- Design
- Randomised, double-blind, placebo-controlled trial
- Population
- 1,028 adults with acute non-traumatic headache in the ED, all given oral paracetamol
- Primary outcome
- ≥30% reduction in pain score at 30 minutes
- Effect
- 78.9% vs 65.1%, difference 13.8% (95% CI 8–19); rescue analgesia 7.1% vs 15.3%
In this double-blind trial published in May 2026, 1,028 adults with acute non-traumatic headache in the emergency department all received 1 g oral paracetamol, then were randomised to 2 g IV magnesium sulphate or placebo over 30 minutes.
Treatment success (at least 30% pain reduction at 30 minutes) was 78.9% with magnesium versus 65.1% with placebo (difference 13.8 points, 95% CI 8–19). Rescue analgesia was needed less often (7.1% vs 15.3%). But differences in pain scores never reached the 1.3-point threshold for clinical importance. Adverse effects, mostly mild, were more frequent (15.4% vs 11.1%).
Magnesium is cheap and widely available. This trial suggests it helps some patients avoid rescue drugs, but the average patient will not feel a large difference.
- Consider IV magnesium 2 g over 30 minutes as an adjunct for acute headache, especially to reduce rescue analgesia.
- Warn patients about flushing and warmth during the infusion.
- Do not let symptomatic relief replace screening for red flags: thunderclap onset, focal signs, fever, neck stiffness.
- Antiemetic-dopamine antagonists remain effective first-line options for migraine in the ED.
Why it matters
It gives honest numbers for a cheap adjunct that is often used on habit.
Don't overread it
A statistically significant response rate is not a clinically important pain reduction — the trial itself says the difference fell below that threshold.
The statistics, in plain English
A responder analysis can show a clear difference in the share of patients crossing a line, even when average pain scores differ only slightly — both are true here.
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.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for emergency & critical care, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free