Journal of Neurology

, Volume 259, Issue 2, pp 251–260

Visual assessment of dopaminergic degeneration pattern in 123I-FP-CIT SPECT differentiates patients with atypical parkinsonian syndromes and idiopathic Parkinson's disease

Authors

    • Department of Nuclear MedicineUniversity Hospital of Cologne
  • Carsten Eggers
    • Department of NeurologyUniversity Hospital of Cologne
  • Harald Schicha
    • Department of Nuclear MedicineUniversity Hospital of Cologne
  • Lars Timmermann
    • Department of NeurologyUniversity Hospital of Cologne
  • Matthias Schmidt
    • Department of Nuclear MedicineUniversity Hospital of Cologne
Original Communication

DOI: 10.1007/s00415-011-6163-1

Cite this article as:
Kahraman, D., Eggers, C., Schicha, H. et al. J Neurol (2012) 259: 251. doi:10.1007/s00415-011-6163-1

Abstract

The aim of this study was to investigate whether visual assessment of 123I-N-ω-fluoropropyl-2β-carbomethoxy-3β-(4-iodophenyl)nortropan (123I-FP-CIT) single photon emission computed tomography (SPECT) in addition to quantitative analyses can help to differentiate idiopathic Parkinson's disease (PD) from atypical parkinsonian syndromes (APS). From a consecutive series of patients examined with 123I-FP-CIT SPECT (n = 190) over a three-year period we identified 165 patients with a clinical diagnosis of PD (n = 120) or APS (n = 45). 123I-FP-CIT SPECT results were analysed visually and quantitatively and compared for PD and APS and for the subgroup of patients with early PD and APS (disease duration <5 years). According to predefined visual patterns of dopaminergic degeneration the results were graded as normal (grade 5) or abnormal (grade 1–4), distinguishing a posterior-anterior degeneration pattern (“egg shape”) from a global and severe degeneration pattern (“burst striatum”). Visual assessment of 123I-FP-CIT SPECT showed significant different dopaminergic degeneration patterns for PD and APS patients. A grade 1 (“burst striatum”) degeneration pattern was predominantly associated with APS patients. In contrast to that, a grade 2 (egg shape) degeneration pattern was the characteristic finding in PD patients. In a subgroup of patients with early disease, visual assessment with identification of the burst striatum degeneration pattern provided 90% positive predictive value and 99% specificity for the diagnosis of APS. Quantitative analysis of striatal binding ratios failed to depict these different degeneration patterns in PD and APS patients. Visual assessment of the pattern of dopaminergic loss in 123I-FP-CIT SPECT shows different patterns of dopaminergic degeneration for PD and APS patients. Therefore, it could provide valuable information to distinguish APS from PD patients, especially in early stages of disease. Within the first 5 years of disease, the occurrence of a burst striatum degeneration pattern has a high positive predictive value of APS.

Keywords

123I-FP-CIT SPECTIdiopathic Parkinson's diseaseAtypical parkinsonian syndromesVisual pattern assessmentDopaminergic degeneration pattern

Introduction

In clinical practice, idiopathic Parkinson's disease (PD) is diagnosed according to the criteria of the UK Parkinson’s Disease Society Brain Bank [13]. Although the diagnosis of PD is straightforward in most of the cases, findings of previous studies on the diagnostic accuracy are contradictory. Clinicopathological studies suggest that the accuracy of the clinical diagnosis in PD is only about 70–80% in comparison with the “Gold-Standard” post-mortem neuropathologic examination [4, 5]. A more recent study has shown an improvement in the diagnostic accuracy up to 90% and higher [6, 7]. However, the major diagnostic errors occur in the differentiation of PD and atypical parkinsonian syndromes (APS) such as multiple system atrophy (MSA), progressive supranuclear palsy (PSP) and corticobasal degeneration (CBD) [4, 6, 7]. Especially in patients with early PD, PD is hard to differentiate from APS, because the full triad of clinical signs and symptoms may not yet be manifested [4, 812]. In the clinical course of the disease PD and APS patients show important differences: APS patients show a faster clinical deterioration and a less favourable outcome with a shorter survival time, whereas PD patients show a slower progress of the disease [1315]. These clinical findings were reflected in results of neuropathologic studies, APS patients revealed a more global and severe striatal dopaminergic degeneration in caudate and putamen, whereas PD patients showed a pronounced striatal dopaminergic degeneration in the putamen [1621]. Thus, for disease management and therapy PD has to be differentiated from APS. In this clinical setting 123I-N-ω-fluoropropyl-2β-carbomethoxy-3β-(4-iodophenyl)nortropan (123I-FP-CIT) single photon emission computed tomography (SPECT) has gained clinical importance for analysis of the striatal dopamine deficit and assessment of the dopaminergic neurodegeneration in PD patients [22, 23]. 123I-FP-CIT SPECT has become an established procedure for differentiating PD from patients with essential tremor and from healthy volunteers. However, this diagnostic technique has so far failed to separate PD from APS [2426].

