Optimizing histopathological evaluation of endoscopic mucosal resection specimens of Barrett’s esophagus–related neoplasia: a randomized study of 3 specimen handling methods

Abstract Background And Aims Endoscopic resection is the cornerstone in the treatment of Barrett’s esophagus (BE)–related neoplasia. However, accurate histopathological evaluation of endoscopic resection specimens can be challenging and the preferred specimen handling method remains unknown. Therefore, the aim of our study was to compare 3 different specimen handling methods for assessment of all clinically relevant histopathological parameters and time required for specimen handling. Methods In this multicenter, randomized study, endoscopic mucosal resection (EMR) specimens of BE related neoplasia with no suspicion of submucosal invasion during endoscopy were randomized to 3 specimen handling methods: (1) pinning on paraffin using needles; (2) direct fixation in formalin without prior tissue handling; (3) the cassette technique (small box for enclosing specimens). The histopathological evaluation scores were assessed by 2 dedicated gastrointestinal pathologists blinded for the handling method. Results Of the 126 randomized EMR specimens, 45 were assigned to pinning on paraffin, 41 to direct fixation in formalin, and 40 to the cassette technique. The percentages of specimens with overall optimal histopathological evaluation scores were similar for the pinning method (98%; 95% confidence interval (CI), 88.0-99.9) and for no handling (90%; 95% CI, 76.9-97.3), but significantly lower (64%; 95% CI, 47.2-78.8) for the cassette technique ( P P Conclusions Both pinning on paraffin and direct fixation in formalin result in optimal histopathological evaluation scores in a high proportion of specimens, while the cassette technique performs significantly worse and its use in clinical daily practice should be discouraged. Given the significantly shorter handling time, direct fixation in formalin appears to be the preferred method over pinning on paraffin. However, the latter needs to be confirmed in larger studies with inclusion of all EMR specimens.