Reduced cardiac 123I-MIBG uptake is a robust biomarker of Lewy body disease in isolated rapid eye movement sleep behaviour disorder

Abstract Cardiac 123I-MIBG scintigraphy is used to assess the function of postganglionic presynaptic cardiac sympathetic nerve endings. 123I-MIBG cardiac uptake is markedly reduced in patients with isolated rapid eye movement sleep behaviour disorder, similar to Parkinson’s disease and dementia with Lewy bodies. As a result, it can be used as an early biomarker of isolated rapid eye movement sleep behaviour disorder. Most patients with isolated rapid eye movement sleep behaviour disorder develop synucleinopathies: Parkinson’s disease, dementia with Lewy bodies or multiple system atrophy. We aimed to investigate whether cardiac postganglionic denervation is present in patients with isolated rapid eye movement sleep behaviour disorder, as well as its possible usefulness as a marker for Lewy body disease status. This retrospective cohort study examined 306 patients (236 men and 70 women; mean age: 68.2 years; age range: 43–87 years) with polysomnography-confirmed isolated rapid eye movement sleep behaviour disorder who were followed for 1–3 months and underwent 123I-MIBG scintigraphy. We retrospectively analysed data from 306 patients with polysomnography-confirmed isolated rapid eye movement sleep behaviour disorder, and their longitudinal outcomes were documented at two centres. Among isolated rapid eye movement sleep behaviour disorder patients, reduced 123I-MIBG uptake was observed in the early and delayed images in 84.4 and 93.4% of patients, respectively, whereas 88.6% of the patients had a high washout rate. This large Japanese two-cohort study (n = 306) found that 91 patients (29.7%) developed an overt synucleinopathy (51 Parkinson’s disease, 35 dementia with Lewy bodies, 4 multiple system atrophy, and 1 cerebellar ataxia) during a mean follow-up duration of 4.72 ± 3.94 years, with a conversion risk of 14.5% at 3 years, 25.4% at 5 years, 41.4% at 8 years and 52.5% at 10 years. On the other hand, among patients with heart-to-mediastinum ratio < 2.2 in the delayed images (n = 286), 85 (29.7%) developed Parkinson’s disease or dementia with Lewy bodies during a mean follow-up duration of 4.71 ± 3.94 years, with a conversion risk of 14.5% at 3 years, 25.6% at 5 years, 42.0% at 8 years and 51.0% at 10 years. Among the 33 patients who underwent repeat 123I-MIBG scintigraphy, there was a progressive decline in uptake over the next 4.2 years, with patients exhibiting reduced uptake progressing to Parkinson’s disease or dementia with Lewy bodies. In contrast, patients without decreased 123I-MIBG uptake progressed to multiple system atrophy. Reduced cardiac 123I-MIBG uptake was detected in over 90% of isolated rapid eye movement sleep behaviour disorder patients, with progression to Parkinson’s disease or dementia with Lewy bodies, rather than multiple system atrophy, over time. Reduced 123I-MIBG uptake is a robust maker for Lewy body disease among isolated rapid eye movement sleep behaviour disorder patients.


