INTRODUCTION — Parkinson disease (PD) is the most common cause of parkinsonism, a syndrome manifested by rest tremor, rigidity, bradykinesia, and postural instability. The disorder was first described by James Parkinson in his 1817 Essay on the Shaking Palsy. Although it has been proposed that PD emerged as a result of the industrial revolution, there is some evidence that a disease known as "kampavata," consisting of shaking (kampa) and lack of muscular movement (vata), existed in the ancient Indian medical system, Ayurveda, as long as 4500 years ago [1]. The Mucuna pruriens plant was used in ancient times to treat the symptoms, and was later discovered to contain levodopa [2].

The pathology of PD was not well understood until the early 20th century, when the German pathologist Frederick Lewy in 1912 reported neuronal cytoplasmic inclusions in a variety of brain regions. In 1919, Tretiakoff observed that the most critical abnormality in PD was the loss of neurons in the substantia nigra pars compacta (SNc) of the midbrain. In the 1950s, investigators discovered the importance of dopamine and its depletion from the basal ganglia as the key to understanding the pathophysiology and pathologic biochemistry of PD [3].

While the cause of PD is still unknown, remarkable advances have been made in understanding the possible underlying mechanisms [4]. This extraordinary progress has been fueled by new discoveries about the anatomy and function of the basal ganglia, by improved characterization of neuropathologic and neurochemical abnormalities in PD, and by studies of genetic and experimental forms of parkinsonism.

This topic will review the etiology and pathogenesis of PD. Clinical issues related to PD are discussed separately. (See "Clinical manifestations of Parkinson disease" and "Diagnosis and differential diagnosis of Parkinson disease" and "Initial pharmacologic treatment of Parkinson disease".)

PATHOPHYSIOLOGY — Dopamine depletion from the basal ganglia results in major disruptions in the connections to the thalamus and motor cortex, and leads to parkinsonian signs such as bradykinesia.

A number of compensatory mechanisms may operate to mask or reduce the deleterious effects of dopamine depletion, particularly in the presymptomatic phase of PD, but they are eventually overwhelmed by disease progression.

Basal ganglia circuits — The basal ganglia, sometimes referred to as the extrapyramidal system, include the substantia nigra, striatum (caudate and putamen), globus pallidus (GP), subthalamic nucleus (STN), and thalamus.

The cortical input to the basal ganglia from the prefrontal supplementary motor area, amygdala, and hippocampus is excitatory, mediated by the neurotransmitter glutamate. Neurons in the substantia nigra pars compacta (SNc) provide major dopaminergic input to the striatum and exert both excitatory and inhibitory influences on the striatal output neurons. The interaction between the afferent and efferent pathways is mediated by striatal interneurons, which utilize acetylcholine as the main neurotransmitter.

The striatal output system is mediated by the inhibitory neurotransmitter gamma-aminobutyric acid (GABA). The connection between the STN and the internal (medial) globus pallidus (GPi) and between STN and the lateral (or external) globus pallidus (GPe) is excitatory, mediated by glutamate.

Five distinct dopamine receptors (D1 through D5) have been cloned and characterized; they are found throughout the basal ganglia and limbic system. The D1 and D2 receptors are highly concentrated in the dorsal (motor) striatum and are the most relevant to the pathophysiology of PD because they are activated by the dopaminergic pathway originating in the SNc and terminating in the caudate and putamen. Receptors designated as D3, D4, and D5 are more abundant in the mesolimbic or emotional part of the brain (D3, D4) and hippocampus/hypothalamus (D5) [5].

Dopamine deficiency in the nigrostriatal pathway, such as seen in PD, causes denervation hypersensitivity of D1 and D2 receptors [6]. When compared with normal matched controls, D2 receptors in the dorsal putamen are increased by 15 percent in patients with PD, whereas D3 receptors in the mesolimbic system are decreased by 40 to 45 percent [7]. These results may explain the basis for the hypersensitivity of the nigrostriatal (D2) dopaminergic receptors that is observed in PD.

There are two output pathways from the striatum (figure 1):

●The indirect pathway is mediated chiefly via dopamine's inhibitory influence on striatal D2 dopamine receptors. In the indirect pathway, the striatum projects to the neurons in the GPe utilizing GABA, and the GPe in turn projects to the STN, which provides excitatory input via glutamate to the GPi and substantia nigra pars reticulata (SNr). GPi neurons are GABAergic and synapse in the ventrolateral nucleus of the thalamus. Thalamic input to the cortex is excitatory.

●The direct pathway is mediated via dopamine's excitatory influence on striatal D1 dopamine receptors. In the direct pathway, the striatum projects directly to the GPi and SNr.

In PD, a reduction of dopamine-producing neurons from the normal complement of approximately 550,000 to the critically low level of 100,000 leads to dopamine depletion in the substantia nigra and in the nigrostriatal pathway to the caudate and putamen. This, in turn, results in relative overactivity of the indirect pathway, functionally disinhibiting the STN. Decreased inhibition of the direct pathway causes additional disinhibition of the output nuclei (GPi and SNr). Increased output from GPi causes increased inhibition of the thalamus and reduced excitatory input to the motor cortex, which is ultimately expressed as bradykinesia and other parkinsonian signs.

In PD, synchronized oscillatory activity in the 10 to 50 Hz band (often termed the beta-band), prevalent in the basal ganglia thalamocortical circuit, may be important in mediating certain parkinsonian features, including bradykinesia and tremor, and can be reduced by dopaminergic treatments [8]. Therefore, surgical treatments of PD, such as lesion placement within or stimulation of GPi or STN, may act by desynchronizing the oscillatory basal ganglia-thalamo-cortical network activity.

Models of basal ganglia dysfunction (figure 1) are useful for conceptualizing how the motor symptoms of PD arise. However, the actual pathophysiology of the basal ganglia associated with PD is much more complex than indicated by the current models [9]. Existing models should be constantly reevaluated as new findings become available.

