You are here

Article Text

Article menu

Download PDFPDF

Neurodegeneration
Research paper
Chronological changes of 123I-MIBG myocardial scintigraphy and clinical features of Parkinson's disease
  1. Koyo Tsujikawa1,
  2. Yasuhiro Hasegawa1,
  3. Satoshi Yokoi2,
  4. Keizo Yasui1,
  5. Ichiro Nanbu3,
  6. Tsutomu Yanagi4,
  7. Akira Takahashi1
  1. 1Department of Neurology, Nagoya Daini Red Cross Hospital, Nagoya, Japan
  2. 2Department of Neurology, Nagoya University Graduate School of Medicine, Nagoya, Japan
  3. 3Department of Radiology, Nagoya Daini Red Cross Hospital, Nagoya, Japan
  4. 4Obu Dementia Care Research and Training Center, Aichi, Japan
  1. Correspondence to Dr Yasuhiro Hasegawa, Department of Neurology, Nagoya Daini Red Cross Hospital, 2-9, Myoken-cho, Showa-ku, Nagoya 446-8650 Japan; yhase{at}nagoya2.jrc.or.jp

Abstract

Objectives The aim of this study was to investigate chronological changes of 123I-meta-iodobenzylguanidine (MIBG) myocardial scintigraphy and its relation to clinical features in patients with Parkinson's disease (PD), and to characterise patients with PD with normal or mildly low MIBG uptakes at their early stages.

Methods The participants were 70 patients with PD who underwent 123I-MIBG myocardial scintigraphy twice or more. A cluster analysis was performed using parameters calculated from heart to mediastinum (H/M) ratio and washout ratio (WR).

Results At baseline, the mean early H/M ratio (H/M(E)), delayed H/M ratio (H/M(D)) and WR were 1.83, 1.69 and 41.7%, respectively. After a mean interval of 3.0 years, follow-up studies showed significantly declined H/M(E) (1.69, p<0.001), declined H/M(D) (1.47, p<0.001) and enhanced WR (43.8%, p=0.007). Our longitudinal observations revealed that there existed heterogeneous changes in MIBG uptakes among patients. The cluster analysis classified the patients into two subgroups: 42 patients with markedly low MIBG uptakes at baseline (group A) and 28 patients with normal or mildly low MIBG uptakes at baseline (group B). Group B showed a significantly higher ratio of females, younger age at onset, lower Hoehn and Yahr stage and less demented, compared with group A.

Conclusions Follow-up studies of MIBG divided the patients with PD into two major subgroups. A subgroup of patients with PD with normal or mildly low MIBG uptakes at the early stages of illness was characterised by female-dominant, young onset, slow progression in motor dysfunctions and preserved cognitive function.

Trial registration number 1033.

  • PARKINSON'S DISEASE
  • NEURORADIOLOGY

https://doi.org/10.1136/jnnp-2015-310327

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Introduction

    123I-meta-iodobenzylguanidine (MIBG) myocardial scintigraphy serves as a non-invasive diagnostic tool for assessing cardiac sympathetic nerve endings. Reduction of MIBG uptakes indicates cardiac sympathetic denervation, which has been observed in almost all patients with Parkinson's disease (PD).1–12 Numerous cross-sectional studies1–9 demonstrate that 123I-MIBG myocardial scintigraphy is helpful in differentiating PD from atypical parkinsonian syndromes (APS) such as progressive supranuclear palsy,3 ,4 ,9 corticobasal degeneration (CBD)4 ,9 and multiple system atrophy (MSA).4 ,6 ,9 A few studies4 ,11 ,12 revealed a generally progressive decline of MIBG uptake and estimated the yearly reduction rate for delayed heart to mediastinum (H/M) ratio to be approximately 0.02–0.03.4 ,11 A small percentage of patients with PD showed normal or mildly low MIBG uptakes in the early stages of the disease.4 ,5 ,10

    The aim of our study is to disclose the main clinical features and disease progression in a selected subgroup of patients with PD with normal or mildly low MIBG uptakes in the early stages. For this purpose, we retrospectively investigated the changes of MIBG uptakes, and evaluated the relationship between clinical course of patients with PD and their chronological changes of MIBG uptakes.

