Acoustic analysis of dysarthria profile in ALS patients

Dysarthria is a leading disability in ALS patients with motor neurone degen eration in the bulbar region. Although different approaches have been tried in the past, currently, no test is available to detect and follow the prog ression of dysarthria. We studied 53 patients with definite (n=27) or proba ble (n=26) ALS (the bulbar onset group n=15, the limb onset group n=38, mea n age 53.66/29-76 years/) according to El Escorial criteria. Each patient w as seen by a neurologist every 10-12 weeks and clinical performance was ass essed using the Norris scale. To evaluate dysarthria we developed a compute r-based acoustic method. All patients had computer-analysed speech sound te sts done three times. The most significantly affected vowels were selected for further studies. A method based on the Euclidian principle was used and the results were compared with 30 age, sex-matched, healthy control subjec ts. Our results demonstrated the existence of a specific dysarthria profile in ALS patients with most significantly affected vowels: 'B', 'O', 'I', 'W ', 'T' in the bulbar group, and: 'B', 'I', 'T', 'W', 'O' in the limb group. This study suggests that it is possible to detect and monitor the progress ion of the disease based on the acoustic analysis of only several sounds. A bnormalities detected in the dysarthria profile may appear prior to any cli nical symptoms of the disease. (C) 1999 Elsevier Science B.V. All rights re served.