Positron emission tomography in the differential diagnosis of parkinsonism.

Positron emission tomography (PET) studies on presynaptic dopaminergic function can reveal hypofunction in early Parkinson's disease (PD) which may help in the early diagnosis especially in patients with mild symptoms. This hypofunction can be detected with fluorodopa (reflecting mainly aromatic amino acid decarboxylase activity of nigrostriatal terminals) or dopamine transporter ligands. These studies can also help to distinguish PD from essential tremor. However, investigations of presynaptic dopaminergic function are not useful in the differential diagnosis of parkinsonian syndromes. PET ligands, such as fluorodeoxyglucose (reflecting glucose metabolism) and dopamine receptor ligands, reflecting striatal neuronal function are better in this respect. Cardiac sympathetic function studies represent a new and interesting approach to improve differential diagnosis of parkinsonian syndromes but more studies are needed in larger patient populations with longer follow-up to evaluate the usefulness of these investigations. Multitracer approach combining ligands reflecting different aspects of dopaminergic neurotransmission and other physiological function will increase differential diagnostic accuracy.

Positron emission tomography (PET) is a technique which employs short-lived radioactive isotopes and by which it is possible to study different physiological brain functions in living humans. PET allows versatile investigation of cerebral blood flow, oxygen consumption, metabolism and the function of various neurotransmitter systems.

In Parkinson’s disease (PD), because of the key biochemical deficit being impairment of the nigrostriatal dopaminergic system, most of PET imaging has been concentrated on examining dopaminergic function. In PD PET can be used a) to study pathophysiology, b) in diagnostics / differential diagnostics, c) in follow-up, d) in detection of asymptomatic cases, e) in development of treatment and in f) monitoring of treatment effects (drugs, stimulations, lesions, transplantation, etc.).

Typical Pet Findings in Parkinson’s Disease

The typical PET findings in PD are listed in Table 1. The most commonly used PET tracer (radiopharmaceutical) to investigate brain dopaminergic function in PD is 3, 4-dihydroxy-6-[18F] fluoro-L-phenyl alanine (18F-FDOPA). The uptake of FDOPA mainly reflects the uptake of FDOPA into presynaptic nigrostriatal terminals and its conversion by aromatic amino acid decarboxylase to [18F]dopamine and storage in vesicles in the pre-synaptic terminals nigrostrital dopa-minergic neurons.

Table 1.

Typical PET changes in brain dopaminergic function and metabolism in patients with early Parkinson’s disease

Asymmetry Impaired dopaminergic function especially in the (posterior) putamen Relative preservation of the caudate nucleus Increased caudate / putamen-ratio Striatal postsynaptic receptor number and metabolism preserved in the early phase

PET: positron emission tomography.

When compared to healthy controls FDOPA uptake is reduced in early unmedicated PD patients to 30-40% in the putamen.1–4 The reduction shows topographical organization, since the posterior part of the putamen (receiving its projections from ventrolateral substantia nigra) is most severely affected. In the caudate nucleus the decline is smaller, the uptake values being 60–70% of control mean.1–4 During disease progression there is further decline in FDOPA uptake, but the reduction seems to be exponential, or at least non-linear5,6 being faster in the beginning of the disease and the slowing down. However, even during progression dopaminergic function in the putamen is more severely affected than in the caudate nucleus. The severity motor symptoms, especially bradykinesia and rigidity are related to the degree of striatal dopaminergic hypofunction striatal as indicated by an association between FDOPA uptake and “off” stage unified Parkinson disease rating scale (UPDRS) motor score (r=−0.50 – −0.68).7–9

There are also other dopaminergic PET tracers. By choosing different radioligand it is possible to study presynaptic dopaminergic function (metabolism and storage by FDOPA), reuptake back to presynaptic terminals by membrane transporter [dopamine transporter (DAT) ligands, e.g. [18F]CFT, [11C] RTI 32, [123I]-beta-CIT-FP], vesicular monoamine transport [vesicular monomaine transporter (VMAT) ligands, e.g. [11C] dihydrotetrabenazine].

Membrane dopamine transporter ligand uptake is reduced usually to 20–30% of the control in the putamen early PD.4,10,11 When compared in same patients, DAT binding seems to be more severely affected than FDOPA uptake.4,12 This may be partially due to compensatory mechanism since by down regulating DAT and enhancing the activity of aromatic L-amino acid decarboxylase (AADC) enzyme nigrostriatal system tries to compensate for reduction in dopamine levels.13 Vesicular monamine trasporter was reported to be reduced in caudate nucleus (−44%), anterior putamen (−68%), and posterior putamen (−77%) from mean control value.14

Dopamine D2 receptors are initially upregulated in early PD, and are relatively preserved also in the more advanced stages of the disease15,16 whereas dopamine D1 receptors are preserved.17,18

Pet in the Differential Diagnosis of Parkinsonism

The possibilities to use PET in the differential diagnostics of parkinsonism are listed in Table 2.

Table 2.

Possibilities to use PET in the differential diagnostics in parkinsonism

With PET it is possible to detect - Impairment of nigrostriatal dopaminergic function - The state of striatal neurons - Changes in cortical and subcortical energy metabolism - Changes in cardiac sympathetic function

PET: positron emission tomography.

Essential tremor may be a differential diagnostic problem with PD. Patients with pure essential tremor generally have normal striatal dopaminergic function in imaging.19–21 However, patients with isolated resting tremor or essential tremor combined with resting tremor may show impairment in striatal dopaminergic function suggesting underlying nigrostriatal pathology and possible “forme frustre” of PD.21

Although patients with PD show typical findings in FDOPA-PET, its ability to differentiate between PD and multiple system atrophy (MSA) or progressive supranuclear palsy (PSP) is relatively poor.1,22 This is understandable if one considers nigral pathology in these disorders. The degree and to some extent also topography of nigral neuronal loss is quite similar in these disorders, especially in PD and MSA (Figure 1).

