Significantly lower annual rates of neoplastic progression in short- compared to long-segment non-dysplastic Barrett's esophagus: a systematic review and meta-analysis

Background Although shorter lengths of Barrett's esophagus (BE) have been associated with a lower risk of neoplastic progression, precise estimates have varied, especially for non-d)splastic BE (NDBE) only. Therefore, current US guidelines do not provide specific recommendations on surveillance intervals based on BE length. We performed a systematic review and meta-analysis of the published literature to examine neoplastic progression rates of NDBE based on BE length. Methods PubMed, Cochrane, Google Scholar, and Embase were comprehensively searched. Studies reporting progression rates in patients with NDBE and >1 year of follow-up were included. The number of patients progressing to esophageal adenocarcinoma (EAC) and high grade dysplasia (HGD)/EAC in individual studies and the mean followup were recorded to derive person-years of follow-up. Pooled rates of progression to EAC and HGD/EAC based on BE length (<3cm vs. 3 cm) were calculated. Results Of the 486 initial studies identified, 10 met the inclusion/exclusion criteria. These included a total of 4097 NDBE patients; 1979 with short-segment BE (SSBE; 10773 person-years of follow-up) and 2118 with long-segment BE (LSBE; 12868 person-years). The annual rates of progression to EAC were significantly lower for SSBE compared with LSBE: 0.06% (95% confidence interval 0.01 %-0.10%) vs. 0.31 % (0.21 %-0.40%), respectively; odds ratio (OR) 0.25 (0.11-0.56); P<0.001, as were the rates for the combined endpoint (HGD/EAC): 0.24% (0.09%-0.32%) vs. 0.76 % (0.43 %-0.89%), respectively; OR 0.35 (0.21-0.58); P<0.001. There was no significant heterogeneity among studies. Conclusion The results demonstrate significantly lower rates of neoplastic progression in NDBE patients with SSBE compared with LSBE. BE length can easily be used for risk stratification purposes for NDBE patients undergoing surveillance endoscopy and consideration should be given to tailoring surveillance intervals based on BE length in future US guidelines.