Idiopathic REM Sleep Behavior Disorder and Parkinson's Disease—Two Sides of One Coin?

THE SEARCH FOR NEUROPROTECTIVE OR DISEASE-MODIFYING AGENTS FOR TREATING PARKINSON DISEASE (PD) IS A MULTIFACETED UNDERTAKING,¹ ONE important aspect of which is the identification of the appropriate patient population. According to the neuroanatomist Heiko Braak,² PD pathology starts with the pre-motor and pre-clinical stages 1 and 2 that remain confined to the medulla oblongata and olfactory bulb. This is followed by the inclusion of the substantia nigra and other nuclear grays of the midbrain and basal forebrain in stages 3 and 4, before the pathology involves cortical structures during the final stage. Early phenotypes of pre-clinical neurodegenerative synucleinopathies have apparently been identified in patients with idiopathic REM sleep behavior disorder (iRBD)^3,4 developing parkinsonism 2 to 29 years after iRBD has been diagnosed. In addition to investigating a population with clinically defined motor stages of PD, researchers now look for these early non-motor, pre-clinical stages of PD, iRBD being the most promising.

Despite well-known sleep disturbances in PD,⁵ sleep has been a rare and mostly neglected topic in Parkinson research. This picture changed when iRBD was clinically associated with neurodegeneration. Since then a number of publications have confirmed the development of neurodegenerative disease in patients with iRBD over time,⁶ supported by olfactory pathology,⁷ reduced dopamine transporter binding,⁸ and cognitive decline.⁹ Few publications, however, followed those RBD patients who did not develop any signs or symptoms of neurodegeneration. Furthermore, some PD cases did not develop RBD despite neurodegeneration with substantia nigra pathology being confirmed.¹⁰ A similarly controversial discussion applies to the cholinergic nuclei of the tegmentum area, possibly involved in the regulation of muscle atonia in REM sleep: those nuclei are assumed to be affected in patients with multiple system atrophy,¹¹ thus not unanimously confirming the relationship of anatomical regions affected by the neurodegenerative process. Therefore, the putative role of RBD in neurodegeneration remains unresolved and a topic of lively discussion.

A further contribution to this dialogue is the publication in this issue of SLEEP of a study by Unger and colleagues.¹² They present magnetic resonance imaging (MRI) evidence for structural alterations in patients with iRBD that underpin the close relationship to PD. The authors compared diffusion tensor imaging (DTI) datasets of 12 patients with video-polysomnography confirmed iRBD against 12 age-matched healthy control subjects using an automated voxel-based method.

DTI is based on the fact that diffusion in the human brain is not free and isotropic but is directed by the fiber architecture. Usually, diffusion following the main fiber orientation is less restricted than the perpendicular diffusion.¹³ Anisotropy can be measured and expressed in various MR-physical parameters; fractional anisotropy (FA), axial diffusivity (AD), and radial diffusivity (RD) were used in the current study. DTI can be used to dissect large white matter tracts in vivo by fiber-tracking (e.g., optic radiation).¹⁴ Fiber directionality can, however, also be seen as a surrogate parameter for fiber integrity and thus an unspecific yet sensitive measure for structural alterations of the human brain. To allow for an unbiased assessment of multiple brain regions the authors used a relatively novel processing method based on an automatically generated pseudo-anatomical skeleton (i.e., tract-based spatial statistics = TBSS). This method is a variation of the common voxel-based morphometry method that carries the advantage of higher sensitivity and accuracy¹⁵ at the cost of restricting the analysis to large white matter branches.

The authors reported highly significant (all P < 0.0001, corrected) diffusion alterations of the fornix, olfactory region, internal capsule, brainstem, right visual stream, and left temporal lobe. In a sample of only 12 patients these findings, even after correction for multiple comparisons, are surprising. These require a very homogeneous, carefully selected group and strong alterations in each individual subject. Technically similar studies have reported much weaker changes (by a factor of 10 to 100) even in diseases with known MRI alterations like temporal lobe epilepsy with hippocampal sclerosis¹⁵ or Alzheimer disease.¹⁶ Although the resolution of DTI studies and the processing method restricted to large white matter tracts makes the exact anatomical interpretation challenging, the reported areas are largely compatible with our understanding of iRBD¹⁷ and early stage PD (pre-clinical stages 1 and 2).³ The observed alterations of the olfactory system are very suggestive for hyposmia, which has been linked to both PD and iRBD.⁷ Also the fornix has been suggested to play a role in PD patients with excessive daytime sleepiness.¹⁸

Unger and colleagues¹² are the first to report strong morphological changes in patients with iRBD. Thus far, structural MRI abnormalities have only been found in lesional cases (i.e., secondary RBD following ischemia or inflammation),¹⁹ whereas, by definition, the MRI is expected to be normal in iRBD. It is known from neuropathological studies that iRBD is associated with different neurodegenerative diseases particularly with Lewy body pathology.¹⁹ Until now, routine MRI seems unable to detect this—in PD, as in iRBD, no MRI abnormality is usually evident. Recently, however, DTI alterations have been described in de novo PD cases by region-of interest based methods.²⁰ Overall, DTI seems to be a promising tool in synucleinopathies; further studies will have to reproduce these findings in different and larger cohorts, and establish the value of such methods over the full clinical course in a longitudinal setting and in individual cases. It also remains to be investigated why many iRBD patients eventually develop PD and others do not; this may offer an alternative opportunity to look into neurodegeneration pathophysiology and eventually prevent disease progression.

DISCLOSURE STATEMENT

The authors have indicated no financial conflicts of interest.