Background: An association between the bacterium Staphylococcus aureus and atopic eczema has been recognized for many years. Although few would dispute the benefit of systemic antibiotics in people with overtly clinically infected eczema, the clinical role of S. aureus in causing inflammatory flares in clinically uninfected eczema is less clear.
Objectives/methods: To see if atopic eczema can be improved by antistaphylococcal agents, we performed a systematic review of randomized controlled trials (RCTs) using Cochrane Skin Group's Specialised Register, the Cochrane Central Register of Controlled Trials, MEDLINE (from 2000), EMBASE (from 1980), the metaRegister of Current Controlled Trials (to March 2009), plus manual searching of references and conference proceedings. RCTs that compared interventions to reduce S. aureus in people with atopic eczema with no treatment, vehicle, or another active compound were included. Publication status and language were not barriers to inclusion.
Results: Twenty-six studies involving 1229 participants were included. The studies were generally short term and of poor quality. There was no significant difference in global outcome for clinically infected eczema when oral antibiotics were compared with placebo [one study, relative risk (RR) 0.40, 95% confidence interval (CI) 0.13-1.24] or when topical steroid antibiotic combinations were compared with steroid alone (two studies, RR 0.52, 95% CI 0.23-1.16). One study of children with infected eczema that added bleach to bathwater showed a significant improvement in eczema severity when compared with bathwater alone, although the difference could have been explained by regression to the mean. Although antistaphylococcal interventions reduced S. aureus numbers in people with clinically uninfected eczema, none of the studies showed any clinical benefit.
Conclusions: We failed to find any evidence that commonly used antistaphylococcal interventions are clinically helpful in people with eczema that is not clinically infected. Their continued use should be questioned in such situations, until better and longer-term studies show clear evidence of clinical benefit.