- Email: [email protected]
Dopamine agonist withdrawal syndrome (DAWS) in a tertiary Parkinson disease treatment center
Journal of the Neurological Sciences 379 (2017) 308–311
Contents lists available at ScienceDirect
Journal of the Neurological Sciences journal homepage: www.elsevier.com/locate/jns
Dopamine agonist withdrawal syndrome (DAWS) in a tertiary Parkinson disease treatment center Shnehal Patel ⁎, Xiomara Garcia, Mohammad Edrees Mohammad, Xin Xin Yu, Katelyn Vlastaris, Kaley O'Donnell, Kathryn Sutton, Hubert H. Fernandez Center of Neurologic Restoration, Cleveland Clinic, 9500 Euclid Ave/U2, Cleveland, OH 44195, United States
a r t i c l e
i n f o
Article history: Received 20 January 2017 Received in revised form 24 April 2017 Accepted 15 June 2017 Available online 16 June 2017 Keywords: Dopamine agonist withdrawal syndrome Dopamine agonist Parkinson disease
a b s t r a c t Introduction: Dopamine agonists are a mainstay of treatment for patients with Parkinson disease (PD). However, side effects limit their use, often necessitating dose change. Upon withdrawal, patients may experience dopamine agonist withdrawal syndrome (DAWS). To date, there is no established protocol for the prevention or treatment of DAWS. Methods: We performed a retrospective chart review of PD patients who were taking a dopamine agonist. Results: In our large cohort of 313 PD patients who were on a dopamine agonist, we found that 39.5% (n = 124) had a change in their dose of medication for various reasons, including 102 patients who experienced a side effect on a dopamine agonist. Twenty out of 102 patients (19.6%) developed symptoms consistent with DAWS, whereas 1 out of 22 patients (4.5%) who had medication dose changed due to any other reason (e.g. dyskinesias, DBS surgery, decreased by another provider, etc.) developed symptoms consistent with DAWS. Our DAWS population had a shorter duration of PD, less exposure to a dopamine agonist, and was on a lower dose compared to those patients who did not develop DAWS. Agitation was the most common DAWS symptom reported in our cohort. Interestingly, in terms of developing DAWS, the prevalence of DAWS (19.0% vs 16.5%; p = 0.76) between partial versus total discontinuation was not significantly different whether the dopamine agonist dose was decreased (21 patients) or completely stopped (103 patients). Conclusion: Contrary to previous reports, we have found that other side effects besides impulse control behavioral disorders also increase risk for developing DAWS. Furthermore, the prevalence of DAWS did not differ between partial versus total discontinuation of the dopamine agonist. © 2017 Elsevier B.V. All rights reserved.
1. Introduction Parkinson disease (PD) is a chronic neurodegenerative condition affecting 7–10 million people in the world. Without a current cure, the mainstay of symptomatic treatment remains dopamine replacement, including dopamine agonists. Although consistently providing meaningful benefit in controlling motor symptoms in PD, patients on dopamine agonists can experience significant side effects which can become intolerable. These side effects include cognitive changes, delusions and hallucinations, excessive daytime sleepiness, “sleep attacks”, leg swelling, weight gain, and impulse control behavioral disorders (ICD), among others. Abbreviations: PD, Parkinson Disease; DAWS, dopamine agonist withdrawal syndrome; ICD, impulse control behavioral disorders; LEDD, levodopa equivalent daily dose; KP, Knowledge Program; HSM, health status measures; PHQ9, Patient Health Questionairre-9; GAD7, Generalized Anxiety Disorder 7; UPDRSII, Unified Parkinson Disease Rating Scale II. ⁎ Corresponding author. E-mail address: [email protected] (S. Patel).
http://dx.doi.org/10.1016/j.jns.2017.06.022 0022-510X/© 2017 Elsevier B.V. All rights reserved.
Present management of side effects requires tapering or withdrawal of the offending dopamine agonist. However, in our clinical practice, this is often easier said than done. PD patients may experience worsening of their motor symptoms and occasionally they have been described to develop signs and symptoms similar to those seen in patients experiencing psychostimulant withdrawal. This recently recognized phenomenon has been termed dopamine agonist withdrawal syndrome (DAWS) [1–3] DAWS is a stereotyped, often severe, cluster of physical and behavioral symptoms occurring with dopamine agonist withdrawal, including panic attacks, depression, diaphoresis, agitation, fatigue, pain, orthostatic hypotension, and drug craving, that are refractory to levodopa supplementation [1–3]. DAWS has been reported in up to 19% of patients undergoing a dopamine agonist taper, with a range of 15 to 19%. Risk factors for DAWS that have been reported include high cumulative dopamine agonist exposure, high baseline dopamine agonist dose, and prior diagnosis of ICD [1–3]. Unfortunately, DAWS can be very challenging to treat. It is typically not responsive to levodopa supplementation, antidepressants and other behavioral
S. Patel et al. / Journal of the Neurological Sciences 379 (2017) 308–311
treatments have been minimally effective. The only presumed definitive treatment has been to reinitiate the dopamine agonist [1–3]. In some patients DAWS is self-limiting with resolution within weeks, allowing complete withdrawal of the dopamine agonist. In other patients it can run over a protracted course of months, preventing complete withdrawal of the dopamine agonist, which in turn causing significant unwanted consequences such as a state of chronic ICD leading to litigation, obesity, financial losses, and social and occupational consequences [1–5]. Unfortunately, our cumulative experience with DAWS is limited by small sample size and short observation period. We aim to report the naturalistic phenomenology and prevalence of DAWS as captured in the electronic recorders at our tertiary movement disorders center, with over 1500 PD patients treated annually.