    Materials and methods

    Patients

    We retrospectively investigated patients with PD who underwent 123I-MIBG myocardial scintigraphy twice or more in Nagoya Daini Red Cross Hospital (Nagoya, Japan). The data were collected between September 2004 and September 2013. The diagnosis of PD was based on the clinical criteria of the UK PD Society Brain Bank.13 Patients with dementia developing within 1 year of illness were excluded. To eliminate the influence of factors that could interfere with MIBG uptake, patients with the following criteria were also excluded: (1) glycated haemoglobin value over 6.2% in blood test or diagnosis of diabetes mellitus; (2) abnormal electrocardiographic findings or history of heart diseases such as myocardial infarction, heart failure and cardiomyopathy; and (3) positive history of medication with tricyclic and tetracyclic antidepressants, calcium channel blockers and sympathomimetics.14 The patients underwent clinical and neurological assessment, routine blood studies and MRI of the head. The disease onset was defined as the time of first subjective motor symptoms. The initial motor symptoms were collected from the clinical records. The Hoehn and Yahr (H&Y) stage was derived from clinical descriptions referring to bilateral motor involvement, postural instability, use of walking aids or wheelchair and confinement to bed. The maximum levodopa daily dose and dopamine agonist usage were also collected from the clinical records. The presence or absence of each non-motor symptom was assessed by bedside examinations of the patients, reviewing the patients’ records and interviewing the patients or their families, or both. Dementia was defined as multiple cognitive impairment centred by memory disturbances, interfering with daily life activities (Diagnostic and Statistical Manual of Mental Disorders (DSM)-IV severity criterion for dementia15). Depression or anxiety, urinary dysfunction and constipation were considered to be present when the patients required medication. This study is approved by the ethics committee of Nagoya Daini Red Cross Hospital.

    123I-MIBG myocardial scintigraphy

    123I-MIBG myocardial scintigraphy was performed at rest. On examination, 111 MBq of 123I-MIBG (FUJIFILM RI Pharma Co, Ltd, Tokyo, Japan) was injected intravenously. Data were collected after 20 min (early image) and 240 min (delayed image) using a dual head γ camera (Millennium VG; GE Healthcare, Wisconsin, USA) with low-energy collimators. The photopeak of 123I was centred at 159 keV with a 20% energy window. Data were acquired for 4 min with a 256×256 matrix for image acquisition. Regions of interest (ROI) were drawn around the whole heart (H) and mediastinum (M). The H/M ratio was calculated from the average number of counts per pixel in the H and M. Tracer uptake was measured within each ROI to calculate ‘early H/M ratio (H/M(E))’, ‘delayed H/M ratio (H/M(D))’ and ‘washout ratio (WR)’. The normal control values of the H/M(E), H/M(D) and WR in this study, obtained in 17 participants who had no history of neurological disorders (8 men and 9 women, mean age 75.0±4.4 years), are 2.43±0.36, 2.47±0.33 and 31.6±6.6, respectively. In this study, all patients with PD underwent 123I-MIBG myocardial scintigraphy twice or more. The last evaluation was used as the follow-up data.

