Retrospective cohort study of hospitalized adults treated with vancomycin or clindamycin for methicillin-resistant Staphylococcus aureus skin infections

Abstract

Background: Vancomycin alternatives, including clindamycin, have in vitro activity against current strains of methicillin-resistant Staphylococcus aureus (MRSA), but clinical evidence of their effectiveness is needed. Objective: The aim of this work was to compare health outcomes for hospitalized adult patients treated with vancomycin and clindamycin for skin and soft- tissue infections caused by MRSA. Methods: This was a retrospective chart review of patients admitted to University Hospital (San Antonio, Texas) with culture-proven MRSA skin or soft-tissue infections from July 1, 2006, to December 31, 2006. Patients were subdivided into groups according to antibiotics received on the first day of hospital admission. The primary outcome was composite failure, which was defined as having an additional positive MRSA culture 5 to 90 days after initial culture or requiring an additional intervention (eg, new course of antibiotics or additional incision and drainage within 90 days after initiation of therapy). Descriptive statistics were used to characterize each group; χ 2, Fisher exact, and Wilcoxon rank sum tests were used to assess differences between the vancomycin and clindamycin groups. Results: Ninety-one patients received vancomycin (n = 40) or clindamycin (n = 51) for a MRSA skin infection. Most vancomycin-treated patients received 1 g IV q12h (92.5% [37/40]), whereas most clindamycintreated patients received 600 mg IV q8h (51.0% [26/51]) or 900 mg IV q8h (27.5% [14/51]). The vancomycin and clindamycin groups had no significant differences with regard to median age (38 vs 37 years, respectively), male sex (62.5% [25/40] vs 74.5% [38/51]), or Hispanic ethnicity (77.5% [31/40] vs 78.4% [40/51]). All MRSA isolates were susceptible to vancomycin and trimethoprimsulfamethoxazole. Few patients who received clindamycin were resistant to clindamycin (3.9% [2/51]). No patients died in the hospital. There were no significant differences between the vancomycin (n = 40) and clindamycin (n = 51) groups with respect to composite failure (15.0% [6/40] vs 7.8% [4/51], respectively), microbiologic failure (2.5% [1/40] vs 3.9% [2/51]), additional inpatient interventions (5.0% [2/40] vs 3.9% [2/51]), or additional outpatient interventions (12.5% [5/40] vs 3.9% [2/51]). Most patients (93.4% [85/91]) received incision and drainage. When those who did not were excluded from the analyses, all trends remained unchanged. Conclusions: In a single institution with a low rate of clindamycin resistance, there were no significant differences between vancomycin and clindamycin for the treatment of these hospitalized patients with MRSA skin infections, on the basis of clinical outcomes data. This finding warrants further investigation in a randomized controlled trial.