The authors concluded that topical azelaic acid was an effective treatment for papulopustular rosacea, and was as effective as metronidazole. This was generally a well-conducted review. However, the conclusion about the relative efficacy of azelaic acid and metronidazole was not based on studies included in this review, and its reliability is therefore unclear.

To evaluate the clinical efficacy of topical 20% azelaic acid cream and 15% azelaic acid gel, compared with their respective vehicles and metronidazole gel, in the treatment of papulopustular rosacea.

MEDLINE, EMBASE, BIOSIS Previews, SciSearch, the Cochrane CENTRAL Register (Issue 3, 2004) conference proceedings (American Academy of Dermatology and the Society for Investigative Dermatology), Current Controlled Trials and ClinicalTrials.gov were searched from inception to July or August 2004. The search terms were reported and no language restrictions were employed. In addition, the references of included studies and key review articles were checked and authors of these papers and other experts were contacted. The manufacturers of the two azelaic acid products under consideration were also contacted.

Study designs of evaluations included in the review

Randomised controlled trials (RCTs) were eligible for inclusion in the review. One included study was a within-subject design, while the remainder were between-subjects designs.

Specific interventions included in the review

There was confusion over the inclusion criteria relating to the interventions. In the text of the review, the authors stated that studies that compared topical 20% azelaic acid cream or 15% azelaic acid gel at any dose with vehicle control formulations were eligible for inclusion in the review. However, the objective and inclusion criteria in the abstract clearly stated that studies that compared azelaic acid formulations with other topical treatments and metronidazole were eligible. The authors specifically excluded two studies that compared azelaic acid formulations with 0.75% metronidazole, but these featured prominently in the review discussion. All studies included in the review compared azelaic acid cream or gel with the respective vehicles over periods of 9 weeks to 3 months. All included studies disallowed concomitant therapy with the potential to affect the course of rosacea.

Participants included in the review

Studies of patients of any age and either gender, with a diagnosis of papulopustular rosacea, were eligible for inclusion. The majority of patients in most included studies were female.

Outcomes assessed in the review

The primary outcome included in the review was physician-assessed changes in the severity of rosacea, as determined by counts of inflammatory lesions. The secondary outcomes were measures of physician or device-assessed changes in rosacea severity, including severity of erythema and telangiectasia. The included studies also reported treatment-related adverse events. Studies used different rating systems to grade erythema severity.

How were decisions on the relevance of primary studies made?

Two reviewers independently assessed the studies for inclusion in the review and resolved any disagreements through discussion.

Two reviewers assessed the studies for validity. The studies were assessed using the Jadad scale, which uses the following criteria: randomisation, double-blinding, and the treatment of withdrawals and drop-outs. The studies were also assessed for a prior power calculation, a description of the inclusion and exclusion criteria, and a description of the methods used to assess adverse events. The maximum possible validity score was 8 points.

Two reviewers independently extracted the data using a pre-designed form. For each study, the mean number of inflammatory lesions before and after treatment and the mean percentage reduction in erythema and telangiectasia scores were presented for each treatment arm; levels of statistical significance were presented for some treatment differences. The authors of the primary studies were contacted for standard deviations of the mean values.

How were the studies combined?

The studies were combined in a brief narrative. A meta-analysis was not possible since standard deviations were obtained for only one of the included studies.

How were differences between studies investigated?

Some differences between the studies were discussed in the narrative synthesis, and were further apparent from the table of included studies.

Five RCTs with a total of 873 patients were included in the review. Two further RCTs with a total of 291 patients were excluded from the review, but were included in study tables and extensively referred to in the discussion and conclusions of the review.

All five included studies were of a high quality, scoring 7 or 8 out of a possible 8 on the quality assessment scale.

Inflammatory lesions.

Four of the five studies reported that participants treated with azelaic acid experienced significant reductions in the mean inflammatory lesion count compared with the vehicle-treated groups (p<0.05). The remaining study found significant reductions from baseline in lesion counts in both the intervention and control groups, but no significant between-group difference.

Erythema.

Four of the five studies found a greater overall reduction in erythema severity in the intervention groups than in the control groups. The remaining study found significant reductions from baseline in severity in both the intervention and control groups, but no significant between-group difference.

Telangiectasia.

None of the five studies found a significant improvement in telangiectasia in either the intervention or the control groups.

Azelaic acid in 20% cream and 15% gel formulations seemed effective in the treatment of papulopustular rosacea, especially in respect of decreases in inflammatory lesion count and erythema severity. Compared with metronidazole, azelaic acid appeared to be an equally effective, if not better, treatment option.

The review question appeared to be clearly defined. However, there was inconsistency between the review question presented in the abstract of the review and the inclusion criteria for eligible comparison therapies reported under 'Study Selection' in the text. The search was extensive, used no language restrictions, and made reasonable attempts to locate unpublished studies; this reduces the possibility that relevant studies were not included in the review. The authors used appropriate methods to reduce bias and error in the study selection, validity assessment and data extraction processes of the review. Appropriate criteria were used for the validity assessment and the results were reported.

The decision to employ a narrative synthesis was appropriate in the absence of necessary statistical information in the primary studies. The authors' conclusion as to the effectiveness of azelaic acid, compared with vehicle, was based on studies included in the review and was therefore likely to be reliable. However, their conclusion with respect to the relative efficacy of azelaic acid and metronidazole appeared to be based on the findings of two excluded studies which were not assessed for validity, thus the reliability of this conclusion cannot be determined.

Practice: The authors did not state any implications for practice.

Research: The authors stated that a system similar to the Psoriasis Area and Severity Index should be developed to evaluate rosacea severity. In addition, future studies should include data on standard deviations to permit the objective comparison of the efficacy of azelaic acid and comparison treatments.

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.