Introduction
Isolated rapid eye movement sleep behaviour disorder (IRBD) is a sleep-associated disorder characterized by abnormal behaviour along with dreams during rapid eye movement (REM) sleep.It was first reported by Schenck et al. 1 in 1986.In previous studies, 81-90% of IRBD patients developed a neurodegenerative disorder. 2Recent studies have suggested that IRBD is a precursor condition to distinct neurodegenerative changes.Longitudinal reports suggest that the rate of phenoconversion from IRBD to Parkinsonism or dementia increases over time. 3 123 -Metaiodobenzylguanidine ( 123 I-MIBG), a physiologic analogue of noradrenaline, is used to determine the location, integrity and function of postganglionic noradrenergic neurons.Cardiac 123 I-MIBG scintigraphy can assess postganglionic cardiac autonomic denervation and has become widely used in Japan since its introduction to clinical cardiology and neurology practice in the 1990s.4 Neurological studies have found reduced cardiac 123 I-MIBG uptake in Parkinson's disease, dementia with Lewy bodies (DLB), pure autonomic failure, and Parkinson's disease-related movement disorders.5 A multicentre study of the diagnostic performance of 123 I-MIBG for differentiation between Alzheimer's disease and DLB found a high correlation between abnormal cardiac sympathetic activity on 123 I-MIBG cardiac scintigraphy and a clinical diagnosis of probable DLB.The diagnostic accuracy is sufficiently high for this technique to be clinically useful for distinguishing between DLB and Alzheimer's disease, particularly in patients with mild dementia.6 A follow-up study revealed a higher correlation between abnormal cardiac sympathetic nerve activity assessed by cardiac 123 I-MIBG scintigraphy and a clinical diagnosis of probable DLB compared to earlier cross-sectional studies.The results also suggested the usefulness of cardiac 123 I-MIBG scintigraphy in the early clinical stages of DLB. 7 On the other hand, severely reduced cardiac 123 I-MIBG uptake in Parkinson's disease patients is associated with a risk of developing subsequent dementia, suggesting a link between cardiac noradrenergic denervation and cognitive function, similar to DLB. 8 In an autopsy-based case report of a patient with IRBD for 20 years, confirmed by polysomnography (PSG), incidental Lewy body disease was diagnosed, with Lewy bodies found in the locus coeruleus and substantia nigra.9 Furthermore, Boeve et al. 10 investigated the neuropathological diagnosis of 15 patients with a history of IRBD (10 PSG confirmed) and found that 12 had Lewy body disease (11 neocortical and 1 peripheral) and 3 had multiple system atrophy (MSA).In 10 of 15 (66.7%) patients with IRBD, dementia or Parkinson's disease manifested after a median of 10 (range 2-29) years.These findings suggest that the underlying cause of IRBD is abnormal α-synuclein.In IRBD, the affected areas include not only the brainstem regions regulating REM sleep but also regions such as the olfactory system, nigrostriatal system and autonomic system. Cariac 123 I-MIBG scintigraphy studies in IRBD patients with prodromal Parkinson's disease or DLB status have shown significantly reduced uptake in IRBD patients compared to controls, suggesting postsynucleinopathy, but this finding is more common in Parkinson's disease patients than in MSA patients 11 and is considered a prodromal phenotype.12 In a meta-analysis of cross-sectional studies of IRBD, cardiac 123 I-MIBG uptake was reduced in Parkinson's disease and DLB to the same extent as in Lewy body disease.13 Although these biomarkers are diagnostic rather than prognostic, there are no reports of large-scale longitudinal data showing an association with phenotypic conversion rates.14 In this study, we retrospectively examined the cardiac 123 I-MIBG uptake and outcomes of consecutive IRBD patients from two centres, as well as the changes in cardiac 123 I-MIBG uptake over time in patients who underwent two or more cardiac 123 I-MIBG tests.

Materials and methods
We retrospectively reviewed our database for consecutive patients with IRBD.This study included a total of 306 patients who were diagnosed with IRBD by PSG and underwent cardiac 123 I-MIBG scintigraphy between August 2004 and July 2022 at the Sleep Medicine Centre of Dokkyo Medical University Hospital (sleep centre) and between August 2011 and July 2023 at the Saitama Medical Centre of Dokkyo Medical University (neurological centre; Fig. 1).IRBD was diagnosed on the basis of a history of dream-enacting behaviours and video PSG demonstration of increased electro-myographic activity during REM sleep that was associated with abnormal behaviours.These criteria are based on the second edition of the International Classification of Sleep Disorders.In the sleep centre, 123 I-MIBG testing was performed on consecutive IRBD cases, whereas in the neurological centre, 123 I-MIBG testing was performed on randomly selected IRBD patients.The cut-off value for early or delayed reduction in cardiac 123 I-MIBG uptake was <2.2, 15 and the degree of 123 I-MIBG uptake and longitudinal prognostic outcomes were retrospectively analysed for the 306 patients.Repeated 123 I-MIBG testing (n = 33) was performed only at the sleep centre, demonstrating normal uptake (≥2.2) on both early and delayed images or only on early images in 19 (57.6%) patients and reduced uptake (<2.2) on both early and delayed images at baseline in 14 (42.4%)patients.The association between changes in 123 I-MIBG uptake and prognosis was also analysed in 33 IRBD patients.
