The clinical syndrome known as parkinsonism was first described in 1817 by the English physician James Parkinson as "the shaking palsy" [5]. This disorder is characterized by the motor symptoms of resting tremor, muscle rigidity, and bradykinesia. Over time the non-motor features of PD have been increasingly identified, including sensory, autonomic, and neuropsychiatric symptoms, some of which appear before motor abnormalities are evident. Although the precise cause of PD is unclear, disease manifestations result from disruptions of dopaminergic neurotransmission within the central, peripheral, and autonomic nervous systems. The defining pathologic feature of PD is the loss of dopamine-producing cells and local deposition of aggregates of the protein alpha-synuclein (Lewy bodies) in the substantia nigra region of the brain [8].

Sporadic (idiopathic) PD is the second most frequent neurodegenerative disorder after Alzheimer disease [8]. The prevalence of diagnosed PD in the United States is estimated to be 500,000, but the Parkinson's Foundation Prevalence Project estimates the actual number is 930,000 [8,156]. Approximately 60,000 new cases are diagnosed each year, the majority of which are people older than 60 years of age. The prevalence of PD increases with age in both men and women; it is 1% in persons 60 years of age or older and up to 4% in those older than 80 years of age. About 10% of patients with PD had onset of illness before 50 years of age. Women develop PD at a lower rate and with later onset than men; delayed onset has been attributed to neuroprotective effects of estrogen on the nigrostriatal dopaminergic system [1,2]. The direct cost of treating PD in the United States is estimated to be $14 billion annually, and the indirect costs add another $6.3 billion [8]. Because of increasing longevity, the prevalence of PD is predicted to exceed 1.2 million by 2030 and to double by 2040 [156].

The incidence of PD varies by age, race, and ethnicity. The ratio of men to women is roughly 2:1 [9]. As noted, the incidence is markedly higher in each decade after 60 years of age, peaking after 80 years of age. The rate is highest among Hispanic individuals, followed by non-Hispanic Whites, Asians, and Black persons [9]. However, there is some variation in the incidence in each racial/ethnic group when divided by sex, with Black men and Asian women at greater risk than their other-sex counterparts. The variable prevalence of PD throughout the world suggests that environmental and genetic factors interact with ethnic differences in disease pathogenesis [2].

The first line of investigation began in the early 1960s, with an unexpected finding that linked cigarette smoking with protection against PD. This association between smoking and neuroprotection from PD has been replicated in numerous epidemiologic, pre-clinical, and case-control studies. These studies also identified coffee drinking (and caffeine) as a factor that reduced risk of developing PD [13].

In these studies, risk of developing PD is shown as odds ratio (OR), where the odds of developing PD in cigarette or coffee users was compared to non-user reference groups. An OR of 1.00 signifies no difference from the reference group, while a number greater than 1 means increased odds of developing PD and a number less than 1 indicates decreased odds of PD.

In one study, smoking, other lifestyle behaviors, family history of PD, and their interaction were examined for possible association with risk of PD diagnosis by comparing 1,808 patients in Denmark with PD diagnosis with 1,876 matched population controls [14]. Strong inverse associations were found between cigarette smoking and risk of PD, even in smokers who quit 10 years before PD diagnosis. Compared with never-smokers without PD family history, the OR was 2.81 in never smokers with family history, versus 1.60 in smokers with family history. Duration had the greatest effect in modulating the smoking-PD relationship. Current smokers who smoked 40 years or more had ORs as low as 0.30. Unlike the correlation between longer smoking and lower PD odds, smoking more than 10 cigarettes per day did not further reduce odds.

Moderate coffee intake (3.1 to 5 cups per day) (vs. no coffee intake) showed an OR of 0.45. Moderate alcohol intake (3.1 to 7 units per week) (vs. no alcohol use) was associated with an OR of 0.60; higher daily alcohol did not further reduce the odds of developing PD. Stronger negative OR for PD was found in smokers with medium-high coffee or moderate alcohol intake than with each alone. Coffee intake association with lower PD odds was found in men and women; only men showed lower risk estimates with caffeine and alcohol, largely attributed to beer consumption [14].

These findings were consistent with numerous prior studies, including publications from the prospective NIH-AARP Diet and Health Study [7,15]. In this study, 306,895 participants (58.8% male, 50 to 71 years of age) were evaluated in 1995–1996 and again in 2000–2006 for development of PD.

