Abstract: Levofloxacin plus rifampicin (L+R) is the treatment of choice for acute staphylococcal prosthetic joint infection
(PJI) managed with debridement and implant retention (DAIR). Long courses have been empirically
recommended, but some studies have suggested that shorter treatments could be as effective. Our aim
was to prove that a short treatment schedule was non-inferior to the standard long schedule. An openlabel,
multicentre, randomised clinical trial (RCT) was performed. Patients with an early post-surgical
or haematogenous staphylococcal PJI, managed with DAIR and initiated on L+R were randomised to receive
8 weeks of treatment (short schedule) versus a long schedule (3 months or 6 months for hip or knee
prostheses, respectively). The primary endpoint was cure rate. From 175 eligible patients, 63 were included
(52% women; median age, 72 years): 33 patients (52%) received the long schedule and 30 (48%)
received the short schedule. There were no differences between the two groups except for a higher rate
of polymicrobial infection in the long-schedule group (27% vs. 7%; P = 0.031). Median follow-up was 540
days. In the intention-to-treat analysis, cure rates were 58% and 73% in patients receiving the long and
short schedules, respectively (difference ?15.7%, 95% CI ?39.2% to 7.8%). Forty-four patients (70%) were
evaluable per-protocol: cure rates were 95.0% and 91.7% for the long and short schedules, respectively
(difference 3.3%, 95% CI ?11.7% to 18.3%). This is the first RCT suggesting that 8 weeks of L+R could be
non-inferior to longer standard treatments for acute staphylococcal PJI managed with DAIR.
