Abstract

Introduction and Background:

Nocturnal pulse oximetry is widely used to screen patients for obstructive sleep apnoea (OSA). It has recently been suggested that one night of pulse oximetry is sufficient in order to diagnose OSA using the McGill Oximetry Score (MOS).   

Aims:

i)     To determine whether there was night to night variation in oximetry in children with possible OSA.

Methods:

A retrospective observational study evaluating night to night consistency of oximetry recordings was conducted of patients referred with suspected OSA. Group 1 included patients with possible OSA but with no other underlying condition. Group 2 included patients with Down Syndrome. Group 3 included patients with underlying conditions other than Down Syndrome e.g. Prader-Willi syndrome. All patients had three nights of oximetry at home and these were independently scored by three investigators. Agreement analysis was done using Cohens kappa coefficient (k).

Results:

330 patients were studied. 35 (10.6%) patients had Down syndrome and 82 (24.8%) had other disorders. When comparing all three groups together, night 1 vs night 2; night 1 vs night 3; and night 2 vs night 3, 61%; 60% and 59% respectively of patients had the same MOS.

In Group 1 moderate agreement was shown between night 1 vs night 2 with k=0.42. In Group 2 only fair agreement was shown when comparing night 1 vs night 2 with k=0.37. Group 3 also showed fair agreement between night 1 vs night 2 with k=0.28.

Conclusion:

This study found significant night to night variability between MOS scores suggesting that a single night of oximetry may not be sufficient to screen for OSA.