To investigate the acoustic characteristics of breath sounds that may be useful in the detection of pneumothorax (PTX) in an animal model.
Eight mongrel dogs under general anesthesia were endotracheally intubated and placed in the left lateral decubitus position. A ventilator delivered tidal volumes of 20 ml/kg at 12 cycles/minute. An electronic stethoscope (Labtron Electromax; Hauppauge, NY) measured the breath sounds in the right mid-clavicular line at the level of the third rib. Breath sounds were acquired in the control (non-PTX) and in an artificially created30% PTX state (confirmed with a thoracoscope), and recorded on a tape recorder (Portastudio 424; Teac Corp.; Japan) for 20 seconds during each test. Recorded sounds were digitized (at 4000 sample/sec) using a laptop computer (Latitude Xpi, P120D; Dell Computers; Dallas, TX) and analog-to-digital card (Card 500; National Instruments; Austin, TX). Digital signal processing software (Matlab; MathWorks; Natick, MA) were employed to calculate and analyze the breath sound spectra. Statistical analysis was performed using the Wilcoxon signed-rank test.
Breath sound amplitudes decreased with pneumothorax in all animals (p<0.01, Figure). The amplitude drop was most pronounced for frequencies between 200-700 Hz.
Pneumothorax decreased the amplitude of breath sounds detected electronically in an animal model. This effect varied with acoustic frequency (“pitch”), with the differences between control and the PTX state being most pronounced within the 200-700 Hz bandwidth. This is consistent with the clinical finding of decreased breath sounds with pneumothorax. Further studies are warranted in human patients.
These findings may lead to the development of devices that can immediately detect pneumothoraces acoustically, facilitating prompt treatment. The accuracy, simplicity of use, portability, and low cost of such devices may make them attractive diagnostic modalities in emergency rooms, ambulances, and intensive care units.
H.A. Hassaballa, None.