One NIH-AARP study examined caffeine intake, risk of PD, and whether smoking affected this relationship [15]. Higher caffeine uses in 1995–1996 were associated with lower risk of PD diagnosis in 2000–2006 for men (OR=0.75) and women (OR=0.60). The linear trend for lower odds with higher caffeine was significant for both sexes [15].

The authors also performed a meta-analysis, which confirmed the inverse association between caffeine intake and risk for PD in men and women. Together with the study findings, this data led the researchers to conclude that gender differences do not influence caffeine risk reduction of PD. Smoking and caffeine may act independently to reduce PD risk [15].

Another NIH-AARP study examined cigarette smoking and risk of PD by comparing those who developed PD to those who did not. Odds for developing PD were 0.78 in past smokers and 0.56 in current smokers, with 0.47 in men and 0.74 in women. For few current smokers at baseline who developed PD, other comparisons were not relevant [16]. The greatest reductions in odds for PD were found with current smoking and higher daily amount/duration of past smoking [16].

In the NIH-AARP study, amount and type of alcohol use was studied for risk of PD. Compared to non-drinkers, the odds ratio for developing PD was 0.73 with 1 to 1.99 drinks of beer per day, 1.22 with liquor, and 0.74 with wine. Beer and liquor consumption showed opposite effects [17].

Higher circulating levels of uric acid have also been associated with decreased incidence of PD and with slower rate of decline in patients with PD. This suggests a link to the neuroprotective effects of caffeine and the purinergic system [7,18].

PD is the most common form of neurodegenerative parkinsonism, a syndrome characterized by progressive deterioration in motor abilities resulting from dopaminergic neuron loss in the substantia nigra pars compacta and ventral tegmental area. Dopamine neuron loss is most prominent in the ventral lateral substantia nigra; 60% to 80% of these neurons are lost when motor symptoms emerge and PD is diagnosed [8,12].

There is a consistent line of evidence linking PD to a neurodegenerative process involving oxidative stress, mitochondrial dysfunction, and neuroinflammation. Environmental and genetic factors induce mitochondrial dysfunction, resulting in abnormal accumulation of miscoded proteins (mostly alpha-synuclein) and generation of oxidative stress in enteric, peripheral, and central nervous systems. In turn, oxidative stress, excitotoxicity, and mitochondrial dysfunction promote the destruction of dopamine neurons and dopaminergic function in midbrain systems [12,38,39,40].

The progression of disease and disability in PD varies and is partially influenced by patient factors such as age. In general, from the mean age at diagnosis of 61 years, the mean time to death is 14 years overall. Survival time is a mean 24 years for patients diagnosed in their 40s and 9.7 years for patients diagnosed in their 70s [4]. With the onset of subclinical non-motor symptoms decades before diagnosis, pathologic processes that underlie PD are probably active over a 40-year period in many patients [45].

Studies of olfaction in PD have shown abnormalities in up to 100% of patients, making olfactory dysfunction the most robust predictor of developing PD. As noted, many patients complain of declining sense of smell long before parkinsonism onset. As it is not disabling and relatively nonspecific, anosmia has mostly failed to gain traction as a predictive and clinical feature of PD [56].

It is important to note that postural instability, while a cardinal motor feature, is seldom a problem early in the course of PD and may not be evident at diagnosis, as it usually appears later in the disease course [60].

Rest tremor, an initial symptom in 70% to 90% of patients, refers to a 4–6 Hz tremor in the fully resting limb, suppressed during movement initiation. Rest tremor is a rhythmic, oscillatory involuntary movement and one of the most characteristic signs in clinical medicine. The most distinguishing rest tremor is the "pill-rolling" type, with rubbing movements of thumb and index fingers against each other. Rest tremor is thought to initiate with nigrostriatal degeneration and subthalamic nucleus or globus pallidus disinhibition, or disrupted thalamo-cortical-cerebellar circuits leading to abnormal thalamic pacemaker cell function [4,7].

The frequency and diversity of non-motor symptoms in PD is substantial, and includes autonomic, neuropsychiatric, olfactory, sensory, and sleep disorders that occur in 80% to 90% of patients (Table 4). Non-motor symptoms can manifest before, during, or after motor symptoms and may result in greater impairment of quality of life. The prevalence of cognitive, autonomic, and mood disorders is very high; progression can result in patients requiring care in a supervised environment [7].

