Mo1273 LOCATION AND PATTERN OF RECURRENCES IN PATIENTS WITH BARRETT’S ESOPHAGUS AFTER ENDOSCOPIC THERAPY: A SYSTEMATIC REVIEW AND CRITICAL ANALYSIS OF THE PUBLISHED LITERATURE

Abstract

Recurrences of intestinal metaplasia and dysplasia/cancer post Barrett’s endoscopic therapy (BET) are well documented. Although it has been suggested that the majority of the recurrences occur in the region of the GEJ and distal esophagus, this has not been systematically evaluated. An electronic literature search was performed in Pubmed, CINAHL, Google scholar and Cochrane for eligible studies that reported recurrences post successful BET i.e. after achieving complete eradication of intestinal metaplasia (CE-IM). Recurrences were classified based on their location (distal 3-cm/proximal esophagus, GEJ/ cardia/neo-squamocolumnar junction (NSCJ) and pattern (visible vs non-visible). Visible recurrences were those that were detected on targeted biopsies of either columnar island, tongue etc., whereas non-visible recurrences were those on random biopsies. The primary and secondary outcomes were to report summary estimates of IM and HGD/cancer recurrences respectively based on the location and pattern. Of the initial 70 studies screened, 19 were included in the final analysis. Eighteen studies reported CE-IM in 2660 patients, of which there were 435 total recurrences (16.4%). Average BE length prior to BET was 4.8 ±1.9 cm. Among the 19 studies, GEJ and cardia was reported separately in 1 study, GEJ/cardia together in 3 studies, GEJ only without mention of cardia in 9 studies, NSCJ in 6 studies and only 3 studies reported actual distance in the esophagus. 13 of 19 studies reported location of the recurrences (n=311): esophagus: 105/311 (34%); GEJ/cardia: 206/311 (66%). Among the esophageal recurrences, distal esophagus 85/105 (81%) was more common than proximal esophagus 20/105 (19%). 14 of 19 studies (n=321) provided data on recurrence patterns; visible 145 (45%) and non-visible 176 (55%). However, only 10 of 19 studies (n=110) provided data on both the location and pattern of recurrence: visible 54 (49.1%) and non-visible 56 (50.9%) (Figure 1). Similarly, data on location and pattern of HGD/EAC recurrences are shown in the flowchart (Figure 2). The published literature on recurrences post successful BET have used varied definitions of the distal vs proximal esophagus and of the GEJ and cardia. Only 10 studies (110 patients) provide information on both the location and pattern of recurrence. These data make it is difficult to make recommendations for biopsy protocols and surveillance strategies after BET. Future prospective studies should utilize standardized definitions and provide data on both the location and pattern of recurrence.View Large Image Figure ViewerDownload Hi-res image Download (PPT)