Summary of the evidence

This is a summary of the evidence. For full details, see the evidence review.

All evidence identified included people with secondary bacterial infection of eczema. All the evidence either was in children, or the population was not reported, so it is unclear whether any studies included an adult population. The evidence for the efficacy, safety and resistance of antimicrobials is based on 1 systematic review and meta-analysis of randomised controlled trials (RCTs; George et al. 2019) and 2 RCTs (Larsen et al. 2007 and Francis et al. 2016). The evidence for choice of antibiotics is based on 1 RCT (Pratap et al. 2013). The evidence for route of administration of antibiotics is based on 2 RCTs (Francis et al. 2016 and Rist et al. 2002).

Antimicrobials

Efficacy of oral antibiotics

Evidence was from 1 systematic review of RCTs.

There were no statistically significant differences in clinical effectiveness, quality of life or microbiological outcomes for oral flucloxacillin compared with placebo in children with infected eczema. Both groups had corticosteroids and were encouraged to use emollients.

Some differences were seen in the presence of clinically apparent infection (definition unclear) at the end of treatment for oral cefadroxil compared with placebo in children with infected eczema (it was unclear whether topical corticosteroids were used in either group). However, there were no statistically significant differences in other clinical-effectiveness outcomes.

There were no differences in adverse events or withdrawals caused by adverse events for oral antibiotics (flucloxacillin or cefadroxil) compared with placebo in children with infected eczema.

Efficacy of topical antibiotics

Evidence for efficacy of topical antibiotics was from 1 systematic review of RCTs.

Some statistically significant differences were seen for the following comparison in children with infected eczema:

  • topical fusidic acid plus a topical corticosteroid (clobetasone butyrate or hydrocortisone) reduced quality of life (using the Children's Dermatology Life Quality Index) compared with placebo plus a topical corticosteroid (clobetasone butyrate or hydrocortisone) at the end of treatment

  • topical fusidic acid plus a topical corticosteroid (clobetasone butyrate or hydrocortisone) was less effective at reducing the extent and severity of eczema (when measured with the Eczema Area and Severity Index) than placebo plus a topical corticosteroid (clobetasone butyrate or hydrocortisone) at the end of treatment.

There were no statistically significant differences in other quality of life, clinical-effectiveness or microbiological outcomes for the same comparison.

There were no statistically significant differences in clinical outcome for topical gentamicin plus a topical corticosteroid (betamethasone valerate) compared with a topical corticosteroid (betamethasone valerate) alone in children with infected eczema.

There were no statistically significant differences in microbiological outcomes for a topical antibiotic (fusidic acid or gentamicin) plus a topical corticosteroid (clobetasone butyrate, hydrocortisone or betamethasone valerate) compared with a topical corticosteroid (clobetasone butyrate, hydrocortisone or betamethasone valerate) alone in people (age not reported) with infected eczema.

There were no differences in adverse events for topical fusidic acid plus a topical corticosteroid (clobetasone butyrate or hydrocortisone) compared with a topical corticosteroid (clobetasone butyrate or hydrocortisone) alone in children with infected eczema.

Efficacy of an antibiotic and corticosteroid combination compared with placebo alone

Evidence for efficacy of an antibiotic and corticosteroid combination compared with placebo alone was from 1 RCT.

Topical fusidic acid plus a topical corticosteroid (betamethasone valerate) was statistically significantly more effective than placebo for several 'responders' (people with a marked improvement or complete clearance of their eczema) and for several people with a successful biological response (baseline pathogen eradication or no visible target lesions) in children aged over 6 years, young people and adults. It was also statistically significantly more effective in terms of total severity score at end of treatment. There were no statistically significant differences in microbiological outcomes for the same comparison.

There were no differences in the number of people reporting adverse events for topical fusidic acid plus a topical corticosteroid (betamethasone valerate) compared with placebo in children with infected eczema. However, statistically significantly fewer people reported adverse drug reactions with topical fusidic acid plus a topical corticosteroid than with placebo.

