Automated analysis of lung sounds as a predictor of VAP
Research type
Research Study
Full title
Automated analysis of lung sounds as a predictor of Ventilator Associated Pneumonia (VAP)
IRAS ID
228937
Contact name
Mike Grocott
Contact email
Sponsor organisation
University of Southampton
Duration of Study in the UK
2 years, 11 months, 3 days
Research summary
Between 40 and 50% of patients on critical care units (CCU) are supported by mechanical ventilation at some point during their CCU stay. Ventilator Associated Pneumonia (VAP) is the second most common cause of hospital acquired sickness in critically ill patients, contributing to significant morbidity and mortality. Lung sounds are simple, non-invasive markers of lung health, but standard lung auscultation is a subjective practice. The interpretation of lung sounds is dependent on the hearing ability of the assessor, and their ability to discriminate between different adventitious sounds. Computer Aided Lung Sound Analysis (CALSA) removes this subjectivity and allows the quantification of the acoustic characteristics of lung sounds such as intensity, frequency content and duration. We hypothesise that information derived from CALSA will aid the monitoring of patients who have been mechanically ventilated in the critical care environment, allowing for early diagnosis of complications, and hence early intervention. However, before this hypothesis can be formally tested, there is firstly a need to establish the feasibility of collecting and analysing digitalised lung sounds from patients who have been mechanically ventilated in a critical care environment. Any patient who has been mechanically ventilated for more than 24 hours will be eligible to participate in this study. Physiotherapists working in a ICU will obtain twice daily recordings of lung sounds using a digital stethoscope in accordance with a standardised protocol, until discharge, or 21 days post admission. Demographic data, medication usage, past medical history and clinical data will be obtained from patient medical notes by a research fellow. Lung sounds will be analysed and correlations with clinical data will be examined. This will be an exploratory, observational study to provide us with some of the information necessary to know if a larger, more definitive study assessing predictive powers of lung sounds, might be feasible.
REC name
South Central - Hampshire A Research Ethics Committee
REC reference
17/SC/0529
Date of REC Opinion
3 Jan 2018
REC opinion
Further Information Favourable Opinion