Treatment of non–purulent cellulitis and cellulitis associated with an abscess is a common reason for hospital admission. Methicillin resistant staph aureus (MRSA) is well documented be a common cause of purulent skin infections, while its relationship to non–purulent infections is much less clear. Based on this data, recent guidelines recommend use of anti–MRSA agents in patients with skin and soft tissue infections (SSTIs). Due to increasing MRSA clindamycin resistance, vancomycin has become the most common empiric intravenous antibiotic prescribed for skin infections on inpatient units. Yet appropriate vancomycin dosing is difficult and may rarely be achieved in patients who receive the medication for very brief periods. In this study, we look at the documented prevalence of MRSA infections and the quality of vancomycin dosing in patients admitted with SSTIs.


We reviewed 108 consecutive medical admissions with the diagnosis of cellulitis. Along with demographic, microbiologic culture data, and outcome measures, we extracted records of antimicrobial use, dosing, and therapeutic range monitoring.


Records of 108 patients with purulent and non–purulent cellulitis were reviewed. 77/108 (71%) patients had blood cultures sent at any time during the admission. Of those, only 4/77 (5%) had a positive result. No MRSA was identified in the positive blood cultures. 45/108 (42%) patients had an abscess. Of those, 31/45 (69%) had cultures of the purulent material. Of the 31, 30 (97%) yielded an identifiable organism, 21 (68%) grew Staph aureus, and 16 (52%) were MRSA. In all, the documented rate of MRSA infection in patients with cellulitis was 16/108 (15%). Eighty–six of the 108 patients (80%) received vancomycin alone or along with other antibiotics. Of the 86, 43 patients (50%) were correctly dosed on admission and only 47 patients (55%) receiving vancomycin had drug levels checked at any time during the hospitalization. Of the 47 with levels checked, only 20 patients (45%) reached target trough levels prior to discharge. Overall, of the 86 patients who received vancomycin for SSTIs, only 20 (23%) had a documented therapeutic serum drug level.


Blood culture yield in patients with SSTIs is low, yet blood cultures are collected on the majority of admitted patients. In patients with skin abscesses, the culture yield of purulent material is very high, and MRSA is the predominant organism. Vancomycin is used in the majority of patients admitted with skin infections, although it is commonly under–dosed and its serum levels often go unmonitored. A therapeutic serum level is reached only in a minority of patients who receive vancomycin.