Effectiveness of Doxycycline in Combination with Other Antibiotics for Gram-Positive Periprosthetic Joint Infections: a Causal Inference Study

Abstract Objectives Doxycycline is occasionally used as step-down therapy in periprosthetic joint infections (PJI), but evidence supporting its efficacy is limited. This study aimed to estimate the effect of doxycycline on 12-month treatment failure in patients with Gram-positive PJI. Methods Adult patients with hip, knee, or shoulder PJI caused by Staphylococcus, Corynebacterium, or Cutibacterium who underwent surgery between April 2013 and April 2023 at Dijon university hospital (France) were included. Demographic, clinical, biological and therapeutic data were collected retrospectively. Treatment failure at 12 months was defined as clinical recurrence, new intraoperative microorganisms, surgical revision for infection, or death. The average treatment effect (ATE) of doxycycline was estimated using causal inference methods. Results 386 patients with PJI (median age 72 years, IQR=65-80) were analyzed. Most infections involved the hip (62%), were caused by S. aureus (64%), and/or polymicrobial (42%). Doxycycline was prescribed in 19% of patients (n=72), for a median of 64 days (IQR=42-84). At 12 months, treatment failure occurred in 35%, without significant difference between exposed and unexposed patients (33% vs. 36%, p=0.68). Overall, doxycycline was not significantly associated with treatment failure (ATE(IPTW) =-0.09; 95%-CI=-0.24 to 0.05; p=0.19). Subgroup analyses suggested that doxycycline reduced treatment failure by 23-26% in S. aureus infections (p0.001), 25-28% in patients without fever (p0.001), and 34-35% when both conditions were present (p0.001). Conclusion Doxycycline in combination with other antibiotics, was not associated with 12-month treatment failure in PJI caused by Staphylococcus, Corynebacterium, or Cutibacterium, with potential benefits in S. aureus infection warranting confirmation in prospective studies.