Abstract:[Objective] To search the factors related to failure of antibiotic bone cement spacer in the treatment of periprosthetic joint in- fection (PJI). [Methods] A retrospective study was conducted on 69 patients who received debridement and antibiotic bone cement spacer in the primary stage for PJI in our hospital from January 2017 to July 2020. The patients were divided into the re-infection group and cured group according to whether they had re-infection after the antibiotic bone cement spacer was removed. Univariate comparison and Cox analy- sis were used to investigate the influencing factors of reinfection after antibiotic bone cement spacer removed. [Results] All patients were fol- lowed up for 0.5 to 52.9 months after the debridement and antibiotic bone cement spacer placement. Among the 69 patients, 15 patients had reinfection, which occurred at 0.5 to 9.1 months after the surgery with a median time of 2.2 months after the surgery. As results of univariate comparison, there were no statistically significant differences in terms of age, gender composition, smoking history, drinking history, ASA grade, aCCI score, diabetes, hypertension, coronary heart disease, rheumatoid arthritis, past infection history and microbial culture results between the two groups (P>0.05); However, the re-infection group had significantly higher ratio of previous same-joint revision history [yes/ no, (5/10) vs (4/50), P<0.05] and the sinus ratio [yes/no, (7/8) vs (9/45), P<0.05] than the cured group. Based on the consequence of Cox analy- sis, the previous revision history (HR=3.755, 95% CI 1.264~11.152, P=0.017) and the presence of sinus tract (HR=2.834, 95% CI 1.018~ 7.885, P=0.046) were independent risk factors for recurrent infection in PJI after removal of the antibiotic bone cement spacer. [Conclu- sion] Previous revision history and the presence of sinus tract are independent risk factors for re-infection in PJI patients after removal of the antibiotic bone cement spacer which should be paid attention to by clinicians.