Acute non-traumatic headache is a high-volume emergency presentation with no settled analgesic ladder, and intravenous magnesium has been used on the strength of small studies and mechanistic reasoning. This randomised, double-blind, placebo-controlled trial gave every patient 1 g oral paracetamol first, then randomised 1,028 adults to 2 g intravenous magnesium sulphate in 150 mL saline or to saline alone, infused over 30 minutes.
On the primary outcome — at least a 30% fall in numerical rating scale score at 30 minutes — magnesium won clearly: 78.9% against 65.1%, a difference of 13.8 percentage points (95% CI 8 to 19). But the authors are candid that every timed difference in the actual pain scores sat below the 1.3-point threshold for clinical importance. Secondary outcomes went the same way: rescue analgesia was needed in 7.1% against 15.3% (difference -8.2%, 95% CI -12 to -4.3) and satisfaction was higher, 91.7% against 85.1%. Adverse events were commoner with magnesium, 15.4% against 11.1% (difference 4.3%, 95% CI 0.1 to 8.4).
The two findings pull in different directions and both are true. More patients crossed the responder threshold, and needed rescue analgesia half as often — those are real, department-relevant outcomes. Yet the average patient's pain score did not move by an amount they would perceive as meaningful. The way to reconcile this is that magnesium probably helps a subset substantially rather than helping everyone a little, and the responder analysis captures that while the mean does not.
How to use it: not as a routine addition to every headache, but as a reasonable second step in a patient who has had simple analgesia and is not settling, before escalating to opioids. The halving of rescue analgesia is the outcome that matters most in a busy department. Be honest about the cost — roughly one extra patient in 25 gets a side effect, mostly flushing, warmth or infusion-site discomfort — and remember that 30 minutes of infusion occupies a bed and a line. And nothing here removes the obligation to think about why the headache is there: this is analgesia for undifferentiated benign headache, after the dangerous causes have been considered.
- Consider 2 g intravenous magnesium over 30 minutes when simple analgesia has not settled an acute non-traumatic headache.
- The best reason to use it is halved need for rescue analgesia, not a large drop in pain score.
- Warn about flushing and warmth; adverse events ran about 4 percentage points higher than placebo.
- Do not add it reflexively to every headache — the average benefit is below what patients perceive.
- Exclude the dangerous causes first; this is symptomatic treatment of benign headache.
The statistics, in plain English
The gap between the two results comes from how they are measured. The primary outcome is a responder analysis — the proportion crossing a 30% improvement line — and dichotomising a continuous score this way can produce a large-looking difference from a small shift in the underlying distribution, because patients sitting just below the line get pushed over it. The mean pain scores, which do not have that amplifying effect, moved by less than the 1.3 points patients reliably notice. Both numbers are honest; the rescue analgesia result, at 7.1% against 15.3%, is the one least vulnerable to this artefact because it counts an actual clinical action.
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