- Design
- Meta-analysis with trial sequential analysis of 4 randomised trials
- Population
- 224 adults aged ≥60 with idiopathic normal-pressure hydrocephalus
- Primary outcome
- Change in gait velocity; secondary function, cognition and harms
- Effect
- Gait SMD 0.73 (0.45–1.01); functional independence OR 3.92; positional headache OR 5.62
Idiopathic normal-pressure hydrocephalus is a potentially reversible cause of gait disturbance, cognitive decline and incontinence in older adults, and this meta-analysis pooled four randomised trials (224 patients) comparing immediate CSF shunting with no or sham shunting.
Shunting improved gait velocity (standardised mean difference 0.73) and functional independence (odds ratio 3.92), and modestly improved cognition (standardised mean difference 0.37). The harms were not trivial: positional headache was far more common (odds ratio 5.62), and the pooled estimate for surgically treated subdural haematoma, though not statistically significant, was too imprecise to exclude a real excess.
The practical reading is that shunting earns its place for the gait and functional disability of carefully selected iNPH, where the diagnosis is secure and a tap test supports it — but the cognitive benefit is smaller than patients may hope, and the bleeding risk has to be part of the conversation.
- CSF shunting improved gait velocity and roughly quadrupled the odds of functional independence.
- The cognitive benefit was real but modest, smaller than the gait gain.
- Positional headache was much more frequent, and a subdural haematoma excess could not be ruled out.
- Reserve shunting for carefully selected patients with a secure iNPH diagnosis.
Why it matters
It confirms a reversible cause of disability is worth treating, while refusing to oversell the cognitive return.
Don't overread it
With only 224 patients the safety estimates are imprecise — a non-significant subdural haematoma signal is not evidence the risk is absent.
The statistics, in plain English
A standardised mean difference of 0.73 for gait is a moderate-to-large effect; the wide confidence interval around the haematoma odds ratio (0.62–26.94) means the trials were simply too small to measure that harm reliably.
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 neurology, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free