Staphylococcus aureus has a peculiar ability to colonize the skin of patients with eczema and atopic dermatitis (AD), and is consistently found in eczematous skin lesions in these patients. A correlation between the severity of the eczema and colonization with S. aureus has been demonstrated, and it has been determined that bacterial colonization is an important factor aggravating skin lesions. Patients colonized with S. aureus have been treated with antibiotics in several open and double-blind placebo-controlled studies, with conflicting results. To investigate the colonizing features of S. aureus in the lesional and nonlesional skin of patients with eczema and AD in China and to compare the therapeutic effect of mupirocin plus hydrocortisone butyrate with vehicle ointment plus hydrocortisone butyrate. A multicentre, double-blind randomized trial was conducted. Eczema Area and Severity Index (EASI) scores were evaluated before the start of the trial and on the 7th, 14th and 28th day of treatment. Swabs for bacterial isolation were taken from lesional skin before the start of the trial and on the 7th, 14th and 28th day of treatment, and from nonlesional skin only before the start of the trial. A combination topical therapy with mupirocin plus hydrocortisone butyrate ointment was used in the experimental group, with vehicle ointment plus hydrocortisone butyrate ointment as a control. Of 327 patients enrolled in the study, 208 had eczema and 119 had AD. Bacteria were isolated from 70.2% of lesional and 32.7% of nonlesional skin samples from patients with eczema, of which S. aureus accounted for 47.3% and 27.9%, respectively. Bacteria were isolated from 74.8% of lesional and 34.5% of nonlesional skin samples from patients with AD, of which S. aureus accounted for 79.8% and 80.5%, respectively. The colonization density of S. aureus was markedly higher in lesional than in nonlesional skin, both in patients with eczema and with AD (P < 0.01, P < 0.05), and was positively correlated with lesion severity. Considering the EASI scores before and after treatment and the final effective rate, good therapeutic effects were obtained in both the combination experimental groups and the control groups (P < 0.01), and there were no differences in the global therapeutic effect between the two groups in patients with eczema and with AD (P > 0.05). However, in patients with eczema with a clinical score of > 8 or in patients with AD with a clinical score of > 7, the therapeutic effect in the experimental groups was superior to that in the control groups (P < 0.05) on the 7th day of treatment. There were no differences between the two groups on the 14th and 28th days of treatment (P > 0.05). Following the improvement of symptoms and signs of eczema and AD, the positive rates of bacteria and S. aureus were reduced on the 7th day of treatment. This study confirmed that lesional skin of patients with eczema and AD was more frequently colonized with S. aureus than was nonlesional skin. The more severe the eczema, the higher the colonization rate of S. aureus, and S. aureus was also more often present in lesional and nonlesional skin in patients with AD than in those with eczema. Staphylococcus aureus infection is related to the pathogenesis of eczema and AD. An antibiotic-corticosteroid combination and corticosteroid alone both gave good therapeutic effect in eczema and in AD, and both reduced colonization by S. aureus. Early combined topical therapy is beneficial to patients with moderate to severe eczema and AD, and it is unnecessary to use antibiotics at later stages of disease or in mild eczema or AD.