One night of oximetry is not enough to clear a child with Down syndrome

Design

Retrospective review of routine screening data with receiver operating characteristic analysis, single UK centre 2020-2024

Population

246 three-night pulse oximetry studies (150 complete) in children aged 1 to under 18 with Down syndrome

Primary outcome

Clinically significant night-to-night variability across published screening thresholds

Effect

53 per cent of complete studies crossed the threshold on some nights only; reassurance thresholds gave negative predictive values of 79 and 81 per cent but applied to 13 per cent of first nights

Two hundred and forty-six three-night pulse oximetry studies in children aged 1 to under 18 with Down syndrome were reviewed, 150 of them complete, against published thresholds predictive of moderate-to-severe obstructive sleep apnoea in this group - a 3 per cent oxygen desaturation index of 6.15 or more, or a delta index of 0.555 or more.

Fifty-three per cent of the complete studies crossed the threshold on some nights and not others. That is not measurement noise at the margins; it is the majority of children producing a different screening answer depending on which night you look at. Receiver operating characteristic analysis identified lower reassurance thresholds - an oxygen desaturation index below 3.61 or a delta index below 0.49 on the first night - below which a subsequent abnormal night was unlikely, with negative predictive values of 79 and 81 per cent. But only 13 per cent of first nights fell below those thresholds, so the reassurance applies to a small minority.

Two further findings deserve carrying forward. Sixty per cent of the diagnostic polygraphy studies showed obstructive sleep apnoea, and 76 per cent of the moderate-to-severe disease was in children over six - an age at which many screening protocols have stopped looking. If your service screens with a single night and concentrates on the preschool years, it is likely to be missing both.

• Do not clear a child with Down syndrome on one normal night of oximetry unless the values are well below threshold.
• Use a lower reassurance cut-off for a single night rather than the diagnostic threshold.
• Keep screening past age six; that is where most of the moderate-to-severe disease sat.
• Repeat testing where symptoms and a single normal night disagree - the symptoms are more likely to be right.
• Record which night of the study each result comes from, so variability is visible rather than averaged away.

The statistics, in plain English

A negative predictive value of about 80 per cent means one child in five below the reassurance threshold still had an abnormal night later - useful, but not a clearance. Because only 13 per cent of first nights fell below those thresholds, the rule identifies a small group and leaves the majority needing further nights. This is a single-centre retrospective review of routine data, so the thresholds derived here need testing elsewhere before being adopted as protocol.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.