By Claire Henchcliffe, MD, PhD

Associate Professor of Neurology and Neuroscience, Weill Cornell Medical College

Dr. Henchcliffe reports she is a consultant for ACADIA Pharmaceuticals and US WorldMeds.

SYNOPSIS: The Montreal Parkinson Risk of Dementia Scale provides a simple eight-item screening tool with high predictive value for developing Parkinson’s disease dementia.

SOURCE: Dawson BK, Fereshtehnejad SM, Anang JBM, et al. Office-based screening for dementia in Parkinson disease: The Montreal Parkinson Risk of Dementia Scale in four longitudinal cohorts. JAMA Neurol 2018; Mar 26. doi:10:1001/jamaneurol.2018.2054. [Epub ahead of print].

The Montreal Parkinson Risk of Dementia Scale (MoPaRDS) was developed based on predictors of developing dementia identified in a prospective study of individuals with Parkinson’s disease (PD) in Montreal. The scale comprises eight items: age, sex, falls and/or freezing, bilateral symptom onset, history suggestive of REM sleep behavior disorder (RBD), visual hallucinations, orthostatic hypotension, and mild cognitive impairment. Dawson et al extracted MoPaRDS scores from four independent cohorts combined into three groups for analyses: 1) the original Montreal cohort with established PD (n = 80); 2) a combined group with established PD, comprising individuals recruited from two local PD clinical trials (n = 52) and a cohort from Tottori, Japan (n = 82); and 3) the Parkinson Progression Marker Initiative (PPMI) for individuals with PD diagnosed within two years and not yet taking medications (n = 393). Age at diagnosis was 66.0 ± 8.2 years (original), 68.5 ± 10.1 years (combined), and 61.3 ± 9.8 years (PPMI). All patients were free of dementia at baseline visits, although mild cognitive impairment was identified in 51% and 46.3% in the original and combined cohorts (results imputed for the Tottori cohort), and in 21.6% in PPMI. A greater percentage were < 70 years of age in the PPMI group than in the other groups. Additionally, fewer exhibited features suggestive of RBD, hallucinations, orthostatic hypotension, presence of falls or freezing, or history of bilateral onset in PPMI compared to the other groups. Investigators analyzed MoPaRDS scores using the primary endpoint of dementia status at the subject’s last office visit, judged by global cognitive decline to Mini-Mental Status Examination (MMSE) < 26, with impairment in > 1 cognitive domain resulting in significant impairment in activities of daily living (ADL). This endpoint was reached in 11.5% of all individuals (but only 3.3% in PPMI alone) over mean 4.4 ± 1.3 years (range, 1-8 years) follow-up. When stratified by MoPaRDS scores, the annual risk of dementia was 0.6% (score, 0-3), 5.8% (score, 4-5), and 14.9% (score, 6-8). Predictive validity, as examined by receiver operating characteristic (ROC) curves, was 0.879 (95% confidence interval [CI], 0.816-0.942), and the determined optimal cutoff score of at least 4 resulted in a positive predictive value of 43.9% (95% CI, 37.8-50.2) and negative predictive value of 96.7% (95% CI, 95.0-97.9).

COMMENTARY

Dementia is more common in PD patients than the general population and has a profound effect, not only on the patient’s function, quality of life, and life expectancy, but also on the caregiver. It is a common cause of placement in nursing homes, and treatment options are limited. It is critical that we better understand the nature of PD dementia and establish improved counseling and treatment pathways. Several studies have determined various clinical and non-clinical risk factors for PD dementia; for example, greater age, presence of hallucinations, or RBD. The MoPaRDS now may provide a step forward in combining a few easily assessed items. Its strength lies in its ease of administration and the simplicity of the data required to generate a MoPaRDS score. Several items can be simply extracted from the clinical chart, such as age, sex, and whether symptom onset was bilateral. Others either already may be obtained or could be added easily to a visit at intervals (for example, measuring orthostatic changes or recording items derived from the Movement Disorders Society Unified Parkinson Disease Rating Scale). Therefore, the resulting MoPaRDS score ultimately may be useful for rapid screening of future dementia risk in the clinic. It also is important that in the PPMI cohort, the baseline MoPaRDS score correlated with potential molecular biomarkers previously identified in cerebrospinal fluid: Ab42/tau and tau concentrations. However, it is too early to recommend routine use in the clinic. There was some variability between cohorts in how particular scores were collected, and certain results were imputed. The definition for the primary endpoint may have missed some cases of dementia. Moreover, the scale performed better in men than women, for unclear reasons. Overall, though, the MoPaRDS appears to be a highly promising approach that merits testing in larger cohorts, and once again serves to draw attention to a challenging feature in PD.