Computerized respiratory sounds in paediatrics: a systematic review

Abstract

Background Diagnosing and monitoring of children with respiratory disorders is often challenging. Respiratory sounds (RS) are simple, non-invasive and universally available measures that are directly related to movement of air, within the tracheobronchial tree. Thus, RS may be valuable indicators of respiratory health, their characteristics in the paediatric population are scattered in the literature and not systematized.

Aim Systematically review the different acoustic RS properties in healthy children and in children with different respiratory disorders. Methods: MEDLINE, EMBASE, AMED and CINHAL databases were searched on Sept 2020. One author extracted data and two independently assessed the quality of the articles using the National Heart Lung and Blood Institute quality assessment tool.

Results Twenty-eight studies were included with a total 2032 participants (44% with a respiratory condition, such as asthma, bronchiolitis, cystic fibrosis, presence of wheezing and non-specified low respiratory tract infections). A high heterogeneity in the procedures, outcomes and outcome measures used was found. Healthy participants showed lower values of F50 (from 194 ± 26 to 521 ± 18Hz) than those with asthma (from 140 ± 8 to 769 ± 85Hz) or bronchiolitis (from 100 to 80Hz). F50 tend to increase with provocation tests (136 ± 9 to 909 ± 81Hz) and decrease with treatments (128 ± 6 to 781 ± 57Hz). Wheeze rates ranged from 0 to 24.7 ± 25% on asthmatic participants. Crackles findings ranged from 6% on people with recurrent wheezing to 30.8% in middle lobe atelectasis.

Conclusion RS show different acoustic properties in healthy children vs with different respiratory disorders and thus may be useful in the diagnostic and monitoring on paediatrics.