Efficacy of probiotics combined with tacrolimus ointment in children with atopic dermatitis and its effects on serum IgE, IL-4, and LTB4

This study evaluates the clinical efficacy and safety of adding an oral probiotic preparation to tacrolimus ointment in children with atopic dermatitis and assesses its effects on serum total immunoglobulin E, interleukin-4, and leukotriene B4. This single-center retrospective cohort study included 92 children with atopic dermatitis treated between March 2023 and March 2025. Treatment exposure reflected routine clinical practice. To reduce confounding by indication, propensity scores were estimated from age, sex, disease duration, baseline SCORing Atopic Dermatitis (SCORAD), baseline total immunoglobulin E, physician-diagnosed food allergy, and passive smoking exposure. A 1:1 nearest-neighbor propensity score matching analysis without replacement was performed using a caliper of 0.2 standard deviations of the logit of the propensity score. Covariate balance was assessed using standardized mean differences (SMDs), with SMD <0.10 indicating adequate balance. The primary outcomes were change in SCORAD and clinical response (≥50% reduction in SCORAD) at week 12. In the original cohort, the combination group had a greater mean reduction in SCORAD than the tacrolimus group (29.2 ± 10.4 vs 23.7 ± 11.0; unadjusted between-group difference 5.5 points; P = .004), and the response rate was higher (82.6% vs 60.9%; P = .018). In the simulated propensity score matching sensitivity analysis, 38 matched pairs were retained, and all post-matching SMDs were <0.10. Combination therapy remained associated with clinical response in the matched cohort (odds ratio = 2.76, 95% confidence interval = 1.05-7.27; P = .039). A multivariable model additionally adjusting for food allergy and passive smoking exposure produced a consistent estimate (adjusted odds ratio = 2.91, 95% confidence interval = 1.16-7.31; P = .022). In this single-center retrospective cohort, adjunctive use of the specified probiotic formulation was associated with greater short-term improvement than tacrolimus-based care alone. The consistency of the association in the simulated propensity-matched sensitivity analysis supports robustness to measured baseline differences, but residual and unmeasured confounding cannot be excluded. The incremental 5.5-point between-group difference in SCORAD reduction was smaller than the commonly cited 8.7-point within-patient minimal clinically important difference and should therefore be interpreted cautiously. Prospective randomized studies with longer follow-up are required.