Impact of Methicillin-Resistant Staphylococcus aureus in Osteomyelitis and Soft Tissue Infections of the Foot

Background: The primary objective of this study is to assess if MRSA in osteomyelitis and soft tissue infections of the foot are associated with an increased time to treatment failure (defined as unanticipated resection of bone or above-ankle amputation). Secondary objectives include impact of antibiotic choice (active or not active against MRSA), MRSA colonization and other pathogens on time to treatment failure.

Methods: Patients admitted to a surgery service and underwent a surgical intervention from January 1, 2010 to December 31, 2014 were included in the study. Patients included were prescribed oral antibiotics while hospitalized, at discharge from the hospital or within four weeks of initial intervention date.

Results: Of 163 patients included in the study, 49 patients (30%) had treatment failure. MRSA was present in 25% (30/122) of all evaluable initial isolates and 20% (10/50) of all evaluable isolates at treatment failure. Median time to treatment failure was 61 days (range: 0-598 days). The presence of MRSA (p=0.27) and choice of MRSA antibiotic (p=0.18) were not associated with increased time to treatment failure by multivariate analysis. Absence of MRSA in nares strongly correlated with absence of MRSA in subsequent cultures (p=0.008). Other variables linked with increased time to treatment failure are shown in Table 1.

Table 1

Factor	P value	95% Confidence Interval	Hazard Ratio
Pseudomonas spp.	0.002	2.46-66.47	12.79
Corynebacterium spp.	<0.001	3.96-27.61	10.45
Escherichia coli	<0.001	2.44-18.84	6.78
Albumin <2.8 g/dL at presentation	0.003	1.38-5.19	2.68
Delayed primary closure	0.004	0.13-0.67	0.29

Conclusion: Presence of MRSA in cultures was not associated with increased time to treatment failure. Empiric anti-MRSA antibiotics, including perioperative/in-hospital vancomycin, may not be required due to low incidence of MRSA and low association with increased time to treatment failure. Other pathogens and variables may be better predictors of time to treatment failure.