[^99mTc]TRODAT-1 SPECT imaging correlates with odor identification in early Parkinson disease

Impairment in olfactory function is an early manifestation of the neuropathologic process in Parkinson disease (PD),¹ and clinical deficits in the sense of smell may precede the development of overt motor symptoms. Several studies have found olfactory deficits in asymptomatic relatives of patients with PD,^2,3 some of whom have subsequently developed clinically manifest PD.⁴ In cross-sections of patients with PD, deficits in odor identification have been noted to be largely independent of motor status, medication use, or disease duration.^5–8 These studies suggest that the olfactory deficit is present, but may be largely static during the progressive course of motor impairment in PD. Studies using PET and SPECT imaging of the dopaminergic system have detected marked abnormalities in dopaminergic function either in the absence of clinical features of PD^4,9 or in the unaffected side of patients with hemiparkinsonism.¹⁰ In contrast to deficits in odor identification, imaging abnormalities appear to correlate with severity of motor dysfunction in PD.^11,12

A prior study¹³ examined odor detection threshold sensitivity, recognition memory, and identification in 13 patients with PD who also underwent SPECT imaging, using the ligand [¹²³I]β-CIT. Results from this study showed a relationship between disease severity (i.e., Hoehn & Yahr ratings) and β-CIT ratios (r = 0.76, p < 0.001). No significant correlations were observed between the olfactory measures and β-CIT uptake or other clinical ratings of disease duration, motor dysfunction, or cognitive status. These findings supported the idea that impaired olfactory function in PD is independent of physiologic measures of dopamine transporter density as well as motor features. In the current study, we evaluated the relationship between [^99mTc]TRODAT-1 (TRODAT) SPECT imaging, odor identification, and motor function specifically in patients with early PD. We sought to determine if an association between olfaction and dopamine transporter imaging is present early in the course of PD at a time when progressive loss of olfactory function may be most likely to occur.

Methods.

Results.

The 24 subjects ranged in age from 39 to 75 years (mean [± SD] = 58.7 ± 12.6) (table 1). Seventeen (71%) were men. All patients had mild symptoms, with a mean UPDRS motor score off medication of 12.3 ± 5.7 (range 4 to 23) and a Hoehn and Yahr stage of 2 or less (mean = 1.4 ± 0.5). Four patients were taking antiparkinsonian medicines at the time of the SPECT scan: one on carbidopa/levodopa only, one on agonist monotherapy, one on both carbidopa/levodopa and a dopamine agonist, and one on selegiline. The mean duration since symptom onset was 20.8 ± 12.1 months (range 3 to 48), and the mean time since diagnosis was 6.8 ± 6.5 months (range 0 to 20).

Mean UPSIT score was 26.7 ± 8.0. Based on age and sex normative data from the UPSIT,¹⁹ eight subjects were categorized as normosmic (i.e., normal sense of smell), 10 as microsmic (i.e., diminished sense of smell), and six as anosmic (i.e., complete inability to smell).

Consistent with the pattern of striatal denervation seen in PD, the mean SUV was higher in the caudate (1.29 ± 0.31) than in the putamen (0.57 ± 0.16). The SUV for the whole striatum was 0.69 ± 0.17. The differences in specific binding among the three ROI were all significant (p < 0.001 for all pairwise comparisons).

Overall, the correlations between TRODAT imaging and olfaction were much stronger than correlations between imaging and UPDRS ratings (table 2). As can be seen in table 2, UPSIT scores were related to dopamine transporter binding in the whole striatum (r = 0.66, p = 0.001). Of the ROIs examined, the strongest correlation between TRODAT values and the UPSIT was seen in the putamen (r = 0.74, p < 0.001) (figure). The correlation between the UPSIT and caudate binding was moderate (r = 0.36), but did not reach significance (p = 0.11). There were no significant correlations between any of the imaging measures and UPDRS total motor scores. There also was no association between UPSIT score and motor function as measured by the UPDRS (r = −0.12, p = 0.57).

• Download figure
• Open in new tab
• Download powerpoint

Figure. Scatterplot showing relationship between University of Pennsylvania Smell Identification Test (UPSIT) scores and [^99mTc]TRODAT-1 SPECT specific binding for the whole putamen (p < 0.001). Specific uptake values are the average of values for the left and right putamen.

Consistent with the results of the correlation analysis, a MANCOVA with impairment rating on the UPSIT (normosmic, or normal sense of smell; microsmic, or reduced sense of smell; and anosmic, or loss of smell function) as the grouping factor, and TRODAT binding in the caudate and putamen as dependent variables and age, duration of illness, and illness severity as covariates revealed a main effect of impairment category (F[2,18] = 4.4, p = 0.027), with normosmic patients showing greater TRODAT binding in the putamen relative to microsmic- (F[1,18] = 8.0, p = 0.01) or anosmic- (F[1,18] = 18.4, p = 0.0004) patients (table 3). Differences in TRODAT binding in the putamen also differed between microsmic and anosmic subgroups (F[1,18] = 4.5, p = 0.048), with anosmic subjects showing significantly reduced dopamine transporter binding compared with microsmic subjects. Identical findings were obtained when examining the whole striatum ROI. No significant differences in caudate binding were observed among the normosmic, microsmic, and anosmic subgroups.

Discussion.

