
Chu A W L, Wen A, Guyatt G H et al. BMJ 2026; 394 :e100163 doi:10.1136/bmj-2026-100163Abstract
Objective To systematically synthesise the benefits and harms of add-on oral antihistamines for managing atopic dermatitis (eczema).
Design Systematic review and network meta-analysis of randomised trials.
Data sources Medline, Embase, CENTRAL, the United States Food and Drug Administration, and the European Medicines Agency databases from database inception to 5 May 2025.
Study selection Randomised trials assessing add-on oral H1 antihistamines, H2 blockers, mast cell stabilisers (nedocromil sodium and sodium cromoglycate), or their combinations. Paired reviewers independently screened records, extracted data, and assessed risk of bias using a modified Cochrane Risk of Bias tool version 2.
Methods Random effects bayesian meta-analyses synthesised atopic dermatitis severity (oSCORAD—objective scoring atopic dermatitis, 0-83; minimal important difference 8.2; lower better), itch severity (numerical rating scale, 0-10; minimal important difference 3; lower better), sleep disturbance, atopic dermatitis related quality of life, and harms. The network meta-analysis GRADE (Grading of Recommendations Assessment, Development and Evaluation) and Core GRADE approaches informed certainty of evidence ratings.
 |
Network plot of randomised trials examining effects of add-on antihistamines and mast cell stabilisers on clinician reported atopic dermatitis severity organised by treatment class and individual treatments. |
Results 47 trials enrolled 6230 children and adults with primarily moderate to severe atopic dermatitis. With moderate certainty, compared with placebo, adding a first generation or second generation H1 antihistamine likely resulted in a small reduction in atopic dermatitis severity (mean difference -1.87, 95% credible interval -3.47 to -0.27) and itch severity (-0.89, -1.41 to -0.37). Although statistically detectable, these effects are well below established minimal important differences (smallest change in an outcome that a patient would identify as important).