Abstract
INTRODUCTION
Level of handicap in Parkinson‘s disease (PD) has rarely been assessed. Previously, we have measured the level of handicap experienced by late stage PD (LS-PD) patients, and we found that handicap was a valuable patient-centred outcome (PCO) in LS-PD and that these patients were severely handicapped due to dementia and axial motor symptoms resistant to levodopa [1, 2]. This raised the question whether handicap would also be a valuable measure of the health condition and participation of an individual in patients with disabling levodopa-induced motor complications (MC).
The World Health Organization (WHO) has recently replaced the term handicap with participation restriction to avoid any negative connotations associated with the term handicap and to emphasize that the ability of an individual to participate in everyday life situations is limited, not only by the effects of disease, but also contextual factors [3]. Nevertheless, the operational definition of handicap still holds true for clinical research since it was found that the 7 domains of participation that can be potentially affected by health states correspond to the 6 common dimensions of handicap [3, 4]. Indeed, instruments that were developed to measure handicap, such as the London Handicap Scale (LHS), accurately measure the WHO definition of participation restriction [4, 5]. Furthermore, handicap is a closely defined and focused concept which retains the subjective perspective and social interaction context of quality of life (QoL), and it is easily understandable to patients and caregivers [6]. Finally, there is good cross-cultural agreement on the construct of handicap allowing for comparison between different populations and different diseases [2, 8]. Reducing handicap is a central aim of therapeutic interventions, and a better understanding of the causes of handicap allows for improved adjustment of interventions and assessment of their effectiveness [6].
Our aim was to study the handicap in advanced stage PD, which is widely defined as PD in patients manifesting levodopa-induced MC [9]. To enrich our sample with patients suffering from very disabling levodopa-induced MC, we enrolled PD patients selected to receive deep brain stimulation (DBS).
PATIENTS AND METHODS
Objective
To establish the pattern and level of handicap in advanced stage PD patients with disabling levodopa-induced MC selected to DBS, and to identify the contributing factors.
Participants
PD patients with disabling levodopa-induced MC who were given DBS therapy in our centre from 2006 to 2015 were consecutively included in the study. Additional inclusion criteria for the study were the completion of LHS. PD was diagnosed according to the UK Parkinson’s Disease Society Brain Bank Criteria [10]. Eligibility criteria for DBS were: a) PD patients with disabling levodopa-induced MC refractory to best medical therapy; b) age below 70 years; c) a positive motor response (>33% improvement) to a levodopa challenge test (a supra-maximal dose of 150% of the patients’ usual morning levodopa equivalent dose); d) lack of postural instability and freezing of gait during on period; e) absence of dementia according to a formal neuropsychological assessment and unstable psychiatric disorder according to a formal psychiatry assessment; and f) lack of significant brain changes, namely atrophy or concurrent brain disorders such as vascular lesions on CT and MRI scan. Participants gave their informed consent to participate and the study was conducted according to the Declaration of Helsinki.
Participants’ assessment and study design
Data was prospectively recorded during routine assessment for DBS, regarding handicap of patients, motor, cognitive, psychiatric and imaging evaluation. The results of this study pertain to the cross-sectional assessment before DBS. Handicap was assessed using the LHS which consists of a self-completed questionnaire and it takes around 10 min to be completed [2]. The questionnaire has six questions, one for each domain of handicap (mobility, physical independence, occupation, social integration, orientation and economic self-sufficiency), and each question contains six sentences hierarchically describing the level of handicap; for each question, the patient must choose the most suitable sentence. Each sentence is assigned a scale weight. The questionnaire comprises a matrix of scale weights which when combined give a total score for handicap, to which a constant value of 0.456 is added. The final score ranges from 0 (maximal handicap) to 1 (no handicap).
Severity of motor symptoms were evaluated using the motor part (part III) of the Unified Parkinson’s Disease Rating Scale (UPDRS) and more recently the motor part (part III) of the Movement Disorders Society-UPDRS (MDS-UPDRS), and the Hoehn & Yahr (HY) scale [11, 12]. Part III of the UPDRS∖MDS-UPDRS and the HY staging were assessed during off period and after the intake of a supra-maximal dose of levodopa (see above) during the best on period. Off period assessment was performed at least 12 hours after the last levodopa dose, 48 hours after the last intake of dopamine agonists, controlled-release levodopa, selegiline or rasagiline and 12 hours after the last intake of entacapone. The levodopa equivalent daily dose (LEDD) was calculated according to reported conversions [13]. Non-motor symptoms were assessed using Part I of UPDRS∖MDS-UPDRS, and disability and independence in activities of daily living (ADL) were evaluated using Part II of UPDRS∖MDS-UPDRS and the Schwab and England Scale (S&E) [14]. The scores of UPDRS parts II and III were converted to the scores of MDS-UPDRS parts II and III according to published formula [15]. Presence and severity of levodopa-induced MC were evaluated using part IV of UPDRS∖MDS-UPDRS, and the severity of dyskinesias was additionally scored using the modified Abnormal Involuntary Movement Scale (mAIMS) in off and on periods [16]. The mAIMS was scored during rest (mAIMS rest) and counting backwards (mAIMS count). Patients were considered to have dyskinesias if they scored ≥1 in items 32 of UPDRS or 4.1 of MDS-UPDRS.
