Efficacité du ciment aux antibiotiques dans l’arthroplastie totale du genou de première intention : revue systématique et méta-analyse

Introduction The application of antibiotics loaded bone cement (ALBC) in the revision of failed total knee arthroplasty (TKA) has been widely accepted to reduce risk of peri-prosthetic infection. However, the prophylactic use of ALBC in primary TKA remains controversial. This study was aimed to identify the prophylactic effect on peri-prosthetic infection and safety of ALBC in primary TKA. Hypothesis The application of ALBC could reduce the risk of peri-prosthetic infection in primary TKA. Materials and methods Electronic platforms including PubMed, EMBASE, and CENTRAL were retrieved to identify studies comparing outcomes of prophylactic ALBC and plain cement in primary TKA. For outcomes reported as dichotomous variable and continuous variable, risk ratio (RR) and weighted mean difference (WMD) as well as their 95 % confidence intervals (95 %CI) were selected as the effect sizes for pooling. While for those outcomes reported the adjusted effect sizes such as odds ratio (OR, derived from multivariate logistic regression), and hazard ratio (HR, derived from multivariate COX proportional hazard model), the reported effect sizes were selected for pooling. Results A total of 17 studies with 2,074,844 patients (1,093,920 in ALBC group and 980,924 in plain cement group) were eligible for final inclusion. No significant difference was found between ALBC and plain cement groups both for the unadjusted (RR = 1.02, 95 %CI: 0.86∼1.21, p = 0.832) and adjusted (OR = 0.94, 95 %CI: 0.76∼1.17, p = 0.596) peri-prosthetic infection rate. ALBC application was related to significantly increased length of hospital stay (WMD = 0.13, 95 %CI: 0.10∼0.17, p < 0.001). There was no significance on the difference of operation related adverse events between two groups (RR = 1.31, 95 %CI: 0.68∼2.52, p = 0.420). Significantly increased risks of acute renal failure and readmission, and temporarily increased ototoxicity in ALBC group were reported in one of the primary study. Discussion There is no sufficient evidence supporting decreased peri-prosthetic infection rate with ALBC application in primary TKA. What's more, it must be taken into consideration about the safety and added cost of additional impregnated antibiotics. Level of evidence III; meta-analysis.