Association of antibiotic-loaded bone cement with infection and revision outcomes after primary total knee arthroplasty: a systematic review and meta-analysis

Antibiotic-loaded bone cement (ALBC) is widely used in primary total knee arthroplasty (TKA), but its effectiveness in preventing periprosthetic joint infection (PJI) remains uncertain. We evaluated infection and revision outcomes with ALBC versus plain bone cement (PBC). PubMed, Web of Science, Scopus, and the Cochrane Library/CENTRAL were searched from inception to May 31, 2026. Embase and trial registries were not searched, and eligibility was restricted to English-language reports. Randomized/quasi-randomized and observational evidence were synthesized separately by outcome, follow-up, effect measure, and adjustment status using random-effects models with REML estimation. Linked reports were mapped before synthesis to avoid overlapping populations within each analysis; a single deduplicated participant count across the review could not be established reliably. Result-level risk of bias was assessed using RoB 2 or ROBINS-I and certainty using GRADE. PROSPERO registration: CRD420251045784. Twenty-eight reports were included. Four randomized/quasi-randomized studies (3,597 procedures) showed no statistically significant difference in incident PJI/deep infection (OR 0.449, 95% CI 0.152–1.328; p = 0.148; I²=38.5%). Four observational studies with compatible adjusted odds ratios at approximately 1 year showed higher infection odds with ALBC (OR 1.507, 95% CI 1.128–2.015; p = 0.006); this association was sensitive to study influence and heterogeneous adjustment. Three observational studies with compatible adjusted conventional Cox estimates showed no statistically significant difference in all-cause revision (HR 0.888, 95% CI 0.754–1.045; p = 0.154; I²=72.0%). All three pooled estimates were of very low certainty. Infection-related revision and reoperation were not combined because endpoints, follow-up, effect measures, estimands, and populations were insufficiently compatible. Current evidence does not provide robust support for routine ALBC use to prevent PJI or revision after primary TKA. The adjusted observational infection association remains vulnerable to residual confounding. Non-significant results do not establish equivalence or exclude clinically important benefit or harm. Descriptive comparisons by antibiotic category do not establish superiority of an antibiotic, dose, or cement formulation. More definitive prospective studies with standardized outcomes and balanced perioperative care are needed.