Network analysis finds oral isotretinoin most effective for acne vulgaris

Référence

Huang CY, Chang IJ, Bolick N, et al. Comparative efficacy of pharmacological treatments for acne vulgaris: A network meta-analysis of 221 randomized controlled trials. Ann Fam Med 2023;21(4):358-369. 

Sommaire

This is an ambitious data-laden network meta-analysis to identify the most effective therapies for a condition many think as trivial or a rite of passage for adolescents. However, the psychological impact, as well as the cosmetic impact, makes acne vulgaris an important condition for primary care clinicians to properly manage. These authors searched PubMed and Embase to identify randomized trials that assessed the effectiveness of therapies for acne. Ultimately, they included 210 papers with 65,601 patients. The median duration of the studies was 12 weeks (range: 2 - 48 weeks). The studies evaluated 6 oral antibiotics, 5 topical antibiotics, oral isotretinoin, 5 topical retinoids, 6 combined oral contraceptives, topical clascoterone, 10 combination therapies, benzoyl peroxide, azelaic acid, and placebo. The most commonly used outcomes were the disease-oriented inflammatory and noninflammatory lesion counts, but 69 studies reported the investigators’ global assessment of success. Sadly, none reported the patients’ assessments. Overall, the studies were of decent quality. After all the statistical gymnastics, oral isotretinoin was the most effective agent at decreasing the number of lesions (48.4% relative reduction), followed by combination therapy (oral antibiotic, topical retinoid, and benzoyl peroxide; 38.2% relative reduction). For patients with inflammatory lesions, oral isotretinoin again performed best (54.2% relative reduction in lesions) followed by a combination of topical antibiotics plus azelaic acid (43.6% relative reduction). Curiously, the combined oral contraceptives were all over the map, possibly due to studies including patients without hormonally sensitive acne. Using the investigators’ assessment of treatment success, combination therapy with topical retinoids and benzoyl peroxide plus oral or topical antibiotics were comparable (odds ratio [OR] 6.04 and 6.76, respectively). The odds of discontinuing treatment due to adverse events was greatest for topical trifarotene (OR 11.4). The authors report moderate to high degrees of heterogeneity among the data (I2 range: 61% - 79%). Finally, the authors report no strong evidence of publication bias.