The stability of lung sounds measurements over time may influence their clinical usefulness. In the present study we investigated the temporal variability of the spectral pattern of normal lung sounds. Breath sounds from five healthy men were recorded on the trachea and at four locations over the chest wall. Each subject was studied twice with a time interval of 1 wk. On each occasion, measurements were done in duplicate, with a 30-min interval between recordings. Sounds were amplified, band-pass filtered (75 to 2,000 Hz) and digitized into a computer, and the average spectra of the inspiratory, expiratory, and background sounds were calculated. The variability of corresponding spectra were calculated between the daily duplicate (same-day variability, SDV) and between the two recording sessions (between-day variability, BDV). SDV was 32.8 +/- 12.0% during inspiration and 40.8 +/- 12.6% during expiration (p = 0.005). BDV was 36.9 +/- 11.3% during inspiration and 42.7 +/- 12.7% during expiration. These values were not significantly different from SDV except for sounds recorded from the interscapular region (SR). At this location the SDV was 28.2 +/- 7.2% during inspiration and 40.8 +/- 14.2% during expiration, and the BDV was 48.2 +/- 18.7% during inspiration and 77.6 +/- 22.3% during expiration (p < 0.05). The increased BDV at SR was found to be a result of slight differences in microphone position from the first session to the next. Similar changes in microphone position at the other recording sites did not alter the variability of lung sounds.(ABSTRACT TRUNCATED AT 250 WORDS)

Fremitus, the transmission of sound and vibration from the mouth to the chest wall, has long been used clinically to examine the pulmonary system. Recently, modern technology has become available to measure the acoustic transfer function (TF) and transit times (TT) of the pulmonary system. Because sound speed is inversely proportional to the square root of gas density in free gas, but not in porous media, we measured the effect of air and Heliox (80% He-20% O2) breathing on pulmonary sound transmission in six healthy subjects to investigate the mechanism of sound transmission. Wide-band noise (75-2,000 Hz) was "injected" into the mouth and picked up over the trachea and chest wall. The averaged power spectra, TF, phase, and coherence were calculated using a fast Fourier transform-based algorithm. The phase data were used to calculate TT as a function of frequency. TF was found to consist of a low-pass filter property with essentially flat transmitted energy to 300 Hz and exponential decline to 600 Hz at the anterior right upper lobe (CR) and flat transmission to 100 Hz with exponential decline to 150 Hz at the right posterior base (BR). TF was not affected by breathing Heliox. The average TT values, calculated from the slopes of the averaged phase, were 1.5 +/- 0.5 ms for trachea to CR and 5.2 +/- 0.5 ms for trachea to BR transmission during air breathing. During Heliox breathing, the values of TT were 1.5 +/- 0.5 ms and 4.9 +/- 0.5 ms from the trachea to CR and from the trachea to BR locations, respectively. These results suggest that sound transmission in the respiratory system is dominated by wave propagation through the parenchymal porous structure.

We studied the mechanisms by which turbulent flow induces tracheal wall vibrations detected as tracheal breath sounds (TRBSs). The effects of flow rate at transitional Reynold's numbers (1300-10,000) and gas density on spectral patterns of TRBSs in eight normal subjects were measured. TRBSs were recorded with a contact sensor during air and heliox breathing at four flow rates (1.0, 1.5, 2.0, and 2.5 l/s). We found that normalized TRBSs were proportional to flow to the 1.89 power during inspiration and to the 1.59 power during expiration irrespective of gas density. The amplitude of TRBSs with heliox was lower than with air by a factor of 0.33 ± 0.12 and 0.44 ± 0.16 during inspiration and expiration, respectively. The spectral resonance frequencies were higher during heliox than air breathing by a factor of 1.75 ± 0.2-approximately the square root of the reciprocal of the air/heliox wave propagation speed ratio. In conclusion, the flow-induced pressure fluctuations inside the trachea, which cause tracheal wall vibrations, were detected as TRBSs consist of two components:(1) a dominant local turbulent eddy component whose amplitude is proportional to the gas density and nonlinearly related to the flow; and (2) a propagating acoustic component with resonances whose frequencies correspond to the length of the upper airway and to the free-field sound speed. Therefore, TRBSs consist primarily of direct turbulent eddy pressure fluctuations that are perceived as sound during auscultation.