von Rahden, BH; Stein, HJ; Weber, A; Vieth, M; Stolte, M; Rösch, T; Schmid, RM; Sarbia, M; Meining, A
Critical reappraisal of current surveillance strategies for Barrett's esophagus: analysis of a large German Barrett's database.
Endoscopic surveillance is recommended for patients with Barrett's esophagus (BE). Based on a large database, gathered from predominantly community-based practices in Germany, we aimed to investigate the time-course of malignant progression and apply these findings to current clinical practice. Data of 1438 patients with BE from a large German BE database were analyzed. Patients with at least one follow-up endoscopy/biopsy were included. Detection of 'malignant Barrett' (either high-grade intra-epithelial neoplasia or invasive adenocarcinoma) was considered as study end-point. Of 1438 patients with BE, 57 patients had low-grade intra-epithelial neoplasia (LG-IN) on initial biopsy and 1381 exhibited non-neoplastic BE. 'Malignant Barrett' was detected in 28 cases (1.9%) during a median follow-up period of 24 months (1-255), accounting for an incidence of 0.95% per patient year of follow-up. The frequency of 'malignant Barrett' was significantly higher (P < 0.001, chi(2)-test) in the LG-IN group (n = 11, 19.3%) compared with the non-neoplastic BE group (n = 17, 1.2%). In the non-neoplastic BE group, 'malignant Barrett' was predominantly found during re-endoscopy within the first year of follow-up (12 of 17; 70.6%), in contrast to the LG-IN group, in which 'malignant Barrett' was observed predominantly after a time exceeding 12 months (8 of 11, 72.7%; P = 0.05, Fisher's exact test). Initial endoscopic evaluations seem to play the most crucial role in managing BE. After 1 year of follow-up, endoscopic surveillance should be focused on patients with LG-IN. In patients with repeatedly proven non-neoplastic BE, elongation of the follow-up intervals to the upper limit of current guidelines, that is, 5 years, might be justified.