- Email: [email protected]
Risk and course of motor complications in a population-based incident Parkinson's disease cohort
Parkinsonism and Related Disorders xxx (2015) 1e6
Contents lists available at ScienceDirect
Parkinsonism and Related Disorders journal homepage: www.elsevier.com/locate/parkreldis
Risk and course of motor complications in a population-based incident Parkinson's disease cohort Anders Bjornestad a, d, *, Elin B. Forsaa a, d, Kenn Freddy Pedersen a, d, Ole-Bjorn Tysnes b, c, Jan Petter Larsen a, Guido Alves a, d a
The Norwegian Centre for Movement Disorders, Stavanger University Hospital, PO Box 8100, N-4068 Stavanger, Norway Department of Neurology, Haukeland University Hospital, PO Box 1400, N-5021 Bergen, Norway Institute of Clinical Medicine, University of Bergen, Norway d Department of Neurology, Stavanger University Hospital, Stavanger, Norway b c
a r t i c l e i n f o
a b s t r a c t
Article history: Received 3 September 2015 Received in revised form 9 October 2015 Accepted 2 November 2015
Background: Motor complications may become major challenges in the management of patients with Parkinson's disease. In this study, we sought to determine the incidence, risk factors, evolution, and treatment of motor fluctuations and dyskinesias in a population-representative, incident Parkinson's disease cohort. Methods: In this prospective population-based 5-year longitudinal study, we followed 189 incident and initially drug-naïve Parkinson's disease patients biannually for detailed examination of dyskinesias and motor fluctuations as defined by the Unified Parkinson's disease Rating Scale. We performed Kaplan eMeier survival and Cox regression analyses to assess cumulative incidence and risk factors of these motor complications. Results: The 5-year cumulative incidence of motor complications was 52.4%. Motor fluctuations occurred in 42.9% and dyskinesias in 24.3%. Besides higher motor severity predicting both motor fluctuations (p ¼ 0.016) and dyskinesias (p < 0.001), lower age at diagnosis predicted motor fluctuations (p ¼ 0.001), whereas female gender predicted dyskinesias (p ¼ 0.001). Actual levodopa dose at onset of motor fluctuations (p ¼ 0.037) or dyskinesias (p < 0.001) rather than initial treatment with levodopa (p > 0.1) independently predicted development of motor complications. Motor fluctuations reversed in 37% and dyskinesias in 49% of patients on oral treatment and remained generally mild in those with persistent complications. No patients received device-aided therapies during the study. Conclusions: More than 50% in the general Parkinson's disease population develop motor complications within 5 years of diagnosis. However, they remain mild in the vast majority and are reversible in a substantial proportion of patients. © 2015 Elsevier Ltd. All rights reserved.
Keywords: Parkinson's disease Motor complications Motor fluctuations Dyskinesias Risk factors
1. Introduction Parkinson disease (PD) is a relentlessly progressive movement disorder of unknown etiology [1]. With PD progression, the efficacy of pharmacotherapy deteriorates [2] and patients may develop motor complications, which can be broadly subdivided into motor
* Corresponding author. The Norwegian Centre for Movement Disorders, Stavanger University Hospital, PO Box 8100, N-4068 Stavanger, Norway. E-mail addresses: [email protected] (A. Bjornestad), elin.bjelland. [email protected] (E.B. Forsaa), [email protected] (K.F. Pedersen), [email protected] (O.-B. Tysnes), [email protected] (J.P. Larsen), [email protected] (G. Alves).
fluctuations and dyskinesias. The risk for and time to emergence of these motor complications vary substantially among patients for reasons that are probably complex, including both disease- and drug-related factors, particularly treatment with levodopa [3,4]. Although motor fluctuations and dyskinesias may compromise quality of life [5], their evolution and prognosis are still incompletely investigated, as most previous studies assessed crosssectional cohorts with established disease [6] or followed patients recruited from movement disorders centers or clinical trials [7]. This may result in both under- and overestimation of the risk of these motor complications. In addition, as treatment options have changed during the last decades, previous estimates might no longer be accurate. Finally, device-aided therapies have emerged as
http://dx.doi.org/10.1016/j.parkreldis.2015.11.007 1353-8020/© 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: A. Bjornestad, et al., Risk and course of motor complications in a population-based incident Parkinson's disease cohort, Parkinsonism and Related Disorders (2015), http://dx.doi.org/10.1016/j.parkreldis.2015.11.007
2
A. Bjornestad et al. / Parkinsonism and Related Disorders xxx (2015) 1e6
second-line treatments in patients with otherwise intractable motor complications [8,9]. Recent studies promote early initiation of these advanced therapies [9,10], which e however e are costly and may cause serious adverse events. Therefore, precise information on the evolution and potential reversal of motor fluctuations and dyskinesias using conventional pharmacotherapy is crucial for optimal patient selection to and timing of second-line treatment in PD. To gain this important knowledge, we recruited a large, population-based, and initially drug-naive PD cohort that we monitored closely for emergence of motor complications for five years from diagnosis. 2. Methods 2.1. Subjects All subjects participate in the Norwegian ParkWest project, a prospective, population-based, multicenter, longitudinal cohort study investigating the incidence, neurobiology and prognosis of PD. In order to establish a population-representative, incident PD cohort, we recruited patients with newly-diagnosed and untreated PD from the general population between November, 2004, and August, 2006, using multiple recruitment strategies, as previously described [11]. Briefly, all patients were screened, followed and treated by neurologists from the study group, who are experienced in movement disorders. Diagnostic procedures at baseline included a full medical history, comprehensive clinical, neuropsychological and neuropsychiatric assessments, laboratory tests, cerebral MRI, and, if indicated clinically, dopamine transporter imaging. Standardized follow-up visits were conducted every six months. Of 212 patients included in the original cohort, five were not drug-naïve at baseline and further 18 rediagnosed during follow-up, leaving 189 patients eligible for this study. All met widely acknowledged research criteria [12,13] for PD at latest follow-up. The study was approved by the Regional Committee for Medical Research Ethics, Western Norway. Signed written consent was obtained from all participants. 2.2. Evaluation program and treatment At baseline, a study neurologist performed general medical and neurological examinations and semi-structured interviews to obtain medical and drug history. Motor severity and disease stage in the drug-naïve state were rated using the Unified PD Rating Scale (UPDRS) [14] motor section (part III) and Hoehn and Yahr staging. Assessment of non-motor features included the Mini-Mental State Examination (MMSE), Montgomery Aasberg Depression Rating Scale (MADRS), Fatigue Severity Scale (FSS), PD Sleep Scale (PDSS), Epworth Sleepiness Scale (ESS), and Starkstein Apathy Scale (SAS). Following baseline examinations, antiparkinsonian treatment was initiated and adjusted throughout follow-up by a study neurologist according to best clinical judgement. Drug changes in between study visits were allowed when deemed clinically necessary. Motor complications were assessed at 6-month intervals by clinical observation and the UPDRS complication of therapy section (part IV) with a score of 1 on items 36, 37, 38 or 39 defining presence of motor fluctuations and a score of 1 on items 32, 33 or 34 defining presence of dyskinesias. Type and dose of medications were recorded at each study visit, and levodopaequivalent doses (LED) were calculated according to published recommendations [15]. Use of advanced therapies was also recorded, defined as subcutaneous apomorphine administration, continuous intestinal levodopa infusion, or subthalamic nucleus deep brain stimulation (STN-DBS), which all are provided at no personal or private insurance company costs in Norway.