After the scans were performed, all IRBD patients were systematically followed with neurological examinations and detailed clinical histories, including for cognitive and motor problems, every 1-3 months by a neurologist (T.M. and M.M.) with expertise in both sleep disorders and neurodegenerative diseases, applying diagnostic criteria for Parkinson's disease, 16 DLB 17 and MSA. 15,18The findings were entered into the medical records of patients.The medical records were reviewed retrospectively between March and July 2023 (end of the current study) to assess the presence and nature of neurodegenerative diseases, such as Parkinsonism or dementia, identified during follow-up with the same neurologist (T.M. and M.M.).The motor component of the Unified Parkinson's Disease Rating Scale-III was also administered, and the Folstein Mini-Mental State Examination was used to evaluate cognitive function, similar to previous studies. 19

I-MIBG cardiac scintigraphy
Early (15-30 min) and delayed (3-4 h) images were obtained after injection of 111 MBq 123 I-MIBG (Daiichi Radioisotope Laboratories Co.; Fujifilm Toyama Chemical Industries; PDRadiopharma Inc., Tokyo, Japan).The scintigraphy findings were immediately interpreted and recorded.Photopeak energy was centred at 159 keV ( 123 I-MIBG) with a window of 10%.The relative organ uptake of 123 I-MIBG was determined by setting the region of interest on the anterior view (Supplementary Fig. 1).The heart-to-mediastinum (H/M) ratio was calculated using the standard method by dividing the mean number of counts per pixel in the circular region of interest of the heart by the mean number of counts per pixel in the rectangular region of interest of the superior mediastinum.Because of the differences in test timings and facilities where 123 I-MIBG cardiac scintigraphy was performed, differences in collimator (GCA-7200, Toshiba; GCA-9300A-HG, Toshiba; SYMBIA E, Siemens Healthcare; E-CAM, Siemens Healthcare) were used for calibration phantoms or standardized using conversion factors established in previous studies. 20The H/M ratios were converted to values equivalent to those of medium energy collimators.The washout rate was calculated from early and delayed heart counts (H early and H delayed , respectively) and mediastinal counts (M early and M delayed , respectively), using the following formula for background (mediastinal counts) and decay corrections: . The coefficient k is a time decay correction factor of 1/0.5 t/13 for time t (h).If the interval between the scans is 3 h, k is 1.17.The washout rates were not standardized.The standard cut-off value was used for the washout rate (34%).None of our patients had taken drugs known to affect the 123 I-MIBG uptake (e.g.tricyclic antidepressants, selective serotonin reuptake inhibitor and selegiline), and none had a history of cardiac disease or untreated diabetes mellitus.The normal H/M ratio for myocardial 123 I-MIBG scintigraphy is ≥ 2.2 for the early and delayed images, whereas the normal washout rate is ≤34%. 20he study protocol was approved by the Ethics Committee of Dokkyo Medical University (approval number R-2-22) and the Ethics Committee of Saitama Medical Centre of Dokkyo Medical University (approval number 1821).The study was conducted in accordance with the ethical standards set out in the 1964 Declaration of Helsinki and subsequent amendments.Written consent was obtained from the participants, but because of the retrospective design, some patients were unable to give written consent and had the opportunity to opt out of the study at any time and seek further explanation.

Statistical analysis
Descriptive demographic, clinical and neuroimaging data are presented as means with standard deviations (SDs) or numbers with percentages.Comparisons between groups were performed using the Kruskal-Wallis test.Neurological disease-free survival rates were estimated using the Kaplan-Meier analysis.Disease-free survival rates were assessed from the date of 123 I-MIBG scintigraphy to the date of Parkinson's disease, DLB or MSA diagnosis, or to the last follow-up for censored observations.Repeated measures paired t-test was used to evaluate the differences in 123 I-MIBG uptake between baseline and repeat evaluations at 4.2 ± 2.8 years.A P < 0.05 was considered statistically significant.