309
3. Statistics Descriptive statistics was used to describe our cohort. Using the SPSS software, Mann-Whitney was used to compare non-normative data such as demographic data between two different populations (e.g. continued dopamine agonist versus dose reduced or discontinued). MannWhitney was also used to compare demographic data of patients with DAWS compared to those without DAWS. Chi-square was used when comparing categorical variables in the population groups (e.g. continued versus decreased versus discontinued dopamine agonist). Chisquare was also used to compare incidence of DAWS between populations based on the type of side effect experienced. Significance was defined as p b 0.05. 4. Results
2. Methods We performed a retrospective chart review of patients diagnosed with PD, by a fellowship-trained movement disorders neurologist, who fulfill the United Kingdom Parkinson's Disease Society Brain Bank criteria, over a 2-year period, from January 2011 to December of 2012, at the Center of Neurological Restoration of the Cleveland Clinic, Cleveland, Ohio. Patients who were not taking a dopamine agonist were excluded from further analysis. Data collected included: gender, race, age at the time of dopamine agonist withdrawal, name of the dopamine agonist, and the duration of PD as well as dopamine agonist use. We then carefully reviewed all notes (in-patient visits, electronic and telephone correspondences) to determine if patient had experienced any symptom that can be considered a side effect of dopamine replacement therapy. Side effects were categorized as follows: (1) ICD–hypersexuality, pathological gambling, binge eating, compulsive shopping, punding; (2) Idiosyncratic dopamine agonist side effects–leg swelling, weight gain, sleep attacks, skin reactions; (3) General dopamine replacement therapy side effects–nausea/ vomiting, orthostasis, cognitive worsening, psychosis, sedation. For all patients, we recorded the total number of PD medications, the daily dose of total dopamine replacement therapy, and also dopamine agonist therapy using published levodopa equivalent daily dose (LEDD) conversion ratios [6]. We further analyzed if patients were continued on the dopamine agonists, or if the dose was tapered or completely stopped. The speed of taper (i.e. b2 weeks, 2–4 weeks, and N4 weeks) was also recorded. Finally, we reviewed if any of these patients experienced symptoms consistent with DAWS including: panic attacks, depression, diaphoresis, agitation, fatigue, pain, drug cravings, or orthostatic hypotension [1–2]. We also collected other demographic variables including education level, job status, marital status, smoking and substance abuse history. We obtained data through EPIC, our electronic medical record system, and our Knowledge Program (KP). KP is a data capture initiative designed to harness routinely collected clinical information to optimize patient care and use of electronic medical record. Patient reported health status measures (HSM) are collected at each patient visit in electronic tablet, patient kiosk, or from patient's home through patients' electronic access (MyChart). These results, along with data from existing clinical systems, are then consolidated into a single data repository, the KP database. The KP database was able to provide further information regarding patient's depression, anxiety, and the activities of daily living measurements using Patient Health Questionairre-9 (PHQ9), Generalized Anxiety Disorder 7 (GAD7), and the Unified Parkinson Disease Rating Scale II (UPDRSII), respectively. We used values that were closest to the time when the patient was either taken off the dopamine agonist or their last visit for patients who were continued on the dopamine agonist. The study was performed in accordance with Cleveland Clinic Institutional Review Board.
Out of 1884 parkinsonian patients who were seen in Cleveland Clinic over the two-year period, 313 met diagnostic criteria for idiopathic PD and were taking a dopamine agonist. Our cohort had a mean: age (at time of analysis) of 65.1 years (SD 9.2); PD duration of 8.45 years (SD 6.23); duration of dopamine agonist use of 58.5 months (SD 170.7); total daily dopamine replacement dose of 770.5 mg LEDD (SD 429.8); and, daily dopamine agonist dosage of 191 mg LEDD (SD 116.8). Types of dopamine agonists used are listed in Table 1. Our cohort was then classified according to dopamine agonist dose modification (i.e. no change versus dose decrease/discontinuation). The demographical data of these two groups are also illustrated in Table 2. Education level, job status, marital status, smoking and substance abuse history were unable to be obtained in over half of the cohort, therefore these variables were not included. The duration of PD, duration of dopamine agonist use, and mean dose of dopamine agonist were greater in patients who had continued the dopamine agonist compared to those who had decreased or stopped the medication. Patients whose duration of PD or dopamine agonist use was not known were not included in the mean analysis. 4.1. Evaluation of DAWS based on dopamine agonist side effects One patient was lost to follow up and was not included. Out of the remaining 312 patients, 50% (n = 156) developed at least 1 side effect. Of these patients, 65% (n = 102) had the dopamine agonist either decreased or completely stopped, whereas 35% (n = 54) continued on the same dose. Of those who had dopamine agonist dose decreased or discontinued, 19.6% (n = 20) developed symptoms consistent with DAWS. Of the patients who did not experience any side effect, there were 14% (n = 22) who had the medication either decreased or completely stopped. The reasons for decrease were varied and included: patient receiving DBS; having dyskinesias; or medication was changed by another provider for unknown reason. Within this group, only 1 patient (4.5%) developed symptoms consistent with DAWS. This patient has his medication stopped during an inpatient admission for an irregular heart rate, which seemed unrelated to the dopamine agonist. 4.2. Side effects and decision to taper The 156 patients who developed side effects were then classified according to the types of adverse event experienced: 32 experienced ICD; Table 1 Type of dopamine agonist used. Dopamine agonists used
N (%)
Pramipexole Ropinirole Rotigotine
33 (21.15%) 99 (63.46%) 24 (15.38)%
310
S. Patel et al. / Journal of the Neurological Sciences 379 (2017) 308–311
Table 2 Demographics of population.
