Background and aims: Eosinophilic esophagitis (EoE) is a mixed inflammatory and fibrostenotic disease. Unlike superficial inflammatory changes, subepithelial fibrosis is not routinely sampled in esophageal biopsies. This study aimed to evaluate the efficacy and safety of deep esophageal sampling with four different types of biopsy forceps. Patients and methods: In this cross-sectional study, esophageal biopsies were taken in 30 adult patients by one expert endoscopist. Biopsies sampled from distal esophagus using a static jaw forceps (Olympus, FB-11K-1) were compared with proximal biopsies sampled with static jaw (Olympus, FB-45Q-1), alligator jaw (Olympus, FB-210K), and large-capacity forceps (Boston Scientific, Radial Jaw 4). One pathologist calculated the surface area of epithelial and subepithelial layers in hematoxylin and eosin (H&E)-stained biopsies. Results: Subepithelial tissue was acquired in 97 % (static jaw FB-11K-1), 93 % (static jaw FB-45Q-1), 80 % (alligator jaw), and 55 % (large-capacity) of samples. Median (interquartile [IQR]) ratios of surface area of epithelial to subepithelial tissue were: static jaw FB-45Q-1, 1.07 (0.65 - 4.465); static jaw FB-11K-1, 1.184 (0.608 - 2.545); alligator jaw, 2.353 (1.312 - 4.465); and large-capacity, 2.71 (1.611 - 4.858). The static jaw models obtained a larger surface area of subepithelial tissue compared with the alligator jaw (P < 0.001 and P = 0.037, for FB-11K-1 and FB-45Q-1, respectively) and the large-capacity forceps (P < 0.001, for both static jaw models). No esophageal perforations occurred. Conclusions: The static jaw forceps models allowed sampling of subepithelial tissue in > 90 % of biopsies and appear to be superior to alligator or large-capacity forceps in sampling larger amounts of subepithelial tissue.