Statistical analysis
Data analysis was performed with R software (version 2.13). A descriptive analysis was performed and the results expressed as mean±standard deviation or percentages. The total score of the LHS was entered as a dependent variable in linear regression analysis. Multiple linear regression analysis, adjusted for age and gender, was performed to calculate which variables contributed significantly to higher levels of handicap (the lower the total score of the LHS the higher the level of handicap). The following independent variables were entered: gender, duration of the disease, scores of the MDS-UPDRS part II on and off and MDS-UPDRS part III on and off, HY stage on and off, S&E score on and off, scores of mAIMS rest on and off and mAIMS count on and off, and the LEDD. We used both-stepwise and backwards regression to select significant variables from the original model. Multicollinearity was tested using the Variance Inflation Factor and no collinearity was observed. Statistical significance was considered for p < 0.05. Coefficients and 95% confidence intervals (CIs) are reported.
RESULTS
One-hundred patients (63 males) were included in the study (Table 1). LHS values followed a Gaussian distribution, with a median total score of 0.509 (0.21–1.00) and a mean of 0.56 (±0.14) (Table 2). No ceiling or floor effects were noted. Female and male patients were equally handicapped. There was a wide range of responses between the different levels of disadvantage for each domain of handicap, which indicated that the LHS could discriminate well between patients. However, the most common categories within each domain of the LHS were “quite a lot” and “very slightly” disadvantage, with fewer subjects scoring in the “almost completely” and “completely” disadvantage categories. Patients were handicapped over several domains of the LHS. The domains for which more patients scored a greater disadvantage (“very much”, “almost completely” and “completely”) were Economic Self-sufficiency (n = 26), Occupation (n = 25) and Mobility (n = 10). The Physical Independence sub-score (r: 0.777, p < 0.001) and the Social Integration sub-score (r: 0.657, p < 0.001) showed the greatest association with the LHS total score, thus these were the most severely affected domains (Table 2). In univariable analysis, the total score of the LHS was similar between patients with and without dyskinesias. In multiple linear regression analysis, we used the LHS total score as the dependent variable, and the result was the same using either both-stepwise or backwards method. In the model (n = 100), the independent variables that were most associated with handicap were the scores of MDS-UPDRS part II off (β= –0.271; p = 0.020), mAIMS count on (β= –0.183; p = 0.042), and S&E on (β= 0.264; p = 0.005) and off (β= 0.226; p = 0.020) (Table 3). This model R2 was 29.6%.
Demographic and clinical characteristics of advanced stage PD patients
LEDD, levodopa equivalent daily dose. IQ, interquartile range. As parts I and IV of UPDRS are not converted to MDS-UPDRS, the table indicates in brackets how many patients have completed parts I and IV of UPDRS and parts I and IV of MDS-UPDRS. Scores of MDS-UPDRS parts II and III (n = 100) include converted scores from UPDRS parts II and III.
Total and sub-scores of London Handicap Scale and correlations with total scores (n = 100)
Patients were handicapped over several domains of the LHS, and all domains correlated with the total score of the LHS. The Physical Independence and the Social Integration were the most severely affected domains
Regression model for the London Handicap Scale (n = 100)
CI: confidence interval The multiple linear regression model explained 29.6% of the variance of the LHS total score. The independent variables most associated with handicap were the scores of MDS-UPDRS part II off, S&E on and off, and the mAIMS count on.
DISCUSSION
Handicap was assessed in advanced stage PD patients with disabling levodopa-induced MC prior to DBS implant, using the LHS. The study population was moderately to highly handicapped, and the level of handicap was strongly determined by the disability and independence in ADL, and the severity of peak-dose dyskinesias during the levodopa challenge test. These independent variables accounted for one-third of the patients’ handicap.
We have previously reported the assessment of handicap, using the same instrument, in a population of LS-PD patients [1]. We found that handicap was a valuable PCO in LS-PD and the LHS was easily completed even in those very disabled patients [1]. We now add findings on the assessment of handicap in PD patients with disabling levodopa-induced MC. Our aim was to study handicap in PD patients with disabling levodopa-induced MC, reason why we chose patients selected to DBS because they are the prototype of patients with severe and disabling MC. We are aware though that DBS patients are a sub-group of PD patients with MC, but we aimed to study that pure sample of advanced stage PD patients [9]. Secondly, we aimed to explore handicap and the LHS as a PCO in DBS patients to test in future whether it may be sensitive to change after surgery.