Figure 1.

Neuronal loss (percentage from control mean) in post mortem samples of the substantia nigra in normal ageing, Parkinson’s disease, Steele-Richardson-Olszewski syndrome (progressive supranuclear palsy), multiple system atrophy.42 PD: Parkinson’s disease, MSA: multiple system atrophy.

In the striatum dopamine receptors are mainly situated on postsynaptic striatal neurons and thus dopamine receptor binding indirectly reflects the state of striatal neurons. In PD the striatum is relatively well preserved whereas in MSA, PSP and corticobasal degeneration (CBD) there is striatal degeneration. Therefore, in PD D2 binding is generally normal (or even shows initial upregulation) but is reduced in MSA and PSP.1,23 In CBD the impairment of dopaminergic system is asymmetrical, usually also affecting the caudate nucleus in addition to the putamen, as can be seen as reduced FDOPA uptake whereas D2 receptor binding is less consistently affected.24–26

Fluorodeoxyglucose (FDG) is a glucose analogue the up-take of which indirectly reflects neuronal and synaptic activity. Striatal FDG uptake in PD is normal (or even hypermetabolism at early stage may be seen). In contrast, in MSA, already at early stage of the disease striatal FDG uptake is reduced indicating striatal neuronal dysfunction and degeneration.23,27,28 Similarly in PSP and CBD striatal hypometabolism is seen, which is usually asymmetrical in CBD and affects especially caudate in PSP.27

In addition to changes in FDG uptake in the striatum, also cortical changes in glucose metabolism can be detected in different parkinsonian syndromes. In PD in non-demented patients cortical metabolism is relatively preserved, but some reduction can be seen in cortical motor areas and parietal cortex.27 In PSP frontal and brainstem, and sometimes caudate and parietal cortex hypometabolism has been reported.27 In CBD impairment in glucose metabolism is asymmetrical (worse on the hemisphere contralateral to predominant symptoms) and is reduced in caudate nucleus, putamen, thalamus and parietal and insular cortex.25,27

PD is associated with several autonomic manifestations, including orthostatic hypotension, gastrointestinal and genitourinary dysfunction. Multiple studies in PD employing different PET (such as [11C]meta-Hydroxyephedrine) or single photon emission computed tomography (SPECT) ligands (such as [123I]-metaiodobenzylguanidine ([123I]-MIBG) have shown striking reductions in uptake of the ligand in myocardium indicating cardiac sympathetic denervation.28–31 Some investigators have found cardiac sympathetic denervation in the early stages of PD,32,33 others reported it only in patients at later stages of the disease.34 The neuropathologic basis of the denervation has been thought to be neurodegenerative, with Lewy body deposition in the cardiac plexus.35 Imaging of cardiac sympathetic function has been suggested to be helpful in separating patients with PD from those with MSA, since impairment is found in PD whereas the uptake is generally normal in MSA.29,33,36,37 However, the separation is not perfect since reduced uptake in both of these disorders has been reported, although the reductions were smaller than those in PD.38 A study comparing cardiac and sympathetic denervation found that a) cardiac sympathetic denervation was found to occur not only in PD but also in other parkinsonian syndromes, such as MSA and PSP, b) cardiac sympathetic denervation was not correlated with striatal denervation, suggesting that the pathophysiologic processes underlying cardiac and striatal denervation occur independently in patients with parkinsonian syndromes.39

The problem of the PET studies in the relatively small number of patients in each study. Often the studies are cross-sectional without longitudinal follow-up. The clinical diagnosis of parkinsonian syndromes is not always correct and the diagnosis may become apparent during longer follow-up. In neuropathologically confirmed series of 100 clinically diagnosed patients the misdiagnosis of PD was 10%.40 In a study investigating dopamine transporter function with [123I]-2-beta-carbomethoxy-3-beta-(4-iodophenyl)-tropane ([123I]-beta-CIT) in different parkinsonian syndromes and using the diagnosis at 6 month clinical follow-up as a “golden standard” found that baseline [123I]-beta-CIT scan result was in disagreement with the final diagnosis in less than 10% of cases whereas the initial clinical diagnosis was in disagreement with the final diagnosis in 20% of cases.41 Of course, it would be desirable to have even longer follow-up with post mortem confirmation of the diagnosis.

Conclusions and Future Directions

PET studies on presynaptic dopaminergic function can reveal hypofunction in early PD which may help in the early diagnosis especially in patients with mild symptoms. These studies can also help to distinguish PD from essential tremor. However, investigations of presynaptic dopaminergic function are not useful in the differential diagnosis of parkinsonian syndromes. PET ligands reflecting striatal neuronal function are better in this respect. Cardiac sympathetic function studies represent a new and interesting approach to improve differential diagnosis of parkinsonian syndromes but more studies are needed in larger patient populations with longer follow-up to evaluate the usefulness of these investigations. In the future new ligands targeted for different protein aggregations seen in these disorders will probably increase specificity. Also new automated region-of-interest and voxel-based analysis methods may be helpful as suggested by initial experience. Multi-tracer approach combining information from different neurotransmitter or other physiological functions may further increase differential diagnostic accuracy. Furthermore, comparison and evaluation of the relative usefulness of different imaging modalities (e.g. PET vs. new MRI techniques) needs to be done to be able to choose the best imaging investigations to help in the clinical differential diagnosis of parkinsonism.