The present findings reveal a strong and consistent relationship between dopamine transporter binding, as measured by TRODAT SPECT, and odor identification in patients with early PD. We found a strong correlation between UPSIT scores and dopamine transporter binding in the putamen. There was a moderate correlation between olfaction and TRODAT binding in the caudate nucleus, however, this correlation was not significant given our sample size. Ratings of motor impairment did not correlate with either TRODAT binding or UPSIT scores.

Our results suggest that olfactory testing may be a useful biomarker early in the course of PD, and may capture some of the same differences in severity of the underlying neurodegenerative process as dopamine transporter imaging in early PD. The finding of a relationship between odor identification performance and dopamine transporter binding stand in contrast to an earlier study¹³ that found that measures of odor detection threshold, recognition memory, and identification were not related to dopamine transporter binding in the striatum as measured by (¹²³I)β-CIT SPECT. The difference between our results and those of the prior study may be due to the fact that our sample included only subjects with early, mild PD, and the other study included patients with a wider range of disease severity. The mean Hoehn and Yahr stage in our cohort was 1.4 (range = 1.0 to 2.0), as opposed to 2.4 (range 1.0 to 4.0) in the other study. It may be that the association between dopamine transporter binding and odor identification is present only early in the disease process, since olfactory abilities may demonstrate a “floor-effect” that potentially masks any relationship with illness progression and dopamine transporter binding levels later in the course of the disease.

We did not find a relationship between either dopamine transporter binding or odor identification and UPDRS motor scores. The lack of association between motor scores and the imaging and olfactory measures suggests that neither TRODAT nor the UPSIT is a useful surrogate marker for motor performance in early PD. There are several potential explanations for the lack of association between motor performance and the other measures. The idea that olfactory deficits manifest early in the course of PD and reach their maximal severity even as motor impairments continue to progress may explain the lack of a relationship between disease severity and degree of olfactory impairment found in many studies, as most have examined patients with PD with greater disease duration and more disability than our cohort.^5,20–22 Moreover, although a number of studies have shown a strong relationship between SPECT imaging and disease severity in cross-sections of patients with PD,^11,23–26 this relationship has not been found when the analysis is confined to patients with early, mild disease such as ours.¹⁰ It is possible that the compensatory mechanisms that allow normal motor function to persist in spite of significant losses in dopamine transporter density in presymptomatic PD^4,9,27 may also attenuate the relationship between motor performance and dopamine transporter density in early, clinically manifest PD.

These data are consistent with previous findings documenting differences in dopamine transporter imaging between relatives of patients with PD with and without significant olfactory impairment.^4,28 Specifically, in this series of studies, a reduction in binding was observed in 4 out of 40 hyposmic relatives of patients with PD, all of whom went on to develop clinical parkinsonism. No similar reductions were seen in the normosmic relatives. These data suggest that olfactory deficits precede the clinical motor signs of PD and may reflect the earliest stage of the evolution of the pathophysiology of PD. The current findings extend this work by documenting differences in dopamine transporter binding in the striatum across different levels of odor identification impairment in patients with early stage clinically manifest PD. Patients with normal UPSIT scores (i.e., normosmic range) showed significantly higher levels of dopamine transporter binding in the putamen and striatum as a whole compared to patients with PD falling in the microsmic or anosmic ranges. The observation that the microsmic and anosmic subgroups also differed significantly from each other in mean TRODAT binding lends further support to the idea that dopamine transporter density and olfactory function are related across a range of levels of olfactory impairment in early PD.

There is evidence from the neuropathologic literature to support our findings. A recent report²⁹ described an increase (possibly compensatory) in dopaminergic neurons in the olfactory bulb in PD, and a number of studies have shown extensive extranigral changes in idiopathic PD that include the olfactory bulb and related portions of the anterior olfactory nucleus.^30,31 Braak and coworkers have put forward a neuropathologic staging system for PD in which they argue that Lewy body pathology in the olfactory bulb and anterior olfactory nucleus actually precedes degeneration of the substantia nigra.^1,32 The developers of this staging system speculate that the anterior olfactory system may represent one of the induction sites of the neuropathologic process in PD. Although Consistent with this model, degeneration in the olfactory system and the nigrostriatal system may proceed in parallel in the early stages of clinical PD, and olfactory loss may be a marker for nigrostriatal dopaminergic cell loss, even though degeneration in the olfactory system is not causally related to the clinical motor features of PD.

A few caveats must be noted. First, the sample of subjects is relatively small, and a larger cohort will be required to fully assess these findings. However, it is notable that the effect size (Cohen’s d) for the relationship between the UPSIT and dopamine transporter binding in the putamen is quite large (d₊ = +2.13, 95% CI = +1.42 < δ < +2.83), and we would expect this effect to be robust in other samples. Second, we only examined the domain of odor identification, and it is important to assess other olfactory abilities such as odor detection and discrimination to see if the relationships we found extend to other measures of olfactory function. For example, a relationship between disease severity and odor discrimination but not odor identification has been reported.⁸ However, it is important to note that most tests of olfactory function are strongly correlated and, in most instances, are likely to be measures of the same physiologic attributes.³³

One small study has suggested that progression in olfactory deficits can be measured in the early stages of PD.²¹ Larger, longitudinal studies are needed to assess the relationship between olfaction and dopamine transporter binding or other measures of the integrity of the nigrostriatal dopaminergic system in patients with recently diagnosed PD and possibly in asymptomatic, at-risk individuals.