Skull-base surgery: longer antibiotic prophylaxis was not linked to less meningitis

Design

Systematic review and meta-analysis with meta-regression

Population

7,981 patients across 14 studies of endoscopic endonasal skull-base surgery

Primary outcome

Postoperative meningitis by prophylaxis duration

Effect

Overall 1.38% (95% CI 0.92–2.08%); ≤1 day 1.24% vs >3 days 2.12%, p=0.76

A meta-analysis of 14 studies and 7,981 patients undergoing endoscopic endonasal skull-base surgery with a single prophylactic antibiotic found postoperative meningitis in 1.4%. Protocols were grouped by duration: up to one day, one to three days, and more than three days.

Meningitis rates were 1.2% with up to one day, 1.2% with one to three days and 2.1% with longer prophylaxis, with no significant difference between groups, and duration showed no association in meta-regression. What did track with meningitis was the proportion of patients with intraoperative CSF leak.

These are study-level observational data with moderate heterogeneity, and centres choosing longer courses may operate on riskier cases. Still, the pattern fits wider surgical evidence that extended prophylaxis does not add protection, and it moves attention to the factor that matters: repairing the leak. For ID physicians advising neurosurgical units, it supports challenging multi-day prophylaxis and nasal-packing antibiotics by default.

• Consider limiting prophylaxis to 24 hours or less after endoscopic endonasal skull-base surgery.
• Assess intraoperative CSF leak as the main meningitis risk factor and ensure a robust repair.
• Investigate fever, headache or neck stiffness after skull-base surgery promptly with CSF analysis.
• Review local protocols that continue prophylaxis until nasal packs are removed.

Why it matters

It challenges the habit of multi-day prophylaxis in a surgery where meningitis is feared but uncommon.

Don't overread it

Study-level observational data cannot prove that short courses are as safe in high-risk leaks.

The statistics, in plain English

The subgroup rates have overlapping confidence intervals, and the meta-regression p-value of 0.49 means no detectable trend with duration. Because the data compare studies rather than randomised patients, a small benefit of longer courses cannot be fully excluded.

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.