Mean age at time of withdrawal or end of study M:F Duration of PD (n = 309) Duration of dopamine agonist use (months) (n = 233) Mean LEDD use Mean dose of dopamine agonist (in LEDD) UPDRS part II Anxiety (GAD7) Depression
Total population (n = 313)
Continued dopamine agonist (n = 188)
Decreased dopamine agonist (n = 124)
p-Value
65.1 188:125 8.45 57.2 770.5 mg 191.2 mg 14.3 4.7 6.54
65.3 123:65 9.17 70.8 802.9 200.2 14.8 4.8 6.6
64.9 64:60 7.31 38.1 723.4 178.4 13.5 4.5 6.5
p = 0.715 p = 0.96 p = 0.005 p = 0.001 p = 0.074 p = 0.021 p = 0.176 p = 0.596 p = 0.715
The duration of dopamine agonist use, average LEDD use, afterage dose of dopamine agonist was at the time of dose change.
33 experienced idiosyncratic dopamine agonist side effects; 105 experienced a typical dopaminergic side effect. The decision to stop the medication completely was made in 59.3% of patients who experienced an ICD, 57.6% of patients who experienced an idiosyncratic dopamine agonist side effect, 44.7% of patients who experienced a generalized DRT side effect. Furthermore, DAWS was experienced in 28.1% of patients who had an ICD, which was greater than all other side effects (p b 0.01). Twenty-one percent of patients with an idiosyncratic dopamine agonist side effect experienced DAWS and 8.5% of patients with a generalized DRT side effect experienced DAWS.
4.3. Evaluation of DAWS based on dopamine agonist dose change Of the 124 patients who had their dopamine agonist dose reduced or discontinued, 83% (n = 103) were completely taken off the medication and the remaining had just a decrease in dose but continued the medication (see Fig. 1). Out of those who were completely off the medication, 16.5% (n = 17) developed symptoms of DAWS and out of the 21 patients that had only a dose decrease, 19.0% (n = 4) developed DAWS (Chi square = 0.091; OR: 0.84 CI: [0.25–2.8]; p = 0.76).
4.4. Phenomenology of DAWS patients In total, out of the 124 patients that had a change in dopamine agonist dose, 16.9% (21 patients) developed symptoms consistent with DAWS. These patients had a mean: age (at time of withdrawal or last follow up) of 64.1 years; PD duration of 3.8 years; duration of dopamine agonist use of 29.0 months; daily total dopamine replacement dose of 792.8 mg; and, daily dopamine agonist dosage of 235.7 mg. Agitation was the most common symptom of DAWS experienced. The number of symptoms experienced was relatively mild with most patients
Fig. 1. DAWS in patients who either decreased or stopped dopamine agonist completely.
(n = 16) having 1 symptom; 4 patients had 2 symptoms and 1 patient had 3 symptoms (see Fig. 2).
5. Discussion To date, this is the largest study to evaluate the use of dopamine agonists and occurrence of DAWS in a PD population. While dopamine agonists are less efficacious than levodopa and more likely to cause generic and idiosyncratic side effects, the biggest advantage of dopamine agonists is that they are less likely to cause dyskinesias. Therefore, clinicians continue to use them as a treatment option for PD. Fifty percent of our population had experienced at least 1 potential side effect either related to dopamine agonists or DRT, in general. Furthermore, 65% of these patients subsequently had their dose reduced or discontinued. Interestingly, those patients who had their dopamine agonist dose reduced or discontinued had a significantly lower PD duration, had less time exposed to a dopamine agonist, and had a lower dose of a dopamine agonist compared to those who were maintained on the same dopamine agonist dose during our study time period. In our practice, clinicians tend to use dopamine agonists as a treatment option in younger patients and therefore our patients had a lower PD duration as well as less time exposed to a dopamine agonist. Side effects to DRTs, especially ICDs, are likely underreported by patients because of shame, guilt or failure to realize that these new developments are a result of their medications [7–8]. DAWS is likely underreported because it is still an evolving concept unfamiliar to all neurologists. Therefore, it is highly likely that the actual incidence of ICDs and DAWS in our cohort is higher than our reported incidence. In our population, out of those patients who had an ICD, 28% of those
Fig. 2. Frequency of DAWS symptoms.
