium at> wrote:
> >>
> >> >Blood oxygen level dependent functional magnetic resonance imaging
> >> >fMRI) scans show active regions of the brain. It has been assumed that
> >the
> >> >excess of oxygen detected in an active region is brought there by
> >increased
> >> >blood flow.
> >> >
> >> >Has this assumption of increased blood flow been proved correct by
> >> >experiment?
> >> >
> >> >Andrew Gyles
> >>
> >> Try looking in places like
> >>
> >> or
> >>
> >> The specific experimental work you want is
> >>
> >> Logothetis NK, Pauls J, Augath M, Trinath T, Oeltermann A.
> >> Neurophysiological investigation of the basis of the fMRI signal.
> >> Nature. 2001 Jul 12;412(6843):150-7.
> >>
> >> The relation to oxygenation is quite certain, since we know
> >> just what molecular properties produce the MRI signal -- it
> >> is the oxygenation state of hemoglobin.  What Logothetis
> >> et al. did was relate it to neurophysiological recordings.
> >> They found that the BOLD fMRI (blood-oxygen-level-dependent)
> >> signals were correlated mostly with the local field potentials, not
> >> the single unit or multiunit spike recordings.  They say "These
> >> findings suggest that the BOLD contrast mechanism reflects the input
> >> and intracortical processing of a given area rather than its spiking
> >> output."
> >
> >In other words, they said that the energy-usage by the cell reflects the
> >oxygen-flow through the blood, while spike(s) are _not_(no big surprise
> >there, I would be much more surprised if the oxygen-level correlated with
> >the spikes...).
> >
> >Brian
> Yes, it is a tautology that oxygen use correlates with metabolism.
> The question is what cellular processes of the neuron consume the most
> energy?  Is it making action potentials?  Or is it synaptic
> transmitter synthesis (and recycling)?  Or is it all the cell
> processes that go along with neuromodulation, including up- and
> down-regulation of membrane proteins?  Or just what?

The difficulty presented by the BOLD fMRI results is that they show a
greater concentration of oxygen in a field containing neurons that we know
to be active, and therefore (one might have expected) consuming more oxygen.

This is explained by saying that the capillaries in the active field expand
and a greater flow of oxygenated blood overcompensates for the increased
oxygen use. Alternatively it is explained by saying that much of the
increased activity is powered by anaerobic respiration, and the purpose of
the increased flow of blood is to bring more glucose to the field and remove
the end-products (presumably including lactic acid).

But any increase of lactic acid in the field cannot be measured by a
scanning system, as far as I know.

I suggest a third alternative: perhaps there is no great increase in blood
flow in the field. Perhaps the "resting" level of aerobic respiration (which
might be substantial) is reduced and the "resting" level of anaerobic
respiration is increased in the field when its neurons become active. The
result would be an increase in oxygen concentration in the field, which
would be detected by the BOLD fMRI scan.

I do not claim that this must be correct, only that it seems a valid
alternative to be considered.

Andrew Gyles

(PS. Thanks for the references; especially the description by Stacey; I had
read the other two.)

More information about the Neur-sci mailing list