Parkinson's disease (PD) is classified as a movement disorder because it is diagnosed based on its cardinal motor features: tremor, rigidity, and bradykinesia. However, PD is also a complex neuropsychiatric condition with high rates of apathy, anxiety, cognitive impairment, depression, psychosis, and impulse control disorders. Increasingly effective treatments for motor symptoms have reduced disability due impaired movement and now neuropsychiatric symptoms have the greatest impact on quality of life, disability, caregiver burden, and risk of nursing home placement.1 Some of these seem to be intrinsic to the neurodegenerative pathology of PD, yet others are associated with the dopaminergic medications prescribed for motor symptoms. Impulse control disorders (ICDs) are the neuropsychiatric symptom most notably associated with dopamine receptor agonist use, being uncommon in its absence.2 This article describes the prevalence, potential mechanism, and clinical management of ICDs in PD.
ICDs are a group of behaviors in which a person repeatedly fails to resist an impulse, drive, or temptation to perform an act that may cause distress or harm to self or others. In PD, the most common ICDs are pathologic gambling, compulsive spending, hypersexuality, and binge eating. The combination of excessive extent, diminished ability to resist, and resultant “harm” in performing these acts is essential for classification as a “disorder,” as otherwise these are potentially normal behaviors when pursued more moderately. Although ICDs share many common traits, the underlying biology and relative contribution of reinforcement learning in perpetuating the behaviors may be different, and as a result, diagnostic categorization of ICDs has been difficult. For example, in the Diagnostic and Statistical Manual of Mental Disorders, fifth edition,3 pathological gambling has been moved to a new category, “substance-related and addictive disorders,” while the other ICDs remain in the “disruptive, impulse-control, and conduct disorders” category.4 To the extent these diagnostic categories are valid, they may have important implications for clinical management.
Akin to the underappreciation of non-motor symptoms before the introduction of levodopa treatment for the motor symptoms of PD, the association between dopamine agonist medications and ICDs went unrecognized for many years. In the 1970s, levodopa revolutionized the treatment of PD, restoring movement and function to patients who had been physically disabled for years. However, it was discovered that patients with longstanding PD eventually developed complications, such as dyskinesias and “wearing off” of levodopa benefit between doses. This led some to incorrectly suspect that levodopa was toxic to dopamine-producing cells, leading to the promotion of levodopa-sparing treatment strategies with dopamine receptor agonists that often resulted in suboptimal care or treatment delay.
Given the concerns about levodopa, there was great enthusiasm in the 1980s when an alternative to levodopa emerged, a class of medications known as dopamine receptor agonists. Dopamine receptor agonists directly stimulate the dopamine receptor, unlike levodopa, which is decarboxylated to form dopamine. Early agonists were ergot derivatives (bromocriptine and pergolide), which had adverse cardiovascular effects and have been replaced by non-ergot medications (pramipexole, ropinirole, and rotigotine). Dopamine agonists became a mainstay of treatment for PD motor symptoms, in part due to speculation that they may not have the same complications as levodopa. However, physicians soon found that excessive daytime somnolence, hallucinations, psychosis, and impulse control disorders were more common in patients treated with a dopamine receptor agonist than those treated with levodopa. Despite the decreased benefit and increased side effects (including ICDs) from dopamine agonists, at least one study suggests there has not been a significant reduction in the use of dopamine receptor agonists during the past decade.5
Dopamine replacement therapy, particularly dopamine agonists, appears to drive the association between ICDs and PD. Case-control studies with newly diagnosed patients with PD who are inexperienced with dopamine find no evidence for an increased prevalence of ICDs compared to healthy controls.6 However, in patients with PD who are receiving dopaminergic treatment for their motor symptoms, the prevalence of ICDs is nearly 14%.7 The prevalence of individual ICDs varies by type with one large study of over 3,000 people with PD reporting compulsive buying at 5.7%, pathologic gambling at 5%, binge-eating disorder at 4.3%, and compulsive sexual behavior at 3.5%, whereas 3.9% engaged in more than one of these behaviors.7,8
Although ICDs have been reported with most forms of dopamine therapy, they are more common in patients receiving treatment regimens including dopamine agonists. In the study, more than 17.1% of patients who were prescribed a dopamine agonist had an ICD compared with 6.9% of patients who were not (odds ratio = 2.7). One of the few longitudinal studies of ICDs showed a dose-effect with both duration of exposure and dose associated with increased risk.9 Patients receiving treatment with both dopamine agonists and levodopa seem to be at the highest risk.7,8 However, there have been reports of a similar prevalence of ICDs in patients treated with dopamine agonists for other conditions, such as restless leg syndrome and hyperprolactinemia,10 where these drugs are prescribed in lower doses than in PD.8 Although dopamine agonists play a central role in ICDs, additional risk factors and idiosyncratic vulnerabilities are likely to be important as well.