Compensatory mechanisms — The brain has a remarkable capacity to compensate for the presynaptic dopamine depletion by increasing the synthesis of dopamine in surviving neurons and by increasing the afferents to the dendrites of dopaminergic neurons. Furthermore, dopaminergic denervation has been shown to result in a proliferation of D2 receptors, as well as a co-localization of D1 and D2 receptors [10]. Similarly, gap junctions, which allow rapid communications between striatal neurons, increase dramatically after dopaminergic denervation [11].

In the brains of patients with PD, the number of tyrosine hydroxylase-staining neurons in the striatum is markedly decreased [12]. Since tyrosine hydroxylase, the rate-limiting enzyme in the synthesis of dopamine, is still present in surviving neurons, the synthesis of dopamine could be increased in these striatal neurons, thereby compensating for the presynaptic dopamine loss [13,14]. Another compensatory mechanism may be downregulation of the dopamine transporter, resulting in less dopamine reuptake and higher synaptic dopamine levels [15].

Three stages of compensation during the presymptomatic period of PD have been proposed [14]:

●An early period during which the dopamine homeostatic compensatory mechanisms discussed above are capable of "masking" the disease

●Increased activity of the basal ganglia output nuclei (eg, internal segment of the GP) as striatal dopamine homeostasis breaks down

●Increased intensity of compensation in structures outside of the basal ganglia (eg, supplementary motor area of the cortex) as parkinsonian motor abnormalities emerge

PATHOLOGY — Depigmentation, neuronal loss, and gliosis, particularly in the substantia nigra pars compacta (SNc) and in the pontine locus ceruleus, are typical abnormalities found in the brains of patients with PD. Neuronal degeneration is also present in the dorsal nucleus of the vagus in the medulla and other brainstem nuclei.

Using a quantitative method, one study of seven patients with PD and seven controls found that the number of pigmented neurons in the substantia nigra, normally 550,000, was reduced by 66 percent in those with PD [16]. In addition, the number of nonpigmented neurons, normally 260,000, was reduced by 24 percent. By the time the first symptoms of PD emerge, approximately 60 percent of the neurons in the SNc have been lost [3].

The ventrolateral portion of the substantia nigra that projects to the dorsal putamen is preferentially affected early in the course of PD, resulting in the gradual loss of dopaminergic neurons in the SNc and a nearly complete depletion of dopamine, particularly in the putamen [17]. This contrasts with normal aging, which is usually associated with neuronal loss in the dorsal tier of the SNc, and depletion of dopamine, predominantly in the caudate nucleus [18].

Because of the apparent discrepancy between loss of striatal dopamine (>80 percent) and the degree of loss of neurons in the substantia nigra (50 to 60 percent), some have suggested that the initial site of pathology is in the striatum and that retrograde degeneration may be responsible for the neuronal loss in the substantia nigra [3]. An alternative explanation is that each dopaminergic neuron has multiple projections that terminate in the striatum, so that death of the cell body has a multiplying effect on loss of terminals.

In addition to the degeneration of the SNc, other nuclei are affected by the pathology of PD, including the internal segment of the globus pallidus (GPi), the center median-parafascicular complex, the pedunculopontine tegmental nucleus, and the glutamatergic caudal intralaminar thalamic nuclei [19]. Moreover, volumetric magnetic resonance imaging (MRI) studies have found significant hippocampal atrophy in patients with PD, with or without cognitive impairment [20].

Lewy bodies and other intracellular inclusions — There is no consensus as to what pathologic criteria are necessary for the diagnosis of PD [4], but most investigators believe that Lewy bodies, named for Frederick Lewy, constitute the pathologic hallmark of PD.

Lewy bodies are round, eosinophilic, intracytoplasmic neuronal inclusions. They are 3 to 25 nm in diameter with a dense granular core (1 to 8 nm) and loosely arranged fibrillary elements extending towards a peripheral "halo." Immunohistochemical studies have demonstrated that Lewy bodies are made up mainly of alpha-synuclein and ubiquitin, and also contain calbindin, complement proteins, microfilament subunits, tubulin, microtubule associated protein 1 and 2, and a parkin substrate protein called Pael-R [21]. However, Lewy bodies do not contain the tau protein [21].

As noted, Lewy bodies stain for alpha-synuclein, a 140 amino acid presynaptic protein found mutated in rare families with PD. Lewy bodies also contain synphilin-1 and other cytoskeletal proteins associated with alpha-synuclein [22]. (See 'SNCA-associated PD' below.)

In patients with PD, Lewy bodies are seen in the substantia nigra, the basal nucleus of Meynert, the locus ceruleus, the cerebral cortex, the sympathetic ganglia, the dorsal vagal nucleus, the myenteric plexus of the intestines, and even in the cardiac sympathetic plexus. Lewy bodies appear to arise from the peripheral portion of other inclusions known as pale bodies, which are found in the substantia nigra and locus ceruleus [23].

Lewy bodies are not specific for PD, since they are found in as many as 10 percent of brains of normal older adults, and in patients with other neurodegenerative diseases, such as neurodegeneration with brain iron accumulation (NBIA), ataxia-telangiectasia, progressive supranuclear palsy, corticobasal degeneration, Down syndrome, and Alzheimer disease. There is growing evidence that Lewy bodies occur not only in synucleinopathies such as PD, or amyloidopathies such as Alzheimer disease, but also in tauopathies, such as frontotemporal dementia [24].

The lack of specificity of the pathologic findings raises the possibility that PD may not be a specific disease entity, but rather a clinically prototypical syndrome with different clinical subtypes and pathogenic causes (table 1) [25].

Inclusions such as Lewy bodies have traditionally been considered toxic. However, some studies suggest that they may actually be neuroprotective, and that compounds that promote the formation of inclusions lessen the pathology of PD [23,26].

Braak staging — In the traditional view, the pathologic process of PD starts with degeneration of dopaminergic neurons in the substantia nigra. This view has been challenged by the neuropathologist Heiko Braak, who has proposed that the pathologic changes of PD start in the medulla of the brainstem and in the olfactory bulb, progressing rostrally over many years to the cerebral cortex in a predictable six-stage process (figure 2) [27,28].

