Spatial distribution of Parkinson's disease mortality in Spain, 1989-1998, as a guide for focused aetiological research or health-care intervention

Jess de Pedro-Cuesta 2 Eduard Rodrguez-Farr 0 1 Gonzalo Lopez- Abente 3 4 0 Department of Pharmacology and Toxicology, Barcelona Institute of Biomedical Research (Instituto de Investigaciones Biomedicas de Barcelona -IIBB), Scientific Research Board-August Pii Sunyer Biomedical Research Institute (Consejo Superior de Investigaciones Cientificas-Institut d'Investigacions Biomediques August Pii Sunyer:CSIC-IDIBAPS) , Rosellon 161, E-08036 Barcelona , Spain 1 Environmental Health Group, Consortium for Biomedical Research in Epidemiology & Public Health-Carlos III Institute of Health (CIBERESP-ISCIII) , Barcelona , Spain 2 Department of Applied Epidemiology, National Centre for Epidemiology, and Consortium for Biomedical Research in Neurodegenerative Diseases (Centro de Investigacion Biomedica en Red sobre Enfermedades Neurodegenerativas - CIBERNED), Carlos III Institute of Health. C/ Sinesio Delgado 6. 28029 Madrid. Spain 3 Consortium for Biomedical Research in Epidemiology & Public Health (CIBERESP) , C/ Sinesio Delgado 6. 28029, Madrid , Spain 4 Environmental and Cancer Epidemiology Unit, National Centre for Epidemiology, Carlos III Institute of Health , C/ Sinesio Delgado 6. 28029, Madrid , Spain Background: Aetiologically, genetic and environmental factors having an uneven spatial distribution may underlie Parkinson's disease (PD). Undiagnosis of PD in selected regions might have limited access to treatment with levodopa and simultaneously, if present at death, determined PD underreporting at the death record. The purpose of this study was to describe and analyse municipal mortality due to PD in Spain in aetiological and interventional perspective. Methods: PD mortality at a municipal level was modelled using the Besag-York- Molli autoregressive spatial model, combining demographic information with cause-of-death diagnostic data (International Classification of Diseases 9th Revision (ICD-9) code 332.0). Municipal relative risks (RRs) were independently estimated for women, men and both sexes, and plotted on maps depicting smoothed RR estimates and the distribution of the posterior probability of RR>1. Results: A south-north gradient, with large geographical areas suggesting clustered towns with high mortality, was seen in Asturias, the Basque Country, Balearic Islands and, particularly, in the Lower Ebro valley around Tarragona. Similarly, there was a suggestion that lowest mortality was clustered in the south-east and south-west. We identified some isolated or clustered municipalities with high mortality that were situated near industrial plants reported to be associated with environmental xenobiotic emissions. However, the same pattern was also observed for some cities with low mortality. Conclusion: Municipal PD mortality in Spain was unevenly distributed. Patterns were roughly similar to reported provincial PD mortality and use of levodopa. While the overall pattern appears to result from spatially selective PD undiagnosis, and can not be ascribed to industrial emissions, it can not be excluded that selected "hot spots" reflect genetic factors and/or environmental exposures inducing parkinsonism. A few municipal populations, located in lowmortality-risk areas in the vicinity of polluting plants or registering high excess PD mortality, might constitute a priority for conducting direct etiological studies. Additionally, interventions aimed to reduce potential PD undiagnosis might be most appropriate in the South. - Background The purpose of this study was to detect spatially uneven mortality from Parkinson's disease (PD) in Spain as a tool potentially useful for design of focused etiological research and interventions aimed to reduce possible PD undiagnosis and undertreatment. The first aim requires a detailed positioning with regard to causality in PD. The second one, will consider comparisons with geographical patterns of levodopa use (LDU). The aetiology of the most common forms of Parkinson's Disease (PD), whether sporadic or familial, is poorly understood. Genetic heterogeneity, with at least eight susceptibility loci, has been implicated in rare, monogenic, familial forms [1]. Nevertheless, low concordance in twins [2], familial-aggregation patterns [3], birth-cohort effects [4], and the results of diverse genome-wide linkage and association studies [1,5,6], albeit debated, support the contention that the large majority of sporadic PD cases result from a synergistic effect of multigenic inheritance and environmental factors. The nature of Lewy bodies, a hallmark of late-onset PD neurodegeneration, consisting of deposits of aggregated misfolded -synuclein and other proteins through the dysfunctioning of -synuclein and other genes, may, in part, have elucidated molecular phenomena responsible for monogenic familial and some sporadic PD forms [7,8]. Exposure to combined environmental chemicals, such as pesticides and metals, may alter -synuclein and dopaminergic function in the substantia nigra [9-12]. In addition, a mitochondrial complex I function defect --described in brain tissue and platelets of patients affected by sporadic PD-- has been proposed as a plausible pathophysiological apoptotic mechanism shared by most PD forms [13-15]. Through activation of apoptotic molecular pathways, mitochondrial poisons, such as 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) and rotenone, lead to degeneration of substantia nigra pars compacta, of a type similar to that seen in PD [16,17]. Likewise, many other oxidative stress-inducing agents lead to activation of mitochondrial apoptotic pathways [10]. Recently-published animal model studies are consistent with a complex aetiology for late-onset PD, i.e., multigenic susceptibility paving the way for a neurodevelopmental basis plus environmental toxic multi-insults or ageing [18]. However, evidence from PD natural history or epidemiology which supports specific steps, is sparse. Clinical and experimental MPTP observations in the early 1980s fuelled the search for environmental toxins potentially implicated in PD; paraquat, a herbicide chemically similar to MPTP, was perhaps the first of a series of such hypotheses relating to various manmade toxins, with pesticides being by far the most frequently investigated [9,10,12,19]. A detailed review of five cohort studies, 38 separate case-control studies and one meta-analysis on pesticide risk, supported by even more recent findings [20], concluded that there does indeed appear to be evidence of a potential role of pesticides in the development of PD, with the current body of evidence being insufficient to establish causation for any specific pesticide [21]. Conversely, an ubiquitous natural toxin, Pertussis toxin, mediated by high age at infection, or dietary exposure to marine food contaminants has been proposed as aetiological environmental factors explaining birth-cohort effects in Iceland and high PD prevalence in geographical isolates, i.e., in Iceland, Greenland (...truncated)