Predicting which patients will develop ICDs is not always possible, and avoiding the use of dopaminergic medications is not feasible in PD. However, in people with a preponderance of risk factors, dopamine agonists should be avoided. Risk factors for the development of ICDs in patients with PD include male sex, younger age, a personal or family history of ICDs or addictions, and premorbid novelty-seeking or impulsive traits.11 Sex differences also appear to modify the risk for developing particular ICDs; hypersexual behavior occurs more in men, whereas both compulsive buying and binge eating are more common in women. The effect of neuropsychiatric symptoms on ICD risk is mixed for cognitive factors such as executive function and reward-punishment learning but appears to be increased in the presence of depressive symptoms, irritability, and appetite changes.12 Based on whole exome sequencing data, more than one-half of the variance in ICD incidence in PD may be related to common genetic variants. Multiple gene interactions may contribute to ICD risk and pathogenesis.13
Although the overall causal mechanism for ICDs likely involves a complex interplay of several networks and neurotransmitter systems, the dopaminergic hypothesis has received the most attention. PD motor symptoms are caused by the loss of dopaminergic neurons in the substantia nigra pars compacta. However, the loss of dopamine neurons is not uniform across the pars compacta, as dopamine neuron loss is greatest in the ventral tier, leaving the dorsal tier less effected. The diminished ventral tier neurons project to the striatum (ie, caudate and putamen, producing motor impairment), whereas the relatively intact dorsal tier neurons innervate the forebrain and limbic regions facilitating responses important to risk versus reward-related behaviors. Thus, when dopaminergic therapy is initiated, motor symptoms caused by neuronal loss in the ventral tier improve the hypodopaminergic state in the striatum, whereas the relatively intact dorsal tier structures (ie, mesocortical and mesolimbic circuits) are at risk of over-stimulation. It is hypothesized that extra-physiologic levels of dopamine in limbic and forebrain regions increase the risk of ICDs.8
Why do dopamine agonists increase the risk of ICDs more than levodopa and other dopamine replacement strategies? The answer may be due to the location and density of different dopamine receptor subtypes within motor and limbic circuits. D1 (D1 and D5) and D2 (D2, D3, D4) receptor subtypes are both located in motor circuits; however, the D3 subtype of the D2 receptor is also highly expressed in limbic circuits. Activation of limbic D3 receptors may increase hedonic drive or pleasure-seeking behaviors in some people.8 Compared with the dopamine that is synthesized from levodopa, the dopamine receptor agonists have increased affinity for D3 receptors. Higher affinity for D3 receptors might also be associated with a higher incidence of hallucinations and delusions with agonists.