    Statistics

    R software V.2.15.1 was used for statistical analysis. Data were expressed as means±SD. Significance was assumed at a p value <0.05. The categorical variable was analysed by Fisher's exact probability test. Paired t test was performed to estimate change in the results of 123I-MIBG myocardial scintigraphy. Mann-Whitney U test was performed to estimate differences in age, duration of illness, H&Y stage and levodopa daily dose between the two groups. To quantify the chronological change of MIBG uptake in each patient, three kinds of regression coefficients were calculated from H/M(E), H/M(D) and WR, and defined as ‘ΔH/M(E)’, ‘ΔH/M(D)’ and ‘ΔWR’, respectively. To classify the patients with PD by the pattern of change in MIBG uptake, we performed a hierarchical cluster analysis with Ward linkage method. This cluster analysis was performed using six parameters: H/M(E) at baseline, H/M(D) at baseline, WR at baseline, ΔH/M(E), ΔH/M(D) and ΔWR. These parameters were used after Z-score normalisation for data analysis. Receiver operating characteristic (ROC) analysis was used to calculate the optimal cut-off values of H/M(E), H/M(D) and WR, for differentiating PD from the normal controls.

    Results

    Patients

    A total of 70 patients with PD (26 men and 44 women) was included in this study. The clinical features of the patients with PD at baseline are described in table 1. All the patients were Japanese. They were well responsive to levodopa and/or a variety of dopamine agonists. MRI of the head showed no abnormal imaging suggestive of APS including putaminal, pontine or cerebellar atrophy. All patients were ambulatory and none were bed-ridden. The mean H&Y stage, the mean levodopa daily dose, the rate of dopamine agonist usage and the rate of non-motor symptoms also suggested that most participants at baseline were in the early stages of PD.

    View this table:
    Table 1

    Clinical features of 70 patients with Parkinson's disease at baseline

    Chronological changes of 123I-MIBG myocardial scintigraphy

    Of 70 patients, the number of patients who underwent 123I-MIBG myocardial scintigraphy twice, three times, four times and five times was 51 (72.9%), 15 (21.4%), 3 (4.3%) and 1 (1.4%), respectively. We defined the last evaluation as follow-up data, which were collected in 3.0±1.8 years after the baseline evaluation (table 2). At baseline, the mean H/M(E), H/M(D) and WR were 1.83±0.40, 1.69±0.48 and 41.7±5.7%, respectively. For the diagnosis of PD at baseline, the optimal cut-off values of H/M(E), H/M(D) and WR were 1.90 (sensitivity of 100%, specificity of 64.3% and area under the curve (AUC) of 0.86), 1.97 (sensitivity of 100%, specificity of 71.4% and AUC of 0.89) and 37.7 (sensitivity of 81.4%, specificity of 85.2% and AUC of 0.88), respectively. Overall, the follow-up data showed a significant and progressive declining in H/M(E) and H/M(D) (1.69±0.37 and 1.47±0.36, both p<0.001). WR significantly increased from baseline to follow-up (43.5±5.1%, p=0.007). At follow-up, the optimal cut-off values of H/M(E), H/M(D) and WR were 1.94 (sensitivity of 100%, specificity of 80.0% and AUC of 0.92), 1.98 (sensitivity of 100%, specificity of 92.9% and AUC of 0.96) and 39.5 (sensitivity of 82.9%, specificity of 94.1% and AUC of 0.93), respectively. To observe the individual changes of 123I-MIBG uptakes in detail, we plotted H/M(E), H/M(D) and WR over disease duration for each patient. The figure 1A shows that chronological change patterns of MIBG uptakes were heterogeneous among patients with PD: changes of H/M(E), H/M(D) and WR were dramatic in some patients, but almost unchanged in the remaining. We calculated three kinds of regression coefficients, ΔH/M(E), ΔH/M(D) and ΔWR, for each patient to quantify his/her own change per year. The mean ΔH/M(E), ΔH/M(D) and ΔWR were −0.04±0.10, −0.07±0.10 and 0.7±2.5%, respectively. The histograms of ΔH/M(E), ΔH/M(D) and ΔWR showed a peak centred at or near zero and extending symmetrically above and below the peak centre (figure 1B).

