Abstract
Accumulating evidence indicates a correlation between gut microbiota (GM) and atopic dermatitis (AD). Nevertheless, the causal relationship between specific pathogenic bacterial taxa and AD remains uncertain. This investigation utilized a two-sample Mendelian randomization (MR) analysis to assess the causal association between GM and AD, identifying the most influential GM taxa. An MR study was conducted utilizing summary statistics derived from genome-wide association studies encompassing 207 GM taxa and their association with AD risk. The genome-wide association studies summary statistics for 207 GM taxa (from phylum to species level) were generated by the Dutch Microbiome Project. The genetic variants (P < 1 × 10-5) correlated with GM (n = 7738) were identified from this investigation. Additionally, 4 supplementary MR approaches, simple mode, MR-Egger, weighted mode, and weighted median, supported the inverse-variance weighted approach. Furthermore, sensitivity analyses were executed using leave-one-out analysis, Cochran Q test, MR-Egger intercept test, MR pleiotropy residual sum and outlier global test, and MR Steiger test. The MR analysis identified 17 distinct bacterial taxa involving 2 orders, 4 families, 5 genera, and 6 species. The inverse-variance weighted method demonstrated that 6 bacterial taxa were positively associated with AD. These taxa included the order Pasteurellales, family Burkholderiales noname, family Pasteurellaceae, genus Burkholderiales noname, species Burkholderiales bacterium_1_1_47, and species Desulfovibrio piger. Eleven bacterial taxa were negatively associated with AD, comprising the order Actinomycetales, family Micrococcaceae, family Oscillospiraceae, genus Rothia, genus Collinsella, genus Oscillibacter, genus Pseudoflavonifractor, species Oscillibacter_unclassified, species Roseburia hominis, species R mucilaginosa, and species Parabacteroides merdae. Moreover, the MR-Egger intercept test and MR pleiotropy residual sum and outlier global test validated that the MR analysis remained unaffected by horizontal pleiotropy (P > .05). Furthermore, the leave-one-out analysis contributed to validating the robustness of the outcomes. Finally, an MR Steiger directionality test confirmed the assessment of potential causal direction (P < .001). This investigation identified specific intestinal flora causally associated with AD risk, offering novel insights for future investigations and innovative avenues for AD diagnosis, therapeutic intervention, and prognostic assessment.