2.3. Statistical analysis SPSS 21.0 was used for statistical analysis. Incidence rates with 95% confidence intervals were calculated and expressed in relation to 1000 person-years of observation. Between-group differences were compared using t-tests, ManneWhitney U-tests and c2-tests as appropriate. We performed KaplaneMeier survival analysis to evaluate cumulative incidence rates of motor fluctuations and dyskinesias, and Cox regression analysis to assess baseline factors associated with evolution of these motor complications. Separate analysis of wearing-off was performed without significant changes in results, thus all motor fluctuations are presented as one group in the following. We considered age, gender, time since motor onset, and motor severity at baseline as primary risk factors of interest. In subsequent models adjusting for these variables, we also included a panel of non-motor symptoms (as these have been associated with motor complications in cross-sectional studies) assessed at baseline using MMSE, MADRS, FSS, PDSS, ESS, and SAS. Finally, we investigated the associations of initial treatment (levodopa vs. dopamine agonists) as well as actual LED at onset of motor complications with the development of motor fluctuations and dyskinesias. Two-tailed p-values <0.05 were considered significant. 3. Results 3.1. Baseline characteristics and subject flow Baseline characteristics are given in Table 1 and the flow of subjects in Fig. 1. Of the 189 patients included at baseline, 183 (97%) completed 1 year, 173 (92%) 3 years and 158 (84%) 5 years of biannual follow-up, generating 1911 observations in total. Few subjects (7/189, 4%) withdrew from the study, whereas remaining loss to follow-up (24/189, 13%) was due to death. Median follow-up time was 5.0 (interquartile range 4.9e5.1) years. 3.2. Frequency of motor complications Point prevalence rates increased progressively, reaching 38.0% for any motor complications, 31.0% for motor fluctuations, and 12.7% for dyskinesias at 5 years of follow-up (Table 2). The 5-year cumulative incidence of any motor complications was 52.4%, with a corresponding incidence rate of 158 (95% CI 132e189) per 1000 person-years of observation. For motor fluctuations, the cumulative Table 1 Baseline characteristics of 189 patients with incident, drug-naïve PD. Characteristics
Values
Number Male, n (%) Age, years Time since diagnosis, months Symptom duration, years UPDRS motor score Hoehn and Yahr stage MMSE score MADRS score FSS score PDSS score ESS score SAS score Body weight, kilograms
189 114 (60.3) 67.7 (9.3) 1.6 (1.7) 2.3 (1.8) 23.4 (11.3) 1.9 (0.6) 27.8 (2.5) 4.7 (5.2) 4.4 (1.6) 118.9 (19.6) 6.0 (3.6) 15.6 (4.8) 76.5 (14.4)
Values are mean (SD) if not otherwise indicated. UPDRS ¼ Unified Parkinson's Disease Rating Scale; MMSE ¼ Mini-Mental State Examination; MADRS ¼ Montgomery Aasberg Depression Rating Scale; FSS ¼ Fatigue Severity Scale; PDSS ¼ Parkinson's Disease Sleep Scale; ESS ¼ Epworth Sleepiness Scale; SAS ¼ Starkstein Apathy Scale.
Please cite this article in press as: A. Bjornestad, et al., Risk and course of motor complications in a population-based incident Parkinson's disease cohort, Parkinsonism and Related Disorders (2015), http://dx.doi.org/10.1016/j.parkreldis.2015.11.007
A. Bjornestad et al. / Parkinsonism and Related Disorders xxx (2015) 1e6
3
Fig. 1. Study flowchart. PD ¼ Parkinson disease. Patients were assessed at 6-month intervals; flowchart is simplified for readability.
Table 2 Prevalence and cumulative incidence rates of motor complications, and type and dose of dopaminergic drugs, during follow-up.
Motor complications overall Point prevalence, % Cumulative incidence, % Motor fluctuations Point prevalence, % Cumulative incidence, % Dyskinesias Point prevalence, % Cumulative incidence, % Concomitant DK and MF Point prevalence, % Cumulative incidence, % Treatment Levodopa, % Dopamine agonist, % Other medications, % Total LED (SD)
Baseline
1 year
2 years
3 years
4 years
5 years
0 0
10.9 14.3
13.6 24.3
22.8 32.8
23.2 40.7
38.0 52.4
0 0
7.7 10.6
10.2 19.0
18.1 26.5
17.1 33.9
31.0 42.9
0 0
3.8 6.3
4.5 9.5
5.2 12.7
9.8 18.0
12.7 24.3
0 0
0.5 2.6
1.1 4.2
0.5 6.3
3.7 11.1
5.7 14.8
0 0 0 0
41.5 44.8 23.5 252 (169)
56.8 47.7 30.7 333 (161)
67.1 48.6 33.5 424 (224)
78.7 53.7 29.9 514 (250)
84.2 53.8 32.9 587 (334)
Patients were assessed at 6-month intervals. Only 12-months findings are displayed for simplicity. DK ¼ dyskinesias. MF ¼ motor fluctuations. LED ¼ levodopa equivalent dose. SD ¼ standard deviation.
incidence at 5 years of follow-up was 42.9%, yielding an incidence rate of 121 (99e148) per 1000 person-years. For dyskinesias, the 5year cumulative incidence was 24.3%, with a corresponding incidence rate of 59 (45e78) per 1000 person-years. 14.8% of the total population and 28.3% of those with motor complications experienced both motor fluctuations and dyskinesias. 3.3. Baseline risk factors As our cohort was drug-naïve at baseline, we first sought to identify clinical and demographic risk factors that could predict motor complications before dopaminergic medication was initiated. In multivariate Cox regression analysis, lower age at diagnosis (hazard ratio (HR) per year 0.97 (0.95e0.99), p ¼ 0.013), female gender (HR 1.84 (1.23e2.77), p ¼ 0.003), and higher UPDRS motor score at baseline (HR per unit 1.04 (1.01e1.06), p ¼ 0.001) independently predicted a higher risk of developing motor complications during follow-up, whereas time since motor onset did not (HR
per year 0.97 (0.86e1.09), p ¼ 0.593). No additional independent risk factors were identified among the assessed non-motor features (all p > 0.05). The risk factor pattern differed between motor fluctuations and dyskinesias when these were tested in separate Cox regression models. Independent baseline risk factors of motor fluctuations were lower age (HR per year 0.96 (0.94e0.98), p ¼ 0.001) and higher UPDRS motor score (HR per unit 1.03 (1.01e1.05), p ¼ 0.016) but not gender (HR females vs. males 1.48 (0.94e2.33), p ¼ 0.087) or time since motor onset (HR per year 1.03 (0.91e1.16, p ¼ 0.675). For example, the cumulative 5-year incidence of motor fluctuations in patients aged >80 was 11% while it was 41% for patients aged 60e79 and 64% for patients below 60 years at baseline (Fig. 2a). In contrast, independent risk factors of dyskinesias were female gender (HR 2.73 (1.49e4.98), p ¼ 0.001, Fig. 2b) and higher baseline UPDRS motor score (HR per unit 1.05 (1.02e1.08), p ¼ 0.001) but not age (HR per year 0.98 (0.95e1.02), p ¼ 0.262) or time since motor onset (HR per year 0.91 (0.75e1.10), p ¼ 0.332). Again, none
Please cite this article in press as: A. Bjornestad, et al., Risk and course of motor complications in a population-based incident Parkinson's disease cohort, Parkinsonism and Related Disorders (2015), http://dx.doi.org/10.1016/j.parkreldis.2015.11.007
4
A. Bjornestad et al. / Parkinsonism and Related Disorders xxx (2015) 1e6
therapies by their study neurologists during follow-up. A subgroup of patients (36/189) never used levodopa. Among these, only 5 (13.9%) experienced motor fluctuations, and 4 (11.1%) had dyskinesias. The corresponding rates in levodopa-treated patients were 49.7% and 27.5%, respectively. In multivariate Cox regression analysis (with adjustment for baseline age, gender, time since motor onset, and motor severity), initial treatment with levodopa was associated with an increased risk of developing motor fluctuations (HR 1.84 (1.09e3.10), p ¼ 0.023) but not dyskinesias (HR 0.88 (0.42e1.85), p ¼ 0.744), compared to initial treatment with dopamine agonists. However, when actual levodopa dose at onset of motor complications was also included in the models, the association between initial levodopa treatment and motor fluctuations was lost (HR 1.37 (0.87e2.87), p ¼ 0.137), whereas actual levodopa dose was independently associated with both motor fluctuations (HR per 100 mg 1.13 (1.01e1.26), p ¼ 0.037) and dyskinesias (HR per 100 mg 1.28 (1.16e1.42), p < 0.001). 3.5. Course of motor complications 3.5.1. Severity Motor complications remained generally mild throughout the study period (esupp table). Motor fluctuations were predictable in the majority of patients, with no more than 5.1% experiencing unpredictable or sudden off periods, respectively, during follow-up. Few patients (5.1%) spent larger proportions (>25%) of their day in off. Similarly, few patients (3.8%) experienced dyskinesias larger proportions of their day and very few rated their dyskinesias as severe (0.6%) or painful (1.8%) within the first 5 years of diagnosis.
