Characteristics of motor speech phenotypes in multiple sclerosis

Abstract

Background: Motor speech disorders in multiple sclerosis (MS) are poorly understood and their quantitative, objective acoustic characterization remains limited. Additionally, little data regarding relationships between the severity of speech disorders and neurological involvement in MS, as well as the contribution of pyramidal and cerebellar functional systems on speech phenotypes, is available.

Methods: Speech data were acquired from 141 MS patients with Expanded Disability Status Scale (EDSS) ranging from 1 to 6.5 and 70 matched healthy controls. Objective acoustic speech assessment including subtests on phonation, oral diadochokinesis, articulation and prosody was performed.

Results: The prevalence of dysarthria in our MS cohort was 56% while the severity was generally mild and primarily consisted of a combination of spastic and ataxic components. Prosodic-articulatory disorder presenting with monopitch, articulatory decay, excess loudness variations and slow rate was the most salient. Speech disorders reflected subclinical motor impairment with 78% accuracy in discriminating between a subgroup of asymptomatic MS (EDSS < 2.0) and control speakers. Speech disorder severity was related to the severity of neurological involvement. Decreased articulation rate was moderately correlated to EDSS as well as all subtests of the multiple sclerosis functional composite. The strongest correlation was observed between irregular oral diadochokinesis and the 9-Hole Peg Test (r = - 0.65, p < 0.001). Irregular oral diadochokinesis and excess loudness variations significantly separated pure pyramidal and mixed pyramidal-cerebellar MS subgroups.

Conclusions: Automated speech analyses may provide valuable biomarkers of disease progression in MS as dysarthria represents common and early manifestation that reflects disease disability and underlying pyramidal-cerebellar pathophysiology.

Keywords: Acoustic analyses; Dysarthria; Multiple sclerosis; Speech; Voice.