Methicillin resistant Staphylococcus aureus versus methicillin sensitive Staphylococcus aureus adult haematogenous septic arthritis

Abstract

Introduction: Septic arthritis is an orthopaedic emergency and Staphylococcus aureus (SA) is the number one cause. Methicillin resistant Staphylococcus aureus (MRSA) is increasing in incidence but how it differs from methicillin sensitive Staphylococcus aureus (MSSA) septic arthritis is unclear. Our aim was to delineate the differences in clinical features and outcomes between patients with MRSA and MSSA septic arthritis.

Materials and methods: We performed a retrospective review of all adult patients presenting to our institute over a 5 year period from June 2000 to June 2005 with haematogenous septic arthritis. We identified 15 cases of MRSA and 43 cases of MSSA septic arthritis. Fisher's exact test and the Student's t-test were used in analysis.

Results: MRSA and MSSA predominantly affected males 60 versus 79%. MRSA cases were older with a mean age of 76 versus 44 years (P < 0.05), and had more comorbidities with a mean of 2.7 versus 1.35 (P < 0.05). In MRSA and MSSA cases the main sources of sepsis were unknown (20 vs. 47%), intravenous lines (47 vs. 2%), soft tissue infection (13 vs. 2%) and intravenous drug use (7 vs. 49%). MRSA was significantly more associated with intravenous line sepsis (P < 0.05), soft tissue infection (P = 0.05) and to be nosocomial (P < 0.05). MSSA was significantly more associated with IVDU (P < 0.05). Presentation was similar in both groups although MRSA patients were significantly more likely to be pyrexial (80 vs. 40%, P < 0.05) and to have glenohumeral involvement (P < 0.05) while MSSA was significantly more likely to affect the knee (P < 0.05). There were no significant differences between the total length of antimicrobial therapy in MRSA and MSSA patients (43 vs. 38 days, P > 0.2), or the number of surgical interventions required (1.8 vs. 1.6, P > 0.2). However MRSA patients were significantly more likely to be placed on inappropriate empirical antimicrobials pending sensitivities (93 vs. 0%, P < 0.05). Outcomes were similar in MRSA and MSSA patients with no significant differences in recurrences (0 vs. 10%, P > 0.2) or sepsis related mortality (13.3 vs. 6.9%, P > 0.2). MRSA, however, did show a strong towards a higher all cause 6 month mortality (26 vs. 7%, P = 0.07).

Conclusion: MRSA septic arthritis tends to affect older patients with multiple comorbidities and has a tropism for the glenohumeral joint while MSSA has a tropism for the knee. We did not find a significant difference in required length of antimicrobials, number or requirement of operative interventions or outcomes in terms of number of recurrences or sepsis related mortality. However MRSA septic arthritis patients were found to have a strong trend towards an increased all cause 6 month mortality and were significantly more likely to receive inappropriate empirical antimicrobials.