Fig. 2. KaplaneMeier plots illustrating the time to (A) incident motor fluctuations stratified by age, and (B) incident dyskinesias stratified by gender.
of the assessed non-motor features at baseline independently predicted motor fluctuations or dyskinesias (all p > 0.05). To explore whether the observed gender difference in the risk of dyskinesias was related to lower body weight (and thus a higher levodopa dose per kg) in females than males, we ran a supplemental model with body weight included as an independent variable, adjusting for age, gender, time since motor onset, and motor severity at baseline. In this model, female gender (HR 2.34 (1.21e4.53, p ¼ 0.011) remained associated with dyskinesias, whereas body weight was not (HR 0.99 per kg (0.96e1.01), p ¼ 0.239). 3.4. Medication effects Patients started dopaminergic treatment following their baseline visits. At 1 year of follow-up, 41.5% of patients received levodopa, 44.8% dopamine agonists, and 23.5% other dopaminergic medications. Among those who started a dopamine agonist, only 26% did not use levodopa at end of follow-up. Overall, 84.2% of those attending the 5-year visit were on levodopa treatment. No patients were considered in need of advanced, device-aided
3.5.2. Reversal To assess the potential reversal of motor complications over time, we considered the 158 patients with complete 5-year followup. Of 78 patients who had experienced motor fluctuations during follow-up, 29 did not so at study end, indicating reversal in 37%. Patients with persistent motor fluctuations were younger than those in whom motor fluctuations reversed on oral treatment (62.3 (SD 9.1) vs. 68.8 (SD 6.9) years, p ¼ 0.001), but did not differ in gender, baseline motor severity, or LED at final visit (all p > 0.05). Dyskinesias were experienced during the study period by 39 of 158 patients with complete follow-up. Nineteen of the 39 had no dyskinesias at final visit, indicating reversal in 49%. Patients whose dyskinesias reversed did not differ from those with persistent dyskinesias in terms of age, gender, baseline motor severity or LED at final visit (all p > 0.05). 4. Discussion Our study highlights a high risk of early motor complications in the general PD population and also provides important insights into their evolution and associated risk factors, with implications for both understanding and management of motor fluctuations and dyskinesias in PD. We are aware of only two previous longitudinal studies of dyskinesias [16] or motor complications [17] in patients with early PD derived from the community. Our study differs importantly from these in several ways, including the entirely drug-naïve cohort at study start, the range of potential demographic and clinical risk factors assessed, the close monitoring of motor complications every six months, and the low attrition rate during follow-up, generating more than 1900 observations over the 5-year study period. In addition, while our patients were recruited unselected from the community, treatment and follow-up was managed
Please cite this article in press as: A. Bjornestad, et al., Risk and course of motor complications in a population-based incident Parkinson's disease cohort, Parkinsonism and Related Disorders (2015), http://dx.doi.org/10.1016/j.parkreldis.2015.11.007
A. Bjornestad et al. / Parkinsonism and Related Disorders xxx (2015) 1e6
by movement disorders neurologists, thus combining the advantages of a population-based design with the benefits of specialist care, including presumed high diagnostic accuracy and up-to-date treatment. We therefore believe that our findings are as valid and representative for the general PD population as possible. Our study demonstrates that motor complications develop early and frequently in the general PD population, affecting more than half of patients within 5 years of diagnosis. The high occurrence of motor complications was driven mainly by a high incidence of motor fluctuations, which were experienced by over 40% during follow-up. This figure is not very different from estimates provided almost four decades ago [18], which may appear surprising given the now widespread use of dopamine agonists and other levodopasparing drugs. Dopamine agonists are, however, generally considered less potent and to have more side effects than levodopa [19]. In line with this, we observed a steady decline in the number of patients who were able to remain on dopamine agonist monotherapy, while the proportion of patients on levodopa increased to nearly 85% by 5 years of follow-up. A number of randomized controlled trials have demonstrated that levodopa treatment shortens the time to onset of motor complications in PD [20,21], thus leading many physicians to prefer dopamine agonists as initial treatment, at least in patients below 65e70 years of age. However, more recent clinical trials with extended follow-up show that the initial treatment has little impact on the long-term risk of motor fluctuations and dyskinesias [22,23]. Indeed, both experimental [24,25] and clinical [26] studies suggest that levodopa dose is more important for the development of motor complications than the duration of levodopa treatment. Extending these observations, we found that initial treatment with levodopa doubled the risk for motor fluctuations, but this association was lost once the actual levodopa dose at onset of motor complications was taken into consideration, with the latter remaining significantly associated with both motor fluctuations and dyskinesias. This implies that actual levodopa dose rather than initial treatment with levodopa is crucial for the development of motor complications. Overall, these data indicate that levodopa treatment in low doses should not be withheld in patients with early PD. In addition to treatment with levodopa, the importance of disease-related factors for the development of motor complications has been debated. Our patients were drug-naïve at baseline, providing a unique opportunity to study the association of diseaserelated factors and motor complications unbiased by dopaminergic treatment. We found that more severe parkinsonism at diagnosis (in the drug-naïve state) independently predicted early development of motor fluctuations and dyskinesias, suggesting that dopamine depletion is a principal underlying cause of both of these motor complications. This observation is in line with and extends recent neuroimaging findings that point towards an important role of presynaptic mechanisms in the etiology of dyskinesias in PD [27,28]. While motor fluctuations developed frequently in our cohort, this was not the case for dyskinesias, which affected less than one quarter of patients during the 5-year study period. The proportion of patients who experienced both motor fluctuations and dyskinesias during follow-up was only 15% of the total cohort and 28% of those with motor complications. This limited overlap of motor fluctuations and dyskinesias is remarkable and may indicate partly distinct underlying biological mechanisms, at least in early PD. This is further supported by our observation that the demographic risk factors differed between motor fluctuations and dyskinesias. We found that lower age but not gender was a strong and independent predictor of motor fluctuations, with a nearly 6-fold increased risk in patients younger than 60 years compared to those aged 80 or
5
older. In contrast, gender was the most important independent predictor of dyskinesias, with an almost 3-fold increased risk in females. This confirms previous findings of gender as a strong, individual predictor of dyskinesias, proposed to overcome potential protective effects of genetic factors such as dopamine receptor D2 polymorphisms [29]. It has been proposed that the genderdifference in risk of dyskinesias observed in our cohort and other studies could be due to lower body weight in females than males [30]. This is, however, not supported by our findings, suggesting that other factors, e.g. genetic predisposition, may be more important. In addition, we found no association between dyskinesias and age. Although this observation differs from many but not all [6] clinical trials, it is in line with previous population-based studies [16,17], possibly indicating that these inconsistencies could result from differences in study design. Current management of motor complications is based on oral drugs and, in severe cases, advanced therapies, such as parenteral apomorphine infusion, intrajejunal levodopa administration, or STN-DBS [8]. The optimal timing of and selection of patients to device-aided therapies is under discussion. The authors of the recent EARLYSTIM trial [9] concluded that STN-DBS is superior to medical therapy alone even in PD patients with early (<3 years) motor complications. On the other hand, our study demonstrates that in the vast majority of patients with early PD, motor complications are mild and non-disabling. Furthermore, in a substantial proportion of patients, motor fluctuations (37%) and dyskinesias (49%) reversed on conventional pharmacotherapy alone. Reversal was probably due to changes in medication type and dosing intervals, although our findings do not allow firm conclusions in this respect. A further caveat of our observation of substantial reversal is that follow-up was limited to the first 5 years of the disease, which may be regarded a relative limitation of our study given the slowly progressive nature of the disease. Hence, we do not know if and to which extent the observed reversal of early motor complications, particularly in elderly patients, might be only transient. As the majority of PD patients develop their disease beyond 60 years of age, these issues should be further investigated in future studies. Funding This study was supported by the Research Council of Norway (grant# 177966), the Western Norway Regional Health Authority (grant# 911218 and 911949), and the Norwegian Parkinson's disease Association. Sponsors had no role in study design; the collection, analysis and interpretation of data; in the writing of the report; or in the decision to submit the article for publication. Financial disclosures Anders Bjornestad reports no disclosures. Elin Bjelland Forsaa reports no disclosures. Kenn Freddy Pedersen reports no disclosures. Ole-Bjørn Tysnes reports no disclosures. Jan Petter Larsen reports no disclosures. Guido Alves has received payment for lecturing from AbbVie and research support from the Norwegian Parkinson's Disease Association. Author’s roles Study concept and design: Drs. Bjornestad, Tysnes, Larsen, Alves. Acquisition, analysis, or interpretation of data: All authors. Drafting of the manuscript: Dr. Bjornestad. Critical revision of the manuscript for important intellectual content: All authors.
