Sound Production in Totally Edentulous Patients Before and After Prosthetic Treatment.

PURPOSE: To study sound pronunciation in patients with total edentulism before and after prosthodontic treatment using complete removable dentures.

MATERIALS AND METHODS: Recordings of phonetically balanced phrases were made using a headset connected to a dictaphone to assess specific pronunciation defects in patients with total edentulism before and after prosthodontic treatment. The obtained sets of sounds were investigated using spectral analysis by calculating the average power spectra. The data of the experimental group was compared to the data of the control group, regarded as the norm. Analysis of the trajectory of the average power spectra in the range of 5300 to 6700 Hz makes it possible to develop a test of the waveform spectrum of a standard pronunciation and that of the phonemes before and after treatment of completely edentulous patients, as well as to compare the phonograms before and after the prosthetics. The shape of the trajectory of the spectra in the specified frequency range was measured by the expectation value and standard deviation value. A confidence interval of 95% was used to test for significant changes in spectra.

RESULTS: By applying the spectral analysis to sound pronunciation, it was possible to objectively, qualitatively, and quantitatively analyze sound characteristics. It can also be used to evaluate the quality of the prosthodontic treatment. For the control group the average power spectrum of sounds was 193.1 ± 11.6 (t = 2.1; < 0.05) for "S;" 74.4 ± 3.7 (t = 5.5; < 0.001) for "F;" and 17.6 ± 2.4 (t = 4.2; < 0.001) for "V." The comparative analysis of the spectra of the experimental group patients showed that before treatment the average power spectrum of sounds was 91.7 ± 27.1 for "S;" 42.6 ± 3.6 for "F;" and 9.2 ± 0.5 for "V." After the treatment the average power spectrum of sounds of the experimental group patients were 220.4 ± 25.9 (t = 2.4; p < 0.01) for "S;" 139.0 ± 27.7 (t = 4.9; p < 0.001) for "F;" and 24.7 ± 4.8 (t = 4.4; p < 0.001) for "V."

CONCLUSIONS: Spectral analysis of the sounds emitted before and after treatment is an informative method for the differential diagnostics of pronunciation defects. Sound analysis can be used efficiently in different stages of prosthodontic treatment to assess quality of treatment.