The use of computer aided lung sound analysis to characterise lung sounds in a healthy population

Introduction: Computer aided lung sound analysis (CALSA) objectively manages lung sound data recorded via a digital stethoscope. Historically it has been assumed that crackles heard on auscultation are indicative of lung pathology; however, there is some limited evidence that crackles may be heard in healthy lungs. The purpose of this study was to use CALSA to confirm whether crackles are present in the healthy population.Methods: One hundred healthy participants were recruited. A digital stethoscope connected to a laptop was used to record lung sounds at seven different locations across the chest wall; participants were seated and mouth breathing at volumes slightly deeper than normal. Lung sound data were analysed by time expanded waveform analysis.Results: Crackles were identified in all participants at all locations across the chest wall. Mean initial deflection width (IDW) across all locations was 2.18±0.62ms. Mean two cycle deflection (2CD) duration (13.90±2.27ms) indicated that coarse crackles were present. A negative correlation between increasing age and increasing 2CD duration was found at all locations, however only two reached statistical significance (p<0.05).Conclusions: Crackles are identifiable in the lungs of healthy individuals, when breathing open-mouthed at slightly deeper than normal volumes. Crackle parameters are not significantly related to age in a healthy population. The presence of crackles during auscultation does not necessarily indicate an underlying lung pathology, so clinicians should not make this assumption on hearing these adventitious sounds