Please cite this article in press as: A. Bjornestad, et al., Risk and course of motor complications in a population-based incident Parkinson's disease cohort, Parkinsonism and Related Disorders (2015), http://dx.doi.org/10.1016/j.parkreldis.2015.11.007
6
A. Bjornestad et al. / Parkinsonism and Related Disorders xxx (2015) 1e6
Statistical analysis: Drs. Bjornestad, Alves. Obtained funding: Dr. Larsen Administrative, technical, or material support: Drs. Tysnes, Larsen. Study supervision: Drs. Tysnes, Larsen. Dr. Bjornestad takes responsibility for the integrity of the data and the accuracy of the data analysis. All authors have approved the final article. Acknowledgments We are grateful to all patients for their participation in this study and thank the members of the Norwegian ParkWest study group and all other personnel involved in this study for their contributions. References [1] E.C. Hirsch, P. Jenner, S. Przedborski, Pathogenesis of Parkinson's disease, Movement Disorders: Official J. Mov. Disord. Soc. 28 (2013) 24e30. [2] S. Fahn, D. Oakes, I. Shoulson, K. Kieburtz, A. Rudolph, A. Lang, C.W. Olanow, C. Tanner, K. Marek, G. Parkinson Study, Levodopa and the progression of Parkinson's disease, N. Engl. J. Med. 351 (2004) 2498e2508. [3] S.H. Fox, A.E. Lang, 'Don't delay, start today': delaying levodopa does not delay motor complications, Brain: A J. Neurol. 137 (2014) 2628e2630. [4] P.A. Lewitt, Relief of parkinsonism and dyskinesia: one and the same dopaminergic mechanism? Neurology 74 (2010) 1169e1170. [5] M. Pechevis, C.E. Clarke, P. Vieregge, B. Khoshnood, C. Deschaseaux-Voinet, G. Berdeaux, M. Ziegler, G. Trial Study, Effects of dyskinesias in Parkinson's disease on quality of life and health-related costs: a prospective European study, Eur. J. Neurol.: Off. J. Eur. Fed. Neurol. Soc. 12 (2005) 956e963. [6] A. Schrag, N. Quinn, Dyskinesias and motor fluctuations in Parkinson's disease. A community-based study, Brain: A J. Neurol. 123 (2000) 2297e2305. [7] G. Parkinson Study, Pramipexole vs levodopa as initial treatment for Parkinson disease: A randomized controlled trial. Parkinson Study Group, JAMA: J. Am. Med. Assoc. 284 (2000) 1931e1938. [8] C.E. Clarke, P. Worth, D. Grosset, D. Stewart, Systematic review of apomorphine infusion, levodopa infusion and deep brain stimulation in advanced Parkinson's disease, Park. Relat. Disord. 15 (2009) 728e741. [9] W.M. Schuepbach, J. Rau, K. Knudsen, J. Volkmann, P. Krack, L. Timmermann, T.D. Halbig, H. Hesekamp, S.M. Navarro, N. Meier, D. Falk, M. Mehdorn, S. Paschen, M. Maarouf, M.T. Barbe, G.R. Fink, A. Kupsch, D. Gruber, G.H. Schneider, E. Seigneuret, A. Kistner, P. Chaynes, F. Ory-Magne, C. Brefel Courbon, J. Vesper, A. Schnitzler, L. Wojtecki, J.L. Houeto, B. Bataille, D. Maltete, P. Damier, S. Raoul, F. Sixel-Doering, D. Hellwig, A. Gharabaghi, R. Kruger, M.O. Pinsker, F. Amtage, J.M. Regis, T. Witjas, S. Thobois, P. Mertens, M. Kloss, A. Hartmann, W.H. Oertel, B. Post, H. Speelman, Y. Agid, C. Schade-Brittinger, G. Deuschl, Group ES, Neurostimulation for Parkinson's disease with early motor complications, N. Engl. J. Med. 368 (2013) 610e622. [10] P.F. Worth, When the going gets tough: how to select patients with Parkinson's disease for advanced therapies, Pract. Neurol. 13 (2013) 140e152. [11] G. Alves, B. Muller, K. Herlofson, I. HogenEsch, W. Telstad, D. Aarsland, O.B. Tysnes, J.P. Larsen, Norwegian ParkWest study g. Incidence of Parkinson's disease in Norway: the Norwegian ParkWest study, J. Neurol. Neurosurg. Psychiatry 80 (2009) 851e857. [12] A.J. Hughes, Y. Ben-Shlomo, S.E. Daniel, A.J. Lees, What features improve the accuracy of clinical diagnosis in Parkinson's disease: a clinicopathologic study,
Neurology 42 (1992) 1142e1146. [13] D.J. Gelb, E. Oliver, S. Gilman, Diagnostic criteria for Parkinson disease, Arch. Neurol. 56 (1999) 33e39. [14] S.E.R. Fahn, S. Fahn, R. Elton, Unified Parkinson's disease rating scale, in: S. Fahn, C.D. Marsden, D. Calne, M. Goldstein (Eds.), Recent Developments in Parkinson's Disease, Macmillan Healthc. Inf., 1987, pp. 153e163. [15] 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.: Off. J. Mov. Disord. Soc. 25 (2010) 2649e2653. [16] J.R. Evans, S.L. Mason, C.H. Williams-Gray, T. Foltynie, C. Brayne, T.W. Robbins, R.A. Barker, The natural history of treated Parkinson's disease in an incident, community based cohort, J. Neurol. Neurosurg. Psychiatry 82 (2011) 1112e1118. [17] N.M. Scott, A.D. Macloed, C.E. Counsell, Levodopa-induced motor complications in an incident Parkinson's disease cohort, J. Neurol. Neurosurg. Psychiatry 84 (e2) (2013) 6e7. [18] C.D. Marsden, J.D. Parkes, Success and problems of long-term levodopa therapy in Parkinson's disease, Lancet 1 (1977) 345e349. [19] A. Antonini, E. Tolosa, Y. Mizuno, M. Yamamoto, W.H. Poewe, A reassessment of risks and benefits of dopamine agonists in Parkinson's disease, Lancet neurol. 8 (2009) 929e937. [20] A.J. Lees, R. Katzenschlager, J. Head, Y. Ben-Shlomo, Ten-year follow-up of three different initial treatments in de-novo PD: a randomized trial, Neurology 57 (2001) 1687e1694. [21] O. Rascol, D.J. Brooks, A.D. Korczyn, P.P. De Deyn, C.E. Clarke, A.E. Lang, A fiveyear study of the incidence of dyskinesia in patients with early Parkinson's disease who were treated with ropinirole or levodopa, N. Engl. J. Med. 342 (2000) 1484e1491. [22] R. Katzenschlager, J. Head, A. Schrag, Y. Ben-Shlomo, A. Evans, A.J. Lees, Parkinson's Disease Research Group of the United K. Fourteen-year final report of the randomized PDRG-UK trial comparing three initial treatments in PD, Neurology 71 (2008) 474e480. [23] P.D.M.C. Group, R. Gray, N. Ives, C. Rick, S. Patel, A. Gray, C. Jenkinson, E. McIntosh, K. Wheatley, A. Williams, C.E. Clarke, Long-term effectiveness of dopamine agonists and monoamine oxidase B inhibitors compared with levodopa as initial treatment for Parkinson's disease (PD MED): a large, openlabel, pragmatic randomised trial, Lancet 384 (2014) 1196e1205. [24] S. Fox, N. Visanji, G. Reyes, P. Huot, J. Gomez-Ramirez, T.H. Johnston, et al., Development of psychosis-like behaviors and motor complications with de novo levodopa treatment in the MPTP primate model of Parkinson's disease, Can. J. Neurol. Sci. 37 (2010) 86e95. [25] T.A.J.T. Mestre, J.M. Brotchie, S.H. Fox, Evolution of the 'short-duration' response to L-DOPA in the MPTP lesioned non-human primate model of Parkinson's disease, Mov. Disord.: Off. J. Mov. Disord. Soc. 25 (2010) S417. [26] R. Cilia, A. Akpalu, F.S. Sarfo, M. Cham, M. Amboni, E. Cereda, M. Fabbri, P. Adjei, J. Akassi, A. Bonetti, G. Pezzoli, The modern pre-levodopa era of Parkinson's disease: insights into motor complications from sub-Saharan Africa, Brain: A J. Neurol. 137 (2014) 2731e2742. [27] J.Y. Hong, J.S. Oh, I. Lee, M.K. Sunwoo, J.H. Ham, J.E. Lee, Y.H. Sohn, J.S. Kim, P.H. Lee, Presynaptic dopamine depletion predicts levodopa-induced dyskinesia in de novo Parkinson disease, Neurology 82 (2014) 1597e1604. [28] P.A. Lewitt, M.M. Mouradian, Predicting the development of levodopainduced dyskinesias: a presynaptic mechanism? Neurology 82 (2014) 1574e1575. [29] M. Zappia, G. Annesi, G. Nicoletti, G. Arabia, F. Annesi, D. Messina, P. Pugliese, P. Spadafora, P. Tarantino, S. Carrideo, D. Civitelli, E.V. De Marco, I.C. CiroCandiano, A. Gambardella, A. Quattrone, Sex differences in clinical and genetic determinants of levodopa peak-dose dyskinesias in Parkinson disease: an exploratory study, Arch. Neurol. 62 (2005) 601e605. [30] J.C. Sharma, C.G. Bachmann, G. Linazasoro, Classifying risk factors for dyskinesia in Parkinson's disease, Park. Relat. Disord. 16 (2010) 490e497.