Disease-specific screening tests or biomarkers for PD are not yet available, and definitive diagnosis is only possible at autopsy by confirmation of striatal dopamine neuron loss and Lewy body pathology [64]. Idiopathic PD is diagnosed through patient history and physical examination, often performed sequentially over time in order to identify signs of progression and the emergence of defining clinical features. History or physical findings inconsistent with features of idiopathic PD are explored further to rule out or confirm an alternate diagnosis. Clinicians with limited experience caring for patients with PD should consider referring a patient with suspected disease to a physician with expertise in movement disorders to confirm diagnosis [2].

The UK Brain Bank diagnostic criteria were established more than 20 years ago (in 1992) and solely address motor symptoms, leading many to consider them outdated. This led to the 2015 publication of new PD diagnostic criteria by the MDS Task Force, comprised of North American and European experts. These criteria better reflect current understanding of PD as a multi-system disorder affecting all parts of the nervous system, often with a genetic component and a very slow progression of neurodegenerative processes reflected in a long prodromal period. These aspects are incorporated in the new criteria [60].

It is also important to explore the possibility of prescription drug-induced parkinsonism, one of few reversible causes of the disorder. This can be identified by a thorough review of the medication history, paying particular attention to potential drug side effects and the time course of usage in relation to onset of parkinsonian symptoms. The drugs implicated in drug-induced parkinsonism are typical antipsychotics (e.g., haloperidol, chlorpromazine), most atypical antipsychotics (e.g., risperidone, olanzapine), and centrally acting agents used to treat gastrointestinal symptoms (prochlorperazine, promethazine, and metoclopramide). Less common causes are tetrabenazine, reserpine, methyldopa, flunarizine, cinnarizine, verapamil, valproic acid, and lithium. In a study of 155 cases of drug-induced parkinsonism diagnosed between 1995 and 2009, 70% developed symptoms within three months of beginning the prescribed medication; the remaining patients developed symptoms within one year on the offending drug [150]. Recovery from drug-induced parkinsonism can be expected following discontinuation of the medication, though many weeks to months may be required for full resolution of symptoms.

The diagnostic assessment of a patient suspected of having PD should include consideration of other, possibly reversible, disorders that may present with motor signs of parkinsonism. These are often referred to as "atypical" PD or "mimics" of PD and include [62,64]:

An important issue, unaddressed by practice guidelines, is medication non-adherence in patients with PD. While the prevalence of non-adherence broadly varies by assessment method, the figures range from 15% to 20% using patient self-report to 67% or more using pharmacy refill data and pill counts [84]. An important dimension in PD treatment is timing adherence, as dopaminergic medications should be taken at precise and evenly spaced intervals, as instructed by the prescribing physician. Non-adherence to timing of dosage is probably very common and contributes to unwanted dopamine variability implicated in earlier onset of motor fluctuations [85]. The overall consequence of non-adherence is unsatisfactory motor control, with diminishing mobility, greater fluctuations, dyskinesias, and declining quality of life [86].

In chronic diseases, highest medication adherence occurs with once-daily formulations, but this sharply decreases with each added daily dose [87]. Polypharmacy in PD is normative, with most patients taking two or more antiparkinsonian drugs and additional medications for non-motor symptoms and comorbidities. In addition to the risk of non-adherence that directly correlates with the number of prescriptions and daily doses per prescription, many patients with PD experience depression and/or cognitive impairment, both of which are strong independent risk factors for medication non-adherence [86].

Medication non-adherence among patients with PD should be recognized as a common, under-reported, detrimental, and costly cause of suboptimal disease control. Reliance on clinical judgment to identify non-adherence is demonstrably inaccurate, and healthcare professionals should use nonjudgmental interviewing skills that encourage patient admission of their non-adherence without fear or concerns of disapproval or termination of care. Barriers to adherence should be explored and clinical resources applied to surmount them. These include simple explanations of how medications optimally work when taken correctly and referral to non-adherence counseling. To avoid unnecessary dose escalations, adverse effects, and increased patient and healthcare costs, non-adherence should be explored before a drug regimen is deemed ineffective [86].