Based on the aforementioned different neuropathological findings in PD and APS patients and the different clinical course, we compared in this retrospective study quantitative analyses and visual assessment of 123I-FP-CIT SPECT in patients with PD and APS acquired consecutively over the last 3 years in our hospital. The main aim of the study was to investigate whether PD and APS patients show different degeneration patterns in the visual assessment of 123I-FP-CIT SPECT. Additionally, we wanted to assess the ability of visual assessment and quantitative analysis of 123I-FP-CIT SPECT to differentiate PD from APS, especially in patients at an early stage of disease.

Materials and methods

Patients

In a consecutive series of patients investigated with 123I-FP-CIT SPECT in our clinic from January 2007 until December 2009 (n = 190), we identified 165 patients (66 females, 99 males, mean age 63 ± 11.3 years) with a clinical diagnosis of PD (n = 120) or APS (n = 45). Another 25 patients with a clinical diagnosis of essential tremor (n = 12), vascular parkinsonism (n = 5), neuroleptic-induced parkinsonism (n = 3), psychogenic tremor (n = 3), normal pressure hydrocephalus (n = 1) and Alzheimer’s disease (n = 1) were excluded from this retrospective analysis. All patients gave informed consent. The patients received a 123I-FP-CIT SPECT in order to differentiate between PD and APS or to prove the diagnosis of PD. Before undergoing 123I-FP-CIT SPECT patients were examined by a movement disorder specialist in the Movement Disorder Clinic at the Department of Neurology, University Hospital of Cologne. PD was diagnosed according to the criteria of the UK Parkinson’s Disease Society Brain Bank in 120 patients [3, 4]. Forty-five patients were diagnosed as APS, with 14 patients fulfilling the diagnostic criteria for MSA [8], 8 patients were diagnosed as PSP [9] and in 7 patients the diagnosis CBD was established [12]. In 16 patients, APS was suspected by the referring movement disorder specialist without being clearly classified as MSA, CBD or PSP. Patients had a mean Hoehn and Yahr stage of 2.2 ± 0.6 [27]. The motor part (part III) of the Unified Parkinson’s Disease Rating Scale was used to assess the severity of disease (mean UPDRS score 31.4 ± 16.0) [3]. The scale was rated during a standardized ‘‘off’’ phase (12 h off drugs). Patients were allowed to continue their antiparkinsonian medication with the exception of selegiline, which was discontinued at least 18 h before 123I-FP-CIT SPECT [28]. Patient characteristics and clinical details are summarised in Table 1.
Table 1

Patient characteristics

 

PD and APS

PD and APS (disease duration <5 years)

All

PD

APS

All

PD

APS

N

165

120

45

118

80

38

Male/female

99/66

72/48

27/18

69/49

46/34

23/15

Mean age (years)

63 ± 11.3

62 ± 11.2*

66 ± 11.0*

64 ± 11.0

62.9 ± 11.4

66.6 ± 9.6

Mean Hoehn & Yahr

2.2 ± 0.6

2.1 ± 0.6*

2.5 ± 0.5*

2.1 ± 0.6

1.9 ± 0.6*

2.5 ± 0.5*

Mean UPDRS Part III

31.4 ± 16.0

30.3 ± 15.7

34.9 ± 16.5

27.4 ± 13.4

24.7 ± 11.8*

33.6 ± 15.2*

Disease duration (years)

5.0 ± 4.1

5.6 ± 4.5*

3.5 ± 2.8*

2.9 ± 1.5

3.0 ± 1.4

2.6 ± 1.6

l-Dopa equivalent dose

577 ± 380

538 ± 377*

688 ± 372*

458 ± 360

366 ± 306*

661 ± 394*

Significant differences between patients with PD and APS are marked with * (p < 0.05)

In order to identify different progression rates of neurodegeneration we defined a subgroup of patients in the early stages of disease according to current literature [29] and our own clinical experience, showing a clinical relevant progress of the disease from Hoehn and Yahr stage I–II within the first 5 years. This subgroup included 118 patients with PD (n = 80) or APS (n = 38) and disease duration shorter than 5 years since onset of clinical symptoms.

