Non-elective and revision arthroplasty are independently associated with hip and knee prosthetic joint infection caused by Acinetobacter baumannii: a Brazilian single center observational cohort study of 98 patients

Background

Prosthetic joint infection (PJI) caused by Acinetobacter baumannii (Ab) has become a growing concern due to its overwhelming ability to express resistance to antibiotics and produce biofilm.

Aim

This study aimed to identify independent risk factors (RFs) associated with Ab-associated PJI and their role in the treatment outcome.

Methods

This was a single-centre, retrospective cohort study of PJI patients diagnosed between January 2014 and July 2018. A PJI diagnosis was made based upon the MSIS 2018 criteria. To estimate RFs associated with Ab-associated PJI, multivariate analyses with a level of significance of p < 0.05 were performed. To evaluate treatment failure, Kaplan–Meier analysis and log-rank test were performed.

Results

Overall, 98 PJI cases were assessed, including 33 with Ab-associated PJI and 65 with PJI involving other microorganisms (non–Ab-associated PJI). Independent RFs associated with Ab-associated PJI were revision arthroplasty [odds ratio (OR) = 3.01; 95% confidence interval (CI) = 1.15–7.90; p = 0.025] and nonelective arthroplasty (OR = 2.65; 95% CI = 1.01–7.01; p = 0.049). Ab-associated PJI was also more likely than non–Ab-associated PJI to be classified as a chronic late infection (OR = 5.81; 95% CI = 2.1–16.07; p = 0.001). Ab-associated PJI was not associated with treatment failure (p = 0.557).

Conclusions

Late chronic infections, surgical revision and nonelective arthroplasty are well-known predictors of PJI but were also independently associated with Ab-associated PJI. Infections caused by Ab and surgical treatment with debridement, antibiotics and implant retention were not associated with PJI treatment failure.