The McGill score as a screening test for obstructive sleep disordered breathing in children with co-morbidities.

Background: The McGill score is used to stratify severity of oximetry in children referred for investigation of obstructive sleep apnoea (OSA) to identify those with more severe disease and prioritize treatment. We hypothesized that its positive predictive value (PPV) and negative predictive value (NPV) in detecting OSA differs significantly between children with medical conditions and otherwise healthy children.

Methods: We performed a two-year retrospective analysis of children referred for investigation of OSA who underwent a cardiorespiratory (CR) polygraphy study. McGill score was calculated from the oximetry trace blinded to polygraphy results. We looked at two definitions of OSA: Obstructive Apnoea Hypopnoea Index (oAHI) ≥1 and ≥ 5. McGill sensitivity, specificity, PPV and NPV were calculated. McGill score = 1 was considered normal or inconclusive, >1 abnormal.

Results: We studied 312 children, 190 males (61%), median age 4.5 (2.4-7.9) years. 129 were otherwise healthy and 183 had associated medical conditions. The PPV of the McGill score was significantly lower in children with medical conditions than otherwise healthy children. The NPV was similar in both groups of children.

Conclusions: The higher number of false positives in children with medical conditions may be due to non-obstructive causes such as central apnoeas. Children with underlying lung disease are also more likely to desaturate following a brief apnoea or hypopnoea. Children with co-morbidities who have an abnormal McGill score should not be assumed to have OSA and need more detailed sleep studies to determine the reason for the oxygen desaturations.