Less common causes of ICDs might include deep brain stimulation (DBS), and the possibility that antidepressant effects of dopamine agonists could trigger mania along with impulse control behaviors in patients with latent bipolar disorder. However, in the case of DBS, there is evidence that subthalamic deep brain stimulation might reduce the risk of ICDs, perhaps mediated by the reduction of dopaminergic therapy postoperatively. The possibility of dopamine agonists triggering a manic switch or ICD behaviors in some patients is interesting for several reasons. First, bipolar disorder has been identified as a risk factor for later development of PD.14,15 Second, antidepressants can induce a manic switch in bipolar patients. Dopamine agonists, like pramipexole, have been shown to have antidepressant effects, and there are case studies of dopamine agonist-induced mania.16,17 Finally, similar to bipolar disorder's manic-depressive paradox, ICDs and apathy (or the anhedonic component of depression) may represent opposite poles on a motivational spectrum associated with high and low dopamine states respectively.18
Challenges in Detection and Diagnosis
Unlike other more conspicuous psychiatric side effects such as hallucinations and delusions, ICDs often elude detection.8 Patients often lack insight that the ICD represents a problem as they may be within the normal repertoire of human behaviors except for their unusual extent or frequency. Even when aware that their behavior is abnormal or excessive, patients often engage in active deception to conceal gambling losses, extramarital affairs, or other illicit activities. When confronted by physicians or family members, patients often minimize the extent of their behaviors due to shame or embarrassment. The behaviors in ICDs are often uncontrollable. Patients and health care providers are accustomed to inquiring about “typical” medication side effects such as nausea, lightheadedness, or rash but less likely to appreciate the role of medications in altering behavior. Consequently, the uncontrollable behavioral impulses triggered by dopamine receptor agonists often remain unrecognized, or do not become apparent until the patient develops serious financial, legal, or social problems.19
When interviewing patients being treated with dopamine agonists, ask directly about ICDs and related behaviors (Table 1). Soliciting collateral information from spouses or caregivers is recommended.2 A final challenge in identifying dopamine agonist-induced ICDs is that they often emerge insidiously, sometimes years after the offending medication is prescribed.8 Thus, before prescribing dopamine receptor agonists, physicians should warn patients and their families or caregivers of the potential for triggering uncontrollable or excessive gambling, sexual interests, spending, or other behavioral addictions. After initiating treatment, physicians should regularly ask patients and their families about any changes in behavior that might indicate development of an ICD, including the presence of ICD-related disorders, described below, which are often comorbid.
Impulse Control and Related Disorders in Parkinson's Disease
The clinical spectrum of ICDs is thought to include related behaviors such as punding (repetitve, purposeless behaviors such as arranging or taking apart objects), hobbyism (compulsive exercise, Internet use, or work on projects), walkabouts (aimless wandering), and hoarding.8,20 Another potentially related phenomenon seen in patients treated for PD is dopamine dysregulation syndrome (DDS). This describes the pathologic overuse of dopaminergic medications, often taking much more than is required for relief of movement symptoms. Like other chemical addictions, this behavior can be motivated both by a desire for an intrinsic reward associated with these drugs as well as a desire to avoid physical and psychological distress associated with withdrawal from them.8 Excessive use of medications that increase dopamine appears to worsen other impulsive behaviors and may also contribute to mania and psychosis.21
To help overcome the challenges to detection mentioned above, numerous screening measures for identifying ICDs in PD have been developed. Although a detailed psychiatric evaluation remains the gold standard for diagnosis, a fast, reliable, and sensitive assessment tool will facilitate efficient screening for ICDs, especially for providers who are not psychiatrists. Although more than 50 scales for the evaluation of ICDs exist, the Questionnaire for Impulsive-Compulsive Disorders in Parkinson's Disease (QUIP) is the most commonly used because it is validated, available in self- or rater-administered versions, and recommended for use in PD by the International Movement Disorder Society.22,23 The QUIP takes about 5 minutes to administer and covers the four major categories of ICD (buying, eating, gambling, sex) as well as the potentially related behaviors of compulsive medication use (DDS), punding, hobbyism, and walkabout. Screening for ICDs over the duration of dopamine agonist treatment is strongly recommended, especially given the high prevalence and opportunity for harm reduction when successfully identified.