Data processing

According to the European Association of Nuclear Medicine procedure guidelines for brain neurotransmission with dopamine transporter ligands, all patients received potassium iodide orally to block the thyroid gland 30 min before intravenous administration of approximately 185 MBq 123I-FP-CIT. This was to prevent free radioactive iodine accumulating in the thyroid [28].

SPECT imaging was performed 3 h p.i. with a triple-headed rotating gamma camera (Picker Prism 3000) using a low-energy, high-resolution parallel hole collimator. One hundred-twenty projections were acquired over 360° in a 128*128 matrix. The unprocessed projection data were controlled with a sinogram and cine display on an Odyssey-FX workstation (Phillips Medical Systems) for possible patient motion and artefacts. The digital images were reconstructed using a low-pass filter (Cut-off 0. 4 Nyquist, order 8) and corrected with the algorithm for attenuation according to Chang [28]. Images were reoriented for transverse, coronal and sagittal planes.

Automated quantitative analysis/BRASS tool

Automated quantitative analysis was carried out with HERMES BRASS on a HERMES workstation (Nuclear Diagnostics, Stockholm, Sweden). The BRASS tool automatically fits the patient’s image data to a reference template created from healthy controls [30]. A predefined three dimensional volume-of-interest (VOI) was then placed for the quantification of specific to non-specific binding in striatum, caudate, putamen and occipital cortex [30]. To compensate for anatomic variation the fitting algorithm includes an automated adjustment of the VOI’s.

Specific 123I-FP-CIT tracer binding ratio (SBR) was calculated for caudate and putamen using the formula: [(mean counts of the target VOI − mean counts of background VOI)/(mean counts of the background VOI)].

Visual assessment

For visual interpretation, two experienced nuclear medicine physicians, who were blinded for clinical diagnosis, examined the hard-copy axial images and classified the SPECT images into five different patterns of striatal 123I-FP-CIT uptake. Before evaluation both examiners were handed an atlas including representative examples for each of the five predefined patterns (Fig. 1 exemplifies normal (grade 5) and abnormal (grade 1–4) dopaminergic binding in the striatum). The institutional visual grading system was established in a cohort of more than 190 patients [31] and based on the visual assessment grading system from Benamer et al. [32].
https://static-content.springer.com/image/art%3A10.1007%2Fs00415-011-6163-1/MediaObjects/415_2011_6163_Fig1_HTML.gif
Fig. 1

Example of different grade of 123I-FP-CIT uptake in visual assessment of 123I-FP-CIT SPECT

Grade 5 (“normal”) shows symmetric tracer uptake bilaterally in putamen and caudate.

Grade 4 (“eagle wing”) shows an almost normal, symmetrical tracer uptake with a discrete reduction in one or both putamina, usually preserving lateral parts of the putamina, creating the shape of a wing.

Grade 3 (“mixed type”) is defined as asymmetric tracer uptake with normal or almost normal uptake in the putamen of one hemisphere and reduced uptake in the contralateral putamen.

Grade 2 (“egg shape”) is defined as bilateral reduction of tracer uptake with almost no or no uptake in the putamen on either side and normal or almost normal uptake in the caudate resulting in an oval, i.e. egg shape of the caudate.

Grade 1 (“burst striatum”) is described as severe bilateral reduction with almost no uptake in either the putamen or caudate.

Statistical analysis

Statistical analyses were carried out using SPSS 17.0 (Chicago, IL, USA). Means and standard deviation were calculated for age, UPDRS score, Hoehn and Yahr grade, disease duration, l-dopa equivalent dose and the results of the quantitative analysis for the ipsi- and contralateral (to the more affected limb) putamen and caudate nucleus.