Efficacy of topical antiseptics

Evidence was from 1 systematic review of RCTs.

The study did not report any data (no event rates), so no conclusions could be made about the differences in clinical effectiveness for triclosan and benzalkonium chloride emollient in bath water compared with non-antimicrobial emollient in bath water in children with infected eczema.

Efficacy of intranasal antibiotic with a bleach bath

Evidence was from 1 systematic review of RCTs.

Intranasal mupirocin (for decolonisation) plus a bleach bath was statistically significantly more effective than placebo in children with infected eczema for:

  • reducing the extent and severity of eczema (when measured with the Eczema Area and Severity Index) at 1 and 3 months after the start of treatment

  • the number of children with a reduced Investigators Global Assessment score at 3 months after the start of treatment.

There were no statistically significant differences in microbiological outcomes, withdrawals due to adverse events or minor adverse events for the same comparison.

Antibiotic resistance

Topical antibiotics compared with placebo

In 1 systematic review, there were no statistically significant differences in antibiotic resistance outcomes for topical fusidic acid plus a topical corticosteroid (betamethasone) compared with placebo plus a topical corticosteroid (clobetasone butyrate or hydrocortisone) in children aged over 6 years, young people and adults.

One systematic review found that topical fusidic acid plus a topical corticosteroid (clobetasone butyrate or hydrocortisone) in children aged over 8 years was associated with the presence of more Staphylococcus aureus (S. aureus) skin isolates resistant to fusidic acid than placebo plus a topical corticosteroid (clobetasone butyrate or hydrocortisone) at 2‑week follow up, but not at 3‑month follow up. There was no difference for S. aureus nose or mouth skin isolates at 2‑week or 3‑month follow up.

One systematic review found that topical fusidic acid plus a topical corticosteroid (clobetasone butyrate or hydrocortisone) was not statistically significantly different to placebo plus a topical corticosteroid (clobetasone butyrate or hydrocortisone) in children aged over 8 years for the presence of S. aureus nose, mouth or skin isolates resistant to oral flucloxacillin or oral erythromycin at 2‑week or 3‑month follow up.

Oral antibiotics compared with placebo

In 1 systematic review, there were no statistically significant differences between oral flucloxacillin and placebo plus a topical corticosteroid (clobetasone butyrate or hydrocortisone) in children for the presence of S. aureus nose, mouth or skin isolates resistant to oral flucloxacillin, oral erythromycin or topical fusidic acid at 2‑week or 3‑month follow up.

Topical antibiotics compared with oral antibiotics

In 1 RCT, treatment with topical fusidic acid was associated with more resistance to fusidic acid in S. aureus skin isolates taken 2 weeks after treatment than treatment with oral flucloxacillin in children with infected eczema.

No antibiotic resistance outcomes were reported for other comparisons.

Choice of antibiotics

Oral antibiotics

No evidence was identified for choice of oral antibiotic.

Topical antibiotics

In 1 RCT, topical fusidic acid plus a topical corticosteroid (halometasone) was statistically significantly more effective than neomycin sulfate plus a topical corticosteroid (betamethasone) in reducing the number of people with a positive bacterial culture at day 10 or end of treatment (20 or 30 days) in adults with infected eczema. There were no statistically significant differences in clinical effectiveness or adverse events for the same comparison.

Course length

No evidence was identified for course length.

Route of administration

Oral antibiotic compared with topical antibiotic

In 1 RCT, there were no statistically significant differences between topical fusidic acid and oral flucloxacillin (both groups had topical corticosteroids) in clinical-effectiveness outcomes, adverse events or healthcare use in children with infected eczema.

In 1 RCT, topical mupirocin was statistically significantly more effective than oral cefalexin at eradicating or improving S. aureus isolates in children aged over 8 years, young people and adults with infected eczema. Patient preference for treatment indicated that more people preferred topical treatment. There were no statistically significant differences in other microbiological outcomes, all clinical-effectiveness outcomes and adverse events for the same comparison.

  • National Institute for Health and Care Excellence (NICE)