# Do not treat ghosts: anti-methicillin-resistant Staphylococcus aureus (MRSA) therapy in osteomyelitis without identified MRSA

## TL;DR

This study compares outcomes of treating lower limb osteomyelitis with or without anti-MRSA therapy when MRSA is not detected, finding no significant differences in treatment success or side effects.

## Contribution

The study provides evidence that anti-MRSA therapy may not be necessary for osteomyelitis when MRSA is not identified in cultures.

## Key findings

- Treatment failure rates were not significantly different between anti-MRSA and non-anti-MRSA therapy groups.
- Adverse events and hospital readmission rates were similar in both treatment groups.
- No significant differences in hospital length of stay or new renal replacement therapy initiation were observed.

## Abstract

To compare the clinical outcomes of patients with lower limb osteomyelitis (LLOM) and negative methicillin-resistant Staphylococcus aureus (MRSA) cultures treated with anti-MRSA therapy (AMT) versus those treated with no-anti-MRSA therapy (NAMT).

Retrospective cohort study.

Hospitalized adult (≥18 yr of age) patients admitted to multiple tertiary referral centers in a single healthcare system between April 1, 2017 and April 1, 2023, with LLOM and planned intravenous antibiotics for at least four weeks.

Electronic medical records were queried for demographic information, admission dates, treatment strategies, imaging and culture results, and discharge diagnoses. Descriptive statistics measured baseline characteristics, imaging, and culture results.

Out of 473 patients, 64 met the inclusion criteria and 409 were excluded. Of the 64 patients, 26 (40%) had AMT and 38 (59%) had NAMT. A larger but statistically insignificant portion of patients in the NAMT cohort failed therapy (23% AMT vs 32% NAMT, P = 0.325). However, hospital readmission for LLOM within 180 days of antibiotic completion (46.2% vs 47%, P = 0.92), hospital length of stay (median (IQR): 6 (5–9) d vs 7 (5–12.5) d, P = 0.285), incidence of new renal replacement therapy initiation (0% vs 2.6%, P = 0.594), creatinine kinase levels (0 vs 2.6%, P = 0.594), and drug-induced immune thrombocytopenia (0% vs 5.3% P = 0.349) were comparable between the two cohorts.

Treatment failure rates and adverse events did not differ significantly among patients with LLOM treated with AMT or NAMT. Further investigation of determinants of clinical failures in LLOM may help optimize overall treatment.