To detect significant differences between the subgroups of PD and APS patients, results of our study were evaluated by using a student t-test for independent samples in the case of normally distributed data and a non-parametric univariate Mann–Whitney U test in the case of not normally distributed data. To assess, how the different visual degeneration patterns in 123I-FP-CIT SPECT are depended on the clinical parameters age, UPDRS score, Hoehn and Yahr grade, disease duration and l-dopa equivalent dose, multivariate analysis with logistic regression was performed. The chi-square test was used for comparison of the distribution of the various visual degeneration patterns in 123I-FP-CIT SPECT between patients with PD and APS. For all tests, a p value of less than 0.05 was considered statistically significant.

Numbers of true positives, false positives, true negatives and false negatives for the clinical diagnosis APS based on the presence or absence of the burst striatum degeneration pattern in visual assessment of 123I-FP-CIT SPECT were calculated. Accordingly, numbers of true positives, false positives, true negatives and false negatives for the clinical diagnosis of PD based on the presence or absence of the egg shape degeneration pattern in 123I-FP-CIT SPECT were calculated. Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), odds ratio and their 95% confidence intervals (CI) were calculated.

Results

The baseline clinical and demographic characteristics of the patients are presented in Table 1. Multivariate analysis with logistic regression showed an association between l-dopa equivalent dose and the different visual degeneration patterns in 123I-FP-CIT SPECT (p = 0.047). In contrast, there was no association between age (p = 0.238), disease duration (p = 0.557), UPDRS score (p = 0.092), and Hoehn & Yahr (p = 0.353) and the different visual degeneration patterns.

Visual assessment

All patients (n = 165) were graded as an abnormal degeneration pattern on visual assessment of 123I-FP-CIT SPECT.

On visual assessment of 123I-FP-CIT SPECT in PD patients (n = 120), 15 (13%) cases were scored as eagle wing, 10 (8%) as mixed type, 87 (72%) as egg shape and 8 (7%) as burst striatum degeneration pattern (Fig. 2, Table 3). Among the 45 patients with APS, 4 (9%) were scored as eagle wing, 6 (13%) as mixed type, 22 (49%) as egg shape and 13 (29%) as burst striatum degeneration pattern (Fig. 2, Table 2). The subgroup of patients in the early stages of disease, defined as patients with disease duration of less than 5 years since onset of symptoms (n = 118), included 80 patients with early PD and 38 patients with early APS. Among the 80 patients with PD, 12 (15%) were scored as eagle wing, 9 (11%) as mixed type, 58 (73%) cases as egg shape and 1 (1%) as burst striatum degeneration pattern (Fig. 3, Table 3). Among the 38 patients with APS, 4 (10%) were scored as eagle wing, 6 (16%) as mixed type, 19 (50%) as egg shape and 9 (24%) as burst striatum degeneration pattern (Fig. 3, Table 3).
https://static-content.springer.com/image/art%3A10.1007%2Fs00415-011-6163-1/MediaObjects/415_2011_6163_Fig2_HTML.gif
Fig. 2

Proportions of visual degeneration patterns for PD and APS patients in the overall group. Significant differences between subgroups are marked with asterisks (p < 0.05)

Table 2

Results of visual assessment of 123I-FP-CIT SPECT in PD and APS patients in the overall group

Diagnosis

Visual analysis (overall)

Burst striatum

Egg shape

Mixed type

Eagle wing

PD (n = 120) No. (%)

8 (7%)

87 (72%)

10 (8%)

15 (13%)

APS (n = 45) No. (%)

13 (29%)

22 (49%)

6 (13%)

4 (9%)

https://static-content.springer.com/image/art%3A10.1007%2Fs00415-011-6163-1/MediaObjects/415_2011_6163_Fig3_HTML.gif
Fig. 3

Proportions of visual degeneration patterns for patients in the subgroup with early PD and APS. Significant differences between subgroups are marked with asterisks (p < 0.05)

Table 3

Results of visual assessment of 123I-FP-CIT SPECT in the subgroup of patients with early PD and APS (disease duration <5 years)

Diagnosis

Visual analysis (early PD and APS)

Burst striatum

Egg shape

Mixed type

Eagle wing

PD (n = 80) No. (%)

1 (1%)

58 (73%)

9 (11%)

12 (15%)

APS (n = 38) No. (%)

9 (24%)

19 (50%)

6 (16%)

4 (10%)

Chi-square test showed statistically significant differences in the distribution of the burst striatum degeneration pattern (p = 0.001) and in the distribution of the egg shape degeneration pattern (p = 0.004) between patients with PD and APS in the overall group (Fig. 2). In contrast, there were no significant differences in the distribution of the mixed type degeneration pattern (p = 0.334) and in the distribution of the eagle wing degeneration pattern (p = 0.518) between PD and APS in the overall group (Fig. 2).

