Design: Multicentre, randomised, open-label, non-inferiority trial
Population: 500 adults after surgery for orthopaedic infection with a local-antibiotic carrier
Primary outcome: Definite treatment failure by 12 months
Effect: 11.1% (≤7 days) vs 14.1% (≥4 weeks); risk difference −3.0 points (95% CI −9.0 to 3.0)

SOLARIO randomised 500 adults who had surgery for orthopaedic infection with a local-antibiotic carrier implanted at operation. They received either seven days or less, or at least four weeks, of systemic antibiotics afterwards.

Definite treatment failure by 12 months occurred in 11.1% with the short course and 14.1% with the long (risk difference −3.0 points, 95% CI −9.0 to 3.0), meeting the 10-point non-inferiority margin. Per-protocol and sensitivity analyses agreed. By six weeks, side effects potentially related to treatment occurred in 17.2% with the short course against 45.2% with the long.

For surgeons and infection teams, this supports stopping systemic antibiotics after a week when local antibiotics have been placed. It does not apply to patients managed without surgery or without local antibiotics, who were not studied.

Where a local-antibiotic carrier was implanted at surgery, consider limiting systemic antibiotics to about a week.
Side effects fell from almost half to about one in six with the shorter course.
The finding applies only to patients who had surgery and local antibiotics — not to medical management alone.
Agree the stop date at the operation, so the course is not extended by default.
Fewer weeks of antibiotics also means fewer line days and less resistance pressure.

Why it matters

Six weeks of antibiotics for bone infection has been routine for decades, and this trial shows most of that course may add harm without benefit.

The statistics, in plain English

Non-inferiority here means the short course was shown not to be worse by more than 10 percentage points; the confidence interval's upper limit was 3.0 points, well within that. The interval also includes the possibility that the short course is slightly better.