Once diagnosed, the key to managing ICDs in patients with PD is close coordination between the patient's neurologist and psychiatrist. If the patient is prescribed a dopamine agonist, the first step is usually to discontinue this medication and replace it with levodopa.24,25 However, abrupt discontinuation can lead to dopamine agonist withdrawal syndrome (DAWS) in up to 19% of patients.26 In addition to potentially worsening motor symptoms, DAWS often includes depression, diaphoresis, agitation, fatigue, pain, orthostatic hypotension, panic attacks, and drug cravings that are refractory to levodopa supplementation. Therefore, dopamine agonists must be tapered carefully. Even when higher doses of levodopa are required to control motor symptoms, impulsive behavior may improve after reduction or removal of the dopamine agonist.8
Case studies detail the use of antipsychotics and other mood-stabilizing medications to manage ICDs in PD; however, a determination of efficacy was confounded by concurrent reduction or elimination of dopamine agonists. There was enthusiasm for using naltrexone for ICDs in PD based on findings in the general population. However, a 2014 randomized control trial did not show clear benefit.27 Amantadine showed benefit for treating pathological gambling in PD in a small placebo-controlled trial,28 but was also found to be associated with having ICDs in a large multi-center study. In 2019, a review conducted by the Movement Disorder Society Evidence-Based Medicine Committee concluded that there was insufficient evidence of efficacy for any medication for the treatment of ICDs in PD.29 The same review found that cognitive-behavioral therapy was likely efficacious, but cautioned that this conclusion was based on only one low-quality study with insufficient evidence on safety.30 Despite the lack of evidence-based treatment for ICDs in PD, early diagnosis remains important, as the consequences can be mitigated by involving the caregiver and family in direct behavioral interventions (eg, access to financial accounts and credit cards can be restricted). Further, ICDs often co-occur with other psychiatric syndromes such as mood and anxiety disorders, which once identified and managed may reduce the overall neuropsychiatric symptom burden.
Stopping dopamine agonists is sometimes not tolerated and many patients have ICDs that persist or are self-sustaining, even in their absence. Management of ICDs in patients with PD often requires reduction of dopamine to the least amount necessary to address motor symptoms, while using behavioral interventions to prevent or mitigate impulsivity and bad outcomes. Often, maladaptive behavioral patterns reinforce themselves intrinsically so that removing triggering medications such as dopamine agonists is no longer sufficient to stop the behaviors once established. Although it is not well established in the literature, the various ICDs may require different treatments. For instance, as mentioned above, pathological gambling was recategorized as a substance-related and addictive disorder; therefore, an addiction treatment paradigm including behavioral therapy and 12-step groups with the goal of promoting abstinence from these behaviors should be considered.8
ICDs are associated with dopaminergic medications used for alleviating parkinsonian motor symptoms. Although the mechanism underlying ICDs in PD is uncertain, increased stimulation of dopamine receptors in limbic and cortical regions likely leads to disinhibition, impulsivity, and a miscalculation of risk-reward weighting. ICDs often have horrific consequences for patients and their families. Yet they can be insidious and difficult to identify, often recognized only after major consequences are suffered. Most patients with PD can be managed with levodopa monotherapy. Therefore, dopamine agonists should not be prescribed without first educating patients, families, and caregivers about the risks of ICDs. Once diagnosed, best management requires a multidisciplinary approach with close coordination between psychiatry and neurology. Elimination of dopamine agonists or reduction to the extent possible while simultaneously intervening—often with the help of family or caregivers—to prevent or limit these disruptive and dysfunctional behaviors, appears to be the best strategy until new evidence-based treatments are identified.
- Balestrino R, Martinez-Martin P. Neuropsychiatric symptoms, behavioural disorders, and quality of life in Parkinson's disease. J Neurol Sci. 2017;373:173–178. doi:10.1016/j.jns.2016.12.060. [CrossRef]
- Weiss HD, Marsh L. Impulse control disorders and compulsive behaviors associated with dopaminergic therapies in Parkinson disease. Neurol Clin Pract. 2012;2(4):267–274. doi:10.1212/CPJ.0b013e318278be9b [CrossRef] PMID:23634371
- American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 5th ed. Arlington, VA: American Psychiatric Publishing; 2013.