Exogenous dopamine administration is ineffective for treatment of PD, because circulating dopamine does not cross the blood-brain barrier so as to reverse brain dopamine depletion. Levodopa is a dopamine prodrug able to cross the blood-brain barrier where it is converted to dopamine by aromatic amino acid decarboxylase (AAAD). The regular administration of oral levodopa leads to repletion of dopamine in the substantia nigra pars compacta, and to storage in presynaptic dopamine neurons for subsequent use. The majority of patients treated with levodopa realize significant and prolonged improvement in motor function, though there are side effects and, in time, many patients experience fluctuations in beneficial effects of the drug. Levodopa was introduced for use in PD in the late 1960s, and remains the criterion-standard treatment [7].

The bioavailability of orally administered levodopa is reduced by extensive metabolism to dopamine in the gut. Only 30% of an oral dose reaches systemic circulation for distribution to the brain. For this reason, Levodopa used to treat PD is always combined with carbidopa, a peripherally acting AAAD inhibitor. Carbidopa inhibits peripheral conversion of levodopa to dopamine, which triples levodopa bioavailability and lowers the dosage requirements. Carbidopa 75–100 mg/day is the dose needed to inhibit peripheral conversion of levodopa to dopamine. Carbidopa also helps to diminish acute peripheral dopamine side effects, such as nausea, vomiting, and hypotension, and improves tolerability [88].

The risk for side effects and toxicity, including troublesome dyskinesia, is high in patients on chronic levodopa therapy. For this reason, careful dose titration and tight adherence to the effective dose is important for PD symptom management. No evidence has been found that using an extended-release levodopa/carbidopa formulation, or adding a catechol-O-methyltransferase (COMT) inhibitor, delays or prevents the development of motor fluctuations [89].

Because levodopa is absorbed in the proximal small intestine, food may delay absorption. Levodopa also competes with dietary proteins for transport into the brain. High-protein meals should be kept separate from levodopa dosing, and daily dietary protein intake should be reduced to approximately 0.8 g/kg (of body weight). Levodopa is metabolized in the gastrointestinal tract, kidneys, and liver, with 70% excreted in the urine. Levodopa half-life is roughly one hour. Dosing should be reduced 10% to 30% when other dopaminergic agents are added to carbidopa/levodopa. Available formulations in the United States are [88]:

COMT is an enzyme that converts levodopa in peripheral circulation to 3-O-methyl-DOPA (3-OMD). This metabolite cannot be converted to dopamine and accumulates in plasma during levodopa therapy. Inhibition of COMT increases the bioavailability of levodopa, allowing a larger amount of the drug to reach the brain and consequently raise dopamine levels. COMT inhibitors are always taken in combination with levodopa because they lack intrinsic dopaminergic activity. They are used in PD to potentiate levodopa effects when "wearing off" or other motor complications appear during carbidopa/levodopa therapy [93].

Amantadine is the only agent demonstrated to suppress levodopa-induced dyskinesia without worsening parkinsonism, and the American Academy of Family Physicians recommends that amantadine should be considered for treatment of dyskinesias in patients with advanced PD [2]. However, use in frail elderly patients with advanced PD may result in confusion, hallucinations, and/or worsening motor symptoms [95].

The acetylcholinesterase inhibitor rivastigmine is the only drug approved by the U.S. Food and Drug Administration (FDA) for the treatment of mild-to-moderate PD dementia. Other approved drugs for dementia, including donepezil, galantamine, and memantine, have been evaluated for the treatment of PD dementia, but their efficacy has not been clearly shown [95].

The initial choice of drug depends on the likelihood of improving motor function (better with levodopa) compared with the risk of motor complications (more common in younger patients, delayed by agonists) and the presence of neuropsychiatric complications (more common in older and cognitively impaired patients, greater with agonists). Levodopa is the mainstay of initial treatment and the most effective drug for improving motor function. One should avoid controlled-release formulations or adding entacapone, as this is not effective for delaying the onset of motor complications. Other treatments include MAO-B inhibitors (e.g., selegiline, rasagiline) or oral or transdermal dopamine agonists (e.g., pramipexole, ropinirole, rotigotine). Initial treatment with an agonist can be recommended in younger patients. Ergot derivatives (e.g., bromocriptine, cabergoline) are not recommended due to the increased risk of fibrotic development [74]. Amantadine and anticholinergic agents are also options. Rehabilitation in early-stage disease has seldom been evaluated, and therefore a recommendation for or against its use cannot be made [74].