In the subgroup of patients with early PD and APS, chi-square test showed statistically significant differences in the distribution of the burst striatum degeneration pattern (p = 0.001) and in the distribution of the egg shape degeneration pattern (p = 0.016) between patients with PD and APS (Fig. 3). In contrast, there were no significant differences in the distribution of the mixed type degeneration pattern (p = 0.489) and in the distribution of the eagle wing degeneration pattern (p = 0.507) between PD and APS patients (Fig. 3).

Table 4 demonstrates the numbers of true positives, false positives, true negatives and false negatives for the clinical diagnosis APS based on the presence or absence of the burst striatum degeneration pattern in visual assessment of 123I-FP-CIT SPECT.
Table 4

Numbers of true positives, false positives, true negatives and false negatives for the clinical diagnosis APS based on the presence or absence of the burst striatum degeneration pattern in visual assessment of 123I-FP-CIT SPECT in the overall group and in the subgroup of patients with early PD and APS

Burst striatum

True positives

False positives

True negatives

False negatives

Overall group (No.)

13

8

112

32

Early group (No.)

9

1

29

79

In the overall group, the sensitivity of the burst striatum degeneration pattern is 29% (95%-CI 16–42%), specificity 93% (95%-CI 88–98%), the PPV 62% (95%-CI 41–83%), the NPV 78% (95%-CI 71–85%) and the diagnostic accuracy 76% (95%-CI 68–84%). For early PD and APS patients the sensitivity of the burst striatum is 24% (95%-CI 10–38%), specificity 99% (95%-CI 97–100%), the PPV 90% (95%-CI- 71–100%), the NPV 73% (95%-CI 65–81%) and the diagnostic accuracy 75% (95%-CI 67–83%). The odds ratios for APS in patients with a burst striatum degeneration pattern were, in the overall group, 5.69 (95%-CI 2.17–14.92) and in the early patients group 25.53 (95%-CI 2.97–202.09), respectively.

Table 5 demonstrates the numbers of true positives, false positives, true negatives and false negatives for the clinical diagnosis PD based on the presence or absence of the egg shape degeneration pattern in visual assessment of 123I-FP-CIT SPECT.
Table 5

Numbers of true positives, false positives, true negatives and false negatives for the clinical diagnosis PD based on the presence or absence of the egg shape degeneration pattern in visual assessment of 123I-FP-CIT SPECT in the overall group and in the subgroup of patients with early PD and APS

Egg shape

True positives

False positives

True negatives

False negatives

Overall group (No.)

87

22

23

33

Early group (No.)

58

19

19

22

In the overall group, the sensitivity of the egg shape pattern is 73% (95%-CI 65–81%), specificity 51% (95%-CI 36–66%), the PPV 80% (95%-CI 72–88%), the NPV 41% (95%-CI 29–53%) and the diagnostic accuracy 67% (95%-CI 60–74%). For early PD and APS patients the sensitivity of an egg shape pattern is 73% (95%-CI 63–83%), specificity 50% (95%-CI 34–66%), the PPV 75% (95%-CI 65–85%), the NPV 46% (95%-CI 31–51%) and the diagnostic accuracy 65% (95%-CI 56–74%). The odds ratios for PD in patients with an egg shape degeneration pattern were in the overall group 2.76 (95%-CI 1.36–5.60) and in the early patients group 2.63 (95%-CI 1.18–5.89), respectively.

Quantitative analysis

As shown in Table 6, SBR of 123I-FP-CIT in putamen and caudate ipsi- and contralateral to the most affected side failed to differentiate PD and APS in either group. A comparison of SBR of 123I-FP-CIT for the two groups revealed no statistically significant difference between PD and APS patients in any of the striatal subregions (overall group >0.05; Table 6).
Table 6

Comparison of striatal binding ratios (SBR) for 123I-FP-CIT SPECT

 

PD and APS

PD and APS (disease duration <5 years)

