1210. Staphylococcal acute post-operative prosthetic joint infection (PJI) treated with 'DAIR' (debridement and implant retention) and impact of rifampin: a retrospective cohort study in France
Session: Poster Abstract Session: Healthcare Epidemiology: MSSA, MRSA and Other Gram Positive Infections
Friday, October 5, 2018
Room: S Poster Hall
Background: Staphylococci are the most frequent bacteria in PJI. In patients with acute PJI (i.e. <1 month following the implantation), DAIR with exchange of removal components followed by a combination of antibiotics including rifampin (RMP) (particularly RMP + fluoroquinolone) are recommended. Unfortunately, some patients could not receive RMP due to drug-drug interaction or stopped it due to an adverse event. Finally, it is unclear if the dose and the duration of RMP influenced the prognosis.

Methods: Retrospective cohort study in 4 hospitals including patients with staphylococcal acute post-operative PJI treated with DAIR in 2011-2016. Univariate and multivariate Cox analysis and Kaplan Meier curves were used to determine the risk factors for treatment failure.

Results: 79 patients were included (median age: 71 years [IQR 53-89]; 55 men [69.6 %]; median ASA score: 2 [IQR 2-3]). Cultures revealed 65 (82%) S. aureus and 15 (19%) coagulase negative staphylococci infections, including 14 methicillin-resistant strains (18 %). Among all isolates, only 2 (3%) were resistant to RMP and 16 (20%) were resistant to fluoroquinolone. The median duration of antimicrobial therapy was 92 days (IQR 31-152). Only 59 patients received RMP (75%), and 35 (44%) the combination RMP + fluoroquinolone. Median duration of RMP was 57 days (IQR 16-86) and median dose 14.6mg/kg/d (IQR 13-17). Forty patients (51%) received RMP in the first 2 weeks and 43 patients (54%) received at least 2 weeks of RMP. Six patients (8%) developed an adverse event leading to RMP interruption. During a median follow-up of 443 days (IQR 220-791), 21 patients (27%) experienced a treatment failure including 12 persistence of the initial pathogen (57%) and 9 superinfections (43%). An ASA score >2 (OR 2.8; 95%CI 1.26-6.15), the use of RMP (OR 0.4; 95% CI 0.71-0.95) and the duration of RMP treatment (OR 0.83; 95%CI 0.75-0.92 per week of treatment) were significant determinants of the outcome (but not methicillin-resistance). Receiving >2 weeks of RMP prevented the failure, but an introduction during the first 2 weeks did not influence the outcome.

Conclusion: In patients with staphylococcal acute PJI, the use of RMP and its duration strongly influenced the prognosis. As 25% of patients could not receive RMP, new drugs with anti-biofilm activity are required.

Agathe Becker, MD1, Claire Triffault-Fillit, MD1, Emmanuel Forestier, MD2, Olivier Lesens, MD, PhD3, Bertrand Boyer, MD4, Stéphane Descamps, MD5, Christian Chidiac, MD, PhD1, Sébastien Lustig, MD, PhD6, Eric Montbarbon, MD7, Cécile Batailler, MD8, Celine Cazorla, MD9 and Tristan Ferry, MD, PhD10, (1)ID Department, Regional Reference Center for Bji, Hospices Civils de Lyon, Lyon, France, (2)ID Department, Centre Hospitalier Métropole Savoie, Chambéry, France, (3)Maladies Infectieuses, CHU de Clermont-Ferrand, Clermont-Ferrand, France, (4)Chirurgie Orthopédique Et Traumatologique, CHU de Saint Etienne, Saint Etienne, France, (5)Chirurgie Orthopédique Et Traumatologique, CHU de Clermont Ferrand, Clermont Ferrand, France, (6)Orthopaedic Surgery, Regional Reference Center for Bji, Hospices Civils de Lyon, Lyon, France, (7)Chirurgie Orthopédique Et Traumatologique, CH de Chambéry, Chambéry, France, (8)Chirurgie Orthopédique Et Traumatologique, CHU de Lyon, Lyon, France, (9)CHU de Saint Etienne, Saint Etienne, France, (10)Inserm 1111, UCBL1, Hospices Civils de Lyon, Lyon, France

Disclosures:

A. Becker, None

C. Triffault-Fillit, None

E. Forestier, None

O. Lesens, None

B. Boyer, None

S. Descamps, None

C. Chidiac, None

S. Lustig, None

E. Montbarbon, None

C. Batailler, None

C. Cazorla, None

T. Ferry, None

Findings in the abstracts are embargoed until 12:01 a.m. PDT, Wednesday Oct. 3rd with the exception of research findings presented at the IDWeek press conferences.