DOXYCYCLINE MONOTHERAPY FOR BONE AND JOINT INFECTIONS: A RETROSPECTIVE COHORT STUDY OF 163 PATIENTS IN A SPECIALIST ORTHOPAEDIC HOSPITAL

Introduction Doxycycline is not internationally used as an oral option for bone and joint infection (BJI) compared with clindamycin, co-trimoxazole, fluoroquinolones, and rifampicin, despite favourable bone penetration and tolerability. This study aimed to evaluate doxycycline monotherapy for curative treatment of BJI, including prosthetic joint infection (PJI), and osteomyelitis (OM) with or without metalwork. Methods This retrospective cohort study included patients aged ≥14 years from a prospective oral antibiotic therapy database at a specialist orthopaedic hospital (May 2017-February 2025). Inclusion criteria were: diagnosed BJI (PJI per European Bone and Joint Infection Society [EBJIS] and International Consensus Meeting [ICM] 2018 criteria, or OM per standard criteria) treated with doxycycline monotherapy (6 weeks total, or 12 weeks following single-stage revision or debridement and implant retention (DAIR)). As per the OVIVA study, primary outcome was definite failure at 42, 180, and 365 days. Secondary outcomes included probable/possible failure, adverse effects, early discontinuation and mortality. Results Of 214 screened patients, 163 were included (51 excluded for not meeting diagnostic criteria or being on combination therapy/suppressive regimen). Diagnoses included PJI (n=101, including 25 megaprostheses), OM with metalwork (n=39) and native OM (n=23). Surgery was performed in 161/163 patients (98.8%). Prior infection episodes affected 39.9% (65/163) of patients. Infections were polymicrobial in 15.2% (25/163) and culture-negative in 5.5% (9/163). Main pathogens were Staphylococcus aureus (68/163, 41.7%), Staphylococcus epidermidis (46/163, 28.2%) and Cutibacterium acnes (34/163, 20.9%). Doxycycline was administered for ≥50% of treatment duration in 94.6% (mean 82% of total treatment duration). At 365 days, definite failure rates were: whole cohort 21.9% (mean time to failure 152 days); PJI cohort 27% (215 days); non-PJI cohort 14% (163 days). Probable/possible failure rates at 365 days were 4.8% overall. Early discontinuation occurred in 6.1% (n=10): 3 for early failure, 7 for adverse effects (pruritus, headache, oesophagitis). Twenty further patients had mild adverse effects (13 gastrointestinal, 5 oral thrush, 4 liver enzyme elevations, 1 photosensitivity). No treatment-related mortality occurred. Conclusions Doxycycline monotherapy demonstrated comparable definite failure rates to historical cohorts (18.6% in previous institutional OVIVA implementation study), with good tolerability. These findings support doxycycline as a safe oral antibiotic option for Gram-positive BJI treatment.