Objective: To evaluate utility of the DRS-II Total Score for stratifying cognitive status in PD patients without frank cognitive impairment.

Background: The DRS-II, a screening tool for dementia, is commonly used to stratify DBS-related cognitive risk. Some institutions employ a cut score, whereby individuals failing to meet this are not offered DBS. However, the utility of this practice has not been empirically supported. Evaluation of raw scores suggests that one subscale (I/P), which gives great weight to a measure of semantic verbal fluency, may contribute differentially to Total Score. Unfortunately, PD patients’ performance on I/P is heavily influenced by medication on/off state and medication response. We sought to evaluate contributors to DRS-II Total Score to evaluate potential for bias, and to identify alternate predictors of overall cognitive status in our presurgical PD group.

Methods: We conducted an archival analysis of medication-on DRS-II performance in PD patients undergoing presurgical workup. Individuals with frank impairment (i.e., <5th percentile) were excluded. After standardizing raw scores, regression analysis and correlation matrices were utilized to illustrate contribution of subscale performances to Total Score. Alternative predictors of overall cognitive health were then identified.

Results: 228 pre-surgical PD-DBS patients met inclusion criteria of DRS-II Total >120 (i.e., not Impaired). Mean age was 64.7 (8.4) years, mean education was 14.1 (2.8) years, and mean disease duration was 9.7 (5.3) years. All DRS subscales correlated significantly with Total score (< .05), but I/P scores were more highly correlated than any other subscale (r = .803). Within I/P, 94% of variance was accounted for by fluency. Sequential multiple regression model supported I/P as optimal predictor (p < .001), explaining 65% of variance in Total Score. Addition of Con and Att contributed significantly (p <.01). Fluency from I/P correlated equivalently with both another semantic task (Animal Naming, r= .588) and letter fluency (FAS, r = .506).

Conclusions: DRS-II Total Score was disproportionately impacted by medication-on I/P performance, which relies heavily on semantic fluency. Verbal fluency is additionally known to be susceptible to medication off states, which may result in further underestimate of capability. Use of DRS-II Total cut scores is not supported.

« Back to 2016 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/use-of-dementia-rating-scale-ii-drs-ii-total-score-in-pd-deep-brain-stimulation-dbs-candidacy-determination/