Laterality of specific binding ratios on DAT-SPECT for differential diagnosis of degenerative parkinsonian syndromes

Motor symptoms of Parkinson’s disease (PD) occur unilaterally and progress with asymmetry, while progressive supranuclear palsy (PSP) and multiple system atrophy of the parkinsonism subtype (MSA-P) lack this tendency. We assessed the laterality of specific binding ratios (SBRs) on dopamine transporter single-photon emission computed tomography (DAT-SPECT) for the differential diagnosis of these diseases in 311 PD, 33 PSP, 20 MSA-P, and 137 control patients. The average SBR in PD was higher than that in PSP (P = 0.035). Compared with Hoehn–Yahr (HY) stages, the average SBR in PD with HY stage I was only higher than that in PSP (P < 0.001). SBR laterality in PD with HY stage I was significantly higher than that in PSP (P = 0.001). This difference was not observed in PD with HY stage II. The average and laterality of SBRs in MSA-P were similar to those in PD and PSP. The asymmetry indices were similar among PD, PSP, and MSA-P. These data suggest that PSP shows a pattern of SBRs different from that in PD, attributed to HY stage I in PD. The limited usefulness of DAT-SPECT may be explained by the low discrimination between PD with bilateral motor symptoms and PSP.


Results
Average SBRs on DAT-SPECT in the PD, PSP, and MSA-P groups. We first compared the demographic data in the PD, PSP, MSA-P, and control groups ( Table 1). The age at the time of DAT-SPECT imaging in the PD group was significantly lower than that in the control group (P = 0.001). There was no difference in the age at the time of DAT-SPECT imaging among the PSP, MSA-P, and control groups. Additionally, there was no difference in disease duration at the time of DAT-SPECT imaging among the PD, PSP, and MSA-P. The proportions of men were comparable among all groups.
To assess the clinical usefulness of DAT-SPECT in the diagnosis of degenerative parkinsonian syndromes, we compared the average SBRs in the PD, PSP, MSA-P, and control groups (Fig. 1). The median values of the average SBRs were 2.37 [interquartile range (IQR) 1.58-3.28] in the PD group, 1.55 (IQR 0.66-2.39) in the PSP group, and 2.64 (IQR 1.50-3.72) in the MSA-P group. The values of the average SBRs in these degenerative parkinsonian syndromes were significantly lower than the value in the control group (median 4.97; IQR 4.01-6.08), (P < 0.001 in each case). Additionally, the value of the average SBR in the PD group was significantly higher than that in Table 1. Summary of patient background of the PD, MSA-P, PSP and control groups. Data are expressed as a median and interquartile range (IQR) or n (%). PD Parkinson's disease, PSP progressive supranuclear palsy, MSA-P multiple system atrophy of the parkinsonism subtype, HY Hoehn-Yahr stage. Duration shows years of DAT-SPECT imaging from onset. *indicates P < 0.01 as compared with the control group (SPSS software, version 23.0). † indicates P < 0.01 and † † indicates P < 0.001 as compared with the PD group with HY stage I.  Average SBRs on DAT-SPECT in the PD with different Hoehn-Yahr stages, PSP and MSA-P groups. As compared with the PD group with Hoehn-Yahr (HY) stage I, the age at the time of DAT-SPECT imaging was significantly higher in the PD group with HY stage IV (P = 0.002) ( Table 1). Disease duration was significantly longer in the PD groups with HY stage III and IV (P < 0.001 in each case). The proportions of men were comparable among different HY stages. When we assessed average SBRs in the different HY stages of PD, the values of the average SBRs decreased with progression in the HY stage (HY I vs. HY II, P = 0.039; HY II vs. HY IV, P = 0.002) (Fig. 2). To assess whether disease progress of PD affects the difference in the average SBRs between PD and PSP, we compared the average SBRs among the PD with different HY stages and PSP groups. The value of the average SBR in the PD group with HY stage I was significantly higher than that in the PSP group (P < 0.001) (Fig. 2). However, there was no significant difference in the average SBRs between the PD group with HY stage II or higher and the PSP group (HY II vs. PSP, P = 0.146; HY III vs. PSP, P = 1.000; HY IV vs. PSP, P = 1.000). Additionally, there was no difference in the values of the average SBRs between the PD group with HY stage I and MSA-P group (P = 1.000), while the value of the average SBR in the PD group with HY stage IV was significantly lower than that in the MSA-P group (P = 0.020).
SBR laterality on DAT-SPECT in the PD, PSP, and MSA-P groups. To assess the usefulness of SBR laterality on DAT-SPECT in the diagnosis of degenerative parkinsonian syndromes, we calculated the absolute values of SBR laterality by simply subtracting the SBR value of one side from that of the contralateral side. The values of SBR laterality were 0.53 (IQR 0.26-0.88) in the PD group, 0.44 (IQR 0.22-0.77) in the PSP group, 0.55 (IQR 0.31-1.17) in the MSA-P group, and 0.29 (IQR 0.16-0.48) in the control group (Fig. 3a). The values of SBR laterality in the PD and MSA-P groups were significantly higher than the value in the control group (PD vs. control, P < 0.001; MSA-P vs. control, P = 0.010). There was no difference in the values of SBR laterality between the PSP and control groups (P = 1.000). Additionally, the values of SBR laterality were comparable among the PD, PSP, and MSA-P groups (P = 1.000 in each case). When we assessed SBR laterality in the different HY stages of PD, the values of SBR laterality decreased between HY stages I and II (P < 0.001) (Fig. 3b). The value of SBR laterality in the PD group with HY stage I was significantly higher than that in the PSP group (P = 0.001). However, the difference in SBR laterality was not observed in the PD group with HY stage II or higher (HY II vs. PSP, HY III vs. PSP, HY IV vs. PSP, P = 1.000 in each case). There was no difference in the values of SBR laterality between the PD group with HY stage I and MSA-P group (P = 0.538). These AI values were significantly higher than the values in the control group (median 5.60; IQR 3.04-10.43) (P < 0.001 in each case). However, no significant difference was observed among the PD, MSA-P, and PSP groups (P = 1.000 in each case). When the AI was assessed in the different HY stages of PD, the values of the AI did not constantly show an decreasing tendency with progression in the HY stage (Fig. 4b). There were no differences in the values of the AI among the PD group with HY stage I, PSP group, and MSA-P group (P = 1.000 in each case).