The handicap of this sample of advanced stage PD patients was moderate to high. Nevertheless, the handicap of advanced stage patients was lower than that of LS-PD patients, who had a mean LHS total score of 0.338 (±0.155) [1]. This suggests that DBS patients, although manifesting very disabling MC, are less handicapped than more advanced PD patients whose clinical picture is dominated by motor and non-motor symptoms poorly responsive to L-dopa [1, 17]. On the other hand, the level of handicap of advanced stage PD patients was greater than that reported for stroke survivors at 6 months, 2 and 5 years whose mean LHS score ranged from 0.73 to 0.93 [18, 19]. Unfortunately, there is no data regarding handicap in other neurodegenerative disorders to compare with, such as Alzheimer’s disease or amyotrophic lateral sclerosis. The Physical Independence and Social Integration domains were the most severely affected. Disadvantage in Physical Independence reflects an inability to look after oneself in tasks such as housework, shopping, looking after money, getting dressed and using the toilet, while disadvantage in Social Integration reflects an inability to meet family, friends and other people during a normal day [2]. Our results show that patients with disabling MC are highly handicapped when it comes to looking after themselves independently and conducting a healthy social life. Our sample consisted of relatively young patients with a disease longer than 10 years, who probably value much carrying an independent life and for whom social isolation and embarrassment is a high burden. Due to the criteria for DBS, these patients could not have dementia, which may explain why Orientation was not a severely affected domain, in contrast to what we found in LS-PD where Orientation was the most handicapped domain [1]. Advanced stage patients also differ from stroke survivors who are most affected in domains that are more motor dependent (Mobility, Physical Independence and Occupation) [19, 20].
Factors associated with greater handicap were the disability and independence in ADL (MDS-UPDRS part II and the S&E scale) and the severity of peak-dose dyskinesias. This reflects how motor impairment impacts on ADL and seems to explain why Physical Independence and Social Integration were the most affected domains of handicap. Patients with severe dyskinesias frequently report social embarrassment. Interestingly, the score of MDS-UPDRS part II in off but not on was correlated with level of handicap, suggesting that the severity of disability when patients are in off period may better translate the severity of PD. Also of interest, neither the MDS-UPDRS motor part nor the response to levodopa or the LEDD correlated with level of handicap, probably because these variables were very homogeneous between patients. Peak-dose dyskinesias predicted greater handicap in advanced stage patients. Indeed, they are a major indication for DBS because of the disability they cause [21, 22]. Unexpectedly, parts IV (motor complications) of UPDRS and MDS-UPDRS were not significantly associated with handicap, probably due to the small sample size of patients completing either the UPDRS or the MDS-UPDRS, as the score of UPDRS part IV are not possible to convert to MDS-UPDRS [15]. However, it may also suggest that dyskinesias should be scored during the levodopa challenge test when selecting patients for DBS. Additionally, LHS score was not predicted by part I (non-motor symptoms) of UPDRS/MDS-UPDRS, which may be related to the absence of dementia, psychosis or unstable psychiatric disorders in patients selected for DBS or, alternatively, due to the fact that also the score of UPDRS part I is not convertible to MDS-UPDRS [15].
QoL and health-related QoL ((HR)QoL), a concept intimately related to, but distinct from, handicap, is similarly poor in PD patients with MC [6, 25]. Data suggest that motor fluctuations might have a stronger impact than dyskinesias on HR(QoL), and that the level of disability associated with dyskinesias might vary according to severity of PD [23, 26]. (HR)QoL scales are not very sensitive to change in disease progression over time in PD, and thus handicap might be an alternative candidate as a PCO to measure change after DBS [27]. Additionally, handicap is a concept easily understood by patients and caregivers, and more closely defined and focused than QoL [6].
Our study has some limitations. If all patients had completed either the UPDRS or the MDS-UPDRS, the study would have more power to test the contribution of parts I and IV of these scales to the handicap of patients. We did not enrolled patients with disabling MC that were excluded from DBS after assessment. In these patients, the severity of handicap and its contributors might have differed from the patients we included. It would also have been interesting to use a (HR)QoL scale to assess our patients, to permit a head-to-head comparison between handicap and (HR)QoL ratings. This will be particularly interesting when assessing the responsiveness of handicap scales to DBS. In our model, the independent variables only accounted for one third of the patient’s handicap, which is a moderate result. Some non-motor symptoms and psychosocial variables could have a greater impact on patients’ handicap and could account for part of its variance, and this could have been further explored, such as having sleep problems or living alone. It is true that all patients performed a formal neuropsychological and psychiatric evaluation, and that Part I of UPDRS/MDS-UPDRS covers several aspects of non-motor symptoms, but even so additional assessment of non-motor and psychosocial aspects would give valuable information.
In conclusion, we were able to use handicap to measure overall health condition and individual’s participation in advanced PD patients with disabling levodopa-induced MC. The LHS was easily completed by patients and we have now values for the LHS in advanced and late-stage PD patients. Handicap was moderate to high and mostly determined by disability in ADL and dyskinesias. Future studies are needed to assess whether handicap is a valuable PCO to measure changes after DBS.
CONFLICT OF INTEREST
The authors have no conflict of interest to report.
Footnotes
ACKNOWLEDGMENTS
The authors thank Ana Teresa Santos for editorial assistance.