S. Patel et al. / Journal of the Neurological Sciences 379 (2017) 308–311
patients subsequently developed a symptom concerning for DAWS. Our findings are consistent with previous reports of prior ICD as a potential risk factor for developing DAWS as well as overall frequency of ICDs in a population taking dopamine agonists. Rabinak and Nirenberg described a cohort of patients who developed a severe cluster of physical and behavioral symptoms occurring with dopamine agonist withdrawal, including panic attacks, depression, diaphoresis, agitation, fatigue, pain, orthostatic hypotension, and drug craving labeling the syndrome as DAWS [1]. They postulated that the patients with DAWS belong to a “mesocorticolimbic variant” of PD. Drugs that increase dopamine in this circuit, such as cocaine and amphetamine, produce significant withdrawal syndrome with symptoms consistent with what they described. Agitation was the most common symptom in our population, which is consistent with other drug withdrawal studies, such as a study by Cantwell and McBride who reported agitation and irritability occurring in 78% of patients undergoing amphetamine withdrawal [9]. There were no reported signs or symptoms of delirium in these patients experiencing agitation. Rabinak and Nirenberg further postulated that DAWS occurred in patients exclusively with ICDs [1]. However, in our large cohort, we showed that although having an ICD is a definite risk factor, it is not exclusive in that other patients who have their dopamine agonist withdrawn due to non-ICD side effects are also at risk. In our population, out of the 102 patients who experienced a dopamine agonist side effect and had a dose change in the medication, 19.6% developed symptoms consistent with DAWS. We can infer that experiencing any side effect, even if not ICD can potentially be a risk factor for DAWS. This has not been emphasized in previous reports. In contrast, only 1 out of 22 (4.5%) patients who had a dose change due to another (non side effect) reason besides side effect developed symptoms consistent with DAWS. The reasons for decreasing medications dose other than side effects varied, such as DBS, patient experiencing dyskinesia, or medication tapered off by another physician or hospital admission. Furthermore, there was no significant difference in the frequency of DAWS (16.5% versus 19%; p = 0.76) among patients who had completely stopped dopamine agonists versus those who had simply lowered and continued on a smaller dose. This mesocorticolimbic variant of PD may be adjusted to a certain amount of dopamine and any loss (whether complete or decreased amount) can result in withdrawal symptoms. Limitations of our retrospective study include: incomplete documentations, missing information that was not recorded or unable to be interpreted due to different jargon, variance in documentation by providers, etc. We relied heavily on note documentation and some clinician notes were more detailed than others. As noted above, ICD are generally underreported and therefore our data can underestimate the true occurrence of ICD. Missing data consisted of not knowing how long patients were diagnosed with PD as initial notes may not have noted it, or how long they have been taking a dopamine agonist. Those patients whose diagnosis date was not available were not included in the descriptive analysis. In summary, this large retrospective cohort chart review characterizes the occurrence of an underreported and potentially life-threatening condition, dopamine agonist withdrawal syndrome. With our large sample we were able to show many variables which may contribute to DAWS. Unlike prior reports, the characteristics of our population differed in that our DAWS population had a lower duration of PD, less exposure to the dopamine agonist, and a lower dose compared to those patients who did not develop DAWS. In addition, we are able to show that, in terms of potential risk for DAWS, simply lowering dose of dopamine agonist versus completely stopping the agent does not make a significant difference. Further studies are needed to identify risk factors within our population as well as treatment and disease course in these patients. Prospective studies are needed to follow the trajectory of these patients, including potential faster progression of disease, increased incidence of deep brain stimulation and cognitive outcomes.
311
Ethics approval and consent to participate This study was approved by Cleveland Clinic Institutional Review Board. Competing interests Authors have no competing interests. Consent for publication Not applicable. Funding Research and education funding from the Parkinson's Pals Organization (T56266). Availability of data and materials The datasets during and/or analyzed during the current study available from the corresponding author on reasonable request. Author's contribution Shnehal Patel—Research project conception, organization and execution, statistical analysis design and execution, writing first draft of manuscript. Xiomara Garcia—Research project execution, statistical analysis design, execution, review and critique, manuscript review and critique. Mohammad Edrees Mohammad—Research project execution manuscript review and critique. Xin Xin Yu—manuscript review and critique. Katelyn Vlastaris—Research project execution. Kaley O'Donnell—Research project execution. Kathryn Sutton—Research project execution. Hubert H Fernandez—Research project conception, organization and execution, statistical analysis review and critique, first draft of manuscript review and critique. Acknowledgements Not Applicable. References [1] C.A. Rabinak, M.J. Nirenberg, Dopamine agonist withdrawal syndrome in Parkinson disease, Arch. Neurol. 67 (2010) 58–63. [2] M.J. Nirenberg, Dopamine agonist withdrawal syndrome: implications for patient care, Drugs Aging 30 (2013) 587–592. [3] M. Pondal, C. Marras, J. Miyasaki, E. Moro, M.J. Armstrong, A.P. Strafella, et al., Clinical features of dopamine agonist withdrawal syndrome in a movement disorders clinic, J. Neurol. Neurosurg. Psychiatry 84 (2013) 130–135. [4] J. Bastiaens, B.J. Dorfman, P.J. Christos, M.J. Nirenberg, Prospective cohort study of impulse control disorders in Parkinson's disease, Mov. Disord. 28 (2013) 327–333. [5] N. Limotai, G. Oyama, C. Go, O. Bernal, T. Ong, S.J. Moum, et al., Addiction-like manifestations and Parkinson's disease: a large single center 9-year experience, Int. J. Neurosci. 122 (3) (2012) 145–153. [6] C.L. Tomlinson, R. Stowe, S. Patel, C. Rick, R. Gray, C.E. Clarke, Systematic review of levodopa dose equivalency reporting in Parkinson's disease, Mov. Disord. 25 (2010) 2649–2653. [7] D. Weintraub, S. Hoops, J.A. Shea, et al., Validation of the questionnaire for impulsivecompulsive disorders in Parkinson's disease, Mov. Disord. 24 (2009) 1461–1467. [8] S. Perez-Lloret, M.V. Rey, N. Fabre, et al., Do Parkinson's disease patients disclose their adverse events spontaneously? Eur. J. Clin. Pharmacol. 68 (2001) 857–865. [9] B. Cantwell, A.J. McBride, Self detoxification by amphetamine dependent patients: a pilot study, Drug Alcohol Depend. 49 (1998) 157–163.
