The authors concluded that, compared with placebo or an alternative antifungal, terbinafine in any formulation is well tolerated and effective as a cure for athlete's foot. Despite several limitations of the review, such as no discussion of study quality and some pooling of very diverse studies, the authors' conclusions appear reasonable.

To compare the safety and efficacy of different formulations of topical terbinafine with placebo, or an alternative antifungal, in the treatment of athlete’s foot (tinea pedis).

BIOSIS Previews, the Cochrane Database of Systematic Reviews, the Cochrane CENTRAL Register, Derwent Drug File, EMBASE, EMBASE Alert, International Pharmaceutical Abstracts, MEDIKAT, MEDLINE, MEDLINE Alert, SciSearch and DARE were searched from 1990 to 2006; the search terms were not reported. In addition, the bibliographies of identified papers were handsearched. The searches were not restricted by language.

Randomised controlled trials (RCTs) comparing terbinafine (cream, emulsion gel, solution, or film-forming solution, FFS) with a placebo, or alternative antifungal, in humans with tinea pedis were eligible for inclusion. Studies involving oral terbinafine were excluded. The included studies compared terbinafine cream (1% or not specified), 1% solution, 1% emulsion gel or 1% FFS with a placebo or active control (clotrimazole, miconazole, bifonazole, butenafine or naftifine). Terbinafine and active controls were administered as a single application, once or twice daily up to 4 weeks. The included studies reported on mycologic cure rates (defined as a negative potassium hydroxide test result and negative fungal culture after 6 weeks), clinical cure rates (for erythema, pustules, incrustation, pruritus, desquamation and vesiculation) measured using symptom scores, and adverse events (including pruritus, dryness, burning, irritation and eczema).

It appears that two reviewers independently screened studies for eligibility.

The authors stated that two reviewers independently assessed study quality, but no other details were provided.

Two reviewers independently extracted the data, ultimately to calculate relative risks (RRs) with 95% confidence intervals (CIs) for dichotomous data (mycologic and clinical cure) after 6 weeks, and adverse events.

The RRs were pooled using a fixed-effect model.  If statistical heterogeneity was evident, a random-effects model was used.  Statistical heterogeneity was assessed using the Cochran Q and I² tests, and represented graphically with forest plots. Subgroup analyses were conducted to assess the differences in effect size between terbinafine formulations.

Nineteen RCTs (n=2,899; 1,266 receiving terbinafine, 402 receiving placebo and 660 receiving active control) were included in the review. The sample sizes ranged from 20 to 685 patients. There was a 19.7% drop-out rate.

Terbinafine versus placebo (9 studies): there was a significant difference between the two treatment groups, favouring terbinafine, for mycologic and clinical cure rates; the pooled RRs were 3.17 (95% CI: 2.67, 3.77, p<0.001) and 2.75 (95% CI: 2.32, 3.24, p<0.001), respectively. There was evidence of significant heterogeneity between studies reporting clinical cure rates (I²=84.7%; p<0.001).

Terbinafine versus active controls (10 studies): there was no significant difference between the two treatment groups for mycologic or clinical cure rates. Significant heterogeneity was evident for both outcomes (I²=60% for mycologic cure rate and I²=68.3% for clinical cure rate).

Terbinafine formulations (9 studies): there was no evidence of a difference between different formulations of terbinafine (subgroup test for heterogeneity, p=0.65) for mycological cure, with the different groups all showing a statistically significant increased cure rate compared with placebo.  However, for clinical cure there was evidence of a difference between the different formulations (p=0.001), where greater effectiveness was seen for FFS 1% terbinafine.

Adverse events (19 studies): the risk of adverse events was similar for terbinafine in any form compared with both placebo and active controls; the RRs were 1.34 (95% CI: 0.74, 2.41, p=0.34) and 1.08 (95% CI: 0.70, 1.68, p=0.72), respectively. There was no evidence of heterogeneity.

Terbinafine is well tolerated in any formulation, including the new single application, and is effective as a cure for tinea pedis irrespective of treatment formulation, duration and frequency of application.

The review question was clear and was supported by appropriate inclusion criteria for the patients, interventions and study design. Relevant literature searches were conducted using a wide range of electronic databases and another appropriate source. The searches were not restricted by language, thus reducing the potential for language bias. It was unclear whether the search covered unpublished material, thus it is possible that potentially relevant papers were missed. Attempts were made to minimise error and bias at each stage of the review process. A validity assessment was mentioned, but no details provided, which means that the reliability of the included studies and their subsequent synthesis is unclear. Appropriate methods were used to assess heterogeneity but, as some analyses showed strong statistical heterogeneity, the pooling of the results might not have been appropriate. However, the authors did attempt to investigate this further by looking at subgroups for the different formulations.  Despite these considerations, the authors' conclusions appear reasonable.

Practice: The authors stated that it is likely that patients will prefer a topical regimen that is easy to follow and involves a single-application treatment.

Research:  The authors did not state any implications for further research.

Novartis Consumer Health; Novartis.

This record is a structured abstract written by CRD reviewers. The original has met a set of quality criteria. Since September 1996 abstracts have been sent to authors for comment. Additional factual information is incorporated into the record. Noted as  [A:....].