Cardiorespiratory polygraphy (CRP) is the predominant technology used to diagnose obstructive sleep apnoea (OSA) in tertiary centres in the UK. Nocturnal pulse oximetry (NPO) is cheaper and more accessible. This study evaluates NPO indices' ability to predict OSA in children with Down syndrome (DS). Indices from simultaneous NPO and CRP recordings were compared in children with DS (aged 2-16 years) referred to evaluate OSA in two tertiary centres across an 8-year period. Receiver operating characteristic curves assessed the diagnostic accuracy of NPO indices, including ODI3 (3% Oxygen Desaturation Index) and ODI4 (4% Oxygen Desaturation Index). Two-by-two tables determined the sensitivities and specificities of cut-off values for predicting OSA. 387 children with DS were included with stand-alone NPO; 177 female (46.7%), median age 6.1 years (range 2.02-15.97). There were 265 children (68.5%) with Obstructive Apnoea-Hypopnoea Index (OAHI) ≥1/hour, 164 with OAHI ≥1<5/hour (42.4%), 51 with OAHI ≥5<10/hour (13.2%) and 50 with OAHI ≥10/hour (12.9%). ODI3 and ODI4 demonstrated the best predictive value for predicting OSA. An ODI3 ≥19/hour and an ODI4 ≥8/hour were associated with the highest combined sensitivity (59.2%/63.8%) and specificity (74.6%/71.3%), respectively. Raised ODI3 and ODI4 predict moderate and severe OSA in children with DS with moderate specificity/sensitivity and have a low sensitivity for detecting mild OSA. The poor predictive performance of oximetry reflects the multifactorial nature of sleep disordered breathing in children with DS. We recommend oximetry is not used for diagnosis of OSA in DS and CRP/polysomnography should be used.
Vennard et al. (Mon,) studied this question.