Smartphone videos to predict the severity of obstructive sleep apnoea
Rahul J. Thomas, Samuel Dalton, Katharine Harman, Julie K. Thacker, Rosemary S C Horne, Margot J. Davey, Gillian M. Nixon
Abstract
OBJECTIVE: Diagnosis of obstructive sleep apnoea (OSA) is made on overnight polysomnography (PSG). Given the widespread availability of smartphone video technology, we aimed to develop and test a standardised scoring system for smartphone videos and compare these scores to PSG results. METHODS: Children aged 1-16 years undergoing PSG for suspected OSA were included. Parents were asked to take 1-2 min videos of the breathing they were concerned about. Videos were scored using a newly developed and tested tool on five components: inspiratory obstructive noises (1-4), presence of obstructive events (0-1), increased work of breathing (0-1), mouth breathing (0-1) and neck extension (0-1). Video scores and the Obstructive Apnoea Hypopnoea Index (OAHI) were compared using Spearman correlation. Sensitivity, specificity, positive predictive value and negative predictive value were calculated for different cut-off scores to achieve the best results. RESULTS: Videos from 43 children (28 men (65.1%), median age 5.7 years (range 2.6-14.0 years), median OAHI 3.8/hour (range 0-82 events/hour) were included. Nine children (20.9%) had a video score of <3, all of whom had an OAHI of ≤5 events/hour. For a video score of ≥3, sensitivity was 100%; specificity was 36%; positive predictive value was 53%; and negative predictive value 100% for moderate to severe OSA (OAHI>5 events/hour) . CONCLUSION: We have developed and validated a simple clinical tool (the Monash Obstructive Sleep Apnoea Video Score) to quantify abnormalities in breathing seen on short video recordings made on a smartphone. A low score rules out moderate-severe OSA and may be valuable in the triage of children with symptoms of OSA.