On the other hand, among patients with H/M < 2.2 in the delayed images (n = 286), 85 (29.7%) developed Lewy body disease (Parkinson's disease or DLB) during a mean follow-up of 4.71 ± 3.94 years, with a conversion risk of 14.5% at 3 years, 25.6% at 5 years, 42.0% at 8 years and 51.0% at 10 years (Fig. 2B).
Of the 215 patients who were followed-up until the end of the study, 51 (23.7%) developed Parkinson's disease, 44 (86.3%) had an abnormal early 123 I-MIBG image at baseline, and 50 (98.0%)had an abnormal delayed 123 I-MIBG image.Of the 35 patients (16.3%) who developed DLB, 32 (91.4%) had an abnormal early 123 I-MIBG image and 34 (97.1%) had an abnormal delayed 123 I-MIBG image at baseline.On the other hand, of the four patients who developed MSA, all (100%) had normal uptake on both early and delayed 123 I-MIBG images (Fig. 3A and B).
In total, 33 patients underwent 123 I-MIBG at least twice during the follow-up, of whom 19 had normal uptake only on early images at baseline, 11 had normal uptake on both early and delayed images, and 14 had decreased uptake on both early and delayed images.During the follow-up (mean: 4.2 ± 2.8 years), the 123 I-MIBG uptake was decreased in 16 of the 19 patients with either early or delayed normal uptake, while 3 continued to have a normal uptake.There was no significant difference in the change over time among the 14 patients with reduced uptake at baseline.Of the 30 patients with reduced uptake at baseline or during follow-up, 11 developed Parkinson's disease, 7 progressed to DLB, and 12 remained disease free.Three cases with normal 123 I-MIBG uptake during follow-up developed MSA (Fig. 4).
In IRBD patients, the H/M ratio for the cardiac 123 I-MIBG uptake on both early and delayed images is weakly negatively correlated with age (Spearman rank test: r = −0.255,P < 0.001; r = −0.287, P < 0.001, respectively), but only the H/M ratio on early images is weakly negatively correlated with disease duration (Spearman rank test: r = −0.39,P = 0.015).On the other hand, the cardiac 123 I-MIBG uptake is weakly positively correlated with Mini-Mental Examination in terms of the H/M ratio on both early and delayed images (Spearman rank test: r = 0.141, P = 0.015; r = 0.195, P < 0.001, respectively), but is no correlated with Unified Parkinson Disease Rating Scale-III when considering the H/M ratio on both early and delayed images (Spearman rank test: r = −0.071,P = 0.220; r = −0.023,P = 0.698, respectively).

Discussion
This study demonstrates that the cardiac 123 I-MIBG uptake is reduced in most IRBD patients and deteriorates markedly over a 3-year period.To our knowledge, this is the first study to assess the cardiac 123 I-MIBG progression in IRBD using imaging markers and to observe outcomes longitudinally.Our findings suggest the existence of a pattern of cardiac 123 I-MIBG progression in IRBD.First, there is measurable degeneration of cardiac noradrenergic fibres from the time of IRBD diagnosis to disease progression among patients who develop Lewy body disease, while the uptake is preserved in patients who progress to MSA with no further decline in the cardiac 123 I-MIBG uptake.
Among patients who progressed from IRBD to Lewy body disease (Parkinson's disease or DLB), those who progressed to DLB were older at the time of diagnosis than those who progressed to Parkinson's disease (Fig. 3C).Even if the incidence of both Parkinson's disease and DLB increases with age, DLB patients are usually older than Parkinson's disease patients at the time of diagnosis. 21Late-onset IRBD was associated with a higher level of neurodegenerative markers and more rapid phenoconversion, especially to probable DLB.Age at the onset of IRBD can help identify clinical features and predict the prognosis of IRBD. 22n a recent multicentre study from 28 centre across 12 countries, including South Korea and China, the 3-year phenoconversion rate was 18.2% 14 compared to 14.5% in the present study.Recent studies, including our previous study from Asia, have shown lower phenoconversion rates 19 of 7.2% at 3 years from South Korea, 23 4.7% at 3 years from China 24 and 19.6% at 5 years from Turkey. 25 These results suggest that methodological differences between studies, such as sample size, follow-up duration and definition of phenoconversion (e.g.whether or not to include mild cognitive impairment), may contribute to the different results.As a result, future studies should also consider racial and  geographical differences and referral bias. 19This study is the first to determine the conversion rate in a homogeneous cohort with IRBD and reduced 123 I-MIBG uptake over time.