Contents lists available at ScienceDirect
Journal of the Neurological Sciences journal homepage: www.elsevier.com/locate/jns
Dopamine agonist withdrawal syndrome (DAWS) in a tertiary Parkinson disease treatment center Shnehal Patel ⁎, Xiomara Garcia, Mohammad Edrees Mohammad, Xin Xin Yu, Katelyn Vlastaris, Kaley O'Donnell, Kathryn Sutton, Hubert H. Fernandez Center of Neurologic Restoration, Cleveland Clinic, 9500 Euclid Ave/U2, Cleveland, OH 44195, United States
a r t i c l e
i n f o
Article history: Received 20 January 2017 Received in revised form 24 April 2017 Accepted 15 June 2017 Available online 16 June 2017 Keywords: Dopamine agonist withdrawal syndrome Dopamine agonist Parkinson disease
a b s t r a c t Introduction: Dopamine agonists are a mainstay of treatment for patients with Parkinson disease (PD). However, side effects limit their use, often necessitating dose change. Upon withdrawal, patients may experience dopamine agonist withdrawal syndrome (DAWS). To date, there is no established protocol for the prevention or treatment of DAWS. Methods: We performed a retrospective chart review of PD patients who were taking a dopamine agonist. Results: In our large cohort of 313 PD patients who were on a dopamine agonist, we found that 39.5% (n = 124) had a change in their dose of medication for various reasons, including 102 patients who experienced a side effect on a dopamine agonist. Twenty out of 102 patients (19.6%) developed symptoms consistent with DAWS, whereas 1 out of 22 patients (4.5%) who had medication dose changed due to any other reason (e.g. dyskinesias, DBS surgery, decreased by another provider, etc.) developed symptoms consistent with DAWS. Our DAWS population had a shorter duration of PD, less exposure to a dopamine agonist, and was on a lower dose compared to those patients who did not develop DAWS. Agitation was the most common DAWS symptom reported in our cohort. Interestingly, in terms of developing DAWS, the prevalence of DAWS (19.0% vs 16.5%; p = 0.76) between partial versus total discontinuation was not significantly different whether the dopamine agonist dose was decreased (21 patients) or completely stopped (103 patients). Conclusion: Contrary to previous reports, we have found that other side effects besides impulse control behavioral disorders also increase risk for developing DAWS. Furthermore, the prevalence of DAWS did not differ between partial versus total discontinuation of the dopamine agonist. © 2017 Elsevier B.V. All rights reserved.
1. Introduction Parkinson disease (PD) is a chronic neurodegenerative condition affecting 7–10 million people in the world. Without a current cure, the mainstay of symptomatic treatment remains dopamine replacement, including dopamine agonists. Although consistently providing meaningful benefit in controlling motor symptoms in PD, patients on dopamine agonists can experience significant side effects which can become intolerable. These side effects include cognitive changes, delusions and hallucinations, excessive daytime sleepiness, “sleep attacks”, leg swelling, weight gain, and impulse control behavioral disorders (ICD), among others. Abbreviations: PD, Parkinson Disease; DAWS, dopamine agonist withdrawal syndrome; ICD, impulse control behavioral disorders; LEDD, levodopa equivalent daily dose; KP, Knowledge Program; HSM, health status measures; PHQ9, Patient Health Questionairre-9; GAD7, Generalized Anxiety Disorder 7; UPDRSII, Unified Parkinson Disease Rating Scale II. ⁎ Corresponding author. E-mail address: [email protected] (S. Patel).
http://dx.doi.org/10.1016/j.jns.2017.06.022 0022-510X/© 2017 Elsevier B.V. All rights reserved.
Present management of side effects requires tapering or withdrawal of the offending dopamine agonist. However, in our clinical practice, this is often easier said than done. PD patients may experience worsening of their motor symptoms and occasionally they have been described to develop signs and symptoms similar to those seen in patients experiencing psychostimulant withdrawal. This recently recognized phenomenon has been termed dopamine agonist withdrawal syndrome (DAWS) [1–3] DAWS is a stereotyped, often severe, cluster of physical and behavioral symptoms occurring with dopamine agonist withdrawal, including panic attacks, depression, diaphoresis, agitation, fatigue, pain, orthostatic hypotension, and drug craving, that are refractory to levodopa supplementation [1–3]. DAWS has been reported in up to 19% of patients undergoing a dopamine agonist taper, with a range of 15 to 19%. Risk factors for DAWS that have been reported include high cumulative dopamine agonist exposure, high baseline dopamine agonist dose, and prior diagnosis of ICD [1–3]. Unfortunately, DAWS can be very challenging to treat. It is typically not responsive to levodopa supplementation, antidepressants and other behavioral
S. Patel et al. / Journal of the Neurological Sciences 379 (2017) 308–311
treatments have been minimally effective. The only presumed definitive treatment has been to reinitiate the dopamine agonist [1–3]. In some patients DAWS is self-limiting with resolution within weeks, allowing complete withdrawal of the dopamine agonist. In other patients it can run over a protracted course of months, preventing complete withdrawal of the dopamine agonist, which in turn causing significant unwanted consequences such as a state of chronic ICD leading to litigation, obesity, financial losses, and social and occupational consequences [1–5]. Unfortunately, our cumulative experience with DAWS is limited by small sample size and short observation period. We aim to report the naturalistic phenomenology and prevalence of DAWS as captured in the electronic recorders at our tertiary movement disorders center, with over 1500 PD patients treated annually.
