Summary
Positron emission tomography (PET) was introduced into Australia in 1992, but it has been available in North America and Europe for over a decade. PET has been used in the assessment of patients with refractory epilepsy, dementia, brain tumours, stroke and movement disorders
Tracers used in neurology
Much of the published work to date has used 18Ffluorodeoxyglucose which enables quantitative measurement of regional glucose metabolism. Other commonly used techniques include 15Olabelled water or carbon dioxide to measure cerebral blood flow and cerebral oxygen utilisation. Neurotransmitter systems within the body have been studied including the dopaminergic system using 18Ffluorodopa and a number of other ligands. The GABA, opiate, and other systems have been studied using 11Clabelled radiopharmaceuticals.
Fig. 1 Temporal lobe epilepsy. The left temporal lobe shows reduced 18Ffluorodeoxyglucose uptake in the interictal period (arrowheads). The patient underwent left temporal lobectomy and is now seizure free.
|
|
Safety and convenience
PET can generally be performed as an outpatient procedure. The isotope is usually injected intravenously. If quantitation of the images is required to measure absolute values of blood flow or metabolism, an arterial line may be inserted. The radiation exposure results in an absorbed dose which is equal to or less than that received in many other routine radio diagnostic procedures.
Focal epilepsy
While most patients with epilepsy are satisfactorily controlled on antiepileptic medication, surgery is being increasingly considered for patients with refractory seizures. PET is of value in the presurgical evaluation of patients with refractory focal epilepsy. In temporal lobe epilepsy, which is the most important form of refractory focal epilepsy, the epileptogenic temporal lobe shows hypometabolism for glucose in approximately 85% of cases (Fig. 1). This finding is quite reliable and, in many centres, has obviated the need for invasive studies using depth electrodes for intracranial EEG monitoring. The addition of PET to the conventional evaluation may shorten hospital stays and make the process more efficient, less invasive and less expensive. These hypotheses are being tested in an Australian trial.
The value of PET in extra temporal epilepsies is less clear. A new development is the observation that PET may reveal a focus in children with extremely severe and apparently generalised epileptic abnormalities. Removal of such a focus may cure the seizures and reverse developmental regression. If these findings are confirmed, PET will have a very important role not only in the treatment but in the prevention of chronic disability due to progressive epilepsy in early childhood.
Dementia
Dementia, in particular Alzheimer's disease, is an increasing health problem in ageing populations. Alzheimer's disease remains a neuropathological diagnosis for which accurate biochemical markers have not yet been found. Whilst the clinical features can lead to a reasonably confident diagnosis of Alzheimer's disease, PET may increase the diagnostic certainty. Alzheimer's disease shows bilateral superior parietal hypometabolism (Fig. 2) extending into the inferior parietal and temporal lobes as the disease progresses. In contrast, multi infarct dementia shows multiple asymmetric regions of cortical and sub cortical hypometabolism. This metabolic information is complementary to that provided by structural imaging studies (CT or MRI). PET, in conjunction with these structural studies, can thus result in an earlier and more specific diagnosis of dementia.
Brain tumours
PET can be used to grade brain tumours, as malignant gliomas have a higher metabolic rate for glucose than benign tumours. This can be useful for planning stereotactic biopsies. Areas of necrosis have low metabolic rates and the surgeon can avoid biopsing these areas as they will not provide diagnostic information. In patients who clinically deteriorate following radiotherapy or chemotherapy, PET can be used to distinguish recurrent tumour from treatment induced necrosis. This distinction can be extremely difficult to make with other techniques and may even require biopsy. With recurrent tumour, PET shows hypermetabolism while necrotic areas have little or no metabolic activity. PET can thus be used to avoid a repeat biopsy and to determine the best treatment option.
Fig. 2 Dementia. The scan on the left, from a patient with clinically diagnosed Alzheimer's disease, shows reduced 18Ffluorodeoxyglucose uptake in both parietal lobes (arrowheads), compared to the normal control scan on the right.
|
|
Cerebrovascular disease
PET studies in stroke have largely been confined to the chronic state. They have included observations of changes in blood flow, oxygen extraction fraction, and oxygen and glucose metabolism distal to major occlusive disease (Fig. 3). In the acute phase of stroke, PET may be able to determine critical metabolic parameters which would predict infarction and suggest appropriate interventions, but this important potential indication requires more study. PET has also been used to study the mechanisms for recovery and rehabilitation after stroke. For example, there is increased activation of the motor cortex ipsilateral to a paralysed limb during the recovery phase of hemiplegia.
Fig. 3 Stroke. Left middle cerebral artery infarct studied one week after clinical onset. The 13Nammonia scan on the left shows increased 'luxury' cerebral perfusion (arrow), whilst the 18Ffluorodeoxyglucose metabolic scan on the right shows decreased glucose metabolism in the corresponding area (arrowheads).
|
|
Movement disorders
The diagnosis of Huntington's disease is supported by the PET finding of hypometabolism of the caudate nuclei. This functional change precedes anatomical evidence of atrophy and is used in some centres for detecting presymptomatic cases. The accuracy of PET for presymptomatic diagnosis is still debated, as are the ethical issues surrounding the presymptomatic diagnosis of this currently untreatable disorder. Characteristic patterns of glucose metabolism have also been described in progressive supra nuclear palsy and in corticobasal degeneration, as well as in certain cerebellar degenerations, although the sensitivity and specificity of these patterns awaits further study.
In Parkinson's disease, the primary biochemical deficit is in the dopaminergic system. This can be directly imaged by 18Ffluorodopa or other dopaminergic ligands. The presynaptic and postsynaptic regions can be specifically imaged and it appears that PET will help to localise the site of dopamine
deficiency in patients with parkinsonism of uncertain aetiology. This may be of great help in selecting therapy e.g. patients with a presynaptic deficit, such as idiopathic Parkinson's disease, should benefit from levodopa, whereas those with a postsynaptic receptor defect may need other strategies.
Normal brain physiology
The localisation of complex cognitive processes within the brain is incompletely understood and present data rely on electrical stimulation studies, lesion studies and animal work. Ten years ago, PET using 18 Ffluorodeoxyglucose was able to show that vision was associated with increased metabolism in the primary visual area (area 17), hearing with activation of the primary auditory areas, etc. More recently, improvements in PET technique and in the design of cognitive activating paradigms (tasks which increase focal brain activity) have led to new information regarding the cerebral localisation of language, memory, olfaction, pain perception and other complex cognitive functions.
Conclusion
PET is a complex and costly technology. After more than a decade of use, PET measurements of cerebral metabolism and blood flow in a wide variety of neurological diseases have established its value in many of them. PET helps in clinical decision making in patients with focal epilepsy, dementia and brain tumours. Applications in stroke and movement disorders are likely to be transferred from research to clinical practice in the near future. PET appears to be cost effective in the assessment of epilepsy and its economic impact in other areas awaits further evaluation. T he next decade is likely to see a burgeoning of research into neurotransmitter abnormalities in the brain using PET following the important studies of the dopamine system in parkinsonian syndromes. Abnormalities of the GABA and opiate systems have been found in patients with epilepsy and it is anticipated that PET will be a most important tool in unravelling the neurochemistry of neurological and psychiatric disorders.
Further reading
Gilman S. Advances in neurology (part 1). N Engl J Med 1992;326:1608-16.
Therapeutics and Technology Assessment Subcommittee. Assessment: positron emission tomography. Report of the Therapeutics and Technology Assessment Subcommittee of the American Academy of Neurology. Neurology 1991;41:163-7.