According to Braak staging, the progression of pathologic changes occurs as follows [27]:

●During presymptomatic stages 1 and 2, the pathologic changes are found in the medulla oblongata and olfactory bulb.

●In stages 3 and 4, the pathology has migrated rostrally to the SNc and other neuronal clusters of the midbrain and basal forebrain, at which time the classic motor symptoms of PD first appear.

●In end-stages 5 and 6, the pathologic process encroaches upon the telencephalic cortex of the temporal and frontal lobes.

However, the validity and predictive utility of Braak staging has been questioned, as there are no cell counts to correlate with the described synuclein pathology and no observed asymmetry in the pathologic findings that correlate with the well-recognized asymmetry of clinical findings [29,30].

In addition, there is controversy as to the classification of dementia with Lewy bodies (DLB), considered by some to be a separate entity from PD. Braak did not include DLB his observations on the progression of PD. (See "Epidemiology, pathology, and pathogenesis of dementia with Lewy bodies" and "Clinical features and diagnosis of dementia with Lewy bodies".)

PATHOGENESIS OF CELL DEGENERATION — Irrespective of the initial trigger (etiology) of the neuronal degeneration in PD, the pathogenesis of neurodegeneration probably involves either programmed cell death (apoptosis) or necrosis [4,31-33].

Apoptosis is characterized by condensation of cytoplasm and chromatin, DNA fragmentation, and cell fragmentation into apoptotic bodies, followed by lysosome-mediated phagocytosis. The other mechanism of cell death, called autophagy, is characterized by accumulation of autophagic vesicles (autophagosomes and autophagolysosomes), and also plays an important role in neurodegeneration in PD [34].

It has been suggested that only 0.5 percent of substantia nigra neurons in normal brains are undergoing apoptosis, but this number is increased fourfold to 2 percent in those with PD. Some experimental models of PD suggest that apoptosis is the primary mechanism of substantia nigra neuronal degeneration in PD, but convincing evidence from careful neuropathologic studies is lacking [35,36].

Although the precise mechanisms of neurodegeneration in PD are not yet understood, they most likely involve a cascade of events that include interaction between genetic factors and abnormalities in protein processing, oxidative stress, mitochondrial dysfunction, excitotoxicity, inflammation, immune regulation, glial-specific factors, lack of trophic factors, and other, yet unknown, mechanisms (figure 3 and figure 4).

One of the emerging hypotheses is that neurodegeneration in PD is due to disruption of intracellular vesicular transport as a result of destabilizing of microtubules [37]. Another area of increasing interest is the role of astrocytes in various neurodegenerative disorders [38].

Some have suggested that PD arises from a prenatal event or process that predisposes some individuals to have fewer dopaminergic neurons at the time of birth because of abnormalities in the genes that code for factors important in the development of the dopaminergic system [39].

As an example, it is possible that prenatal or early postnatal exposure to certain dopaminergic neuronal toxins, such as the 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP)-like herbicide paraquat and the manganese-containing fungicide maneb, may reduce the number of dopamine neurons in the substantia nigra in early development and enhance vulnerability to these toxins when individuals are subsequently exposed in adulthood [40]. This is consistent with the so-called "multiple-hit hypothesis."

Protein misfolding, aggregation, and toxicity — Mutations in the gene on chromosome 4q21.3-q22 that codes for alpha-synuclein (SNCA) have emerged as one of the most important elements of cell death in various neurodegenerative disorders, together known as synucleinopathies [4,41]. These include not only PD, but also dementia with Lewy bodies (DLB), multiple system atrophy, and neurodegeneration with brain iron accumulation (NBIA).

Alpha-synuclein is quite abundant in the central nervous system (CNS), accounting for 1 percent of total CNS protein. Its physiologic role is not fully understood, though it appears to be involved in synaptic function and plasticity [42]. A number of observations suggest that abnormal alpha-synuclein processing plays a role in the pathogenesis of PD:

●Mutations in the SNCA gene may cause the natively unfolded alpha-synuclein protein to alter its secondary structure and self-aggregate after being targeted for proteasomal degradation by ubiquitin [43]. Misfolding of proteins and subsequent formation of insoluble aggregates can occur due to gene mutations that result in abnormal structure of the gene product, or as a result of age-related phenomenon [44].

●An aggregated and insoluble form of alpha-synuclein is a major component of Lewy bodies, intracellular inclusions that are the pathologic hallmark of PD [45]. (See 'Lewy bodies and other intracellular inclusions' above.)

●A number of observations in transgenic and normal mice and in humans suggest that misfolded forms of alpha-synuclein can somehow be transmitted from diseased neurons to healthy ones [46].

•In normal mice, a single injection of synthetic misfolded alpha-synuclein fibrils into the striatum leads to cell-to-cell transmission of pathologic alpha-synuclein and Lewy body-like pathology with progressive loss of dopaminergic neurons in the substantia nigra pars compacta (SNc) and impaired motor coordination [47,48]. Pathologic alpha-synuclein appears to act as a template that corrupts normal alpha-synuclein, so it too becomes pathologic and thereby spreads the disease from an affected neuron to a normal one, which in turn becomes diseased.

•Human fetal nigral neurons transplanted into the brains of patients with PD developed Lewy bodies that are demonstrable at autopsy in long-term survivors [49,50].

●The hydrophobic portion of alpha-synuclein can spontaneously form fibrillar protein aggregates, and SNCA mutations may promote the development of these aggregates [51]. Furthermore, alpha-synuclein protoaggregates or oligomers can disrupt cell membranes, including dopamine vesicles and mitochondria, possibly by causing pores in the membranes [51]. A protofibrillar form of alpha-synuclein appears to be more toxic than the normal or fibrillar form. Several anti-synuclein strategies are currently being investigated in clinical trials as potential disease-modifying therapies [52].