- Potenza MN. Non-substance addictive behaviors in the context of DSM-5. Addict Behav. 2014;39(1):1–2. doi:10.1016/j.addbeh.2013.09.004 [CrossRef]. PMID:24119712
- Dubaz OM, Wu S, Cubillos F, Miao G, Simuni TParkinson's Foundation Quality Improvement Initiative Investigators. Changes in prescribing practices of dopaminergic medications in individuals with Parkinson's disease by expert care centers from 2010 to 2017: the Parkinson's foundation quality improvement initiative. Mov Disord Clin Pract. 2019;6(8):687–692. doi:10.1002/mdc3.12837 [CrossRef]. PMID:31745479
- Weintraub D, Papay K, Siderowf AParkinson's Progression Markers Initiative. Screening for impulse control symptoms in patients with de novo Parkinson disease: a case-control study. Neurology. 2013;80(2):176–180. doi:10.1212/WNL.0b013e31827b915c [CrossRef]. PMID:23296128
- Weintraub D, Koester J, Potenza MN, et al. Impulse control disorders in Parkinson disease: a cross-sectional study of 3090 patients. Arch Neurol. 2010;67(5):589–595. doi:10.1001/archneurol.2010.65 [CrossRef]. PMID:20457959
- Taylor J, Pontone GM. Impulse control disorders in Parkinson's disease. Invited article. Psychiatr Times. 2015;32(8).
- Corvol JC, Artaud F, Cormier-Dequaire F, et al. DIGPD Study Group. Longitudinal analysis of impulse control disorders in Parkinson disease. Neurology. 2018;91(3):e189–e201. doi:10.1212/WNL.0000000000005816 [CrossRef]. PMID:29925549
- De Sousa SMC, Baranoff J, Rushworth RL, Butler J, Sorbello J, Vorster J, Thompson T, McCormack AI, Inder WJ, Torpy DJ. Impulse control disorders in dopamine agonist-treated hyperprolactinemia: prevalence and risk factors. J Clin Endocrinol Metab. 2020;105(3):dgz076. doi:10.1210/clinem/dgz076 [CrossRef].
- Liu B, Luo W, Mo Y, Wei C, Tao R, Han M. Meta-analysis of related factors of impulse control disorders in patients with Parkinson's disease. Neurosci Lett. 2019;707:134313. doi:10.1016/j.neulet.2019.134313 [CrossRef]
- Pontone G, Williams JR, Bassett SS, Marsh L. Clinical features associated with impulse control disorders in Parkinson disease. Neurology. 2006;67(7):1258–1261. doi:10.1212/01.wnl.0000238401.76928.45 [CrossRef]
- Kraemmer J, Smith K, Weintraub D, et al. Clinical-genetic model predicts incident impulse control disorders in Parkinson's disease. J Neurol Neurosurg Psychiatry. 2016;87(10):1106–1111. doi:10.1136/jnnp-2015-312848 [CrossRef]. PMID:27076492
- Huang MH, Cheng CM, Huang KL, et al. Bipolar disorder and risk of Parkinson disease: a nationwide longitudinal study. Neurology. 2019;92(24):e2735–e2742. doi:10.1212/WNL.0000000000007649 [CrossRef]. PMID:31118242
- Pontone GM, Koch G. An association between bipolar disorder and Parkinson disease: when mood makes you move. Neurology. 2019;92(24):1125–1126. doi:10.1212/WNL.0000000000007641 [CrossRef]. PMID:31118246
- Barone P, Poewe W, Albrecht S, et al. Pramipexole for the treatment of depressive symptoms in patients with Parkinson's disease: a randomised, double-blind, placebo-controlled trial. Lancet Neurol. 2010;9(6):573–580. doi:10.1016/S1474-4422(10)70106-X [CrossRef]. PMID:20452823
- Taylor J, Anderson WS, Brandt J, Mari Z, Pontone GM. Neuropsychiatric complications of Parkinson disease treatments: Importance of multidisciplinary care. Am J Geriatr Psychiatry. 2016;24(12):1171–1180. doi:10.1016/j.jagp.2016.08.017 [CrossRef].
