Although other nonmotor phenomena representing possible prodromal symptoms of Parkinson's disease have been described in some detail, the occurrence and characteristics of cognitive decline in this early phase of the disease are less well understood. The aim of this review is to summarize the current state of research on cognitive changes in prodromal PD. Only a small number of longitudinal studies have been conducted that examined cognitive function in individuals with a subsequent PD diagnosis. However, when we consider data from at-risk groups, the evidence suggests that cognitive decline may occur in a substantial number of individuals who have the potential for developing PD. In terms of specific cognitive domains, executive function in particular and, less frequently, memory scores are reduced. Prospective longitudinal studies are thus needed to clarify whether cognitive, and specifically executive, decline might be added to the prodromal nonmotor symptom complex that may precede motor manifestations of PD by years and may help to update the risk scores used for early identification of PD. © 2017 International Parkinson and Movement Disorder Society.

Many cognitive screening instruments have been developed during the last decades to detect mild cognitive dysfunction and dementia, and there is an ongoing discussion as to which tool should be used in which setting and which challenges have to be considered. Among other aspects, dependence on age is a recognized problem in screening tools which still has not found its way into common scoring procedures. Another aspect which has been handled very heterogeneously is which domain is represented in which proportion in the total score. Furthermore, screening ethnic minority patients has been identified as an important but so far widely unresolved matter. In this review, four cognitive screening tools that all follow a common, stringent concept and pay regard to some critical aspects are described: the DemTect, a "generic" tool; the PANDA for Parkinson's disease patients; the EASY, a non-verbal, culture-fair screening test for patients with migration background; and the MUSIC for patients with multiple sclerosis. All of these screening instruments have an age-correction, provide a total score in which the different subtests are weighted according to their individual sensitivity and specificity, and include tasks that are specifically aligned to the cognitive profile of the target group, including the EASY with non-verbal, culture-fair tasks to overcome language and cultural barriers. The development, main characteristics, data, and limitations of these tools are presented and discussed against the background of the current landscape of cognitive screening tools.

BackgroundGiven the high prevalence of cognitive impairment in Parkinson’s disease (PD), cognitive screening is important in clinical practice. The Montreal Cognitive Assessment (MoCA) is a frequently used screening test in PD to detect mild cognitive impairment (PD-MCI) and Parkinson’s disease dementia (PD-D). However, the proportion in which the subtests are represented in the MoCA total score does not seem reasonable. We present the development and preliminary evaluation of an empirically based alternative scoring system of the MoCA which aims at increasing the overall diagnostic accuracy.MethodsIn study 1, the MoCA was administered to 40 patients with PD without cognitive impairment (PD-N), PD-MCI, or PD-D, as defined by a comprehensive neuropsychological test battery. The new MoCA scoring algorithm was developed by defining Areas under the Curve (AUC) for MoCA subtests in a Receiver Operating Characteristic (ROC) and by weighting the subtests according to their sensitivities and specificities. In study 2, an independent sample of 24 PD patients (PD-N, PD-MCI, or PD-D) was tested with the MoCA. In both studies, diagnostic accuracy of the original and the new scoring procedure was calculated.ResultsDiagnostic accuracy increased with the new MoCA scoring algorithm. In study 1, the sensitivity to detect cognitive impairment increased from 62.5% to 92%, while specificity decreased only slightly from 77.7% to 73%; in study 2, sensitivity increased from 68.8% to 81.3%, while specificity stayed stable at 75%.ConclusionThis pilot study demonstrates that the sensitivity of the MoCA can be enhanced substantially by an empirically based weighting procedure and that the proposed scoring algorithm may serve the MoCA’s actual purpose as a screening tool in the detection of cognitive dysfunction in PD patients better than the original scoring of the MoCA. Further research with larger sample sizes is necessary to establish efficacy of the alternate scoring system.

The DemTect, a frequently used cognitive screening tool for the German-speaking population, has been proven to be age-, but not education-dependent. To date, scoring routines for persons under and over 60 years of age have been available. In order to describe the age-effect more specifically, the DemTect was administered to persons under 40 (n = 105; median 33 years [18 - 39]) and over 80 years of age [n = 68; median 83 years [80 - 93}). After transformation of the raw scores, which are based on the mean and standard deviation of the respective reference group, an adequate attribution of scores could be achieved and no differences between the groups can be observed in the total score or the subtest scores. The median of the transformed total score was 16 [5 – 18] for the younger and 15 [5 – 18] for the older age group. The new scoring routines supplement the scoring procedures of the DemTect for these age groups.

Normative data corrected for gender and other sociodemographic variables are relevant, since they may elucidate a markedly different cognitive profile compared to raw scores. Our study also suggests that verbal memory decline is stronger in women than in men with PD. Future studies are needed to replicate these findings, examine the progression of gender-specific cognitive decline in PD and define different underlying mechanisms of this dysfunction.

No abstract