Predictive Factors to Guide Empiric Antimicrobial Therapy of Acute Hematogenous Osteomyelitis in Children

Abstract

Background: Acute hematogenous osteomyelitis (AHO) is a serious infection in children. Pediatric Infectious Diseases Society guidelines recommend empiric methicillin-resistant Staphylococcus aureus (MRSA) therapy in regions where MRSA accounts for more than 10–20% of all staphylococcal osteomyelitis. We sought to examine factors present at the time of admission which may predict etiology and guide empiric treatment for pediatric AHO in a region with endemic MRSA. Methods: We reviewed admissions with International Classification of Diseases 9/10 codes for AHO from 2011 to 2020 in otherwise healthy children. Medical records were reviewed for clinical and laboratory parameters present on the day of admission. Logistic regression was used to determine clinical variables independently associated with (1) MRSA infection and (2) non-Staphylococcus aureus infection. Results: A total of 545 cases were included. An organism was identified in 77.1% of cases and S. aureus was the most common (66.2%); 18.9% of all AHO cases were MRSA. Organisms besides S. aureus were identified in 10.8% of cases. CRP >7 mg/dL, subperiosteal abscess, history of any prior skin or soft tissue infection (SSTI) and need for intensive care unit admission were independently associated with MRSA infection. Vancomycin was used as an empiric treatment in 57.6% of cases. If the above criteria were relied upon to predict MRSA AHO, empiric vancomycin use could have been reduced by 25%. Conclusions: Critical illness, CRP >7 mg/dL at the time of presentation, subperiosteal abscess and history of SSTI are suggestive of MRSA AHO, and could be considered when planning empiric therapy. Further work is needed to validate these findings before wider implementation.