Investigation of prevalence of MRSA in referrals to neurosurgery: implications for antibiotic prophylaxis.

In order to establish the appropriateness of our current prophylactic antibiotic regimen we analysed the prevalence of MRSA in emergency referrals to our unit. MRSA screening records for all emergency admissions for a 3-month period were analysed. One-hundred-and-seventy-five patients were admitted as transfers from another hospital. Evidence of screening was found in 61% (107 patients). Of the screened patients, 15% (16) were MRSA positive. Source of referral or length of inpatient stay after referral to the time of transfer were not predictive for MRSA status. Gentamicin is active against more than 95% of MRSA strains cultured in our hospital and against 87% of MRSA strains cultured in the neurosurgery unit. A number-needed-to-treat (NNT) analysis showed that, with MRSA prevalence at 15%, cefuroxime plus gentamicin at induction could prevent one MRSA infection per 421 treated patients compared with cefuroxime alone. Vancomycin had minimal additional benefit over cefuroxime plus gentamicin (NNT: 1684). We conclude that MRSA carriage is common in patients referred as emergencies from other hospitals. Cefuroxime plus gentamicin can be used as antibiotic prophylaxis in this group. Vancomycin can be reserved for patients known to be colonized with MRSA (NNT: 51).