Pillar suturing and pharyngoplasty added no clear benefit to tonsillectomy in children's sleep apnoea

Design

Systematic review and meta-analysis of randomised and non-randomised comparative studies

Population

11 studies of children with obstructive sleep apnoea having tonsillectomy with or without adjuncts

Primary outcome

Postoperative apnoea-hypopnoea index and other polysomnography measures

Effect

Mean difference 2.22 events/hour (95% CI -4.36 to 8.80); haemorrhage OR 0.50 (0.13 to 1.89)

A meta-analysis of 11 randomised and non-randomised comparative studies assessed whether adding pillar suturing or pharyngoplasty to tonsillectomy, with or without adenoidectomy, improves outcomes in children with obstructive sleep apnoea.

Postoperative apnoea-hypopnoea index did not differ significantly (mean difference 2.22 events per hour, 95% CI -4.36 to 8.80). Neither did the odds of an index below 1 event per hour (OR 0.75, 0.33 to 1.72), surgical success or OSA-18 scores. Postoperative haemorrhage was comparable (OR 0.50, 0.13 to 1.89), and results held after excluding children with Down syndrome.

The wide intervals mean a small benefit or harm cannot be excluded, and many included studies were non-randomised. The authors conclude the evidence does not support routine adjunctive suturing and favour individualised decisions. For most children, tonsillectomy alone remains the straightforward operation.

• Consider tonsillectomy alone, without routine pillar suturing or pharyngoplasty, in children with sleep apnoea.
• Apnoea-hypopnoea index, success rate and symptom scores did not differ in pooled data.
• Haemorrhage rates were comparable, but the intervals were wide.
• Reserve adjunctive procedures for individual anatomical reasons and document the rationale.

Why it matters

It questions an adjunct added on instinct, with no signal in sleep study results or symptom scores.

Don't overread it

Many studies were non-randomised and small; the intervals cannot exclude a modest effect.

The statistics, in plain English

A mean difference of 2.22 events per hour with an interval from -4.36 to 8.80 is compatible with a meaningful benefit, no effect or a meaningful harm. Wide intervals reflect small and varied studies, so the finding is an absence of proof, not proof of equivalence.

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.