Osteocutaneous flap (OCF) mandible reconstruction is at high risk for surgical site infection. This study aimed to describe diagnosis, management, and outcome of OCF-related osteomyelitis. All patients managed at our institution for an OCF-related osteomyelitis following mandible reconstruction were included in a retrospective cohort study (2012-2019). Microbiology was described according to gold-standard surgical samples, considering all virulent pathogens, and potential contaminants if present on at least two samples. Determinants of treatment failure were assessed by logistic regression and Kaplan-Meier curve analysis. The 48 included patients (median age 60.5 (IQR, 52.4-66.6) years) benefited from OCF mandible reconstruction mostly for carcinoma ( ; 56.3 %) or osteoradionecrosis ( ; 25.0 %). OCF-related osteomyelitis was mostly early ( months post-surgery; ; 89.6 %), presenting with local inflammation ( ; 59.6 %), nonunion (wound dehiscence) or sinus tract ( ; 59.6 %), and/or bone or device exposure ( ; 44.7 %). Main implicated pathogens were Enterobacteriaceae ( ; 61.0 %), streptococci ( ; 53.7 %), Staphylococcus aureus ( ; 24.4 %), enterococci ( ; 22.0 %), non-fermenting Gram-negative bacilli ( ; 19.5 %), and anaerobes ( ; 19.5 %). Thirty-nine patients (81.3 %) benefited from surgery, consisting of debridement with implant retention (DAIR) in (64.1 %) cases, associated with 93 (IQR, 64-128) days of antimicrobial therapy. After a follow-up of 18 (IQR, 11-31) months, (50.0 %) treatment failures were observed. Determinants of treatment outcomes were DAIR (OR, 3.333; 95 % CI, 1.020-10.898) and an early infectious disease specialist referral (OR, 0.236 if weeks; 95 % CI, 0.062-0.933). OCF-related osteomyelitis following mandibular reconstruction represents difficult-to-treat infections. Our results advocate for a multidisciplinary management, including an early infectious-disease-specialist referral to manage the antimicrobial therapy driven by complex microbiological documentation.
Copyright: © 2022 Clément Javaux et al.