Discussion in that article:



In this study, we sought to determine how often and where αSN pathology occurs in neurologically unimpaired elderly individuals. We were especially interested in comparing the incidence of αSN pathology in the spinal cord or PANS with that of medullary nuclei where the process of αSN aggregation is supposed to begin. αSN pathology in the shape of LN and LB was seen in 17% of individuals in our cohort, which is a higher percentage than previously reported in the substantia nigra using conventional staining methods for the detection of LB. Our main result was the high incidence of αSN pathology in the spinal cord and PANS suggesting a very early involvement of these structures next to the dorsal motor nucleus of the vagus nerve, the locus coeruleus and the olfactory nerve.

Until now the spinal cord and the PANS have not been investigated for the presence of LB or αSN immunoreactive lesions in aged neurologically unimpaired individuals. To our knowledge, the only available neuropathological studies on these structures concern subjects with clinically and morphologically well-documented cases of PD.

They all underscore the constant involvement of structures of the central and peripheral nervous system involved in autonomic regulation. In PD, LB have been amply documented in the hypothalamus [16], the intermediolateral nucleus of the thoracic spinal cord [6,11,12,17], the stellate ganglion [5], the sympathetic ganglia [17,18], the sacral parasympathetic neurones [8,9,20,21], as well as the visceral autonomic nervous system [7,22,23], and they have been considered responsible for some of the autonomic dysfunctions in PD.

Profound cardiac sympathetic denervation has been shown in PD patients by sympathetic neuroimaging [24], and post mortem immunostaining of cardiac sympathetic nerves for tyrosine hydroxylase and neurofilament antigens [25,26]. Hishikawa et al. [27] used αSN-specific antibodies to study the distribution of LB and coil-like glial inclusions in patients with Lewy body disease including PD cases. αSN positive LB were found in various grey area of the spinal cord including the posterior, lateral and anterior horns. The authors, however, did not include asymptomatic elderly control cases in their study [27].

Our finding of a near universal occurrence of αSN positive lesions in spinal cord autonomic nuclei and PANS in clinically asymptomatic ILBD cases not only underscores the findings of the studies just mentioned on subjects with PD but also points to a pathological involvement of these structures at an earlier stage probably far below disease threshold for motor symptoms of PD. It remains uncertain if our subjects with an αSN pathology in the central and peripheral autonomic nervous system have presented subjective complaints or even clinical evidence of autonomic disturbances during the latest period of their life. Although this possibility cannot be ruled out, this issue has to remain unsettled due to the mode of selection of our cases and the difficulties of retrospective anamnestic enquiry.

It was long assumed that the manifestation of autonomic failure in PD is confined to later stages of the disease. However, more recent clinical evidence supports the opposite view that manifestations of autonomic failure, including gastrointestinal, urogenital, cardiovascular, sudomotor and thermoregulatory symptoms may compromise patient's daily life activities in all stages of the disease [28,29], and are poorly correlated with parameters felt to reflect disease severity [30].

Autonomic failures have even been reported as the initial disease manifestation in some PD cases suggesting very early involvement of central and peripheral autonomic structures in PD [31]. For instance, gastrointestinal dysfunctions, especially colonic dysmotility, which are common in PD, may constitute the initial clinical feature [32].

Recent epidemiological evidence revealed that colonic dysmotility may even occur long before typical PD symptomatology. Thus, an association between the frequency of bowel movement and the risk of developing PD has been demonstrated [33].

Stomach and intestinal functions are controlled largely by the autonomic and enteric nervous system. The bulk of parasympathetic innervation of the gastrointestinal tract is derived from the dorsal motor nucleus of the vagus nerve which supplies the stomach, small intestine and proximal colon.

Parasympathetic supply to the middle and distal colon is via sacral nerves and ultimately comes from the parasympathetic autonomic nuclei of the sacral cord. Early pathological changes in the dorsal motor nucleus of the vagus nerve have been well documented [2], and we now demonstrate for the first time that sacral parasympathetic autonomic supply is involved in a very early prodromal phase of the disease.

The observed changes in both autonomic centres are likely to be responsible for parasympathetic denervation of the gut and are therefore likely to account for early gastrointestinal disturbances in PD. However, we also found early αSN abnormalities within the gut itself which may also contribute to early autonomic dysfunction in PD.

A systematic study on a cohort of cases with ILBD ranging from 54 to 86 years at death was recently performed using αSN immunohistochemistry [2].

The main result of this study was that the very earliest αSN pathology of PD displays a marked preference for specific classes of brainstem neurones including the noncatecholaminergic projection cells of the dorsal motor nucleus of the glossopharyngeal and vagal nerves, the intermediate reticular zone, the pigmented cells of the locus coeruleus and for the olfactory bulb and tract.

Based on observations of post mortem tissues from diagnosed sporadic PD cases as well as from patients without clinical signs of PD, a successive propagation of αSN pathology from the medulla oblongata to telencephalic structures in end-stage PD has been proposed [3]. In the present study we found a distribution pattern of αSN positive lesions which is well compatible with an initiation of αSN pathology in the lower brainstem and a subsequent spreading pattern to more upward located regions of the central nervous system.

However, stageing was difficult in some of our cases as αSN pathology did not follow the predicted caudo-rostral spreading pathway suggested by Braak et al. [3]. Thus, in six ILBD cases less vulnerable structures were affected in the absence of LB and LN in regions considered more vulnerable to αSN pathology.

Nevertheless, the findings by Del Tredici et al. [2], Braak et al. [3], as well as our own results, are challenging the traditional view that SNpc is the region mainly and most early affected in PD, but are compatible with the well-known fact that nigral damage is always accompanied by extensive extranigral pathology, including that in lower brainstem regions.

Our results do not allow to pinpoint the precise induction site of αSN pathology among the lower parts of the central nervous system including the spinal cord and the medulla as there were no cases among the studied cohort with isolated spinal or medullary pathology. However, the average densities of αSN positive lesions were generally higher in the lower brainstem, particularly in the region of the dorsal motor nucleus of the vagus nerve, than in the spinal cord nuclei.

This suggests that the disease process may start in lower brainstem and subsequently pursue both an ascending course to more rostrally situated brainstem nuclei and a descending course towards the spinal cord. To solve this issue will probably need a much larger cohort of elderly as well as younger ILBD subjects to be included in future studies.