SleepImage® as a tool for investigating sleep in children at high risk of comorbid sleep disorders.

OBJECTIVE: Children with neurodisability experience high rates of both respiratory and non-respiratory sleep disorders and greater challenges tolerating polysomnography (PSG). This study aimed to assess tolerability of a home sleep assessment device (SleepImage (SI), MyCardio, Denver, USA), compare SI respiratory and non-respiratory outcomes to PSG, assess the diagnostic accuracy of SI for OSA, and assess the correlation of the SI Sleep Quality Index (SQI) with PSG outcomes. STUDY DESIGN: Paired diagnostic accuracy study. SETTING: Sleep laboratory in a specialist children's hospital. PATIENTS: Children with neurodisability aged 3-18 years undergoing diagnostic PSG between April and August 2025. INTERVENTIONS: SI was worn concurrently with attended PSG. Total sleep time ≥4 h on SI and PSG and SI signal quality >80% defined acceptable recordings for comparative analysis. MAIN OUTCOME MEASURES: Tolerability and diagnostic yield of SI, severity of obstructive sleep apnoea, sleep quality. RESULTS: 48 children were recruited, and 20 had acceptable recordings for comparison (median (IQR) age 11.8 (7.0-13.1) years, 45% female). Not including SI recordings affected by technical factors, SI tolerance was 70%. SI diagnostic yield was 49%. The median (IQR) Obstructive Apnoea Hypopnoea Index on PSG was 1.8 (0.8-6.6)/h compared to the comparable SI index (sAHIOBS) of 4.3 (2.8-6.4)/h (p = 0.31), with SI overestimating OSA severity at milder severities but under-estimating it in severe OSA. SI had a negative predictive value for moderate-severe OSA of 100% (95% CI 71.5% - 100%). SQI correlated negatively with arousal index (r = -0.45, p = 0.046). CONCLUSION: SI is tolerated in most children with neurodisability but diagnostic yield is limited by poor signal quality. It is useful for ruling out moderate-severe OSA, and can quantify non-OSA causes of impaired sleep quality.