Discussion
A previous study showed that striatal 123 I-Ioflupane accumulation is reduced in PD, PSP, and MSA-P, and the striatal reduction is more significant in PSP with no statistical difference between that in PD and MSA-P 15,16 . In this study, we found that the average SBR in MSA-P showed no difference from that in PD and PSP. Additionally, we found that the average SBR in PSP is significantly lower than that in PD. On assessing the values of the average SBRs in different HY stages, there was a difference in the values between PSP and PD with HY stage I, but the difference was not observed in PD with HY stage II or higher. These findings suggest that the value of the  . In support of this idea, the value of SBR laterality in PD with HY stage I was only higher than that in PSP. These findings suggest that patients with unilateral parkinsonism have a high probability of PD or MSA-P on DAT-SPECT, and the SBRs on DAT-SPECT are not useful for the differential diagnosis of patients with bilateral parkinsonism. Low discrimination in patients with bilateral parkinsonism may yield a limited usefulness of DAT-SPECT in the differential diagnosis of PD and PSP. In most patients with PD, there is a substantial asymmetry of clinical motor symptoms from disease onset. A previous pathological study of 21  www.nature.com/scientificreports/ contralateral to the initially affected side 17 . However, it is unclear as to the mechanism generating asymmetry of clinical motor symptoms. One possibility is that the number of nigral neurons on one side is different from that on the other side from birth, and the side with the reduced number of neurons reaches the threshold of clinical manifestations earlier 18 . Second possibility is that nigral neurons on one side are more vulnerable than those on other side 18 . Once the degenerative process starts, cell death occurs on the vulnerable side earlier. The present data showed the value of SBR laterality in the control group was the lowest in the subject groups. This finding may support the latter possibility from the point of striatal 123 I-Ioflupane accumulation on DAT-SPECT. Additionally, the present data showed that the values of SBR laterality were significantly compressed from HY stage II. It raises an idea that nigrostriatal dopaminergic neurons contralateral to the initially unaffected side may accelerate degeneration in HY stage II. This idea may explain the difficulty of DAT-SPECT for the differential diagnosis of patients with bilateral parkinsonism. The AI is commonly used as an indicator to evaluate SBR laterality 14 . However, the present data demonstrated that the values of the AI were not useful to differentiate PD with HY stage I from PSP, and they failed to show a decreasing tendency with progression of HY stage. In contrast, SBR laterality reflected progress of PD, because it matched with unilateral decrease of striatal 123 I-Ioflupane accumulation in HY stage I and loss of unilaterality in more advanced HY stages. This might be attributed to the definition of the AI. The value of the AI was calculated using the following formula: | SBRleft -SBRright | / [(SBRleft + SBRright)/2] × 100 7,19 . As the value of the SBR on one side approaches zero in the advanced stage, the value of the AI increases, generating variability in the values of the AI.
It should be noted that the age at the time of DAT-SPECT imaging in the PD group was significantly younger than that in the control group. Generally, the values of SBR tend to decrease with aging 20 . This finding suggests that the difference in average SBRs between the PD and control groups in this study might be smaller than that in the age-matched comparison. However, there was no significant difference in the age among the PD, PSP and MSA-P groups, showing that the age did not affect the comparison of SBR values among these diseases. In addition, the age at DAT-SPECT imaging in the PD groups with HY stage I was significantly younger than that in the PD group with HY stage IV. This may attenuate the difference in SBR values between HY stages I and IV. However, there was no difference in the age among the PD groups with HY stages I, II, and III, showing that the age did not affect the comparison of SBR values among these HY stages. There were several limitations of this study. The imaging system of DAT-SPECT used in this study did not set the region of interest at the subregional levels, such as caudate, anterior putamen, and posterior putamen, although subregional analysis may provide more detailed information regarding the clinical usefulness of DAT-SPECT. The diagnosis of PD was not confirmed by the pathological analysis. Additionally, the sample sizes of the PSP and MSA-P groups in this study were relatively small, providing a limitation of this study. Further examination using larger cohorts are needed to clarify the clinical usefulness of DAT-SPECT.