    View this table:
    Table 2

    123I-MIBG myocardial scintigraphy

    Figure 1
    Figure 1

    Chronological changes of 123I-MIBG myocardial scintigraphy. (A) Chronological changes of H/M(E), H/M(D) and WR. (B) Histograms of ΔH/M(E), ΔH/M(D) and ΔWR. ‘ΔH/M(E)’, ‘ΔH/M(D)’ and ‘ΔWR’ represent the regression coefficients of H/M(E), H/M(D) and WR, respectively. 123I-MIBG, 123I-meta-iodobenzylguanidine; H/M, heart/mediastinum ratio; H/M(E), early H/E; H/M(D), delayed H/E; WR, washout ratio.

    Cluster analysis

    To classify the 70 patients with PD with respect to MIBG uptake, we performed a hierarchical cluster analysis using six parameters: H/M(E) at baseline, H/M(D) at baseline, WR at baseline, ΔH/M(E), ΔH/M(D) and ΔWR. The results are illustrated as the dendrogram in figure 2A. The vertical axis of the dendrogram is a measure of dissimilarity between chronological changes of MIBG uptakes. We divide the patient population into two subgroups: ‘group A’ and ‘group B’. Group A consists of 42 (60%) patients with PD with significantly low MIBG uptakes at baseline (table 2); group B consists of 28 (40%) patients with PD with normal or mildly low MIBG uptakes at baseline (table 2). The follow-up study revealed that MIBG uptakes remained low in group A and normal or mildly low in group B (figure 2B). A small number of patients in group B showed normal MIBG uptakes throughout the study (figure 2B).

    Figure 2
    Figure 2

    Hierarchical cluster analysis. (A) The dendrogram created by the hierarchical cluster analysis. (B) Chronological changes of H/M(E), H/M(D) and WR in cluster (A) and cluster (B). H/M, heart/mediastinum ratio; H/M(E), early H/E; H/M(D), delayed H/E; WR, washout ratio; PD, Parkinson's disease.

    Comparison of clinical characteristics between the two clusters

    To clarify the clinical characteristics and disease progression in group B, we compared the clinical parameters between the two groups (table 3). The ratio of women to men was higher in group B (78.6%) than in group A (52.4%; p=0.043). The mean ages at onset, at baseline and at follow-up were younger in group B than those in group A (table 3). At baseline, the mean duration of illness was shorter in group B (1.3±1.1 years) than in group A (2.3±2.8 years). There was no significant difference between the two groups (p=0.670). The initial motor symptoms in group A were not significantly different from those in group B (p=0.537). An analysis on the H&Y stage showed slower progression of motor disability in group B than in group A (figure 3A). At baseline, the mean levodopa daily dose was lower in group B than in group A. There was no difference between the clusters at the final stage of following up study. There were no significant differences in the usage rate of dopamine agonist at baseline and during the follow-up. As to the presence or absence of non-motor symptoms (figure 3B), the prevalence rates of each non-motor symptom increased from the baseline throughout the follow-up in both groups. Patients with dementia appeared to be less common in group B (3.6% at baseline, 14.3% at follow-up) than in group A (28.6% at baseline, 38.1% at follow-up) (p=0.020 for baseline, p=0.035 for follow-up). The prevalence rate of depression or anxiety, as well as urinary dysfunction, was smaller in group B than in group A, without significant difference. At baseline, the prevalence rate of constipation was significantly lower in group B (35.7%) than in group A (66.7%; p=0.021). At the final stage of the following up study, there was no significant difference in the prevalence rate of constipation between the two groups (p=0.196). Only four patients with PD in group B showed normal MIBG uptakes throughout the study; all were female and exhibited tremor onset.

    View this table:
    Table 3

    Comparison of characteristics between the two groups

    Figure 3
    Figure 3

    Comparison of clinical characteristics between two clusters. (A) The accumulation of Hoehn and Yahr stage. (B) The prevalence rate of dementia, depression or anxiety, urinary dysfunction and constipation (*p<0.05, Fisher's exact probability test). NS, not significant.

