[Multi-dimensional analysis of vocal function. A case-control study].

Abstract

The aim of this study was to select the most appropriate methods for analyzing vocal function from an array of existing tests. A case-control study (64 patients and 64 controls) was made of teaching staff in Logroño, Spain. The 64 patients had 25 nodules, 7 polyps, 15 hyperfunctional dysphonia, 6 hypofunctional dysphonia, 6 submucosal suffusion and 5 chronic laryngitis. The study methods included anamnesis, ENT examination, videostrobolaryngoscopy, perceptual evaluation of hoarse voices, basic aerodynamic tests, the frequency range of phonation, physical analysis of the voice signal of the vowel /a/ and the sentence ìmi padre era castellanoî, and spectrographic analysis of Spanish vowels according to Yanagiharaís classification using Sound Scope 1.0 software. We found no single test that consistently differentiated between hoarse and normal voices; values always overlapped. In contrast the phonation quotient (PQ), maximum phonation time (MPT) of the vowel /a/, S/Z quotient (Eckel and Boone), frequency range of phonation, APQ (average perturbation quotient), HNR (harmonic-to-noise ratio), BR (breathiness, or the high-frequency content of the source wave), RAP (relative average perturbation), standard deviation (SD) of the fundamental frequency (F0) of the vowel /a/, intensity and SD of the sentence ìmi padre era castellanoî, and the sound spectrographic analysis of Spanish vowels disclosed statistically significant differences between hoarse and normal voices.