A new systematic review and meta-analysis indicates that current evidence does not provide robust support for the routine use of antibiotic-loaded bone cement to prevent periprosthetic joint infection or revision surgery following primary total knee arthroplasty. Although antibiotic-loaded bone cement is widely utilized during primary knee replacements, its true clinical effectiveness compared with plain bone cement has remained uncertain. Across twenty-eight included reports synthesizing randomized trials and observational studies, pooled analyses yielded very low certainty evidence regarding whether the prophylactic addition of antibiotics provides meaningful protective benefits.
To evaluate post-surgical infection and revision outcomes, researchers conducted comprehensive database searches across PubMed, Web of Science, Scopus, and the Cochrane Library through May 31, 2026. The systematic search deliberately excluded Embase and prospective clinical trial registries, and the eligibility criteria restricted inclusion strictly to English-language reports. Before performing any statistical pooling, investigators mapped linked publications to prevent overlapping patient populations from distorting individual analyses. Despite these mapping efforts, the authors noted that a single deduplicated participant count could not be established reliably across the overall review.
Methodological quality and evidence confidence were systematically evaluated across the included literature. Result-level risk of bias was assessed using the Cochrane RoB 2 tool for randomized trials and the ROBINS-I instrument for non-randomized observational designs. The authors also applied the Grading of Recommendations Assessment, Development and Evaluation framework to determine overall outcome certainty. To avoid confounding study designs, investigators synthesized randomized or quasi-randomized trials separately from observational evidence, structuring distinct meta-analyses according to specific clinical outcomes, follow-up durations, reported effect measures, and statistical adjustment statuses using random-effects models with restricted maximum likelihood estimation.
In the synthesis of randomized and quasi-randomized evidence, investigators evaluated four studies encompassing 3,597 surgical procedures to measure incident periprosthetic joint infection or deep infection. This pooled randomized analysis demonstrated no statistically significant difference in infection rates between patients receiving antibiotic-loaded bone cement and those receiving plain bone cement, yielding an odds ratio of 0.449 with a 95 percent confidence interval spanning 0.152 to 1.328 (p = 0.148; I² = 38.5%). According to GRADE criteria, this pooled randomized finding was judged to be of very low certainty, preventing definitive conclusions regarding clinical efficacy.
Conversely, when researchers synthesized observational evidence, four studies reporting compatible adjusted odds ratios at approximately one year of follow-up demonstrated higher odds of infection among patients receiving antibiotic-loaded cement, with an odds ratio of 1.507 and a 95 percent confidence interval of 1.128 to 2.015 (p = 0.006). However, the authors emphasized that this observed statistical association was highly sensitive to individual study influence and heterogeneous covariate adjustment. Consequently, the authors cautioned that this paradoxical observational finding remains vulnerable to residual confounding rather than reflecting a demonstrated causal risk increase from antibiotic cement.
The review also assessed surgical durability by evaluating all-cause revision across three observational studies with compatible adjusted conventional Cox proportional hazards estimates. This pooled analysis found no statistically significant difference in all-cause revision between antibiotic-loaded cement and plain bone cement, reporting a hazard ratio of 0.888 with a 95 percent confidence interval from 0.754 to 1.045 (p = 0.154; I² = 72.0%). Like the infection outcomes, this pooled revision estimate was rated as having very low certainty. Endpoints for infection-related revision and reoperation could not be combined because clinical endpoints, follow-up intervals, effect measures, estimands, and study populations were insufficiently compatible.
In their conclusions, the authors emphasized that non-significant statistical findings do not establish clinical equivalence or exclude clinically important benefits or harms. Furthermore, descriptive comparisons across antibiotic categories in the published literature do not establish the clinical superiority of any specific antibiotic agent, dosage regimen, or bone cement formulation. Given the very low certainty across all pooled analyses and the persistent vulnerability to confounding in observational datasets, the researchers concluded that routine prophylactic use lacks robust empirical support, calling for more definitive prospective studies featuring standardized outcome definitions and carefully balanced perioperative care.
Subject of Research: Efficacy of antibiotic-loaded bone cement in primary total knee arthroplasty
Article Title: Association of antibiotic-loaded bone cement with infection and revision outcomes after primary total knee arthroplasty: a systematic review and meta-analysis
Article References: Lin, H., Tan, Z., & Si, H. (2026). Association of antibiotic-loaded bone cement with infection and revision outcomes after primary total knee arthroplasty: a systematic review and meta-analysis. Journal of Orthopaedic Surgery and Research. https://doi.org/10.1186/s13018-026-07345-7
Image Credits: AI Generated
DOI: 10.1186/s13018-026-07345-7
Keywords: total knee arthroplasty, antibiotic-loaded bone cement, plain bone cement, periprosthetic joint infection, all-cause revision, systematic review, meta-analysis, deep infection, orthopaedic surgery, random-effects models, REML estimation, RoB 2, ROBINS-I, GRADE certainty, residual confounding, primary TKA, reoperation
News Source: Ophelia Keating. (November 20, 2026). Review Questions Routine Antibiotic Bone Cement Use in Primary Knee Replacements. Scienmag.