Methods
Patients. Subjects were 672 patients, who visited Osaka Medical College Hospital and were examined using DAT-SPECT between March 14, 2014 and December 31, 2018. When patients were regularly given drugs, including selective serotonin reuptake inhibitors, tricyclic antidepressants and opioids, DAT-SPECT was performed after a 2-week drug washout period to reduce a possibility that striatal 123 I-Ioflupane accumulation were interfered with these drugs. Of them, 311 patients with PD (158 men) were used for this study ( Table 1) To focus on the differential diagnosis of degenerative parkinsonian syndromes, we excluded patients with MSA-C from this study. There were no significant differences in the age and percentages of men among the PD, PSP and MSA-P groups. As a control group without parkinsonism, 137 patients were used (62 men, median 76.0 years; IQR 72.0-80.0 years). They included 48 patients with essential tremor (median of average SBR 5.08; IQR 4.08-5.90), 23 patients with orthopaedic diseases (average SBR 4.34; IQR 3.56-6.20), 11 patients with Alzheimer-type dementia (average SBR 5.03; IQR 4.03-5.38), 9 patients with mental diseases (average SBR 4.84; IQR 3.93-6.21), 11 patients with normal aging (average SBR 4.73; IQR 4.28-6.75), 10 patients with endocrine diseases and metabolic diseases (average SBR 5.06; IQR 3.39-6.48), and 25 patients with other reasons (average SBR 5.05; IQR 3.87-6.11). There were no significant differences in average SBRs among these control groups (P = 0.968, Kruskal-Wallis test). Additionally, there was no difference in the values of SBR laterality among the control groups (P = 0.651, Kruskal-Wallis test). These patients underwent DAT-SPECT to rule out degenerative parkinsonian syndromes. Inclusion criteria of this study were: (1) age > 30 years; (2) diagnosis according to the UK Parkinson's Disease Society Brain Bank diagnostic criteria for PD 22 , National Institute of Neurological Disorders and Stroke-Society for probable PSP 23 and Gilman's consensus criteria for probable MSA-P 24 ; (3) no evidence of another disease to explain the symptoms. Additionally, the history of unilateral onset was confirmed in most patients with PD. Exclusion criteria were: (1) obvious history of current treatment with psychoactive drugs, such as antiepileptic drugs, tricyclic antidepressant drugs, and serotonin reuptake inhibitors; (2) abnormalities on brain magnetic resonance imaging (MRI), such as diffuse vascular lesions, hydrocephalus, neoplasm, and metal deposition; (3) history of deep brain stimulation.
We retrospectively collected information on age, sex, family history, medication history, neurological symptoms, clinical course, responsiveness to levodopa/decarboxylase inhibitors, presence of cognitive dysfunction,

Scientific RepoRtS
| (2020) 10:15761 | https://doi.org/10.1038/s41598-020-72321-y www.nature.com/scientificreports/ brain MRI, and diagnosis and HY stage from medical records. The diagnosis was not confirmed by the pathological analysis. This study was conducted according to the 2013 Helsinki Declaration, and the Osaka Medical College Ethics Committee approved the study protocol and the need for informed consent was waived because this was a retrospective study and the data were collected without individual patient identifiers (Approval number # 2166).

Assessment of DAT-SPECT images. At 3-4 h interval after intravenous administration of 123 I-Ioflupane,
head SPECT was performed for 30 min using the GCA 9300R SPECT camera (Canon Medical Systems, Tokyo, Japan). For image analysis, DaTView (Aze Ltd., Tokyo) was used, and values of left and right SBRs were obtained for quantitative evaluation. Region of interest was set to fully cover the striatum. The standard volume of striatum was assumed as 11.2 ml. The area except the striatum was used as the non-specific background. After subtracting the background, the value of SBR was estimated. The evaluation parameters of DAT-SPECT in this study were: (1) average SBR; (2) SBR laterality; and (3) asymmetry index (AI). SBR laterality values were calculated by the following formula: |SBRleft -SBRright|. AI values were calculated by using the following formula: |SBRleft -SBRright| / [(SBRleft + SBRright)/2] × 100 7,19 .
Statistical analysis. Statistical analysis was performed using the SPSS software program (version 23.0 for MS windows. Chicago, USA). Multiple comparisons above three groups were performed by the Kruskal-Wallis test, and then the Bonferroni correction was performed for pairwise comparisons. The nominal data (sex difference) were analysed by chi-square test. Data were expressed as a median and IQR, and P value < 0.05 was considered statistically significant.