Reduced 123 I-MIBG uptake is commonly observed in IRBD patients with moderate subjective disease, and its frequency does not significantly increase with prolonged disease duration.This finding suggests that reduced 123 I-MIBG uptake is the earliest sign of Lewy body disease and can indicate the presence of incidental Lewy bodies. 26,27Among these prodromal symptoms, IRBD has a strong association with the risk of Lewy body disease, suggesting that this is a useful diagnostic tool for prodromal-stage Lewy body disease. 28rom a pathophysiological point of view, the reduced cardiac 123 I-MIBG uptake suggests that the abnormality is not confined to the brain, and that the sympathetic function is more extensively impaired even in the early stages of Lewy body disease. 29,30Indeed, Lewy bodies are widely distributed in the hypothalamus, sympathetic nervous system, parasympathetic nervous system, cardiac plexus, gastrointestinal tract enteric nervous system, pelvic plexus and adrenal medulla. 2 From a genetic perspective, the finding that symptomatic carriers of PARK1 mutations show a similar degree of delayed H/M ratio as DLB and IRBD patients, diseases characterized by the presence of Lewy body, supports the validity of cardiac 123 I-MIBG scintigraphy as a biomarker of Lewy body spectrum disorders.These findings are based on analysis of various idiopathic and genetic variants of Parkinson's disease associated with varying degrees of Lewy body pathology. 313][34] Objective markers, bowel dysfunction and cardiac sympathetic denervation worsen over 3 years in the majority of IRBD patients, and the lack of correlation between markers and nigrostriatal dopaminergic dysfunction suggests that progressive gastrointestinal and cardiac dysfunction in IRBD are mainly caused by non-dopaminergic mechanisms. 35rodromal MSA in IRBD progresses rapidly, often without substantial autonomic dysfunction, and olfactory and cognitive functions are maintained throughout the prodromal phase. 36A novel finding of our study is the normal cardiac 123 I-MIBG uptake values in the four patients with prodromal MSA at baseline and during follow-up.Our results suggest that one pathway of prodromal MSA, which emerges from prodromal IRBD, is different from that of Lewy body disease. 37These findings have implications for the differentiation of early stages of MSA and for the development of diagnostic criteria for prodromal MSA.In such cases, in combination with the 123 I-MIBG test, markers that may be used include an olfactory identification test, 38 differential symptoms that can be used in the clinical diagnosis of MSA, 39 stridor or sleep-disordered breathing 40 and MRI findings 41 (Supplementary Table 1).
The main limitations of the present study were its retrospective design and a lack of pathological data, as Alzheimer's disease and Lewy body pathologies frequently coexist in DLB and Alzheimer's disease. 2,42The abnormal findings of 123 I-MIBG cardiac scintigraphy strongly support the presence of Lewy body disease and cardiac sympathetic denervation, which is in line with a previous clinicopathological validation study. 30Our repeated observations of cases suggest that cases in which 123 I-MIBG uptake decreases over time progress to Parkinson's disease/DLB, while cases in which uptake does not decrease over time progress to MSA.In IRBD patients with normal uptake at baseline, it is necessary to perform 123 I-MIBG repeatedly and observe changes in the degree of 123 I-MIBG uptake.However, normal cardiac 123 I-MIBG uptake cannot exclude the pathogenesis of Lewy body disease, which can be more accurately diagnosed by observing changes over time or by evaluating other biomarkers. 43Furthermore, the timing of IRBD onset may be subject to recall bias.Based on our results, in which 123 I-MIBG uptake was reduced in >90% of IRBD patients at the time of PSG confirmation, the timing of the initial decline in 123 I-MIBG uptake in prodromal RBD is unclear. 44The future question is whether RBD-negative cases with reduced 123 I-MIBG uptake will develop RBD or progress to neurodegenerative diseases without developing RBD.The latter scenario may affect the sensitivity of the detection of progression of IRBD with reduced 123 I-MIBG uptake to Lewy body disease.Even at the end of the observation period in this study, cases with decreased 123 I-MIBG uptake that are still disease free are at risk of progression to Parkinson's disease/DLB if followed for a longer period of time.Therefore, we would expect the results to be more robust with a longer follow-up.
