Distribution of pathology and clinico-pathological correlation
Previous reports point to the SN as the primary site of neurodegeneration in PD, Tretiakoff  has been credited for this finding (in 1919). Since then, our understanding of the pathological changes in the SN has grown significantly. LRP is first found ventrally in the SN pars compacta (SNc), and then spreads to paranigral, medial, and dorsal areas. There is neuronal loss in conjunction with LRP, specifically of melanized and dopaminergic neurons in area A9 of the SNc . These pathological changes in A9, which result in decrease in dopamine, tyrosine hydroxylase, and dopamine transporter immunoreactivity in the striatum, correlate with the duration and severity of motor symptoms . It is not clear, however, whether these neurochemical changes are directly related to LRP.
It has become evident that neurodegeneration in PD goes well beyond the SN, and in fact affects not only subcortical dopaminergic nuclei but also basal cholinergic structures, the dorsal motor nucleus of the vagus nerve, olfactory structures, raphe nuclei, the locus coeruleus, pedunculo- pontine nuclei, hypothalamic nuclei, spinal nuclei, and cortical structures. LRP, ARP, and vascular change are the principal pathologies thought to result in neuronal loss in these structures. Specifically, LRP has been documented in the striatum of patients with a variety of synucleinopathies . It has been suggested that greater LRP burden in the SN is indicative of a predominant motor syndrome as in PD or PD-D, and conversely a greater LRP burden in the striatum is more indicative of an illness of predominantly dementing type such as DLB . Furthermore, LRP often involves the peripheral nervous system, including autonomic nuclei, ganglia, and nerves subserving the vasculature, intestines, heart, and adrenal and salivary glands . Of note, there have been reports of LBs in two patients with PD who had undergone transplantation of fetal mesencephalic dopaminergic neurons with long-term survival. Grafted cells in these two patients were not functionally impaired .
LRP has been described in both limbic (layer II) and neocortical structures (layers V and VI). Pathological change in specific structures within these areas has been studied meticulously in order to correlate clinical symptomatology with cortical LRP burden, and to differentiate between synucleinopathies. Of all the cortical areas studied, it appears that LRP burden in the parahip- pocampus is most indicative of dementia, although in DLB cases the pathology was more heterogeneous . Similarly, in the entorhinal and anterior cingulate cortex (Brodmann area 24) LRP burden correlates well with cognitive deficits in patients with PD-D . A high LRP burden in the parahippocampus and amygdala correlates well with early visual hallucinations. Furthermore, LRP in the inferior temporal cortical structures is associated with earlier onset of hallucinations, and therefore with a greater likelihood of developing DLB .
The diverse pathological findings previously described, which individually correlate with some clinical symptoms, have been consolidated in different ways in order to account for specific clinical syndromes. McKeith et al.  have devised and revised criteria for the pathological diagnosis of DLB, taking into account the extent of LRP in specific brain regions, as well as ARP, and rating pathological burden in a semi-quantitative fashion. In determining which clinical syndrome is most likely indicated by the pathological change being observed, the type of pathological change is as important as the distribution of such change. Therefore, the McKeith criteria assign a probability that the pathological change is consistent with the clinical syndrome of DLB, based not only on the LRP burden (which is directly related to the likelihood of the pathology representing clinical DLB) and the ARP (which is inversely related to the likelihood of the pathology representing clinical DLB ), but also on the areas involved. These criteria group cases into brainstem-predominant, limbic or transitional, and diffuse neocortical, and the likelihood that the pathological changes represent the clinical syndrome of DLB is greater with more rostral and lesser with more caudal changes [32, 33].