Background: It is currently fashionable to consider atopic dermatitis (AD), like other inflammatory dermatoses, as immunologic in pathogenesis (“inside-outside” hypothesis). Accordingly, topical glucocorticoids and other immunosuppressive agents are mainstays of therapy, but the risk of toxicity from these agents is not insignificant, particularly in children. Alternatively, because stratum corneum (SC) permeability barrier function is also abnormal in AD, it has been hypothesized that the barrier abnormality could drive disease activity. Yet commonly used emollients and moisturizers do not correct the SC ceramide deficiency, the putative cause of the barrier abnormality. Objectives: We assessed the efficacy of a newly developed, ceramide-dominant, physiologic lipid-based emollient, when substituted for currently used moisturizers, in 24 children who were also receiving standard therapy for stubborn-to-recalcitrant AD. Methods: All subjects continued prior therapy (eg, topical tacrolimus or corticosteroids), only substituting the barrier repair emollient for their prior moisturizer. Follow-up evaluations, which included severity scoring of atopic dermatitis (SCORAD) values and several biophysical measures of SC function, were performed every 3 weeks for 20 to 21 weeks. Results: SCORAD values improved significantly in 22 of 24 patients by 3 weeks, with further progressive improvement in all patients between 6 and 20 or 21 weeks. Transepidermal water loss levels (TEWL), which were elevated over involved and uninvolved areas at entry, decreased in parallel with SCORAD scores and continued to decline even after SCORAD scores plateaued. Both SC integrity (cohesion) and hydration also improved slowly but significantly during therapy. Finally, the ultrastructure of the SC, treated with ceramide-dominant emollient, revealed extracellular lamellar membranes, which were largely absent in baseline SC samples. Conclusion: These studies suggest that (1) a ceramide-dominant, barrier repair emollient represents a safe, useful adjunct to the treatment of childhood AD and (2) TEWL is at least as sensitive an indicator of fluctuations in AD disease activity as are SCORAD values. These studies support the outside-inside hypothesis as a component of pathogenesis in AD and other inflammatory dermatoses that are accompanied by a barrier abnormality. (J Am Acad Dermatol 2002;47:198-208.)