●Recombinant alpha-synuclein preformed fibrils (PFF) similar to those found in PD cause poly(adenosine 5’-diphosphate-ribose) [PAR] polymerase-1 (PARP-1) activation and cell death when injected into mouse brain, and PAR generated by PARP-1 activation binds to alpha-synuclein PFF and accelerates fibrillization, misfolding, and in vivo spread [53]. Further, transmission and neurotoxicity of pathologic alpha-synuclein in this model system were attenuated by PARP-1 deletion and by clinically available PARP inhibitors (developed as anticancer drugs).

It should be noted that the Braak hypothesis and the notion of aggregated synuclein as the primary pathogenic mechanism of neurodegeneration have been increasingly challenged. Some investigators have proposed that abnormal soluble oligomers and fibrils of alpha-synuclein that aggregate into Lewy bodies are merely byproducts and may actually serve a protective rather than toxic function [54,55].

Defective proteolysis — Cellular protein homeostasis is normally maintained primarily by three coordinated pathways (molecular chaperones, the ubiquitin-proteasome system, and the autophagy-lysosomal pathway) that mediate the repair or removal of abnormal proteins [56-59]. While the data are not entirely consistent, it appears that all three pathways are involved in the processing of alpha-synuclein. When these systems are inhibited or impaired, abnormal proteins such as mutated alpha-synuclein can misfold, aggregate, and clog the normal molecular traffic of the cell, leading to cell death.

Of particular interest is the finding in mice that the proteins parkin, PTEN-induced putative kinase 1 (PINK1), and DJ-1 (encoded by the PARK7 gene) bind to each other to form a complex that promotes degradation of unfolded or misfolded proteins via the ubiquitin-proteasome system [60]. This observation is notable because gene mutations of parkin (PARK2), PINK1, and DJ-1 (PARK7) are individually associated with autosomal recessive forms of PD. Furthermore, Atp13a2 deficiency can cause lysosomal dysfunction and enhance the accumulation and toxicity of alpha-synuclein in vitro [57,61]. This finding may reflect the pathogenesis of neurodegeneration associated with loss-of-function ATP13A2 gene mutations that cause an early-onset form of parkinsonism. (See 'Genetics' below.)

Mitochondrial dysfunction — The role of mitochondria in the pathogenesis of PD was first suggested by discovery of the association between the meperidine analogue MPTP and parkinsonism [62,63]. The oxidation of MPTP produces 1-methyl-4-phenylpyridium (MPP+), which is taken up by dopaminergic terminals, selectively inhibits mitochondrial complex I activity, disrupts calcium homeostasis, and induces endoplasmic reticulum stress, resulting in cell damage [63,64]. Direct evidence of mitochondrial dysfunction is supported by the finding that complex I activity is decreased by 32 to 38 percent in the substantia nigra of patients with sporadic PD [65,66], and by the finding that mitochondrial membrane potential and intracellular adenosine triphosphate (ATP) levels are significantly decreased in skin fibroblasts of patients with PD who carry the leucine-rich repeat kinase-2 (LRRK2) G2019S mutation [67].

The following cascade of intracellular events has been postulated to lead to neurodegeneration [68,69]:

●A cellular insult (eg, oxidative stress, excitotoxicity, DNA damage) increases cytosolic calcium and oxidative radicals, and activates nuclear enzyme poly(ADP-ribose) polymerase-1 (PARP-1) leading to the formation of poly(ADP-ribose PAR)

●This leads to decreased mitochondrial membrane potential, which in turn opens mitochondrial permeability transition pores (PTP)

●Release of nicotinamide adenine dinucleotide (NAD+) through the PTP leads to NAD+ depletion

●Release of mitochondrial apoptosis initiating factors promotes release of cytochrome c, which leads to activation of the "executioner" enzyme caspase and to apoptosis

Based on epidemiologic studies supporting an association between exposure to pesticides and PD (see 'Risk factors' below), and evidence of mitochondrial complex I deficiency in PD, an animal model of PD pathogenesis has been proposed [70]. The lipophilic pesticide rotenone is a potent inhibitor of mitochondrial complex I. Continuous infusion of rotenone into the jugular vein of rats for several weeks produced highly selective degeneration of the nigrostriatal dopaminergic pathway, associated clinically with bradykinesia and rigidity, and pathologically with fibrillary cytoplasmic inclusions staining for ubiquitin and alpha-synuclein.

Unlike MPTP, rotenone does not require the dopamine transporter for neuronal uptake. The degeneration actually begins in the dopaminergic nerve terminals in the central and dorsolateral striatum, leading to retrograde apoptotic death of the cell body. Rotenone infusion is associated with marked and uniform depletion of complex I in the brain and generation of reactive oxygen species, or oxygen-free radicals, resulting in oxidative damage [70]. This can also lead to release of cytochrome c from the mitochondria into the cytoplasm and aggregation of proteins, such as alpha-synuclein, and formation of cytoplasmic inclusions.

The rotenone model provides support for the theory that neurodegeneration results from an interaction between environmental exposure and mitochondrial dysfunction. Further evidence for the role of mitochondria in the pathogenesis of PD is provided by the increasing number of genes (such as PINK1) coding for mitochondrial proteins, and implicated in cellular protection against oxidative damage, that have been associated with the PD phenotype [71,72]. (See 'PINK1-associated PD' below.)

Oxidative stress — The oxidative stress hypothesis postulates that inappropriate production of reactive oxygen species leads to neurodegeneration [66,73]. Dopamine is normally metabolized not only by monoamine oxidase-mediated enzymatic oxidation, but also by auto-oxidation to neuromelanin.

Intraneuronal neuromelanin appears to have a dual role [74]. First, it may be neuroprotective, preventing toxic accumulation of metabolites of catechol amines and scavenging reactive metals, pesticides, and other oxidants. Second, dying neurons may release neuromelanin, leading to chronic inflammation.

