Search for a command to run...
We read with great interest the prospective cohort study by Yang and colleagues examining the duration of surgical antibiotic prophylaxis (SAP) and the risk of surgical site infection (SSI) in orthopedic surgery[1]. The authors highlight a familiar dilemma: when procedures are lengthy, drains are left in situ, or implants are used, clinicians may be tempted to continue antibiotics beyond the traditional “prophylaxis window.” This tendency is understandable, yet it sits in tension with major guidance documents, which generally emphasize correct agent selection and execution, and discontinuation within 24 hours for most operations[2,3]. Yang et al, therefore, tackle an important and highly actionable question. That said, interpreting an apparent benefit of prolonged SAP in an observational cohort requires particular caution. SAP duration is rarely assigned at random; it may reflect surgeons’ anticipation of risk or early postoperative signals. Revision operations, open injuries, soft-tissue compromise, contamination class, prolonged operative time, transfusion, and early postoperative fever can all prompt escalation from prophylaxis to empiric treatment. If these indications are incompletely captured, residual confounding may bias estimates toward benefit. We suggest clarifying how “prophylaxis” was distinguished from therapeutic antibiotics, reporting whether the stop decision occurred pre-operatively or post-operatively, and considering time-dependent methods (e.g., landmark analyses or models that treat duration as a time-varying exposure) to reduce immortal-time and indication bias. To reduce bias, it would be helpful to (1) define SAP duration as a time-varying exposure to mitigate immortal-time bias, (2) report propensity-score or inverse-probability weighting diagnostics for covariate balance, and (3) perform sensitivity analyses excluding patients in whom antibiotics were extended because of suspected early infection or unplanned re-intervention. Clear reporting of SSI ascertainment and follow-up windows would further strengthen interpretability. In addition, the manuscript would benefit from a more granular description of SAP fidelity. Guidance stresses weight-based dosing, timely incision-to-dose interval, and intraoperative re-dosing when procedures are prolonged or blood loss is substantial[2,3]. Duration is only one component, and prolongation cannot compensate for delayed administration or missed re-dosing. Reporting antibiotic class (including any anti-MRSA coverage), dose, timing relative to incision, re-dosing criteria, protocol adherence, and adjunctive local measures (e.g., vancomycin powder when used) would help readers interpret whether duration is acting as a proxy for broader care quality. Moreover, the orthopedic evidence base on “how long is enough” remains mixed across procedure types. For fracture surgery, a Cochrane review supports perioperative prophylaxis to reduce infection, but it does not establish that routine postoperative extension provides incremental benefit over appropriately delivered perioperative dosing[4]. Given the heterogeneity of orthopedic procedures, we suggest stratified analyses (elective arthroplasty vs trauma; primary vs revision; clean vs contaminated; implant vs no implant) and interaction testing to identify subgroups – if any – in which longer SAP appears beneficial. Sensitivity analyses using propensity-score weighting or restriction to patients meeting standardized “uncomplicated course” criteria could further strengthen causal interpretation. We also encourage the authors to predefine clinically meaningful cut-offs (e.g., ≤24, 24–48, and >48 hours) and to report absolute risk differences alongside relative measures, because small relative changes may translate into different clinical decisions across baseline-risk strata. Finally, infection prevention in orthopedics is increasingly bundle-based rather than antibiotic-centric. Randomized evidence shows that screening and decolonization of Staphylococcus aureus carriers can reduce postoperative infections[5], and broader decolonization strategies have reduced healthcare-associated infections without relying on prolonged systemic antibiotics[6]. Framing extended SAP as one possible component within a broader bundle (nasal decolonization, chlorhexidine bathing, glycemic control, normothermia, meticulous hemostasis, and optimized wound care) may offer a clearer pathway for pragmatic implementation. From a stewardship perspective, the goal is to minimize unnecessary duration while preserving efficacy[7]. In the context of the global burden of bacterial antimicrobial resistance[8], adopting longer SAP as routine practice should require robust evidence of net benefit, explicit reporting of antibiotic-related harms, and discussion of local resistance ecology. We believe these clarifications and analyses would strengthen interpretation and improve the translational value of Yang et al.’s important work.