Adding oral antihistamines to the treatment of atopic dermatitis may not produce clinically important improvements in disease severity or itch, while first-generation agents likely increased cognitive impairment.

In a systematic review and network meta-analysis of 47 randomized trials involving 6,230 pediatric and adult patients with moderate-to-severe atopic dermatitis, investigators evaluated add-on oral H1 antihistamines, H2 blockers, mast-cell stabilizers, and combinations of these treatments using data from published and unpublished randomized controlled trials in any language in the Medline, Embase, CENTRAL, the US Food and Drug Administration, and the European Medicines Agency databases from inception through May 5, 2025. Eligible trials evaluated treatment for at least 3 days, and background therapies such as moisturizers or topical corticosteroids were permitted when administered to all participants.

Investigators conducted Bayesian random-effects network meta-analyses using direct, indirect, and network estimates. They assessed certainty using the Grading of Recommendations Assessment, Development, and Evaluation approach and interpreted continuous outcomes against established minimal important differences.

Because trials at high risk of bias showed larger treatment effects, the investigators restricted their analysis of atopic dermatitis severity to trials at low risk of bias. Second-generation H1 antihistamines reduced objective Scoring Atopic Dermatitis scores by approximately 2 points compared with placebo, substantially below the minimal important difference. First-generation H1 antihistamines did not produce a clinically important difference in disease severity.

Second-generation H1 antihistamines reduced itch by less than 1 point on a 10-point scale compared with placebo, below the minimal important difference. First-generation agents did not produce a clinically important difference. The investigators reported that increasing antihistamine doses may not improve itch, although the evidence for this comparison was very uncertain.

For sleep disturbance, the included trials had low-certainty evidence, suggesting that antihistamines may not differ significantly from placebo. No trial evaluated a first-generation antihistamine for this outcome. Data also suggested that antihistamines may have provided little to no reduction in atopic dermatitis exacerbations and levocetirizine may have little to no difference in reducing atopic dermatitis–related quality of life compared with placebo.

First-generation H1 antihistamines likely increased cognitive impairment, corresponding to an estimated 66 additional cognitive impairment events per 1,000 patients. Among second-generation agents, cetirizine and levocetirizine resulted in an estimated 17 and 16 additional events per 1,000 patients, respectively, whereas loratadine did not increase cognitive impairment.

First-generation H1 antihistamines may have increased treatment discontinuation because of adverse events, with an estimated 29 additional discontinuations per 1,000 patients based on low-certainty evidence. First- and second-generation H1 antihistamines did not increase serious adverse events during the treatment periods.

Prespecified subgroup analyses found no credible modification of treatment effects by age, trial design, treatment duration, baseline atopic dermatitis severity, or use of background treatment. Sensitivity analyses adjusting for baseline severity and applying alternative minimal important difference thresholds supported the primary findings.

The investigators noted several limitations. Trial populations differed in baseline disease severity, subgroup analyses may have been limited by small sample sizes, and some crossover trials reported outcomes across the entire trial rather than by crossover period, introducing the possibility of carryover effects. The evidence also spanned nearly 60 years, during which atopic dermatitis assessment and treatment changed, and many trials preceded the availability of contemporary biologic agents and Janus kinase inhibitors.

“Our review’s moderate and high certainty findings of harms and no clinically important benefit of oral antihistamines added to placebo with or without background topical treatments can support stronger, more consistent recommendations in updated evidence based clinical practice guidelines,” wrote lead study author Alexandro W. L. Chu, of McMaster University and the University of Toronto in Canada, and colleagues.

Julie Wang reported research support paid to her institution from the National Institute of Allergy and Infectious Diseases, DBV Technologies, and Siolta, as well as consulting fees from DBV Technologies and Novartis outside the submitted work. The study authors reported no other conflicts of interest.

Source: The BMJ