    Discussion

    The present study demonstrated that the patients with PD had different sequential changes in MIBG uptakes. As a whole, MIBG uptakes progressively declined during clinical course of PD, but the rates of change varied among patients. Our data showed that the patients with PD were divided into two subtypes with different change patterns of MIBG uptakes. One subgroup (group A) showed markedly low MIBG uptakes at the early stages, and was clinically characterised by male-dominant and older onset. Another subgroup (group B) showed normal or mildly low MIBG uptakes at the early stages and was clinically characterised by female-dominant and younger onset. The motor manifestation and dementia were less progressive in group B than in group A.

    Previous studies showed that age at onset,6 H&Y stage6 and cognitive impairment severity16 negatively correlated with MIBG uptakes. Little is known about the sex difference in MIBG uptakes. A recent report17 demonstrated that the Scales for Outcomes in Parkinson's disease-Autonomic (SCOPA-AUT) scale more significantly correlated with MIBG uptakes in men than in women. These data seemed to be compatible with our results.

    A hierarchical cluster analysis was performed to address the issue of the heterogeneity of MIBG uptakes in patients with PD. The classification according to the arbitrary cut-off value could lead to an inherent bias in the conclusions. Our cluster analysis can help to minimise this effect.

    ROC analyses demonstrated that the AUC values of H/M(E), H/M(D) and WR, respectively, increased from baseline to follow-up. These data suggest that a re-examination of 123I-MIBG myocardial scintigraphy in approximately 3 years might be beneficial in reinforcing the diagnosis of PD.

    Our results confirm that reduced MIBG uptakes progresses to varying degrees among patients with PD. The heterogeneity of chronological change of MIBG uptakes in patients with PD may be interpreted in terms of pathogenesis as follows. One possible explanation is that the anatomical distribution of Lewy bodies (LBs) could influence the variety of MIBG uptakes among patients with PD. A postmortem examination in PD demonstrated markedly decreased tyrosine hydroxylase immunoreactive nerve fibres in the epicardial space.7 It is assumed that cardiac sympathetic denervation is closely related to the presence of LBs, and that reduced MIBG uptakes is a potential biomarker for the presence of LBs.9 The difference of MIBG uptakes between group A and group B might reflect the presence or absence of LBs in the cardiac sympathetic nerve. A small number of patients with PD with normal MIBG uptakes throughout the study could have had less evolution of LBs. An alternative possibility is the genetic factor. A previous study18 showed normal MIBG uptakes in 8 of 14 patients with PD with genetic mutation in parkin, DJ-1, PINK-1 or LRRK2. Genetic analysis was not performed in this study. Further studies, including pathological investigation and genetic analysis, are required in order to clarify the mechanism of chronological change of MIBG uptakes in PD.

    There are some limitations to this study. First, in all patients, PD was ascertained clinically and with neuroimaging; we cannot deny that they had APS. On autopsy, some patients with a diagnosis of clinical PD may be pathologically proven to have PSP, CBD or MSA. Therefore, some patients, especially patients without low MIBG uptakes, might have been misdiagnosed. Second, all clinical data were collected retrospectively, which may cause underestimating of the prevalence of clinical parameters. For instance, the prevalence of dementia in this study was lower than reported in some prospectively neuropsychological investigations of PD.19 ,20 Third, levodopa could affect MIBG uptakes to some degree. A previous report21 showed that the mean H/M(D) in seven patients with PD with H/M(D) larger than 1.8, reduced from 2.50 to 2.28 after levodopa administration. However, this reduction rate by levodopa was smaller than chronological change of H/M(D) in our group B, and levodopa seemed to have little influence on MIBG uptakes.

    In summary, the present study demonstrated that the course of MIBG uptakes in PD is variable, ranging from a relatively stable type to a rapidly progressive one. The cluster analysis divided patients with PD into two subgroups with different change patterns of MIBG uptakes. A subgroup of patients with PD with normal or mildly low MIBG uptakes in the early stages of illness was characterised by female dominancy, younger onset and slow progression in motor dysfunctions, and by relatively well-preserved cognitive function.