- Leroi I, Andrews M, McDonald K, et al. Apathy and impulse control disorders in Parkinson's disease: a direct comparison. Parkinsonism Relat Disord. 2012;18(2):198–203. doi:10.1016/j.parkreldis.2011.10.005 [CrossRef]. PMID:22035735
- Weiss HD, Pontone GM. Dopamine receptor agonist drugs and impulse control disorders. JAMA Intern Med. 2014;174(12):1935–1937. doi:10.1001/jamainternmed.2014.4097 [CrossRef]. PMID:25329734
- Weintraub D, David AS, Evans AH, Grant JE, Stacy M. Clinical spectrum of impulse control disorders in Parkinson's disease. Mov Disord. 2015;30(2):121–127. doi:10.1002/mds.26016 [CrossRef]. PMID:25370355
- O'Sullivan SS, Evans AH, Lees AJ. Dopamine dysregulation syndrome: an overview of its epidemiology, mechanisms and management. CNS Drugs. 2009;23(2):157–170. doi:10.2165/00023210-200923020-00005 [CrossRef]. PMID:19173374
- Weintraub D, Hoops S, Shea JA, et al. Validation of the questionnaire for impulsive-compulsive disorders in Parkinson's disease. Mov Disord. 2009;24(10):1461–1467. doi:10.1002/mds.22571 [CrossRef]. PMID:19452562
- Evans AH, Okai D, Weintraub D, et al. Members of the International Parkinson and Movement Disorder Society (IPMDS) Rating Scales Review Committee. Scales to assess impulsive and compulsive behaviors in Parkinson's disease: critique and recommendations. Mov Disord. 2019;34(6):791–798. doi:10.1002/mds.27689 [CrossRef]. PMID:31136681
- Lee JY, Jeon B, Koh SB, et al. REIN-PD Investigators. Behavioural and trait changes in parkinsonian patients with impulse control disorder after switching from dopamine agonist to levodopa therapy: results of REIN-PD trial. J Neurol Neurosurg Psychiatry. 2019;90(1):30–37. doi:10.1136/jnnp-2018-318942 [CrossRef]. PMID:30361296
- Mamikonyan E, Siderowf AD, Duda JE, et al. Long-term follow-up of impulse control disorders in Parkinson's disease. Mov Disord. 2008;23(1):75–80. doi:10.1002/mds.21770 [CrossRef]. PMID:17960796
- Rabinak CA, Nirenberg MJ. Dopamine agonist withdrawal syndrome in Parkinson disease. Arch Neurol. 2010;67(1):58–63. doi:10.1001/archneurol.2009.294 [CrossRef]. PMID:20065130
- Papay K, Xie SX, Stern M, et al. Naltrexone for impulse control disorders in Parkinson disease: a placebo-controlled study. Neurology. 2014;83(9):826–833. doi:10.1212/WNL.0000000000000729 [CrossRef]. PMID:25037206
- Thomas A, Bonanni L, Gambi F, Di Iorio A, Onofrj M. Pathological gambling in Parkinson disease is reduced by amantadine. Ann Neurol. 2010;68(3):400–404. doi:10.1002/ana.22029 [CrossRef]. PMID:20687121
- Seppi K, Ray Chaudhuri K, Coelho M, et al. the collaborators of the Parkinson's Disease Update on Non-Motor Symptoms Study Group on behalf of the Movement Disorders Society Evidence-Based Medicine Committee. Update on treatments for nonmotor symptoms of Parkinson's disease-an evidence-based medicine review. Mov Disord. 2019;34(2):180–198. doi:10.1002/mds.27602 [CrossRef]. PMID:30653247
- Okai D, Askey-Jones S, Samuel M, et al. Trial of CBT for impulse control behaviors affecting Parkinson patients and their caregivers. Neurology. 2013;80(9):792–799. doi:10.1212/WNL.0b013e3182840678 [CrossRef]. PMID:23325911
Impulse Control and Related Disorders in Parkinson's Disease
|Impulse control disorders in Parkinson's disease
Hoarding (excessive accumulation of and inability to discard objects)
Hobbyism (compulsive exercise, Internet use, or work on projects)
Punding (repetitive, purposeless behaviors such as arranging or taking apart objects)
Walkabouts (aimless wandering)
Dopamine dysregulation syndrome (pathologic overuse of dopamine medications)