Functional Imaging Modalities
1) PET is the gold standard of functional neuroimaging modalities.
2)Only PET can measure cerebral glucose metabolism.
3) A larger number of radioligands, especially those used for neuroreceptor characterization, are available for PET.
4) PET offers excellent spatial resolution ( 4 mm).
5) PET is very expensive and requires relatively rapid access to a cyclotron, which produces the positron-emitting radionuclides. Most PET centers have a cyclotron on the premises. However, because of the high cost of obtaining and maintaining a func-tional cyclotron, some centers do not have a cyclotron on the premises. 18F has a longer half-life (110 minutes) than other ra-dionuclides such as 11C (20 minutes), 13N (10 minutes) and 15O (2 minutes), so 18FDG may be prepared using a cyclotron at one site and then transported to other sites as long as they are within hours (generally 1–3 half-lives) of the site of synthesis.
1) SPECT is more affordable than PET and does not require a cyclotron for production of nuclides.
2)Currently, SPECT provides inferior spatial resolution ( 6–8 mm) compared with PET.
3) Typically, SPECT spatial resolution worsens as one attempts to image deeper brain structures, although this is becoming less of an issue with the introduction of newer generation SPECT cameras.
1) PET provides superior spatial resolution, especially for deeper brain structures.
2) PET offers a broader array of radioligands for use in receptor studies and is the only modality that allows for the measure-ment of metabolism.
3) SPECT is less expensive than PET and is more widely available.
Although functional neuroimaging modalities include PET, SPECT and functional MRI (to be discussed later), in clinical situ-ations only PET and SPECT are generally used at this time. These modalities are now commonly used as aids for diagnosis and monitoring treatment in cardiology and oncology (including brain tumors). Also, the use of ligands for neuroreceptor characteriza-tion is being increasingly used for the diagnosis and assessment of basal ganglia diseases such as Parkinson’s disease. Still, most applications of functional neuroimaging in psychiatry occur in the field of research. However, a clinical role for functional neuroim-aging in dementia and seizures is evolving and showing promise.
As characteristic functional neuroimaging profiles emerge for var-ious forms of dementia, the role of PET and SPECT in the evalua-tion of dementia is expanding. For example, Alzheimer’s disease is associated with characteristic hypoperfusion in bilateral temporo-parietal regions. Some studies have indicated that functional neu-roimaging can offer better than 90% sensitivity and specificity in distinguishing Alzheimer’s disease from other kinds of dementia (Silverman et al., 2001; Bonte et al., 2001). Since the introduction of cholinesterase inhibitors for treatment of dementia, early diagnosis of Alzheimer’s disease may have greater import than in the past. Some PET studies of healthy older subjects with normal cognitive function who are homozygous for the apolipoprotein E epsilon 4 allele (a gene associated with the development of Alzheimer’s dis-ease) reveal that these subjects have temporo-parietal hypoper-fusion before the onset of disease (Small et al., 1995; Reiman et al., 1996). Initiation of treatment, with cholinesterase inhibitors or other drugs being developed, in these individuals prior to the onset of symptoms of dementia may prove to be of great value.
Some seizures, especially complex partial seizures, are not al-ways detected by electroencephalogram (EEG). EEG measures cortical surface electrical activity but is less efficacious if the sei-zure focus is deeper in the brain. PET and SPECT images typi-cally demonstrate ictal hypermetabolism and interictal hypome-tabolism (Krausz et al., 1996, Theodore and Gaillard 2000). This allows for the detection of seizure foci during the predominant interictal period. To evaluate a possible seizure disorder, func-tional neuroimaging is usually performed in conjunction with EEG. PET is also useful for more precise localization of seizure foci in a patient with a known seizure disorder, if neurosurgical intervention is indicated.
Expanding upon the basic principles underlying standard structural MRI used in clinical situations, functional magnetic resonance imaging (fMRI) allows investigators to assess brain function. Using either a standard or, preferably, a high-speed MR scanner with specific image-acquisition parameters, indices of cerebral blood flow and blood volume can be measured. While structural MRI studies rely on excitation and relaxation of hydro-gen atoms in water, fMRI takes advantage of the paramagnetic properties of hemoglobin to measure blood flow and volume.
fMRI has distinct advantages when compared with PET and SPECT, and these include:
· fMRI does not expose subjects to radiation and its relative safety has been documented.
· The spatial resolution of fMRI is at least equal to if not supe-rior to that of PET.
· The temporal resolution of fMRI is vastly superior to that of PET or SPECT. While the temporal resolution of an 15O PET study is 1 to 2 min, multiple time points can be assessed per second with fMRI.
The major disadvantage of fMRI when compared with PET or SPECT is that no satisfactory techniques for receptor neuroimaging have yet been developed for fMRI.
While fMRI currently has little utility, it may be use-ful in clinical situations in the future. One area of research that shows particular promise is the use of intravenous paramagnetic contrast agents with fMRI. This method, termed tracer kinetic technique, produces maps of cerebral blood volume (CBV) (Bel-liveau et al., 1991). These CBV maps are relatively well matched to PET images of FDG uptake (Gonzalez et al., 1995) and HM-PAO SPECT images of cerebral blood flow (Johnson et al., 1995). Because these fMRI CBV maps can be obtained after only 1 to 2 min of imaging without exposure to radiation, this technique may be used instead of PET or SPECT for evaluation of dementia or seizures in the future.