These metabolic pathways generate byproducts, including hydrogen peroxide, superoxide anions, and hydroxyradicals. Free radicals, through interaction with membrane lipids, cause toxic lipid peroxidation, which has been found to be increased in the substantia nigra of PD brains. The oxidative products may cause neurotoxicity and, therefore, may play an important role in the development of PD.

It is possible that increased oxidative stress also contributes to misfolding of proteins. This notion is supported by the finding that nitric oxide, a free radical increased in the brains of patients with PD, attacks disulfide isomerase, an aggregation-preventing chaperone protein localized to the endoplasmic reticulum and normally responsible for unfolding and transport of proteins [75].

Iron metabolism — Elemental iron plays a critical role in oxidative metabolism, and it also serves as a cofactor in the synthesis of neurotransmitters [76]. It is increased by approximately 50 percent in substantia nigra of PD brains relative to controls [77], suggesting that abnormal iron metabolism plays a pathologic role in the development of PD [78]. One study found that mice lacking the tau protein developed parkinsonism due to toxic iron accumulation and neuronal loss in the substantia nigra [79]. In addition, loss of tau in neuronal culture caused intracellular iron retention. Brain permeable iron chelators prevent experimentally induced degeneration of nigrostriatal dopamine neurons [79,80].

Immunologic and inflammatory mechanisms — Immunologic mechanisms have been implicated in the pathogenesis of PD [81,82]. Supporting evidence comes from the finding of elevated levels of the proinflammatory cytokines TNF-alpha, interleukin-1 beta, and interferon-gamma in patients with PD.

The role of inflammatory processes in the pathogenesis of PD is further supported by the following observations:

●Cyclooxygenase-2, the rate-limiting enzyme in prostaglandin E2 synthesis, appears to be upregulated in patients with PD and in the MPTP mouse model of PD; cyclooxygenase-2 inhibition prevents the formation of potentially toxic dopamine-quinones in MPTP mice and presumably in patients with PD [83].

●In a positron emission tomography (PET) study that used markers for activated microglia and for dopamine transporter, microglial activity in patients with PD correlated with decreased density of dopamine transporter [84].

●Infiltration of CD4+ T lymphocytes contributed to neuronal cell death in a mouse model of PD [85].

EPIDEMIOLOGY

Incidence and prevalence — PD is a growing source of disability and mortality among neurologic disorders. Based on a meta-analysis of 47 studies, the worldwide prevalence of PD is estimated to be 0.3 percent in the general population 40 years of age and older [86]. In the year 2016, the estimated global prevalence of PD was 6.1 million people, increased from 2.5 million in 1990 [87]. A similar trend has been observed in age-adjusted mortality from PD [88]. The increase over time is largely although not entirely accounted for by increasing numbers of older people.

Estimates of the incidence of PD (number of new cases per year) range from 8 to 18.6 per 100,000 person-years [89].

A male preponderance of PD has been observed in many but not all epidemiologic studies [86,89,90], suggesting that men have a higher risk than women for developing PD.

Associated factors — Epidemiologic studies may provide important clues to potential risk factors associated with PD. Data for most putative PD risk factors are conflicting, but there is consistent evidence that older age and a family history of PD are associated with an increased risk of developing PD, while cigarette smoking is associated with a decreased risk [91,92].

Protective factors — An inverse correlation between PD and smoking is supported by the findings of large cohort studies and meta-analyses [4,91,93-96]. In a 2012 meta-analysis of 26 observational studies, the risk of PD was more than twofold lower for current smokers compared with never smokers (relative risk [RR] 0.44, 95% CI 0.39-0.50) [91]. In addition, the risk of PD was lower for ever smokers compared with never smokers (RR 0.64, 95% CI 0.60-0.69). A neuroprotective effect of nicotine has been proposed as one possible explanation for these observations [97]. The relatively high frequency of nonsmokers among patients with PD does not explain the relatively low prevalence of most cancers, except for melanoma and possibly breast cancer, reported in patients with PD [98].

An alternative hypothesis is that patients who develop PD are less likely to smoke in the first place, or more likely to quit smoking than those who do not develop PD. This alternative explanation posits that since dopamine is an integral component of the brain's reward system, people who will later develop signs of PD do not engage in reward-seeking behaviors, such as smoking, because dopamine is significantly depleted in the basal ganglia years before symptoms of PD appear [99-101]. Others hypothesize that the tendency to smoke represents a risk-taking behavior, and that general risk tolerance as a genetically determined trait is causally related to PD [102].

Other factors associated with a reduced risk of PD in at least some studies include coffee and caffeine intake, exercise, and certain medications.

●Caffeine – Coffee and caffeine intake have been associated with lower risk of PD in meta-analyses and large cohort studies [91,95,103,104].

●Exercise – Aerobic exercise and physical activity may be protective against development of PD [105-111]. In a meta-analysis of eight prospective studies in more than 500,000 individuals, moderate to vigorous physical activity was associated with an approximately 30 percent reduction in the RR of PD [112]. However, an alternative explanation for the association is reversed causality, given that reduced physical activity may be a preclinical sign of PD.

●Ibuprofen – There is evidence from several meta-analyses that ibuprofen may be associated with a reduced risk of PD [113-116]. The data regarding other nonsteroidal anti-inflammatory drugs (NSAIDs) are conflicting, with some meta-analyses finding that NSAIDs are associated with a reduced risk of PD [91,114], and others finding no significant association [113,115,116].

●Statins – The relationship between statin use and lipid levels with PD is unsettled. Several studies suggest that statins are associated with a lower risk of PD [117,118]. Other studies suggest that the apparent protective effect of statin use is explained, at least in part, by failure to adjust for confounders [119], or that statin use is associated with a higher risk of PD incidence and progression [120,121].

●Glycolysis-enhancing drugs – Terazosin, doxazosin, and alfuzosin are alpha-1-adrenergic receptor antagonists used to treat hypertension and benign prostatic hyperplasia that bind to phosphoglycerate kinase 1 (PGK1) and increase energy metabolism. In cellular and animal models, enhancing glycolysis reduces PD progression [122]. In several large database studies, use of terazosin, doxazosin, or alfuzosin has been associated with reduced risk of PD or PD progression compared with nonuse or use of tamsulosin, an alpha-1-adrenergic receptor antagonist that does not have PGK1 activity [122,123]. Further studies are needed to explore a possible protective effect prospectively.

