Adult native septic arthritis: 10 years in review in order to establish local guidelines on empiric antibiotic therapy

Abstract number: P1581

Clerc O., Senn L., Prod'hom G., Greub G., Zanetti G.

Objective: Antibiotic stewardship includes development of practice guidelines incorporating local microbiology and resistance patterns. In case of septic arthritis (SA), addition of vancomycin to the empiric therapy and broad-spectrum antibiotherapy in some clinical settings are subjects of discussion. Our objective was to review the local epidemiology of native septic arthritis in adults, in order to establish local guidelines for empiric therapy.

Methods: Retrospective study based on positive synovial fluid cultures and hospital discharge diagnoses of SA obtained from 1999 to 2008 in patients geqslant R: gt-or-equal, slanted16 years. Medical records were reviewed to assess the diagnosis and complete relevant clinical information.

Results: During this ten-year period, we identified 233 SA on native joints in 231 patients. 107 episodes (46%) were obtained through positive synovial fluid cultures, and 126 episodes (54%) through the discharge diagnosis. 147 SA (63%) were large joint infections (LJI). 35 SA (15%) occurred in intravenous drug users. Preexisting arthropathy was present in 51% of cases. 42% of patients with small joint infection (SJI) were diabetic, vs. 23% with LJI (p = 0.003). When available, synovial fluid direct examination was positive in 35% of cases. Etiologic agents are reported in the table. Five of the 11 MRSA SA (45%) occurred in known carriers. SJI were more frequently polymicrobial (24% vs. 1%, p < 0.001).

For LJI, an empiric treatment with amoxicillin/clavulanate (A/C) would have been appropriate in 85% of cases. MRSA (8 cases) and tuberculous (7 cases) arthritis would have been the most frequently untreated pathogens. Addition of vancomycin to A/C in MRSA carriers would rise the adequacy to 87%. In contrast, A/C would cover only 75% of SJI (82% if restricted to non-diabetic patients). MRSA (3 cases) and P. aeruginosa (9 cases, 7 monomicrobial) would be the main untreated pathogens. An anti-pseudomonal penicillin would have been appropriate in 94% of cases of SJI (P = 0.002 vs. A/C, p = 0.19 if diabetic patients not included).

Conclusions: Treatment with A/C seems adequate for empiric coverage of LJI in our setting. Broad-spectrum antibiotherapy was significantly superior for SJI in diabetic patients, due to different causative bacteria. In an area of low MRSA incidence, our results do not justify a systematic empiric therapy for MRSA, which should be considered in a known carrier.

Etiologic agentLarge joints
No. of cases (%) N = 147
Small joints
No. of cases (%) N = 86
Polymicrobial2 (1.4%)21 (24.4%)
Staphylococcus aureus
  – MSSA78 (53%)26 (30.2%)
  – MRSA8 (5.4%)3 (3.5%)
Streptococcus spp.20 (13.6%)13 (15.1%)
P. aeruginosa4 (2.7%)7 (8.1%)
Other Gram-negative14 (9.5%)6 (7%)
Coagulase-negative staphylococci3 (2%)3 (3.5%)
Others13 (8.8%)2 (2.3%)
Unknown5 (3.4%)5 (5.8%)

Session Details

Date: 10/04/2010
Time: 00:00-00:00
Session name: Abstracts 20th European Congress of Clinical Microbiology and Infectious Diseases
Location: Vienna, Austria, 10 - 13 April 2010
Presentation type:
Back to top