    References

    1. ↵
      1. Hakusui S,
      2. Yasuda T,
      3. Yanagi T, et al
      . A radiological analysis of heart sympathetic functions with meta-[123I]iodobenzylguanidine in neurological patients with autonomic failure. J Auton Nerv Syst 1994;49:81–4. doi:10.1016/0165-1838(94)90023-X
      OpenUrlCrossRefPubMed
    2. ↵
      1. Hirayama M,
      2. Hakusui S,
      3. Koike Y, et al
      . A scintigraphical qualitative analysis of peripheral vascular sympathetic function with meta-[123I]iodobenzylguanidine in neurological patients with autonomic failure. J Auton Nerv Syst 1995;53:230–4. doi:10.1016/0165-1838(95)00002-F
      OpenUrlCrossRefPubMed
    3. ↵
      1. Yoshita M
      . Differentiation of idiopathic Parkinson's disease from striatonigral degeneration and progressive supranuclear palsy using iodine-123 meta-iodobenzylguanidine myocardial scintigraphy. J Neurol Sci 1998;155:60–7. doi:10.1016/S0022-510X(97)00278-5
      OpenUrlCrossRefPubMedWeb of Science
    4. ↵
      1. Satoh A,
      2. Serita T,
      3. Seto M, et al
      . Loss of 123I-MIBG uptake by the heart in Parkinson's disease: assessment of cardiac sympathetic denervation and diagnostic value. J Nucl Med 1999;40:371–5.
      OpenUrlAbstract/FREE Full Text
    5. ↵
      1. Orimo S,
      2. Ozawa E,
      3. Nakade S, et al
      . (123)I-metaiodobenzylguanidine myocardial scintigraphy in Parkinson's disease. J Neurol Neurosurg Psychiatry 1999;67:189–94. doi:10.1136/jnnp.67.2.189
      OpenUrlAbstract/FREE Full Text
    6. ↵
      1. Hamada K,
      2. Hirayama M,
      3. Watanabe H, et al
      . Onset age and severity of motor impairment are associated with reduction of myocardial 123I-MIBG uptake in Parkinson's disease. J Neurol Neurosurg Psychiatry 2003;74:423–6. doi:10.1136/jnnp.74.4.423
      OpenUrlAbstract/FREE Full Text
    7. ↵
      1. Orimo S,
      2. Ozawa E,
      3. Oka T, et al
      . Different histopathology accounting for a decrease in myocardial MIBG uptake in PD and MSA. Neurology 2001;57:1140–1. doi:10.1212/WNL.57.6.1140
      OpenUrlCrossRefPubMed
    8. ↵
      1. Orimo S,
      2. Ozawa E,
      3. Nakade S
      . [123I] meta-iodobenzylguanidine myocardial scintigraphy differentiates corticobasal degeneration from Parkinson's disease. Intern Med 2003;42:127–8. doi:10.2169/internalmedicine.42.127
      OpenUrlCrossRefPubMed
    9. ↵
      1. Orimo S,
      2. Amino T,
      3. Uchihara T, et al
      . Decreased cardiac uptake of MIBG is a potential biomarker for the presence of Lewy bodies. J Neurol 2007;254(Suppl 14): IV21–IV8. doi:10.1007/s00415-007-4005-y
      OpenUrlCrossRef
    10. ↵
      1. Sawada H,
      2. Oeda T,
      3. Yamamoto K, et al
      . Diagnostic accuracy of cardiac metaiodobenzylguanidine scintigraphy in Parkinson disease. Eur J Neurol 2009;16:174–82. doi:10.1111/j.1468-1331.2008.02372.x
      OpenUrlCrossRefPubMed
    11. ↵
      1. Watanabe M,
      2. Takeda T,
      3. Nakamagoe K, et al
      . Sequential imaging analysis using MIBG scintigraphy revealed progressive degeneration of cardiac sympathetic nerve in Parkinson's disease. Eur J Neurol 2011;18:1010–13. doi:10.1111/j.1468-1331.2010.03211.x
