- Design
- Cochrane systematic review and meta-analysis of RCTs
- Population
- 12 trials, 1,860 adults after ischaemic or haemorrhagic stroke
- Primary outcome
- Motor function; serious adverse events
- Effect
- Low-bias trials SMD 0.02 (−0.10 to 0.14); serious adverse events RR 1.03 (0.83–1.27)
This updated Cochrane review included 12 randomised trials with 1,860 participants, mostly testing levodopa, one testing ropinirole, alongside equivalent rehabilitation. Trials ran in Europe, the USA and Asia between 2001 and 2025.
There was no difference in continuous motor function scores (SMD 1.42, 95% CI −0.20 to 3.05, very low certainty) or in favourable motor outcomes (RR 0.93, low certainty). Heterogeneity was extreme, but restricting to the two trials at low risk of bias — which held most of the patients — removed it and gave an SMD of 0.02, essentially zero. Serious adverse events did not differ, nor did any other outcome.
Levodopa has been prescribed off-label to aid stroke rehabilitation on the strength of small early trials. The largest, best-conducted trials now point to no benefit, which argues for stopping routine use and putting the effort into rehabilitation itself.
- Do not start levodopa or a dopamine agonist to enhance motor recovery after stroke.
- Review existing off-label prescriptions given for this purpose and consider stopping them.
- Focus resources on early, intensive, task-specific rehabilitation.
- Continue dopaminergic drugs where there is a separate indication, such as Parkinson's disease.
Why it matters
It removes the evidence base for a common off-label add-on to stroke rehabilitation.
Don't overread it
The review cannot exclude benefit in a specific subgroup, but none has been shown.
The statistics, in plain English
A standardised mean difference of 0.02 (−0.10 to 0.14) in the low-bias trials means the two groups ended up almost identical. The headline SMD of 1.42 looks large only because a few small, biased trials inflated it, as the I² of 99% signals.
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