The main strengths of the present study are the large sample size and use of standardized post-processing of cardiac 123 I-MIBG scintigraphy based on a widely available, registered tool in Japan.The diagnostic accuracy of IRBD was enhanced by the use of PSG.The medical records provided longitudinal neurodiagnostic follow-up data for every 1-3-month interval.A number of studies have suggested that the presence of non-motor problems, including autonomic abnormalities, such as IRBD and orthostatic hypotension, are associated with diffuse and malignant subtypes of Parkinson's disease with high rates of progression and a high incidence of dementia.IRBD and autonomic dysfunction are strongly correlated.A major hurdle for developing and testing such curative therapies results from the fact that most dopaminergic neurons are already lost at the time of the clinical diagnosis, rendering them inaccessible to therapy.Understanding the early pathological changes that precede Lewy body pathology and cell loss in Parkinson's disease or DLB will likely support the development of novel diagnostic and therapeutic strategies and help to differentiate Lewy body disease, non-Lewy body disease and MSA alterations.Cardiac 123 I-MIBG is highly efficient at distinguishing at-risk individuals who are diagnosed subsequently with Lewy body disease from those who develop non-Lewy body disease or MSA.The current findings from IRBD patients suggest that the temporal changes in I-MIBG uptake does not decline during follow-up may progress to MSA.Therefore, the 123 I-MIBG uptake, in combination with other promising biomarkers, can predict the likelihood of conversion to MSA.The present longitudinal study indicates that cardiac 123 I-MIBG uptake is a robust biomarker for predicting the transition from IRBD to Lewy body disease.These findings can be used to stratify patients for enrolment in future neuroprotective clinical trials.

Figure 2
Figure 2 Phenoconversion for neurodegenerative disorders in IRBD.Rates of neurological disease-free survival using the Kaplan-Meier method, according to the time of IRBD diagnosis among the entire cohort of 306 patients (A) and 286 patients with IRBD and H/M ratio < 2.2 (B).IRBD, isolated rapid eye movement sleep behaviour disorder.

Figure 3
Figure 3Phenoconversion age and early and delayed 123 I-MIBG uptake at baseline in patients with IRBD.(A) Among IRBD patients who exhibited phenoconversion, the 123 I-MIBG early uptake at baseline was relatively preserved in the MSA group compared to the Parkinson's disease, DLB and disease-free groups (early image; Kruskal-Wallis test, disease free versus MSA, P = 0.0032; Parkinson's disease versus MSA, P = 0.0177; DLB versus MSA, P = 0.0051, respectively).(B) 123 I-MIBG delayed uptake at baseline among IRBD patients who exhibited photoconversion (disease free versus MSA, P = 0.0034; Parkinson's disease versus MSA, P = 0.0307; DLB versus MSA, P = 0.0021).(C) Following phenoconversion from IRBD, patients with DLB at conversion were older than those with Parkinson's disease (P = 0.0201).

Figure
Figure Longitudinal assessment of 123 I-MIBG scintigraphy in IRBD.Repeated measurements of early and delayed 123 I-MIBG scintigraphy in IRBD.Among the three patients who progressed to MSA, the accumulation was maintained after repeated testing from baseline.IRBD, isolated rapid eye movement sleep behaviour disorder; MSA, multiple system atrophy.