Caudate nucleus

Putamen

Caudate nucleus

Putamen

Contralateral

Ipsilateral

Contralateral

Ipsilateral

Contralateral

Ipsilateral

Contralateral

Ipsilateral

PD

1.17 ± 0.50

1.32 ± 0.51

0.73 ± 0.46

0.86 ± 0.46

1.31 ± 0.47

1.47 ± 0.49

0.84 ± 0.46

0.99 ± 0.44

APS

1.06 ± 0.60

1.21 ± 0.62

0.75 ± 0.54

0.82 ± 0.56

1.14 ± 0.60

1.28 ± 0.62

0.82 ± 0.55

0.88 ± 0.55

Values are mean ± standard deviation, ipsi- and contralateral (to the clinically affected body side). None of the differences are statistically significant (p > 0.05)

Discussion

Previous studies in patients with suspected PD reported that 123I-FP-CIT SPECT enables patients with PD to be differentiated from those with essential tremor or from healthy volunteers by visual assessment and quantitative analysis, but cannot be used to differentiate between patients with PD and APS [2426]. This is also supported by our findings demonstrating that for the majority of patients the differentiation between PD and APS patients was not possible by visual assessment and quantitative analysis of 123I-FP-CIT SPECT. The majority of patients in both groups had either an egg shape, a mixed type or an eagle wing degeneration pattern in the visual assessment. Such a finding did not allow the physician to distinguish between APS and PD patients. However, special attention had to be paid to patients with a burst striatum degeneration pattern indicative of a severe dopaminergic deficit. The occurrence of a burst striatum in the visual assessment of 123I-FP-CIT SPECT within the first 5 years after onset of clinical symptoms was highly indicative of APS (positive predictive value >90%). In contrast, a burst striatum is an unusual finding in PD patients as neurodegeneration usually proceeds slower in this patient group. Thus, the findings of our retrospective study suggest that the visual assessment of 123I-FP-CIT reflects different patterns of neurodegeneration in PD and APS patients. A burst striatum indicates a more severe status of neurodegeneration than an egg shape, a mixed type or an eagle wing pattern. This is especially valid for patients with a short disease period and a time period of less than 5 years since onset of clinical symptoms. Visual assessment of 123I-FP-CIT SPECT could provide reliable information for differentiating between patients with PD and APS in cases of clinically uncertain PD in the early stages of disease.

Different patterns of neurodegeneration in PD and APS patients are known from post-mortem studies that showed topographical differences in nigrostriatal degeneration in PD and APS [1621, 33]: APS patients revealed a more global and severe striatal dopaminergic degeneration in caudate and putamen whereas PD patients showed a pronounced striatal dopaminergic degeneration in the putamen [1621]. Thus, we hypothesized that these neuropathological findings might be the reason for different 123I-FP-CIT uptake patterns in PD and APS patients, as has been shown in PET studies using 18F-dopa as a marker of nigrostriatal degeneration [34]. The phrase, “from a comma to a dot” is a well-known description for the ongoing changes in the course of PD and reflects the dopaminergic degeneration which proceeds from the posterior to the anterior striatum [19, 21, 33, 35]. This predominant dopaminergic degeneration in the putamen in PD patients leads to a characteristic egg shape degeneration pattern in visual assessment of 123I-FP-CIT SPECT. The results of our study show that an egg shape pattern as a result of the posterior-anterior degeneration was the characteristic finding in patients with PD. However, this egg shape pattern was also the most prevalent finding in the APS group with 22/45 (49%) APS patients having an egg shape type of neurodegeneration. But APS patients can show a more extensive and global cell striatal degeneration in the caudate and the putamen, which was reflected in a severe bilateral reduction of 123I-FP-CIT uptake producing a burst striatum pattern on 123I-FP-CIT SPECT [1618]. A burst striatum degeneration pattern was predominately associated with APS patients in our study. Simultaneously, it reflects the faster clinical progress in APS patients compared with PD patients. In case of an egg-shape degeneration pattern a reliable differentiation between APS and PD is not possible, even if the statistical analysis revealed a significant difference between APS and PD patients, because 49% APS patients in the overall group and 50% of the early APS patients showed a grade 2 degeneration pattern.