For treatment of PD, deep brain stimulation targets the subthalamic nucleus of the internal capsule or the globus pallidus. Clinical trials of deep brain stimulation have reported a 40% to 60% reduction in the severity of motor symptoms and up to 50% reduction in medication use [159]. Short- and long-term studies have been conducted to assess the effectiveness of subthalamic nucleus stimulation for levodopa-refractory signs and symptoms. The overall improvement in activities of daily living and motor UPDRS scores averaged 50% compared to pre-surgery. Severity of levodopa-induced dyskinesias have been reduced by an average 69%. Surgical implantation of electrodes deep in the brain has a 1% to 6% risk of postoperative intracranial hemorrhage, infection, or stroke. Late-onset adverse events include migration or misplacement of the leads (5.1%), lead fractures (5%), and skin erosion (1.3%) [115,159].

For treatment of PD, deep brain stimulation targets the subthalamic nucleus of the internal capsule or the globus pallidus. Clinical trials of deep brain stimulation have reported a 40% to 60% reduction in the severity of motor symptoms and up to 50% reduction in medication use [159]. Short- and long-term studies have been conducted to assess the effectiveness of subthalamic nucleus stimulation for levodopa-refractory signs and symptoms. The overall improvement in activities of daily living and motor UPDRS scores averaged 50% compared to pre-surgery. Severity of levodopa-induced dyskinesias have been reduced by an average 69%. Surgical implantation of electrodes deep in the brain has a 1% to 6% risk of postoperative intracranial hemorrhage, infection, or stroke. Late-onset adverse events include migration or misplacement of the leads (5.1%), lead fractures (5%), and skin erosion (1.3%) [115,159].

While deep brain stimulation was formerly offered only in late-phase disease (mean: 13 to 14 years post-diagnosis), several considerations have now moved the timing of surgery earlier [117]. Deep brain stimulation produces improvement in symptoms responsive to dopaminergic drugs, but in late-stage disease, symptom responsiveness to brain stimulation is less predictable and often unsatisfactory. Performing deep brain stimulation at advanced stages of illness can alleviate some motor dysfunction features but does not much benefit ongoing sense of well-being or functional status in relation to family, occupation, and social roles. In addition, older patients are more likely to develop surgical complications and/or worsening of axial motor functions.

Psychosis in PD is common and multifactorial in etiology. Up to 60% of patients with PD develop psychosis. Following its onset, PD psychosis remains a persistent, lifelong problem for most patients [118]. Pharmacologic management is challenging, in part because dopaminergic agents required for motor control can exacerbate psychotic symptoms, and antipsychotic agents can exacerbate motor symptoms [119]. The onset of psychosis in PD predicts a poor prognosis, including increased likelihood of nursing home placement and early mortality [120].

The early clinical manifestations of PD-associated psychosis differ from other psychotic disorders in that hallucinations are common and patients initially remain lucid and connected with reality. Visual hallucinations are the most prevalent form. Functional MRI performed on patients with PD who are experiencing visual hallucinations show several abnormalities: altered cortical visual processing; decreased occipital response and increased caudate and frontal cortical activation to visual stimuli; overactive visual association cortex; and decreased primary visual cortex activity [119].

Auditory, tactile, olfactory, and gustatory hallucinations do occur, though less commonly and usually in combination with visual hallucinations. Confusion states, delusions, paranoia, agitation, and delirium may also develop.

The stage of PD at which psychotic features emerge has some diagnostic import. In newly suspected or recently diagnosed (within three months) cases of PD, the appearance of psychotic symptoms suggests early-onset dementia with Lewy bodies, but could also indicate an alternative neuropsychiatric diagnosis, such as Alzheimer disease with extrapyramidal symptoms or underlying functional (psychiatric) psychosis. Differences in the initial presentation of PD-associated psychosis do not substantively change the management approach (with some caveats) [119].

Risk factors for PD-associated psychosis include cognitive impairment, dementia, advanced age, sleep disturbances, and disease duration/severity [121]. Psychosis is unrelated to total dose or duration of dopaminergic medication, and no differences have been found in the incidence rate among patients receiving levodopa versus those on dopamine agonists or anticholinergic drugs [122].