309
3. Statistics Descriptive statistics was used to describe our cohort. Using the SPSS software, Mann-Whitney was used to compare non-normative data such as demographic data between two different populations (e.g. continued dopamine agonist versus dose reduced or discontinued). MannWhitney was also used to compare demographic data of patients with DAWS compared to those without DAWS. Chi-square was used when comparing categorical variables in the population groups (e.g. continued versus decreased versus discontinued dopamine agonist). Chisquare was also used to compare incidence of DAWS between populations based on the type of side effect experienced. Significance was defined as p b 0.05. 4. Results
2. Methods We performed a retrospective chart review of patients diagnosed with PD, by a fellowship-trained movement disorders neurologist, who fulfill the United Kingdom Parkinson's Disease Society Brain Bank criteria, over a 2-year period, from January 2011 to December of 2012, at the Center of Neurological Restoration of the Cleveland Clinic, Cleveland, Ohio. Patients who were not taking a dopamine agonist were excluded from further analysis. Data collected included: gender, race, age at the time of dopamine agonist withdrawal, name of the dopamine agonist, and the duration of PD as well as dopamine agonist use. We then carefully reviewed all notes (in-patient visits, electronic and telephone correspondences) to determine if patient had experienced any symptom that can be considered a side effect of dopamine replacement therapy. Side effects were categorized as follows: (1) ICD–hypersexuality, pathological gambling, binge eating, compulsive shopping, punding; (2) Idiosyncratic dopamine agonist side effects–leg swelling, weight gain, sleep attacks, skin reactions; (3) General dopamine replacement therapy side effects–nausea/ vomiting, orthostasis, cognitive worsening, psychosis, sedation. For all patients, we recorded the total number of PD medications, the daily dose of total dopamine replacement therapy, and also dopamine agonist therapy using published levodopa equivalent daily dose (LEDD) conversion ratios [6]. We further analyzed if patients were continued on the dopamine agonists, or if the dose was tapered or completely stopped. The speed of taper (i.e. b2 weeks, 2–4 weeks, and N4 weeks) was also recorded. Finally, we reviewed if any of these patients experienced symptoms consistent with DAWS including: panic attacks, depression, diaphoresis, agitation, fatigue, pain, drug cravings, or orthostatic hypotension [1–2]. We also collected other demographic variables including education level, job status, marital status, smoking and substance abuse history. We obtained data through EPIC, our electronic medical record system, and our Knowledge Program (KP). KP is a data capture initiative designed to harness routinely collected clinical information to optimize patient care and use of electronic medical record. Patient reported health status measures (HSM) are collected at each patient visit in electronic tablet, patient kiosk, or from patient's home through patients' electronic access (MyChart). These results, along with data from existing clinical systems, are then consolidated into a single data repository, the KP database. The KP database was able to provide further information regarding patient's depression, anxiety, and the activities of daily living measurements using Patient Health Questionairre-9 (PHQ9), Generalized Anxiety Disorder 7 (GAD7), and the Unified Parkinson Disease Rating Scale II (UPDRSII), respectively. We used values that were closest to the time when the patient was either taken off the dopamine agonist or their last visit for patients who were continued on the dopamine agonist. The study was performed in accordance with Cleveland Clinic Institutional Review Board.
Out of 1884 parkinsonian patients who were seen in Cleveland Clinic over the two-year period, 313 met diagnostic criteria for idiopathic PD and were taking a dopamine agonist. Our cohort had a mean: age (at time of analysis) of 65.1 years (SD 9.2); PD duration of 8.45 years (SD 6.23); duration of dopamine agonist use of 58.5 months (SD 170.7); total daily dopamine replacement dose of 770.5 mg LEDD (SD 429.8); and, daily dopamine agonist dosage of 191 mg LEDD (SD 116.8). Types of dopamine agonists used are listed in Table 1. Our cohort was then classified according to dopamine agonist dose modification (i.e. no change versus dose decrease/discontinuation). The demographical data of these two groups are also illustrated in Table 2. Education level, job status, marital status, smoking and substance abuse history were unable to be obtained in over half of the cohort, therefore these variables were not included. The duration of PD, duration of dopamine agonist use, and mean dose of dopamine agonist were greater in patients who had continued the dopamine agonist compared to those who had decreased or stopped the medication. Patients whose duration of PD or dopamine agonist use was not known were not included in the mean analysis. 4.1. Evaluation of DAWS based on dopamine agonist side effects One patient was lost to follow up and was not included. Out of the remaining 312 patients, 50% (n = 156) developed at least 1 side effect. Of these patients, 65% (n = 102) had the dopamine agonist either decreased or completely stopped, whereas 35% (n = 54) continued on the same dose. Of those who had dopamine agonist dose decreased or discontinued, 19.6% (n = 20) developed symptoms consistent with DAWS. Of the patients who did not experience any side effect, there were 14% (n = 22) who had the medication either decreased or completely stopped. The reasons for decrease were varied and included: patient receiving DBS; having dyskinesias; or medication was changed by another provider for unknown reason. Within this group, only 1 patient (4.5%) developed symptoms consistent with DAWS. This patient has his medication stopped during an inpatient admission for an irregular heart rate, which seemed unrelated to the dopamine agonist. 4.2. Side effects and decision to taper The 156 patients who developed side effects were then classified according to the types of adverse event experienced: 32 experienced ICD; Table 1 Type of dopamine agonist used. Dopamine agonists used
N (%)
Pramipexole Ropinirole Rotigotine
33 (21.15%) 99 (63.46%) 24 (15.38)%
310
S. Patel et al. / Journal of the Neurological Sciences 379 (2017) 308–311
Table 2 Demographics of population.
