"Does speech in patients with different Parkinson's disease subtypes decline over time?"

This study investigates whether speech changes over time in patients with different motor subtypes of Parkinson's disease (PD).

This longitudinal study included 60 individuals with PD, classified into tremor-dominant (n = 42) and non-tremor-dominant (n = 18) subtypes according to the Movement Disorder Society Unified Parkinson's Disease Rating Scale. Participants were assessed at three points: baseline, 12 months, and 24 months. Speech samples were collected from diadochokinesis (/pataka/) and 60-second monologue tasks. The analyzed acoustic variables included syllable number, duration, phonation time, articulation rate, average syllable duration (ASD), and pause ratio.

There were no statistically significant differences between groups regarding age (p = 0.298), disease duration (p = 0.098), age at symptom onset (p = 0.879), HY (p = 0.895), and UPDRS (p = 0.224) at baseline. The tremor- and non-tremor-dominant subtypes did not show significant differences over time when analyzed separately in articulation time metrics. However, significant differences were observed between the subtypes in all variables of the diadochokinesia task, including the number of syllables (p < 0.001), phonation time (p < 0.001), articulation rate (p = 0.002), and ASD (p = 0.028). Regarding the trend analysis, for the monologue, the following were observed: dominant non-tremor [number of syllables (R² = 0.978), phonation time (R² = 0.860) and pause ratio (R² = 0.766)] and tremor [number of syllables (R² = 0.989) and phonation time (R² = 0.974)]. For the diadochokinesis: dominant non-tremor [number of syllables (R2 = 0.934) and phonation time (R² = 0.977)] and tremor [ASD (R² = 0.787)].

Over three years, the speech revealed that individuals with tremor and non-tremor subtypes exhibited stable patterns over time. Distinct speech differences emerged between the two groups, and articulation time metrics proved to be an effective tool for distinguishing between subtypes.This task emerged as a relevant approach from which biomarkers can be derived.