Please cite this article in press as: A. Bjornestad, et al., Risk and course of motor complications in a population-based incident Parkinson's disease cohort, Parkinsonism and Related Disorders (2015), http://dx.doi.org/10.1016/j.parkreldis.2015.11.007
Contents lists available at ScienceDirect
Parkinsonism and Related Disorders journal homepage: www.elsevier.com/locate/parkreldis
Risk and course of motor complications in a population-based incident Parkinson's disease cohort Anders Bjornestad a, d, *, Elin B. Forsaa a, d, Kenn Freddy Pedersen a, d, Ole-Bjorn Tysnes b, c, Jan Petter Larsen a, Guido Alves a, d a
The Norwegian Centre for Movement Disorders, Stavanger University Hospital, PO Box 8100, N-4068 Stavanger, Norway Department of Neurology, Haukeland University Hospital, PO Box 1400, N-5021 Bergen, Norway Institute of Clinical Medicine, University of Bergen, Norway d Department of Neurology, Stavanger University Hospital, Stavanger, Norway b c
a r t i c l e i n f o
a b s t r a c t
Article history: Received 3 September 2015 Received in revised form 9 October 2015 Accepted 2 November 2015
Background: Motor complications may become major challenges in the management of patients with Parkinson's disease. In this study, we sought to determine the incidence, risk factors, evolution, and treatment of motor fluctuations and dyskinesias in a population-representative, incident Parkinson's disease cohort. Methods: In this prospective population-based 5-year longitudinal study, we followed 189 incident and initially drug-naïve Parkinson's disease patients biannually for detailed examination of dyskinesias and motor fluctuations as defined by the Unified Parkinson's disease Rating Scale. We performed Kaplan eMeier survival and Cox regression analyses to assess cumulative incidence and risk factors of these motor complications. Results: The 5-year cumulative incidence of motor complications was 52.4%. Motor fluctuations occurred in 42.9% and dyskinesias in 24.3%. Besides higher motor severity predicting both motor fluctuations (p ¼ 0.016) and dyskinesias (p < 0.001), lower age at diagnosis predicted motor fluctuations (p ¼ 0.001), whereas female gender predicted dyskinesias (p ¼ 0.001). Actual levodopa dose at onset of motor fluctuations (p ¼ 0.037) or dyskinesias (p < 0.001) rather than initial treatment with levodopa (p > 0.1) independently predicted development of motor complications. Motor fluctuations reversed in 37% and dyskinesias in 49% of patients on oral treatment and remained generally mild in those with persistent complications. No patients received device-aided therapies during the study. Conclusions: More than 50% in the general Parkinson's disease population develop motor complications within 5 years of diagnosis. However, they remain mild in the vast majority and are reversible in a substantial proportion of patients. © 2015 Elsevier Ltd. All rights reserved.
Keywords: Parkinson's disease Motor complications Motor fluctuations Dyskinesias Risk factors
1. Introduction Parkinson disease (PD) is a relentlessly progressive movement disorder of unknown etiology [1]. With PD progression, the efficacy of pharmacotherapy deteriorates [2] and patients may develop motor complications, which can be broadly subdivided into motor
* Corresponding author. The Norwegian Centre for Movement Disorders, Stavanger University Hospital, PO Box 8100, N-4068 Stavanger, Norway. E-mail addresses: [email protected] (A. Bjornestad), elin.bjelland. [email protected] (E.B. Forsaa), [email protected] (K.F. Pedersen), [email protected] (O.-B. Tysnes), [email protected] (J.P. Larsen), [email protected] (G. Alves).
fluctuations and dyskinesias. The risk for and time to emergence of these motor complications vary substantially among patients for reasons that are probably complex, including both disease- and drug-related factors, particularly treatment with levodopa [3,4]. Although motor fluctuations and dyskinesias may compromise quality of life [5], their evolution and prognosis are still incompletely investigated, as most previous studies assessed crosssectional cohorts with established disease [6] or followed patients recruited from movement disorders centers or clinical trials [7]. This may result in both under- and overestimation of the risk of these motor complications. In addition, as treatment options have changed during the last decades, previous estimates might no longer be accurate. Finally, device-aided therapies have emerged as
http://dx.doi.org/10.1016/j.parkreldis.2015.11.007 1353-8020/© 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: A. Bjornestad, et al., Risk and course of motor complications in a population-based incident Parkinson's disease cohort, Parkinsonism and Related Disorders (2015), http://dx.doi.org/10.1016/j.parkreldis.2015.11.007
2
A. Bjornestad et al. / Parkinsonism and Related Disorders xxx (2015) 1e6
second-line treatments in patients with otherwise intractable motor complications [8,9]. Recent studies promote early initiation of these advanced therapies [9,10], which e however e are costly and may cause serious adverse events. Therefore, precise information on the evolution and potential reversal of motor fluctuations and dyskinesias using conventional pharmacotherapy is crucial for optimal patient selection to and timing of second-line treatment in PD. To gain this important knowledge, we recruited a large, population-based, and initially drug-naive PD cohort that we monitored closely for emergence of motor complications for five years from diagnosis. 2. Methods 2.1. Subjects All subjects participate in the Norwegian ParkWest project, a prospective, population-based, multicenter, longitudinal cohort study investigating the incidence, neurobiology and prognosis of PD. In order to establish a population-representative, incident PD cohort, we recruited patients with newly-diagnosed and untreated PD from the general population between November, 2004, and August, 2006, using multiple recruitment strategies, as previously described [11]. Briefly, all patients were screened, followed and treated by neurologists from the study group, who are experienced in movement disorders. Diagnostic procedures at baseline included a full medical history, comprehensive clinical, neuropsychological and neuropsychiatric assessments, laboratory tests, cerebral MRI, and, if indicated clinically, dopamine transporter imaging. Standardized follow-up visits were conducted every six months. Of 212 patients included in the original cohort, five were not drug-naïve at baseline and further 18 rediagnosed during follow-up, leaving 189 patients eligible for this study. All met widely acknowledged research criteria [12,13] for PD at latest follow-up. The study was approved by the Regional Committee for Medical Research Ethics, Western Norway. Signed written consent was obtained from all participants. 2.2. Evaluation program and treatment At baseline, a study neurologist performed general medical and neurological examinations and semi-structured interviews to obtain medical and drug history. Motor severity and disease stage in the drug-naïve state were rated using the Unified PD Rating Scale (UPDRS) [14] motor section (part III) and Hoehn and Yahr staging. Assessment of non-motor features included the Mini-Mental State Examination (MMSE), Montgomery Aasberg Depression Rating Scale (MADRS), Fatigue Severity Scale (FSS), PD Sleep Scale (PDSS), Epworth Sleepiness Scale (ESS), and Starkstein Apathy Scale (SAS). Following baseline examinations, antiparkinsonian treatment was initiated and adjusted throughout follow-up by a study neurologist according to best clinical judgement. Drug changes in between study visits were allowed when deemed clinically necessary. Motor complications were assessed at 6-month intervals by clinical observation and the UPDRS complication of therapy section (part IV) with a score of 1 on items 36, 37, 38 or 39 defining presence of motor fluctuations and a score of 1 on items 32, 33 or 34 defining presence of dyskinesias. Type and dose of medications were recorded at each study visit, and levodopaequivalent doses (LED) were calculated according to published recommendations [15]. Use of advanced therapies was also recorded, defined as subcutaneous apomorphine administration, continuous intestinal levodopa infusion, or subthalamic nucleus deep brain stimulation (STN-DBS), which all are provided at no personal or private insurance company costs in Norway.