      OpenUrlCrossRefPubMed
    12. ↵
      1. Orimo S
      . The clinical significance of MIBG myocardial scintigraphy in Parkinson disease. Brain Nerve 2012;64:403–12 (in Japanese).
      OpenUrlPubMed
    13. ↵
      1. Gibb WR,
      2. Lees AJ
      . The relevance of the Lewy body to the pathogenesis of the idiopathic Parkinson's disease. J Neurol Neurosurg Psychiatry 1988;51:745–52. doi:10.1136/jnnp.51.6.745
      OpenUrlAbstract/FREE Full Text
    14. ↵
      1. Stefanelli A,
      2. Treglia G,
      3. Bruno I, et al
      . Pharmacological interference with 123I-metaiodobenzylguanidine: a limitation to developing cardiac innervation imaging in clinical practice? Eur Rev Med Pharmacol Sci 2013;17:1326–33.
      OpenUrlPubMed
    15. ↵
      American Psychiatric Association. Diagnostic and Statistical Manual. 4th edn. Text Revision (DSM-IV-TR), 2000.
    16. ↵
      1. Kitayama M,
      2. Wada-Isoe K,
      3. Irizawa Y, et al
      . Association of visual hallucinations with reduction of MIBG cardiac uptake in Parkinson's disease. J Neurol Sci 2008;264:22–6. doi:10.1016/j.jns.2007.07.017
      OpenUrlCrossRefPubMed
    17. ↵
      1. Guidez D,
      2. Behnke S,
      3. Halmer R, et al
      . Is reduced myocardial sympathetic innervation associated with clinical symptoms of autonomic impairment in idiopathic Parkinson's disease? J Neurol 2014;261:45–51. doi:10.1007/s00415-013-7135-4
      OpenUrlCrossRefPubMed
    18. ↵
      1. Quattrone A,
      2. Bagnato A,
      3. Annesi G, et al
      . Myocardial 123metaiodobenzylguanidine uptake in genetic Parkinson's disease. Mov Disord 2008;23:21–7. doi:10.1002/mds.21701
      OpenUrlCrossRefPubMed
    19. ↵
      1. Buter TC,
      2. van den Hout A,
      3. Matthews FE, et al
      . Dementia and survival in Parkinson's disease. Neurology 2008;70:1017–22. doi:10.1212/01.wnl.0000306632.43729.24
      OpenUrlCrossRefPubMed
    20. ↵
      1. Hely MA,
      2. Reid WG,
      3. Adena MA, et al
      . The Sydney multicentre study of Parkinson's disease: the inevitability of dementia at 20 years. Mov Disord 2008;30:837–44. doi:10.1002/mds.21956
      OpenUrl
    21. ↵
      1. Kishi M,
      2. Sakakibara R,
      3. Terada H, et al
      . Does levodopa affect metaiodobenzylguanidine myocardial accumulation in Parkinson's disease? Mov Disord 2011;26:563–4. doi:10.1002/mds.23464
      OpenUrlPubMed

    Footnotes

    • Contributors KT, YH, SY and KY conceived and designed the original protocol. All authors were involved in amending the protocol. KT coordinated the study throughout. Data entry was carried out by KT. KT cleaned the data and ran analysis with input from IN. KT wrote the first draft of the manuscript with YH and KY. KT, YH, KY, TY and AT contributed to subsequent and final drafts. YH is guarantor of the paper.

    • Competing interests None declared.

    • Ethics approval The ethics committee of Nagoya Daini Red Cross Hospital.

    • Provenance and peer review Not commissioned; externally peer reviewed.

    Linked Articles