Quantitative analysis of 123I-FP-CIT SPECT with the BRASS tool in our study failed to reveal these differences. SBR was higher in PD patients than in APS patients but the difference was not statistically significant. In line with previous studies we could not differentiate PD and APS patients by quantitative analysis of 123I-FP-CIT SPECT [2426]. There are several reasons which may explain this finding. The majority of patients had an egg shape, a mixed type or an eagle wing pattern (112/120 PD and 32/45 APS patients). In comparison to these patients, the group of patients with a burst striatum (8/120 PD and 13/45 APS patients) was much smaller and statistical analysis was dominated by the aforementioned patients. Therefore, different patterns of dopaminergic degeneration in PD and APS could not be depicted by quantification. Furthermore, quantification relies on the specific uptake ratio of the radioligand in either the caudate or the putamen in relation to the background signal, whereby the distribution of the radioligand is not taking into account. Visual assessment of 123I-FP-CIT SPECT allows a more subtle analysis of differences in visual degeneration pattern between PD and APS in comparison to the gross anatomy of regions using the BRASS tool. An experienced observer has the ability to identify extremely small differences in visual degeneration pattern which cannot be measured by quantification.

The clinical diagnosis of PD is straightforward most of the time, but clinicopathological studies revealed that up to 20% of patients still have an alternative parkinsonian syndrome [4, 7]. Even if the diagnostic accuracy improves with disease duration as the typical clinical symptoms and signs become evident, it is complicated by the presence of common symptoms for different PD, especially in the early stages of disease [4, 812, 36, 37]. Thus, the current clinical diagnostic criteria for early PD and early APS are insufficient. However, for setting a treatment strategy and defining prognosis an early differentiation between PD and APS is important. In comparison with PD, APS show a more severe and faster clinical deterioration with worse outcome and shorter survival time [1315, 38]. Furthermore, patients with APS are poorly responsive to antiparkinsonian treatment and there are no reports of such treatment influencing survival time [12, 3840]. Even if patients with APS are responsive to antiparkinsonian treatment, the degree of response is insufficient to alter overall survival [13, 3840]. This contrasts with PD, where a much higher proportion of patients respond to levodopa, with recent studies suggesting that levodopa reduces mortality in PD [41]. Thus, improving accuracy in differentiating APS from PD should help to avoid administration of ineffective and inappropriate antiparkinsonian therapy with possibly serious, adverse events and to decrease therapy costs [4245]. Our data suggests that visual assessment of 123I-FP-CIT SPECT could provide information to help differentiating APS from PD patients, especially in the early stages of disease and in clinically uncertain cases.

A major shortcoming of our study is that visual assessment has, in general, a greater bias of subjectivity and the problem of reliable reproducibility in comparison to quantitative assessment. It depends on the experience and knowledge of the observer to assess the distribution of the radioligand uptake and to analyse the dopaminergic degeneration pattern. Therefore it is reasonable to assume a bias for over- or undergrading in the classification of the visual analysis. With a larger number of observers and their consensus diagnosis or systematic teaching it would be possible to diminish diagnostic uncertainty. An automated pattern recognition technique for evaluation of 123I-FP-CIT could be a possible solution for a more objective pattern assessment [46]. Additionally, a pictorial atlas with normal and pathologic axial images for visual assessment of 123I-FP-CIT could improve diagnostic accuracy [47]. A second limitation of our study is that no pathological data were available and that the gold standard for image validation was the initial clinical diagnosis and not a neuropathological diagnosis. To reduce bias two independent observers read the clinical records critically for checking the clinical criteria of one of the parkinsonian syndrome. No clinical follow-up was carried out to refine the clinical diagnosis. Another potential limitation is that we as a specialist movement disorder service see more frequently parkinsonian syndromes with atypical presentations and unusual cases than in general hospitals leading to a possible bias in the frequency of APS in our data.

In conclusion, 123I-FP-CIT SPECT enables dopaminergic degeneration in PD and APS to be visualized. We were able to demonstrate that PD and APS show different patterns of dopaminergic degeneration on 123I-FP-CIT SPECT. Furthermore, visual assessment of 123I-FP-CIT SPECT could provide valuable information to distinguish APS from PD patients in clinically uncertain cases, as this is often the case in the early stages of disease. Within the first 5 years of disease, a burst striatum degeneration pattern had a high positive predictive value of APS. The presence of this global and severe dopaminergic deficit indicating rapidly progressing neurodegeneration supports the clinical diagnosis of APS.

Conflict of interest

Lars Timmermann received honoria for lectures on symposia with sponsoring of GE medical. None of the other authors has any financial disclosures regarding the present manuscript.

Copyright information

© Springer-Verlag 2011