2.3. Statistical analysis SPSS 21.0 was used for statistical analysis. Incidence rates with 95% confidence intervals were calculated and expressed in relation to 1000 person-years of observation. Between-group differences were compared using t-tests, ManneWhitney U-tests and c2-tests as appropriate. We performed KaplaneMeier survival analysis to evaluate cumulative incidence rates of motor fluctuations and dyskinesias, and Cox regression analysis to assess baseline factors associated with evolution of these motor complications. Separate analysis of wearing-off was performed without significant changes in results, thus all motor fluctuations are presented as one group in the following. We considered age, gender, time since motor onset, and motor severity at baseline as primary risk factors of interest. In subsequent models adjusting for these variables, we also included a panel of non-motor symptoms (as these have been associated with motor complications in cross-sectional studies) assessed at baseline using MMSE, MADRS, FSS, PDSS, ESS, and SAS. Finally, we investigated the associations of initial treatment (levodopa vs. dopamine agonists) as well as actual LED at onset of motor complications with the development of motor fluctuations and dyskinesias. Two-tailed p-values <0.05 were considered significant. 3. Results 3.1. Baseline characteristics and subject flow Baseline characteristics are given in Table 1 and the flow of subjects in Fig. 1. Of the 189 patients included at baseline, 183 (97%) completed 1 year, 173 (92%) 3 years and 158 (84%) 5 years of biannual follow-up, generating 1911 observations in total. Few subjects (7/189, 4%) withdrew from the study, whereas remaining loss to follow-up (24/189, 13%) was due to death. Median follow-up time was 5.0 (interquartile range 4.9e5.1) years. 3.2. Frequency of motor complications Point prevalence rates increased progressively, reaching 38.0% for any motor complications, 31.0% for motor fluctuations, and 12.7% for dyskinesias at 5 years of follow-up (Table 2). The 5-year cumulative incidence of any motor complications was 52.4%, with a corresponding incidence rate of 158 (95% CI 132e189) per 1000 person-years of observation. For motor fluctuations, the cumulative Table 1 Baseline characteristics of 189 patients with incident, drug-naïve PD. Characteristics
Values
Number Male, n (%) Age, years Time since diagnosis, months Symptom duration, years UPDRS motor score Hoehn and Yahr stage MMSE score MADRS score FSS score PDSS score ESS score SAS score Body weight, kilograms
189 114 (60.3) 67.7 (9.3) 1.6 (1.7) 2.3 (1.8) 23.4 (11.3) 1.9 (0.6) 27.8 (2.5) 4.7 (5.2) 4.4 (1.6) 118.9 (19.6) 6.0 (3.6) 15.6 (4.8) 76.5 (14.4)
Values are mean (SD) if not otherwise indicated. UPDRS ¼ Unified Parkinson's Disease Rating Scale; MMSE ¼ Mini-Mental State Examination; MADRS ¼ Montgomery Aasberg Depression Rating Scale; FSS ¼ Fatigue Severity Scale; PDSS ¼ Parkinson's Disease Sleep Scale; ESS ¼ Epworth Sleepiness Scale; SAS ¼ Starkstein Apathy Scale.
Please cite this article in press as: A. Bjornestad, et al., Risk and course of motor complications in a population-based incident Parkinson's disease cohort, Parkinsonism and Related Disorders (2015), http://dx.doi.org/10.1016/j.parkreldis.2015.11.007
A. Bjornestad et al. / Parkinsonism and Related Disorders xxx (2015) 1e6
3
Fig. 1. Study flowchart. PD ¼ Parkinson disease. Patients were assessed at 6-month intervals; flowchart is simplified for readability.
Table 2 Prevalence and cumulative incidence rates of motor complications, and type and dose of dopaminergic drugs, during follow-up.
Motor complications overall Point prevalence, % Cumulative incidence, % Motor fluctuations Point prevalence, % Cumulative incidence, % Dyskinesias Point prevalence, % Cumulative incidence, % Concomitant DK and MF Point prevalence, % Cumulative incidence, % Treatment Levodopa, % Dopamine agonist, % Other medications, % Total LED (SD)
Baseline
1 year
2 years
3 years
4 years
5 years
0 0
10.9 14.3
13.6 24.3
22.8 32.8
23.2 40.7
38.0 52.4
0 0
7.7 10.6
10.2 19.0
18.1 26.5
17.1 33.9
31.0 42.9
0 0
3.8 6.3
4.5 9.5
5.2 12.7
9.8 18.0
12.7 24.3
0 0
0.5 2.6
1.1 4.2
0.5 6.3
3.7 11.1
5.7 14.8
0 0 0 0
41.5 44.8 23.5 252 (169)
56.8 47.7 30.7 333 (161)
67.1 48.6 33.5 424 (224)
78.7 53.7 29.9 514 (250)
84.2 53.8 32.9 587 (334)
Patients were assessed at 6-month intervals. Only 12-months findings are displayed for simplicity. DK ¼ dyskinesias. MF ¼ motor fluctuations. LED ¼ levodopa equivalent dose. SD ¼ standard deviation.
incidence at 5 years of follow-up was 42.9%, yielding an incidence rate of 121 (99e148) per 1000 person-years. For dyskinesias, the 5year cumulative incidence was 24.3%, with a corresponding incidence rate of 59 (45e78) per 1000 person-years. 14.8% of the total population and 28.3% of those with motor complications experienced both motor fluctuations and dyskinesias. 3.3. Baseline risk factors As our cohort was drug-naïve at baseline, we first sought to identify clinical and demographic risk factors that could predict motor complications before dopaminergic medication was initiated. In multivariate Cox regression analysis, lower age at diagnosis (hazard ratio (HR) per year 0.97 (0.95e0.99), p ¼ 0.013), female gender (HR 1.84 (1.23e2.77), p ¼ 0.003), and higher UPDRS motor score at baseline (HR per unit 1.04 (1.01e1.06), p ¼ 0.001) independently predicted a higher risk of developing motor complications during follow-up, whereas time since motor onset did not (HR
per year 0.97 (0.86e1.09), p ¼ 0.593). No additional independent risk factors were identified among the assessed non-motor features (all p > 0.05). The risk factor pattern differed between motor fluctuations and dyskinesias when these were tested in separate Cox regression models. Independent baseline risk factors of motor fluctuations were lower age (HR per year 0.96 (0.94e0.98), p ¼ 0.001) and higher UPDRS motor score (HR per unit 1.03 (1.01e1.05), p ¼ 0.016) but not gender (HR females vs. males 1.48 (0.94e2.33), p ¼ 0.087) or time since motor onset (HR per year 1.03 (0.91e1.16, p ¼ 0.675). For example, the cumulative 5-year incidence of motor fluctuations in patients aged >80 was 11% while it was 41% for patients aged 60e79 and 64% for patients below 60 years at baseline (Fig. 2a). In contrast, independent risk factors of dyskinesias were female gender (HR 2.73 (1.49e4.98), p ¼ 0.001, Fig. 2b) and higher baseline UPDRS motor score (HR per unit 1.05 (1.02e1.08), p ¼ 0.001) but not age (HR per year 0.98 (0.95e1.02), p ¼ 0.262) or time since motor onset (HR per year 0.91 (0.75e1.10), p ¼ 0.332). Again, none
Please cite this article in press as: A. Bjornestad, et al., Risk and course of motor complications in a population-based incident Parkinson's disease cohort, Parkinsonism and Related Disorders (2015), http://dx.doi.org/10.1016/j.parkreldis.2015.11.007
4
A. Bjornestad et al. / Parkinsonism and Related Disorders xxx (2015) 1e6
therapies by their study neurologists during follow-up. A subgroup of patients (36/189) never used levodopa. Among these, only 5 (13.9%) experienced motor fluctuations, and 4 (11.1%) had dyskinesias. The corresponding rates in levodopa-treated patients were 49.7% and 27.5%, respectively. In multivariate Cox regression analysis (with adjustment for baseline age, gender, time since motor onset, and motor severity), initial treatment with levodopa was associated with an increased risk of developing motor fluctuations (HR 1.84 (1.09e3.10), p ¼ 0.023) but not dyskinesias (HR 0.88 (0.42e1.85), p ¼ 0.744), compared to initial treatment with dopamine agonists. However, when actual levodopa dose at onset of motor complications was also included in the models, the association between initial levodopa treatment and motor fluctuations was lost (HR 1.37 (0.87e2.87), p ¼ 0.137), whereas actual levodopa dose was independently associated with both motor fluctuations (HR per 100 mg 1.13 (1.01e1.26), p ¼ 0.037) and dyskinesias (HR per 100 mg 1.28 (1.16e1.42), p < 0.001). 3.5. Course of motor complications 3.5.1. Severity Motor complications remained generally mild throughout the study period (esupp table). Motor fluctuations were predictable in the majority of patients, with no more than 5.1% experiencing unpredictable or sudden off periods, respectively, during follow-up. Few patients (5.1%) spent larger proportions (>25%) of their day in off. Similarly, few patients (3.8%) experienced dyskinesias larger proportions of their day and very few rated their dyskinesias as severe (0.6%) or painful (1.8%) within the first 5 years of diagnosis.
