Post-tonsillectomy haemorrhage: 2.65% overall, mostly secondary, higher in adults

Design

Retrospective single-centre cohort, 2015 to 2024

Population

945 patients having tonsillectomy, mean age 8.7 years (85.7% children)

Primary outcome

Post-tonsillectomy haemorrhage and its predictors

Effect

2.65% (95% CI 1.7 to 3.9); adults 7.4% vs children 1.9%; adjusted OR for adult age 2.93 (1.17 to 7.36)

This retrospective single-centre study covered 945 tonsillectomies from 2015 to 2024 at a Lebanese tertiary hospital (mean age 8.7 years; 85.7% children). Haemorrhage was defined as documented bleeding needing medical evaluation, classed as primary (within 24 hours) or secondary.

It occurred in 25 patients (2.65%, 95% CI 1.7 to 3.9). Ninety-two per cent was secondary, typically around postoperative day 7. Adults bled more often than children (7.4% against 1.9%, P = 0.001), and adult age was the only independent predictor (adjusted OR 2.93, 95% CI 1.17 to 7.36). Sex, operative time, surgeon, concurrent procedures, pathology and routine laboratory values were not independently associated. Five patients (20% of those who bled) needed transfusion.

There were only 25 events, so the confidence interval for age is wide and other risks may have been missed. The finding suits what many surgeons expect: counsel adults specifically, and make sure families know how to act on bleeding in the second week.

• Counsel adult patients that bleeding risk is higher than in children, around three times the odds here.
• Warn families that bleeding most often comes about a week after surgery, not in the first day.
• Give a clear plan for bleeding at home, including where to go and how quickly.
• Arrange access to emergency review during the second postoperative week.
• Audit your own rates by age group and by primary or secondary timing.

Why it matters

The danger window falls after discharge, when most patients are at home, so counselling matters more than inpatient observation.

Don't overread it

A retrospective single-centre cohort with 25 events cannot rule out other risk factors or give rates for other settings.

The statistics, in plain English

With 25 bleeds in 945 patients, estimates for individual risk factors are imprecise; an interval of 1.17 to 7.36 means the true effect of adult age could be small or large. Absence of an association with surgeon or technique may only reflect too few events to detect one.

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.