Little strong evidence to base treatment of symptomatic lichen planus

Objectives To assess the effectiveness and safety of any form of palliative therapy against placebo for the treatment of symptomatic oral lichen planus. Data sources Search strategy: Medline 1966–98, EMBASE 1980–98, Cochrane Library, handsearching of conference proceedings and specific journals, researchers in the field, drug manufacturers. Selection criteria were any placebo-controlled trial of palliative therapy for symptomatic oral lichen planus, using a randomised or quasi-randomised design that measured changes in symptoms and/or clinical signs. Data extraction and synthesis Change in symptoms (pain, discomfort) and clinical signs (visual impression, lesion measurements) at the end of therapy. Odds ratio of improvement versus no improvement for each trial outcome and pooling where appropriate. Results Nine therapies identified grouped into four separate classes (cyclosporines, retinoids, steroids and phototherapy) for comparison. No therapy was replicated exactly. Large odds ratios with very wide confidence intervals indicating a statistically significant treatment benefit were seen in all trials. However, this has to be tempered by considerations of the small study sizes, lack of replication, difficulty in measuring outcome changes and the very high likelihood of publication bias. Only systemic agents were associated with treatment toxicities, all other side-effects were mild and mainly limited to local mucosal reactions. Conclusions The review provides only weak evidence for the superiority of the assessed interventions over placebo for palliation of symptomatic OLP. The results highlight the need for larger placebo-controlled RCTs with more carefully selected and standardised outcome measures before between-treatment comparisons can be properly interpreted.