Fig. 2. KaplaneMeier plots illustrating the time to (A) incident motor fluctuations stratified by age, and (B) incident dyskinesias stratified by gender.
of the assessed non-motor features at baseline independently predicted motor fluctuations or dyskinesias (all p > 0.05). To explore whether the observed gender difference in the risk of dyskinesias was related to lower body weight (and thus a higher levodopa dose per kg) in females than males, we ran a supplemental model with body weight included as an independent variable, adjusting for age, gender, time since motor onset, and motor severity at baseline. In this model, female gender (HR 2.34 (1.21e4.53, p ¼ 0.011) remained associated with dyskinesias, whereas body weight was not (HR 0.99 per kg (0.96e1.01), p ¼ 0.239). 3.4. Medication effects Patients started dopaminergic treatment following their baseline visits. At 1 year of follow-up, 41.5% of patients received levodopa, 44.8% dopamine agonists, and 23.5% other dopaminergic medications. Among those who started a dopamine agonist, only 26% did not use levodopa at end of follow-up. Overall, 84.2% of those attending the 5-year visit were on levodopa treatment. No patients were considered in need of advanced, device-aided
3.5.2. Reversal To assess the potential reversal of motor complications over time, we considered the 158 patients with complete 5-year followup. Of 78 patients who had experienced motor fluctuations during follow-up, 29 did not so at study end, indicating reversal in 37%. Patients with persistent motor fluctuations were younger than those in whom motor fluctuations reversed on oral treatment (62.3 (SD 9.1) vs. 68.8 (SD 6.9) years, p ¼ 0.001), but did not differ in gender, baseline motor severity, or LED at final visit (all p > 0.05). Dyskinesias were experienced during the study period by 39 of 158 patients with complete follow-up. Nineteen of the 39 had no dyskinesias at final visit, indicating reversal in 49%. Patients whose dyskinesias reversed did not differ from those with persistent dyskinesias in terms of age, gender, baseline motor severity or LED at final visit (all p > 0.05). 4. Discussion Our study highlights a high risk of early motor complications in the general PD population and also provides important insights into their evolution and associated risk factors, with implications for both understanding and management of motor fluctuations and dyskinesias in PD. We are aware of only two previous longitudinal studies of dyskinesias [16] or motor complications [17] in patients with early PD derived from the community. Our study differs importantly from these in several ways, including the entirely drug-naïve cohort at study start, the range of potential demographic and clinical risk factors assessed, the close monitoring of motor complications every six months, and the low attrition rate during follow-up, generating more than 1900 observations over the 5-year study period. In addition, while our patients were recruited unselected from the community, treatment and follow-up was managed
Please cite this article in press as: A. Bjornestad, et al., Risk and course of motor complications in a population-based incident Parkinson's disease cohort, Parkinsonism and Related Disorders (2015), http://dx.doi.org/10.1016/j.parkreldis.2015.11.007
A. Bjornestad et al. / Parkinsonism and Related Disorders xxx (2015) 1e6
by movement disorders neurologists, thus combining the advantages of a population-based design with the benefits of specialist care, including presumed high diagnostic accuracy and up-to-date treatment. We therefore believe that our findings are as valid and representative for the general PD population as possible. Our study demonstrates that motor complications develop early and frequently in the general PD population, affecting more than half of patients within 5 years of diagnosis. The high occurrence of motor complications was driven mainly by a high incidence of motor fluctuations, which were experienced by over 40% during follow-up. This figure is not very different from estimates provided almost four decades ago [18], which may appear surprising given the now widespread use of dopamine agonists and other levodopasparing drugs. Dopamine agonists are, however, generally considered less potent and to have more side effects than levodopa [19]. In line with this, we observed a steady decline in the number of patients who were able to remain on dopamine agonist monotherapy, while the proportion of patients on levodopa increased to nearly 85% by 5 years of follow-up. A number of randomized controlled trials have demonstrated that levodopa treatment shortens the time to onset of motor complications in PD [20,21], thus leading many physicians to prefer dopamine agonists as initial treatment, at least in patients below 65e70 years of age. However, more recent clinical trials with extended follow-up show that the initial treatment has little impact on the long-term risk of motor fluctuations and dyskinesias [22,23]. Indeed, both experimental [24,25] and clinical [26] studies suggest that levodopa dose is more important for the development of motor complications than the duration of levodopa treatment. Extending these observations, we found that initial treatment with levodopa doubled the risk for motor fluctuations, but this association was lost once the actual levodopa dose at onset of motor complications was taken into consideration, with the latter remaining significantly associated with both motor fluctuations and dyskinesias. This implies that actual levodopa dose rather than initial treatment with levodopa is crucial for the development of motor complications. Overall, these data indicate that levodopa treatment in low doses should not be withheld in patients with early PD. In addition to treatment with levodopa, the importance of disease-related factors for the development of motor complications has been debated. Our patients were drug-naïve at baseline, providing a unique opportunity to study the association of diseaserelated factors and motor complications unbiased by dopaminergic treatment. We found that more severe parkinsonism at diagnosis (in the drug-naïve state) independently predicted early development of motor fluctuations and dyskinesias, suggesting that dopamine depletion is a principal underlying cause of both of these motor complications. This observation is in line with and extends recent neuroimaging findings that point towards an important role of presynaptic mechanisms in the etiology of dyskinesias in PD [27,28]. While motor fluctuations developed frequently in our cohort, this was not the case for dyskinesias, which affected less than one quarter of patients during the 5-year study period. The proportion of patients who experienced both motor fluctuations and dyskinesias during follow-up was only 15% of the total cohort and 28% of those with motor complications. This limited overlap of motor fluctuations and dyskinesias is remarkable and may indicate partly distinct underlying biological mechanisms, at least in early PD. This is further supported by our observation that the demographic risk factors differed between motor fluctuations and dyskinesias. We found that lower age but not gender was a strong and independent predictor of motor fluctuations, with a nearly 6-fold increased risk in patients younger than 60 years compared to those aged 80 or
5
older. In contrast, gender was the most important independent predictor of dyskinesias, with an almost 3-fold increased risk in females. This confirms previous findings of gender as a strong, individual predictor of dyskinesias, proposed to overcome potential protective effects of genetic factors such as dopamine receptor D2 polymorphisms [29]. It has been proposed that the genderdifference in risk of dyskinesias observed in our cohort and other studies could be due to lower body weight in females than males [30]. This is, however, not supported by our findings, suggesting that other factors, e.g. genetic predisposition, may be more important. In addition, we found no association between dyskinesias and age. Although this observation differs from many but not all [6] clinical trials, it is in line with previous population-based studies [16,17], possibly indicating that these inconsistencies could result from differences in study design. Current management of motor complications is based on oral drugs and, in severe cases, advanced therapies, such as parenteral apomorphine infusion, intrajejunal levodopa administration, or STN-DBS [8]. The optimal timing of and selection of patients to device-aided therapies is under discussion. The authors of the recent EARLYSTIM trial [9] concluded that STN-DBS is superior to medical therapy alone even in PD patients with early (<3 years) motor complications. On the other hand, our study demonstrates that in the vast majority of patients with early PD, motor complications are mild and non-disabling. Furthermore, in a substantial proportion of patients, motor fluctuations (37%) and dyskinesias (49%) reversed on conventional pharmacotherapy alone. Reversal was probably due to changes in medication type and dosing intervals, although our findings do not allow firm conclusions in this respect. A further caveat of our observation of substantial reversal is that follow-up was limited to the first 5 years of the disease, which may be regarded a relative limitation of our study given the slowly progressive nature of the disease. Hence, we do not know if and to which extent the observed reversal of early motor complications, particularly in elderly patients, might be only transient. As the majority of PD patients develop their disease beyond 60 years of age, these issues should be further investigated in future studies. Funding This study was supported by the Research Council of Norway (grant# 177966), the Western Norway Regional Health Authority (grant# 911218 and 911949), and the Norwegian Parkinson's disease Association. Sponsors had no role in study design; the collection, analysis and interpretation of data; in the writing of the report; or in the decision to submit the article for publication. Financial disclosures Anders Bjornestad reports no disclosures. Elin Bjelland Forsaa reports no disclosures. Kenn Freddy Pedersen reports no disclosures. Ole-Bjørn Tysnes reports no disclosures. Jan Petter Larsen reports no disclosures. Guido Alves has received payment for lecturing from AbbVie and research support from the Norwegian Parkinson's Disease Association. Author’s roles Study concept and design: Drs. Bjornestad, Tysnes, Larsen, Alves. Acquisition, analysis, or interpretation of data: All authors. Drafting of the manuscript: Dr. Bjornestad. Critical revision of the manuscript for important intellectual content: All authors.