Risk factors — A family history of PD is an important risk factor for developing PD [91]. The risk for PD associated with specific genes and genetic loci is discussed below. (See 'Genetics' below.)

A number of reports, including several large population-based case-control studies, have found an association between depression and the subsequent development of PD [124-129]. These findings suggest that depression is either a risk factor for PD or a prodromal symptom of PD. Similarly, meta-analyses of observational studies suggest a preceding history of constipation is either an early manifestation of PD or a risk factor for PD [91,130].

A large number of environmental and other potentially modifiable risk factors have been identified in epidemiologic studies. Examples include the following:

●Exposure to pesticides [91,131-137]

●Exposure to nitrogen dioxide in air pollution [138]

●High consumption of dairy products [139,140]

●History of traumatic brain injury [141-143]

●Reduced levels of dietary and sunlight-derived vitamin D [144-146]

●History of midlife migraine with aura [147]

●Living in urban or industrial areas with high release of copper, manganese, or lead [148]

●Exposure to hydrocarbon solvents, particularly trichloroethylene [149]

●Living in rural areas [91]

●Farming or agriculture work [91]

●The use of well water [91,150]

●High dietary intake of iron, especially in combination with high manganese intake [151]

●Excess body weight, type 2 diabetes, and metabolic syndrome [152-155]

Paradoxically, other data suggest that mortality from PD is increased among people with higher socioeconomic occupations (eg, education, computer and mathematical, legal, architecture and engineering) where exposure to toxins is unlikely, while PD mortality is decreased among people with lower socioeconomic occupations (eg, mining and drilling, transportation and material moving, construction) where exposure to toxins is more likely [156,157].

Although the prevalence of most cancers among patients with PD is relatively low [98], there is evidence that a history of melanoma or prostate cancer is associated with an increased risk for PD, and vice versa [158-162].

GENETICS — Although the majority of cases of PD appear to be sporadic, there is increasing evidence that genetic factors play a role in the pathogenesis of PD, particularly when the age at symptom onset is younger than 50 years [163-165].

The most cogent evidence for a genetic contribution to the pathogenesis of PD comes from the study of large multicase kindreds with a PD phenotype of monogenetic origin. These familial forms of parkinsonism (designated PARK1 through PARK13) have been associated with causative mutations in a number of nuclear genes (table 2) [166]. Autosomal dominant, autosomal recessive, and possible X-linked forms of PD have been identified [163,166].

The reported frequency of monogenetic forms of PD varies considerably across studies, depending on the genes tested, the racial and ethnic background of the cohort, and age at disease onset. In a population-based cohort in the United Kingdom that included analysis of four genes (PRKN, PINK1, LRRK2, and SNCA) in over 2000 patients with PD, monogenetic forms of PD accounted for 1.4 percent of all PD cases and approximately 3 percent of young-onset PD (≤50 years old at onset) [167]. In a multicenter cross-sectional study of 953 patients with young-onset PD that included analysis of six relevant genes (including glucocerebrosidase), the mutation carrier frequency was 17 percent [168].

Several of the most important genetic risk factors and monogenetic forms of PD are discussed in greater detail in the following sections.

SNCA-associated PD — Alpha-synuclein (SNCA) gene missense mutations (PARK1) or multiplications (PARK4) are associated with autosomal dominant parkinsonism; the phenotype varies from classic PD to dementia with Lewy bodies (DLB) [166,169,170]. However, SNCA mutations or multiplications appear to be rare causes of PD [171].

The mechanism of SNCA-associated neurodegeneration is not yet fully understood, but is probably due to a toxic gain-of-function effect [166]. As discussed earlier, there is evidence that impaired processing of alpha-synuclein leads to abnormal protein aggregation and misfolding, Lewy body formation, cellular oxidative stress, and energy depletion. (See 'Protein misfolding, aggregation, and toxicity' above.)

The SNCA gene was the first to be associated with parkinsonism when a genome scan in the Greek-Italian Contursi kindred identified a genetic marker on chromosome 4q21-q23 linked to the PD phenotype [172]. This finding led to the discovery of a mutation in the single base pair (Ala53Thr) of the SNCA gene, designated PARK1 in the hierarchy of genes associated with PD [172,173].

In the Contursi kindred, the clinical features of the disease are similar to otherwise typical PD, except for younger age at onset (mean 46), greater cognitive decline, and more rapid progression, with a mean time from onset to death of nine years [174-176]. The pattern of nigrostriatal degeneration, with preservation of D2 receptors, as demonstrated by positron emission tomography (PET), is similar to that seen in idiopathic PD.

Subsequently, a second mutation in the SNCA gene involving an alanine for proline substitution at amino acid 30 (Ala30Pro) was found in a German family [169], and a third mutation involving a glutamic acid-to-lysine substitution at position 46 (E46K) was identified in a Spanish family [170].

With reports of additional families, the phenotype of PARK1 has expanded to include not only typical PD features, but also dementia, hallucinations, central hypoventilation, orthostatic hypotension, myoclonus, and urinary incontinence, with pathologic involvement of the brainstem pigmented nuclei, hippocampus, and temporal neocortex. Thus, the clinical and pathologic features of families with SNCA mutations and parkinsonism overlap the features of multiple system atrophy and DLB.

Similarly, overexpression of SNCA may lead to neurodegenerative disease with features that overlap those of PD, DLB, and multiple system atrophy, as observed in families with parkinsonism and whole gene duplication or triplication of SNCA [177]. Duplication of SNCA appears to be associated with late-onset parkinsonism and dysautonomia, while triplication (designated PARK4) leads to early-onset PD and dementia [178]. SNCA triplication was found in one large family with autosomal dominant, young-onset parkinsonism, dysautonomia, cardiac denervation, DLB, and glial cytoplasmic inclusions at autopsy, features typical of multiple system atrophy [179].

