Dementia: Update for the Practitioner

Imaging modalities can be organized and described based on what we intend to measure: brain volume, oxidative metabolism, glucose metabolism, brain chemistry, or radioligands. Note that there is no function or structure in this scheme.


	Imaging Modalities 1. Imaging Brain Volume: MRI 2. Imaging Oxidative Metabolism: MRI, PET, SPECT a. Acute b. Chronic 3. Imaging Glucose Metabolism: PET 4. Imaging Brain Chemistry: MRI 5. Imaging Radioligands: PET

	Imaging modalities can be organized and described based on what is to be measured. Courtesy of Dr. Scott Small

Brain volumetric studies and morphometric studies are synonymous terms and rely on MRI techniques. There is a growing body of research and algorithms to quantitate changes in brain volume. With volumetrics, we can only look at large areas of the brain, even though the MRI sequences we use generate very high-resolution spatial maps. The spatial resolution is submillimeter, but the anatomical resolution is macroscopic.


	Imaging Brain Volume: MRIs


	Brain volumetric studies rely on MRI techniques. Courtesy of Dr. Scott Small. Source of data: Jeffrey R. Petrella et al., "Neuroimaging and Early Diagnosis of Alzheimer's Disease: A Look to the Future," Radiology 226, no. 2 (2003): 315-36.

Initially, the people who were doing volumetrics thought that they were measuring changes in cell number, or cell loss. In a number of recent studies, this seems not to be true; instead we are looking at interstitial volume, or cell volume, in the absence of cell loss, which makes it an even more interesting technology. We can now quantitate a volume of the hippocampus and/or the entorhinal cortex using MRI and see systematic differences between controls and Alzheimer's patients.


	Imaging Brain Volume
	Physiological Source Cell loss Interstitial volume Cell volume	Anatomical Resolution Macroscopic

	Volumetrics measure cell volume with macroscopic anatomical resolution. Courtesy of Dr. Scott Small

Acute oxidative metabolism is related to the hemodynamic variables I mentioned above: cerebral blood flow, cerebral blood volume, and deoxyhemoglobin. Imagine a group of neurons that is fed by a vascular tree. If we now increase metabolism by firing a neuron, blood flow increases and blood volume increases. This makes sense considering simple fluid mechanics.

When we increase the firing rate, we increase deoxyhemoglobin, but the simultaneous overflow of fresh blood effectively reduces the proportional amount of deoxyhemoglobin. In fMRI, which is used by psychologists to map brain function, we can use this deoxyhemoglobin effect to see where the brain fires when it is performing an activation task.

fMRI allows us to look at changes in spike activity and changes in synaptic potentials at a macroscopic level. In a typical fMRI experiment that relies on deoxyhemoglobin, a subject is measured at what is called a resting state, say with the eyes closed. The subject then opens his or her eyes, lights or an image flashes before them, and the signal change in the occipital cortex is detected. The difference between the signal at rest and the signal upon seeing flashing lights tells us something about the firing neurons in that part of the brain.

We can use these acute changes in metabolism to look at hippocampal function. In a study we published several years ago that has now been replicated, we had subjects try to memorize a series of faces that were shown to them, and we could see activation in the hippocampus with fMRI. We saw a decreased activation in Alzheimer's patients.


	Imaging Acute Metabolism


	Acute oxidative metabolism is measured using fMRI to see where the brain fires when it is performing an activation task. Here, subjects try to memorize a series of faces shown to them, and the resulting activation in the hippocampus was lower in Alzheimer's patients. Courtesy of Dr. Scott Small. Source of data: Scott A. Small et al., "Differential Regional Dysfunction of the Hippocampal Formation Among Elderly with Memory Decline and Alzheimer's Disease," Annals of Neurology 45, no. 4 (1999): 466–72.

Though this was encouraging, fMRI as it is currently practiced is not ready for clinical use. Recall that we quantitate change by comparing an acute change in metabolism to a chronic resting state. We now know from very elegant studies that if you change the chronic resting state, you will also change the amplitude of the response. This is like a ceiling effect. If a patient's resting state is closer to that ceiling, the amplitude will be diminished. One could falsely conclude that the subject who has a higher resting state and a lower change in amplitude has decreased activity, but that would not be true.


	Imaging Acute Metabolism: Limitations


	Further, fMRI as currently practiced is not ready for clinical use because of a "ceiling effect." Courtesy of Dr. Scott Small. Source of data: E. R. Cohen et al., "Effect of Basal Conditions on the Magnitude and Dynamics of the Blood Oxygenation Level-Dependent fMRI Response," Journal of Cerebral Blood Flow and Metabolism 22, no. 9 (2002): 1042–53.

An interesting study looked at visual stimulation in the occipital cortex in young subjects, older subjects, and subjects with dementia. They found that by simply looking at amplitude, they would falsely conclude that all subjects suffer from age-related cortical blindness. We now know that this is probably a problem with the technique, not with the human cortex. Once techniques are calibrated, fMRI may be useful as a diagnostic tool.

In imaging chronic oxidative metabolism, we look at changes that occur over hours and days and that reflect biochemistry. SPECT and PET are gold standards in imaging, and look at biochemistry and chronic bioelectricity. SPECT is used to measure cerebral blood flow in the chronic state, and PET is used in imaging glucose metabolism.

If we can rely on imaging oxidative metabolism, we can capture the biochemistry of disease and perhaps chronic changes in bioelectricity. Importantly, SPECT and PET can achieve this with macroscopic resolution, and only MRI has the microscopic resolution that is important for clinical utility. fMRI is safer, cheaper, and more readily available than PET and is gaining popularity as the way to capture both chronic and acute changes in metabolism.

Spectroscopy is an emerging technique that allows us to look at a number of different chemicals. Two of these, decreasing N-acetyl aspartate (NAA) and increasing myo-inositol, seem to be interesting players in Alzheimer's disease. The physiological source for spectroscopy is cell number and cell integrity. The anatomical resolution is macroscopic, at a centimeter resolution. This is poor resolution but it gives interesting insight into brain biochemistry.

Another modality that has gotten a lot of attention in the press is the use of radioligands against different components of the brain. In this area, PET will remain supreme. There are radioligands available against a number of receptors, including acetylcholine and glutamate receptors. A study from Gary Small's group using radioligands against amyloid plaques showed an increased signal in Alzheimer's patients.


	Imaging Brain Chemistry: Spectroscopy
	N-acetylaspartate
	Glutamine
	Glutamate
	GABA	Decreased NAA
	Myo-inositol	Increased myo-inositol
	Choline
	Creatine
	Lipids
	Lactate

	Spectroscopy allows us to look at a number of different chemicals, two of which, decreasing N-acetyl aspartate (NAA) and increasing myo-inositol, may be involved in Alzheimer's disease. Courtesy of Dr. Scott Small

PREVIOUS | NEXT