Parkinsonian motor features distinguish the agrammatic from logopenic variant of primary progressive aphasia

Abstract

Background: Few studies have assessed parkinsonian motor features across variants of primary progressive aphasia (PPA). Our objective was to compare degree of parkinsonian motor features between two PPA variants.

Methods: Parkinsonian motor features were assessed with the unified Parkinson's disease rating scale in prospectively recruited PPA subjects, classified based on recently published criteria. Comparisons across groups were performed with Fisher's exact test for binary data and Mann-Whitney U test for continuous data.

Results: Twenty-three PPA subjects were diagnosed with logopenic (n = 11) or nonfluent/agrammatic (n = 12) aphasia. There were no significant differences in illness duration (agrammatic = 3.4 years; logopenic = 3.3 years) or age at onset (nonfluent/agrammatic = 67.3; logopenic = 62.0), but those with logopenic aphasia were more impaired on mini-mental state examination (21.7/30 points vs. 26.1/30; p = 0.04). In contrast, those with logopenic aphasia had fewer parkinsonian motor features than those with agrammatic aphasia (5.7/132 vs. 16.8/132; p = 0.003) which was driven by bradykinesia (p = 0.02) and speech facial/expression (p = 0.04); less so rigidity (p = 0.06). Dysarthria was more frequent in the nonfluent/agrammatic subgroup.

Conclusions: Nonfluent/agrammatic subjects have more parkinsonian motor features than logopenic subjects, likely reflecting underlying tau pathology in the agrammatic variant.

Figure 1

Representative FDG-PET images for nonfluent/agrammatic and logopenic primary progressive aphasias.