ABSTRACT
Nonspeech (or paraspeech) parameters are widely used in clinical assessment of speech impairment in persons with dysarthria (PWD). Virtually every standard clinical instrument used in dysarthria diagnostics includes nonspeech parameters, often in considerable numbers. While theoretical considerations have challenged the validity of these measures as markers of speech impairment, only a few studies have directly examined their relationship to speech parameters on a broader scale. This study was designed to investigate how nonspeech parameters commonly used in clinical dysarthria assessment relate to speech characteristics of dysarthria in individuals with movement disorders. Maximum syllable repetition rates, accuracies, and rates of isolated and repetitive nonspeech oral-facial movements and maximum phonation times were compared with auditory-perceptual and acoustic speech parameters. Overall, 23 diagnostic parameters were assessed in a sample of 130 patients with movement disorders of six etiologies. Each variable was standardized for its distribution and for age and sex effects in 130 neurotypical speakers. Exploratory Graph Analysis (EGA) and Confirmatory Factor Analysis (CFA) were used to examine the factor structure underlying the diagnostic parameters. In the first analysis, we tested the hypothesis that nonspeech parameters combine with speech parameters within diagnostic dimensions representing domain-general motor control principles. In a second analysis, we tested the more specific hypotheses that diagnostic parameters split along effector (lip vs. tongue) or functional (speed vs. accuracy) rather than task boundaries. Our findings contradict the view that nonspeech parameters currently used in dysarthria diagnostics are congruent with diagnostic measures of speech characteristics in PWD.
ABSTRACT
BACKGROUND: Globus pallidus internus deep brain stimulation (GPi-DBS) can be an effective treatment for primary dystonia. However, speech disorders have previously been reported as a common possible side effect of the treatment. OBJECTIVES: To study possible deterioration of speech after GPi-DBS and describe this in different dimensions. METHODS: Speech was systematically evaluated in 15 patients with predominant torticollis and GPi-DBS. Each patient was tested twice within one day in two stimulation conditions: ON-DBS vs. OFF-DBS. Speech analyses comprised both function-oriented (perceptual scales, acoustic analyses) and communication-related measures (intelligibility, naturalness). A control sample of 15 healthy speakers underwent the same speech assessment. RESULTS: On the group level, patients with dystonia showed mild but significant impairment on the overall dysarthria scale, the intelligibility score, and the naturalness ratings in both stimulation conditions (Mann-Whitney, P < .05). No stimulation-induced deterioration was found. A slight increase in articulation rate was measured in the ON condition. On the single-case level, effects of GPi-DBS on speech were heterogenous. In one patient we observed a deterioration of speech (dysarthria), in a second patient with a history of childhood stuttering we found an aggravation of dysfluency. Impressive benefits could be documented in another patient who also suffered from spasmodic dysphonia. CONCLUSIONS: The study provides evidence that speech impairment is not a necessary side-effect of GPi-DBS in primary dystonia. Both, recurring of stuttering and a worsening of dysarthria may be seen in individual patients. The positive effects of GPi-DBS on the symptoms of spasmodic dysphonia merits further research as DBS is not commonly applied in this population.
