Spectral characteristics of chest wall breath sounds in normal subjects.

Autor: Gavriely N; Department of Physiology and Biophysics, Bruce Rappaport Faculty of Medicine, Rappaport Family Institute for Research in the Medical Sciences, Haifa, Israel., Nissan M, Rubin AH, Cugell DW
Jazyk: angličtina
Zdroj: Thorax [Thorax] 1995 Dec; Vol. 50 (12), pp. 1292-300.
DOI: 10.1136/thx.50.12.1292
Abstrakt: Background: This study was carried out to establish a reliable bank of information on the spectral characteristics of chest wall breath sounds from healthy men and women, both non-smokers and smokers.
Methods: Chest wall breath sounds from 272 men and 81 women were measured using contact acoustic sensors, amplifiers, and fast Fourier transform (FFT) based spectral analysis software. Inspiratory and expiratory sounds were picked up at three standard locations on the chest wall during breathing at flows of 1-2 l/s and analysed breath by breath in real time.
Results: The amplitude spectrum of normal chest wall breath sounds has two linear parts in the log-log plane--low and high frequency segments--that are best characterised by their corresponding regression lines. Four parameters are needed and are sufficient for complete quantitative representation of each of the spectra: the slopes of the two regression lines plus the amplitude and frequency coordinates of their intersection. The range of slopes of the high frequency lines was -12.7 to -15.2 dB/oct during inspiration and -13.4 to -20.3 dB/oct during expiration. The frequency at which this line crossed the zero dB level--that is, the amplitude resolution threshold of the system--was designated as the maximal frequency (Fmax) which varied from 736 to 999 Hz during inspiration and from 426 to 796 Hz during expiration with higher values in women than in men. The mean (SD) regression coefficient of the high frequency line was 0.89 (0.05).
Conclusions: These data define the boundaries of normal chest wall breath sounds and may be used as reference for comparison with abnormal sounds.
Databáze: MEDLINE