Mean age at time of withdrawal or end of study M:F Duration of PD (n = 309) Duration of dopamine agonist use (months) (n = 233) Mean LEDD use Mean dose of dopamine agonist (in LEDD) UPDRS part II Anxiety (GAD7) Depression
Total population (n = 313)
Continued dopamine agonist (n = 188)
Decreased dopamine agonist (n = 124)
p-Value
65.1 188:125 8.45 57.2 770.5 mg 191.2 mg 14.3 4.7 6.54
65.3 123:65 9.17 70.8 802.9 200.2 14.8 4.8 6.6
64.9 64:60 7.31 38.1 723.4 178.4 13.5 4.5 6.5
p = 0.715 p = 0.96 p = 0.005 p = 0.001 p = 0.074 p = 0.021 p = 0.176 p = 0.596 p = 0.715
The duration of dopamine agonist use, average LEDD use, afterage dose of dopamine agonist was at the time of dose change.
33 experienced idiosyncratic dopamine agonist side effects; 105 experienced a typical dopaminergic side effect. The decision to stop the medication completely was made in 59.3% of patients who experienced an ICD, 57.6% of patients who experienced an idiosyncratic dopamine agonist side effect, 44.7% of patients who experienced a generalized DRT side effect. Furthermore, DAWS was experienced in 28.1% of patients who had an ICD, which was greater than all other side effects (p b 0.01). Twenty-one percent of patients with an idiosyncratic dopamine agonist side effect experienced DAWS and 8.5% of patients with a generalized DRT side effect experienced DAWS.
4.3. Evaluation of DAWS based on dopamine agonist dose change Of the 124 patients who had their dopamine agonist dose reduced or discontinued, 83% (n = 103) were completely taken off the medication and the remaining had just a decrease in dose but continued the medication (see Fig. 1). Out of those who were completely off the medication, 16.5% (n = 17) developed symptoms of DAWS and out of the 21 patients that had only a dose decrease, 19.0% (n = 4) developed DAWS (Chi square = 0.091; OR: 0.84 CI: [0.25–2.8]; p = 0.76).
4.4. Phenomenology of DAWS patients In total, out of the 124 patients that had a change in dopamine agonist dose, 16.9% (21 patients) developed symptoms consistent with DAWS. These patients had a mean: age (at time of withdrawal or last follow up) of 64.1 years; PD duration of 3.8 years; duration of dopamine agonist use of 29.0 months; daily total dopamine replacement dose of 792.8 mg; and, daily dopamine agonist dosage of 235.7 mg. Agitation was the most common symptom of DAWS experienced. The number of symptoms experienced was relatively mild with most patients
Fig. 1. DAWS in patients who either decreased or stopped dopamine agonist completely.
(n = 16) having 1 symptom; 4 patients had 2 symptoms and 1 patient had 3 symptoms (see Fig. 2).
5. Discussion To date, this is the largest study to evaluate the use of dopamine agonists and occurrence of DAWS in a PD population. While dopamine agonists are less efficacious than levodopa and more likely to cause generic and idiosyncratic side effects, the biggest advantage of dopamine agonists is that they are less likely to cause dyskinesias. Therefore, clinicians continue to use them as a treatment option for PD. Fifty percent of our population had experienced at least 1 potential side effect either related to dopamine agonists or DRT, in general. Furthermore, 65% of these patients subsequently had their dose reduced or discontinued. Interestingly, those patients who had their dopamine agonist dose reduced or discontinued had a significantly lower PD duration, had less time exposed to a dopamine agonist, and had a lower dose of a dopamine agonist compared to those who were maintained on the same dopamine agonist dose during our study time period. In our practice, clinicians tend to use dopamine agonists as a treatment option in younger patients and therefore our patients had a lower PD duration as well as less time exposed to a dopamine agonist. Side effects to DRTs, especially ICDs, are likely underreported by patients because of shame, guilt or failure to realize that these new developments are a result of their medications [7–8]. DAWS is likely underreported because it is still an evolving concept unfamiliar to all neurologists. Therefore, it is highly likely that the actual incidence of ICDs and DAWS in our cohort is higher than our reported incidence. In our population, out of those patients who had an ICD, 28% of those
Fig. 2. Frequency of DAWS symptoms.