Mounting evidence supports the role of common SNCA polymorphisms in sporadic PD [180-183]. In particular, several large GWAS and a meta-analysis of GWAS studies in PD found that SNCA was a risk locus for PD [182-185].

LRRK2-associated PD — The most common form of monogenic PD is PARK8, caused by mutations in the leucine-rich repeat kinase-2 (LRRK2) gene on chromosome 12p11.2-q13.1 [186,187].

The LRRK2 gene product is a protein called dardarin (from the Basque word "dardara," meaning tremor) that probably functions as a cytoplasmic kinase involved in phosphorylation of proteins, such as alpha-synuclein and microtubule-associated protein tau [188-190]. Dardarin is a large molecule, encoded by 51 exons and containing 2527 amino acids. This contrasts with the much smaller alpha-synuclein protein, which contains 140 amino acids.

Dardarin is closely associated with a variety of membrane and vesicular structures, membrane-bound organelles and microtubules, suggesting its role in vesicular transport and in membrane and protein turnover, including lysosomal degradation pathway.

LRRK2-associated PD may account for a significant proportion of familial PD cases, and a smaller proportion of sporadic PD cases. It is estimated to cause up to 8 percent of autosomal dominant PD in the Basque population and up to 50 percent of familial PD in people of North African and Middle Eastern origin [189,191-194]. In addition, LRRK2 mutations have been found in 0.4 to 1.9 percent of patients with idiopathic PD [195-199], although such cases could also be explained by reduced penetrance in familial disease [192].

Genetic screening studies suggest that the G2019S mutation, the most common of the LRRK2 mutations, accounts for 3 to 13 percent of autosomal dominant PD in Europe [171,193,200-202], and 10 to 18 percent of autosomal dominant PD in Ashkenazi Jews [194,203]. The G2019S mutation has also been identified in asymptomatic carriers, suggesting reduced or age-dependent penetrance [192,201]. Evidence of age-dependent penetrance was found in a study of 19 families with the G2019S mutation, where the cumulative incidence of PD at ages 60, 70, and 80 years was 15, 21, and 32 percent [204].

The LRRK2-associated PD phenotype is often, but not always, associated with late-onset (mean age 65 years) disease [190,201,205-207]. However, the typical features of PD associated with LRRK2 G2019S mutations are indistinguishable from idiopathic PD [176,199]. The course of the disease is relatively benign, usually presenting with unilateral hand or leg tremor without cognitive deficit. Patients respond well to levodopa and have a slower decline in motor function compared with patients without an LRRK2 mutation [208]. Several studies have suggested that LRRK2-PD patients may have an increased risk of certain types of cancer compared with idiopathic PD and healthy controls, such as leukemia, colon cancer, and possibly breast cancer [209-212].

Other clinical phenotypes associated with LRRK2 mutations have included parkinsonism with dementia or amyotrophy or both, typical essential tremor, dysautonomia, familial progressive supranuclear palsy, familial multiple system atrophy, corticobasal degeneration, and primary progressive aphasia [213,214].

Autopsy findings in patients with LRRK2-associated PD are heterogeneous and range from pure nigral degeneration without Lewy bodies, to pathology consistent with typical PD, diffuse Lewy body disease, and neurofibrillary tangle and other tau pathology [213,215,216].

Parkin-associated PD — Patients with parkin (PARK2) gene mutations usually have a family history consistent with autosomal recessive inheritance. The disease is characterized by early onset of symptoms (before age 50), a slowly progressive course, a symmetric presentation at onset of parkinsonian signs and symptoms (unlike classic PD, which is asymmetric), and early dystonia and postural instability. Additional features include leg tremor, freezing, festination, retropulsion, hyperreflexia, sensory axonal neuropathy, and autonomic involvement [217-220]. Dementia is uncommon (<3 percent) among PARK2 mutation carriers [221]. However, on an individual basis, patients with early-onset PD who have parkin mutations are clinically indistinguishable from those with early-onset PD who lack parkin mutations [222].

There is typically a good response to levodopa with early development of motor fluctuations and dyskinesia. Fluorodopa PET shows a marked reduction in the fluorodopa uptake, similar to idiopathic PD, but asymptomatic carriers may also have abnormal PET studies [223].

Parkin, the protein product of the PARK2 gene, is expressed in the substantia nigra and other brain regions, as well as in Lewy bodies. Normal parkin strongly binds to microtubules and is involved in ubiquitination and subsequent degradation of certain proteins by proteasomes [37,224]. However, mutated parkin protein loses this proteasome-enhancing activity, and the result is a hastening of neuronal cell death because the disabled proteasome cannot clear the cell of accumulating aggregated protein. Neurodegeneration associated with the parkin mutation is not usually accompanied by formation of Lewy bodies, although there are exceptions [225].

Over 180 mutations, variants, and polymorphisms have been identified in the PARK2 gene [221]. However, it is not clear which of these changes leads to functional deficits. Several studies have demonstrated that heterozygous PARK2 variants may not necessarily be disease-causing [226-228].

The incomplete penetrance and variability of clinical and pathologic expression of PARK2-associated PD may be due to an interaction between the mutated parkin gene and other genes, including SNCA [229].

PINK1-associated PD — Mutations of mitochondrial PTEN-induced putative kinase 1 gene (PINK1; PARK6) are associated with autosomal recessive familial PD, age younger than 50 at onset, slow progression, and excellent response to levodopa [230-232], similar to parkin (PARK2) and DJ-1 (PARK7) mutations. PINK1 mutations have been found worldwide with a frequency that ranges from 1 to 8 percent of patients, most of whom had early-onset and/or familial PD [166,231,233-235].

Postmortem brain examinations of patients with PD and PINK1 mutations are limited to a few cases. One showed Lewy body pathology in the brainstem and Meynert nucleus while the locus ceruleus and the amygdala were spared [236]; another case revealed nigral degeneration without Lewy bodies [237].