Please cite this article in press as: A. Bjornestad, et al., Risk and course of motor complications in a population-based incident Parkinson's disease cohort, Parkinsonism and Related Disorders (2015), http://dx.doi.org/10.1016/j.parkreldis.2015.11.007
6
A. Bjornestad et al. / Parkinsonism and Related Disorders xxx (2015) 1e6
Statistical analysis: Drs. Bjornestad, Alves. Obtained funding: Dr. Larsen Administrative, technical, or material support: Drs. Tysnes, Larsen. Study supervision: Drs. Tysnes, Larsen. Dr. Bjornestad takes responsibility for the integrity of the data and the accuracy of the data analysis. All authors have approved the final article. Acknowledgments We are grateful to all patients for their participation in this study and thank the members of the Norwegian ParkWest study group and all other personnel involved in this study for their contributions. References [1] E.C. Hirsch, P. Jenner, S. Przedborski, Pathogenesis of Parkinson's disease, Movement Disorders: Official J. Mov. Disord. Soc. 28 (2013) 24e30. [2] S. Fahn, D. Oakes, I. Shoulson, K. Kieburtz, A. Rudolph, A. Lang, C.W. Olanow, C. Tanner, K. Marek, G. Parkinson Study, Levodopa and the progression of Parkinson's disease, N. Engl. J. Med. 351 (2004) 2498e2508. [3] S.H. Fox, A.E. Lang, 'Don't delay, start today': delaying levodopa does not delay motor complications, Brain: A J. Neurol. 137 (2014) 2628e2630. [4] P.A. Lewitt, Relief of parkinsonism and dyskinesia: one and the same dopaminergic mechanism? Neurology 74 (2010) 1169e1170. [5] M. Pechevis, C.E. Clarke, P. Vieregge, B. Khoshnood, C. Deschaseaux-Voinet, G. Berdeaux, M. Ziegler, G. Trial Study, Effects of dyskinesias in Parkinson's disease on quality of life and health-related costs: a prospective European study, Eur. J. Neurol.: Off. J. Eur. Fed. Neurol. Soc. 12 (2005) 956e963. [6] A. Schrag, N. Quinn, Dyskinesias and motor fluctuations in Parkinson's disease. A community-based study, Brain: A J. Neurol. 123 (2000) 2297e2305. [7] G. Parkinson Study, Pramipexole vs levodopa as initial treatment for Parkinson disease: A randomized controlled trial. Parkinson Study Group, JAMA: J. Am. Med. Assoc. 284 (2000) 1931e1938. [8] C.E. Clarke, P. Worth, D. Grosset, D. Stewart, Systematic review of apomorphine infusion, levodopa infusion and deep brain stimulation in advanced Parkinson's disease, Park. Relat. Disord. 15 (2009) 728e741. [9] W.M. Schuepbach, J. Rau, K. Knudsen, J. Volkmann, P. Krack, L. Timmermann, T.D. Halbig, H. Hesekamp, S.M. Navarro, N. Meier, D. Falk, M. Mehdorn, S. Paschen, M. Maarouf, M.T. Barbe, G.R. Fink, A. Kupsch, D. Gruber, G.H. Schneider, E. Seigneuret, A. Kistner, P. Chaynes, F. Ory-Magne, C. Brefel Courbon, J. Vesper, A. Schnitzler, L. Wojtecki, J.L. Houeto, B. Bataille, D. Maltete, P. Damier, S. Raoul, F. Sixel-Doering, D. Hellwig, A. Gharabaghi, R. Kruger, M.O. Pinsker, F. Amtage, J.M. Regis, T. Witjas, S. Thobois, P. Mertens, M. Kloss, A. Hartmann, W.H. Oertel, B. Post, H. Speelman, Y. Agid, C. Schade-Brittinger, G. Deuschl, Group ES, Neurostimulation for Parkinson's disease with early motor complications, N. Engl. J. Med. 368 (2013) 610e622. [10] P.F. Worth, When the going gets tough: how to select patients with Parkinson's disease for advanced therapies, Pract. Neurol. 13 (2013) 140e152. [11] G. Alves, B. Muller, K. Herlofson, I. HogenEsch, W. Telstad, D. Aarsland, O.B. Tysnes, J.P. Larsen, Norwegian ParkWest study g. Incidence of Parkinson's disease in Norway: the Norwegian ParkWest study, J. Neurol. Neurosurg. Psychiatry 80 (2009) 851e857. [12] A.J. Hughes, Y. Ben-Shlomo, S.E. Daniel, A.J. Lees, What features improve the accuracy of clinical diagnosis in Parkinson's disease: a clinicopathologic study,
Neurology 42 (1992) 1142e1146. [13] D.J. Gelb, E. Oliver, S. Gilman, Diagnostic criteria for Parkinson disease, Arch. Neurol. 56 (1999) 33e39. [14] S.E.R. Fahn, S. Fahn, R. Elton, Unified Parkinson's disease rating scale, in: S. Fahn, C.D. Marsden, D. Calne, M. Goldstein (Eds.), Recent Developments in Parkinson's Disease, Macmillan Healthc. Inf., 1987, pp. 153e163. [15] 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.: Off. J. Mov. Disord. Soc. 25 (2010) 2649e2653. [16] J.R. Evans, S.L. Mason, C.H. Williams-Gray, T. Foltynie, C. Brayne, T.W. Robbins, R.A. Barker, The natural history of treated Parkinson's disease in an incident, community based cohort, J. Neurol. Neurosurg. Psychiatry 82 (2011) 1112e1118. [17] N.M. Scott, A.D. Macloed, C.E. Counsell, Levodopa-induced motor complications in an incident Parkinson's disease cohort, J. Neurol. Neurosurg. Psychiatry 84 (e2) (2013) 6e7. [18] C.D. Marsden, J.D. Parkes, Success and problems of long-term levodopa therapy in Parkinson's disease, Lancet 1 (1977) 345e349. [19] A. Antonini, E. Tolosa, Y. Mizuno, M. Yamamoto, W.H. Poewe, A reassessment of risks and benefits of dopamine agonists in Parkinson's disease, Lancet neurol. 8 (2009) 929e937. [20] A.J. Lees, R. Katzenschlager, J. Head, Y. Ben-Shlomo, Ten-year follow-up of three different initial treatments in de-novo PD: a randomized trial, Neurology 57 (2001) 1687e1694. [21] O. Rascol, D.J. Brooks, A.D. Korczyn, P.P. De Deyn, C.E. Clarke, A.E. Lang, A fiveyear study of the incidence of dyskinesia in patients with early Parkinson's disease who were treated with ropinirole or levodopa, N. Engl. J. Med. 342 (2000) 1484e1491. [22] R. Katzenschlager, J. Head, A. Schrag, Y. Ben-Shlomo, A. Evans, A.J. Lees, Parkinson's Disease Research Group of the United K. Fourteen-year final report of the randomized PDRG-UK trial comparing three initial treatments in PD, Neurology 71 (2008) 474e480. [23] P.D.M.C. Group, R. Gray, N. Ives, C. Rick, S. Patel, A. Gray, C. Jenkinson, E. McIntosh, K. Wheatley, A. Williams, C.E. Clarke, Long-term effectiveness of dopamine agonists and monoamine oxidase B inhibitors compared with levodopa as initial treatment for Parkinson's disease (PD MED): a large, openlabel, pragmatic randomised trial, Lancet 384 (2014) 1196e1205. [24] S. Fox, N. Visanji, G. Reyes, P. Huot, J. Gomez-Ramirez, T.H. Johnston, et al., Development of psychosis-like behaviors and motor complications with de novo levodopa treatment in the MPTP primate model of Parkinson's disease, Can. J. Neurol. Sci. 37 (2010) 86e95. [25] T.A.J.T. Mestre, J.M. Brotchie, S.H. Fox, Evolution of the 'short-duration' response to L-DOPA in the MPTP lesioned non-human primate model of Parkinson's disease, Mov. Disord.: Off. J. Mov. Disord. Soc. 25 (2010) S417. [26] R. Cilia, A. Akpalu, F.S. Sarfo, M. Cham, M. Amboni, E. Cereda, M. Fabbri, P. Adjei, J. Akassi, A. Bonetti, G. Pezzoli, The modern pre-levodopa era of Parkinson's disease: insights into motor complications from sub-Saharan Africa, Brain: A J. Neurol. 137 (2014) 2731e2742. [27] J.Y. Hong, J.S. Oh, I. Lee, M.K. Sunwoo, J.H. Ham, J.E. Lee, Y.H. Sohn, J.S. Kim, P.H. Lee, Presynaptic dopamine depletion predicts levodopa-induced dyskinesia in de novo Parkinson disease, Neurology 82 (2014) 1597e1604. [28] P.A. Lewitt, M.M. Mouradian, Predicting the development of levodopainduced dyskinesias: a presynaptic mechanism? Neurology 82 (2014) 1574e1575. [29] M. Zappia, G. Annesi, G. Nicoletti, G. Arabia, F. Annesi, D. Messina, P. Pugliese, P. Spadafora, P. Tarantino, S. Carrideo, D. Civitelli, E.V. De Marco, I.C. CiroCandiano, A. Gambardella, A. Quattrone, Sex differences in clinical and genetic determinants of levodopa peak-dose dyskinesias in Parkinson disease: an exploratory study, Arch. Neurol. 62 (2005) 601e605. [30] J.C. Sharma, C.G. Bachmann, G. Linazasoro, Classifying risk factors for dyskinesia in Parkinson's disease, Park. Relat. Disord. 16 (2010) 490e497.
Please cite this article in press as: A. Bjornestad, et al., Risk and course of motor complications in a population-based incident Parkinson's disease cohort, Parkinsonism and Related Disorders (2015), http://dx.doi.org/10.1016/j.parkreldis.2015.11.007