S. Patel et al. / Journal of the Neurological Sciences 379 (2017) 308–311
patients subsequently developed a symptom concerning for DAWS. Our findings are consistent with previous reports of prior ICD as a potential risk factor for developing DAWS as well as overall frequency of ICDs in a population taking dopamine agonists. Rabinak and Nirenberg described a cohort of patients who developed a severe cluster of physical and behavioral symptoms occurring with dopamine agonist withdrawal, including panic attacks, depression, diaphoresis, agitation, fatigue, pain, orthostatic hypotension, and drug craving labeling the syndrome as DAWS [1]. They postulated that the patients with DAWS belong to a “mesocorticolimbic variant” of PD. Drugs that increase dopamine in this circuit, such as cocaine and amphetamine, produce significant withdrawal syndrome with symptoms consistent with what they described. Agitation was the most common symptom in our population, which is consistent with other drug withdrawal studies, such as a study by Cantwell and McBride who reported agitation and irritability occurring in 78% of patients undergoing amphetamine withdrawal [9]. There were no reported signs or symptoms of delirium in these patients experiencing agitation. Rabinak and Nirenberg further postulated that DAWS occurred in patients exclusively with ICDs [1]. However, in our large cohort, we showed that although having an ICD is a definite risk factor, it is not exclusive in that other patients who have their dopamine agonist withdrawn due to non-ICD side effects are also at risk. In our population, out of the 102 patients who experienced a dopamine agonist side effect and had a dose change in the medication, 19.6% developed symptoms consistent with DAWS. We can infer that experiencing any side effect, even if not ICD can potentially be a risk factor for DAWS. This has not been emphasized in previous reports. In contrast, only 1 out of 22 (4.5%) patients who had a dose change due to another (non side effect) reason besides side effect developed symptoms consistent with DAWS. The reasons for decreasing medications dose other than side effects varied, such as DBS, patient experiencing dyskinesia, or medication tapered off by another physician or hospital admission. Furthermore, there was no significant difference in the frequency of DAWS (16.5% versus 19%; p = 0.76) among patients who had completely stopped dopamine agonists versus those who had simply lowered and continued on a smaller dose. This mesocorticolimbic variant of PD may be adjusted to a certain amount of dopamine and any loss (whether complete or decreased amount) can result in withdrawal symptoms. Limitations of our retrospective study include: incomplete documentations, missing information that was not recorded or unable to be interpreted due to different jargon, variance in documentation by providers, etc. We relied heavily on note documentation and some clinician notes were more detailed than others. As noted above, ICD are generally underreported and therefore our data can underestimate the true occurrence of ICD. Missing data consisted of not knowing how long patients were diagnosed with PD as initial notes may not have noted it, or how long they have been taking a dopamine agonist. Those patients whose diagnosis date was not available were not included in the descriptive analysis. In summary, this large retrospective cohort chart review characterizes the occurrence of an underreported and potentially life-threatening condition, dopamine agonist withdrawal syndrome. With our large sample we were able to show many variables which may contribute to DAWS. Unlike prior reports, the characteristics of our population differed in that our DAWS population had a lower duration of PD, less exposure to the dopamine agonist, and a lower dose compared to those patients who did not develop DAWS. In addition, we are able to show that, in terms of potential risk for DAWS, simply lowering dose of dopamine agonist versus completely stopping the agent does not make a significant difference. Further studies are needed to identify risk factors within our population as well as treatment and disease course in these patients. Prospective studies are needed to follow the trajectory of these patients, including potential faster progression of disease, increased incidence of deep brain stimulation and cognitive outcomes.
311
Ethics approval and consent to participate This study was approved by Cleveland Clinic Institutional Review Board. Competing interests Authors have no competing interests. Consent for publication Not applicable. Funding Research and education funding from the Parkinson's Pals Organization (T56266). Availability of data and materials The datasets during and/or analyzed during the current study available from the corresponding author on reasonable request. Author's contribution Shnehal Patel—Research project conception, organization and execution, statistical analysis design and execution, writing first draft of manuscript. Xiomara Garcia—Research project execution, statistical analysis design, execution, review and critique, manuscript review and critique. Mohammad Edrees Mohammad—Research project execution manuscript review and critique. Xin Xin Yu—manuscript review and critique. Katelyn Vlastaris—Research project execution. Kaley O'Donnell—Research project execution. Kathryn Sutton—Research project execution. Hubert H Fernandez—Research project conception, organization and execution, statistical analysis review and critique, first draft of manuscript review and critique. Acknowledgements Not Applicable. References [1] C.A. Rabinak, M.J. Nirenberg, Dopamine agonist withdrawal syndrome in Parkinson disease, Arch. Neurol. 67 (2010) 58–63. [2] M.J. Nirenberg, Dopamine agonist withdrawal syndrome: implications for patient care, Drugs Aging 30 (2013) 587–592. [3] M. Pondal, C. Marras, J. Miyasaki, E. Moro, M.J. Armstrong, A.P. Strafella, et al., Clinical features of dopamine agonist withdrawal syndrome in a movement disorders clinic, J. Neurol. Neurosurg. Psychiatry 84 (2013) 130–135. [4] J. Bastiaens, B.J. Dorfman, P.J. Christos, M.J. Nirenberg, Prospective cohort study of impulse control disorders in Parkinson's disease, Mov. Disord. 28 (2013) 327–333. [5] N. Limotai, G. Oyama, C. Go, O. Bernal, T. Ong, S.J. Moum, et al., Addiction-like manifestations and Parkinson's disease: a large single center 9-year experience, Int. J. Neurosci. 122 (3) (2012) 145–153. [6] C.L. Tomlinson, R. Stowe, S. Patel, C. Rick, R. Gray, C.E. Clarke, Systematic review of levodopa dose equivalency reporting in Parkinson's disease, Mov. Disord. 25 (2010) 2649–2653. [7] D. Weintraub, S. Hoops, J.A. Shea, et al., Validation of the questionnaire for impulsivecompulsive disorders in Parkinson's disease, Mov. Disord. 24 (2009) 1461–1467. [8] S. Perez-Lloret, M.V. Rey, N. Fabre, et al., Do Parkinson's disease patients disclose their adverse events spontaneously? Eur. J. Clin. Pharmacol. 68 (2001) 857–865. [9] B. Cantwell, A.J. McBride, Self detoxification by amphetamine dependent patients: a pilot study, Drug Alcohol Depend. 49 (1998) 157–163.