The association between sleep disturbance and PD psychosis is sufficiently robust to suggest REM sleep behavior disorder manifests from an evolving synucleinopathy in patients with PD-associated psychosis or dementia. Both factors may develop from a single epiphenomenon, such as neurodegeneration. Evidence also suggests contribution to PD psychosis from non-dopaminergic neurotransmitters, including serotoninergic or cholinergic systems [119].

Visual hallucinations require medication adjustment and possibly specific therapies if they are troublesome, threatening, or associated with behavioral change [4]. Triggering factors, such as infection, metabolic disorders, fluid/electrolyte imbalance, and sleep disorder, should be controlled. In addition, steps should be taken to reduce polypharmacy. Tricyclic antidepressants and anxiolytics/sedatives should be reduced or stopped. Antiparkinsonism drugs should also be reassessed. Anticholinergics and amantadine should be halted, while dopamine agonists and MAO-B and COMT inhibitors should be reduced or halted. The levodopa dose may be reduced [74,123].

Unfortunately, most commonly used antipsychotic drugs have side effects that exacerbate PD. Consequently, atypical antipsychotics are often key in the management of PD-associated psychosis. Almost all antipsychotic drugs can exacerbate PD. Clozapine is the only antipsychotic with high-level evidence of efficacy; in some patients, it also improves motor function [124]. Clozapine is widely recommended as the first-line choice, but it is associated with potentially fatal agranulocytosis, which develops in 1% of patients and makes routine blood neutrophil counts mandatory. Less serious side effects include sedation, tachycardia, orthostatic hypotension, and sialorrhea. Low-dose clozapine (less than 50 mg) also has efficacy, with less frequent and more tolerable side effects and rare agranulocytosis [125,126].

Impulse control disorders and aberrant behaviors can develop during dopamine agonist treatment in PD and worsen patient and caregiver quality of life. Often, patients lack insight into the negative consequences of their behavior. Risk factors include male sex, younger age at onset, personality traits of high impulsivity and novelty seeking, and personal or family history of addictive disorders. In predisposed patients, overstimulation of mesocorticolimbic dopamine receptors by dopamine agonists leads to impulse control disorders and compulsive medication use. Impulse control disorders are more likely in early PD with normal-range medication dosing, while compulsive medication use is more commonly associated with fluctuations in advanced disease. Affected patients often lack noteworthy psychiatric histories and cognitive impairment, making identification difficult. Management requires reducing dopaminergic therapy, and psychosocial support is often necessary. SSRIs may help, while atypical antipsychotics have limited benefit. Prevention is based on the identification of at-risk individuals and active monitoring [135].

A concern in using opioids to treat pain in patients with PD is potential exacerbation of constipation, a common, burdensome symptom of autonomic dysfunction. To possibly mitigate this issue, an oral formulation combining prolonged-release oxycodone with naloxone has been evaluated. In an eight-week trial, patients with PD-associated chronic pain received low-dose oxycodone/naloxone (5 mg/2.5 mg) twice daily. Of the 87.5% who completed the trial, significant pain reduction was achieved, no adjustment of dopaminergic therapy was required, no significant changes were observed in bowel function and constipation symptoms, no changes were observed in sleep symptoms, and improvements were recorded in clinician impression of therapeutic effect [141].

With disease progression, patients with PD become more reliant on medication to maintain their ability to function. In addition to regular monitoring for drug-specific side effects, clinicians should be careful not to abruptly withdraw dopaminergic medication [89]. Patients and family should be educated on the importance of medication compliance and regular dosing so as to avoid rapid changes in efficacy. Special attention is required during periods of intercurrent illness, such as gastroenteritis or abdominal surgery, which may result in interruption of dosage or poor intestinal absorption. These measures help to avoid the potential development of acute akinesia or neuroleptic malignant syndrome. "Drug holidays" are not recommended due to the risk of developing neuroleptic malignant syndrome.

Considering the risks of sudden changes in dopaminergic medication, patients with PD admitted to hospitals or care facilities should receive their medication at the appropriate times or be allowed self-medication. Medication adjustment should be reserved for specialists in PD management [89].