Skin testing-guided elimination diet has proved unsuccessful for adult eosinophilic oesophagitis (EoE), whereas empiric six-food elimination diet (SFED) achieves an efficacy of 70%. To compare the efficacy of food-specific serum IgE-targeted elimination diet (sIgE-ED) and SFED. Prospective study in adult patients with EoE. Food-specific serum IgE, skin prick test (SPT) and atopy patch test (APT) to foods included in SFED were performed. Those with ≥1 positive IgE test, defined by ≥0.1 kU/l, followed a 6-week sIgE-ED, whereas non-IgE-sensitized patients underwent a 6-week SFED. Responders to diet (<15 eos/HPF) underwent individual reintroduction of foods followed by histological assessment. Forty-three EoE patients were included (26 sIgE-ED and 17 SFED). Regarding sIgE-ED, the mean number of eliminated foods per patient was significantly lower than in SFED (3.81 vs 6; P < 0.001), being wheat (85%), nuts (73%) and cow's milk (61%) the most commonly foods withdrawn. No difference in histological response was observed between sIgE-ED and SFED (73% vs 53%, P = 0.17). Causative foods identified by food challenge were cow's milk (64%), wheat (28%), egg (21%) and legumes (7%), with a single food trigger in 71% of patients. sIgE exhibited the higher accuracy to predict offending foods in IgE-sensitized patients (sensitivity 87.5%, specificity 68% (κ = 0.43)), with k values of 1 for cow's milk. APT results were all negative. Histological remission was accomplished in 73% of patients undergoing sIgE-ED, which was nonsignificantly superior to SFED. sIgE effectively identified cow's milk as a food trigger in IgE-sensitized patients.