PINK1 mutations may cause disease through a loss-of-function effect resulting in mitochondrial dysfunction [238,239].

DJ-1-associated PD — Mutations of the mitochondrial DJ-1 gene (PARK7) are associated with autosomal recessive inheritance, age younger than 40 at onset, slow progression, and good response to levodopa [240,241]. Wildtype DJ-1 is thought to be neuroprotective against oxidative stress.

Other monogenic forms of PD — Other monogenic forms of PD are reviewed in the table (table 2) [242-247]. Novel genetic forms of parkinsonism are likely to emerge from future research, thereby shrinking the large majority of PD cases currently regarded as sporadic.

Glucocerebrosidase gene — Heterozygous mutations in the glucocerebrosidase (GBA) gene, which when homozygous are the cause of Gaucher disease, are an important genetic risk factor for PD [4,248-251].

Across various populations, the risk of PD among GBA variant carriers is increased by two- to sevenfold over noncarriers [248,252-255]. The estimated penetrance of heterozygous GBA mutations for PD over a lifetime varies fairly widely, from approximately 10 to 30 percent depending on the population studied [256-260]. Genotype-phenotype relationships have not been well established, but GBA variant carriers as a group appear to have more cognitive decline compared with patients with idiopathic PD.

Mutations in GBA were initially linked with PD and other movement disorders in people of Ashkenazi Jewish decent, who have an increased prevalence of GBA mutations [261]. One of the largest studies analyzed data from 16 centers around the world, including 5691 patients with PD and 4898 controls [248]. All centers screened for two relatively frequent GBA mutations (N370S and L444P). Compared with controls, either GBA mutation was more common in patients with PD, both among Ashkenazi Jewish subjects (15 percent, versus 3 percent of controls) and among non-Ashkenazi Jewish subjects (3 percent, versus <1 percent of controls). Overall, the likelihood of finding any GBA mutation was significantly higher in patients with PD than in controls (odds ratio 5.43, 95% CI 3.89-7.57).

Among all patients with PD, the clinical profile was generally similar. However, when compared with patients who had PD but lacked a GBA mutation, those with PD who carried a GBA mutation were significantly more likely to have the following features [248]:

●Younger age at onset

●Less prominent tremor, bradykinesia, and rigidity

●Lower frequency of asymmetric onset

●Higher frequency of a family history of PD

●Greater likelihood of cognitive impairment

Other reports have also found that patients with PD who are GBA mutation carriers have a younger age at onset, faster motor progression, and an increased prevalence of cognitive dysfunction compared with noncarriers [261-267].

Other lysosomal genes — There is emerging evidence that mutations in other lysosomal enzyme-encoding genes also affect PD risk through effects on alpha-synuclein [268-271]. For example, several genetic variants in the acid sphingomyelinase gene SMPD1, the gene associated with Niemann-Pick type A, have been identified with increased prevalence in patients of Ashkenazi Jewish ancestry (and other populations) with PD compared with controls [272-275]. Furthermore, experimentally, SMPD1 knockdown results in alpha-synuclein accumulation [272].

Genetic testing — Molecular genetic testing is available for many of the genetic forms of PD, including SNCA, parkin (PARK2), PINK1, DJ-1 (PARK7), and LRRK2 (PARK8) gene mutations [276], but the interpretation of these tests is difficult, and there are no established disease-modifying therapies yet available for PD. Genetic testing is therefore not recommended in routine clinical practice at this time [166,277,278].

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topics (see "Patient education: Parkinson disease (The Basics)")

●Beyond the Basics topics (see "Patient education: Parkinson disease symptoms and diagnosis (Beyond the Basics)")

PATIENT PERSPECTIVE TOPIC — Patient perspectives are provided for selected disorders to help clinicians better understand the patient experience and patient concerns. These narratives may offer insights into patient values and preferences not included in other UpToDate topics. (See "Patient perspective: Parkinson disease".)

SUMMARY

●In Parkinson disease (PD), dopamine depletion in the substantia nigra and in the nigrostriatal pathway to the caudate and putamen ultimately results in increased inhibition of the thalamus and reduced excitatory input to the motor cortex (figure 1), which is expressed as bradykinesia and other parkinsonian signs. (See 'Basal ganglia circuits' above.)

●Compensatory mechanisms may operate to mask or reduce the deleterious effects of dopamine depletion, particularly in the presymptomatic phase of PD. (See 'Compensatory mechanisms' above.)

●Typical abnormalities found in the brains of patients with PD include depigmentation (from loss of neuromelanin), neuronal loss, and gliosis, particularly in the substantia nigra pars compacta (SNc) and in the pontine locus ceruleus. Lewy bodies are round, eosinophilic, intracytoplasmic inclusions in the nuclei of neurons that constitute the pathologic hallmark of PD. (See 'Pathology' above.)

●The precise mechanisms of neurodegeneration in PD are not yet understood, but they most likely involve a cascade of events that include interaction between genetic and environmental factors, and abnormalities in protein processing, oxidative stress, mitochondrial dysfunction, excitotoxicity, inflammation, immune regulation, and other mechanisms (figure 3 and figure 4). (See 'Pathogenesis of cell degeneration' above.)

●The prevalence of PD is approximately 0.3 percent in the general population 40 years of age and older. The prevalence of PD rises with age. Worldwide, there are approximately 6.1 million people with PD. (See 'Epidemiology' above.)

●There is consistent evidence that older age is associated with increased risk of PD, while cigarette smoking is a protective factor. Evidence supporting other potential risk factors for PD, including occupational exposures (eg, pesticides, herbicides, heavy metals), dietary factors, and body weight is inconclusive. (See 'Associated factors' above.)

●Although the majority of cases of PD appear to be sporadic, there are genetic forms of parkinsonism (designated PARK1 through PARK13) related to nuclear and mitochondrial genes. Novel genetic forms of parkinsonism are likely to emerge from future research, thereby shrinking the large majority of PD cases currently regarded as sporadic. (See 'Genetics' above.)