Factors Associated With Recurrence of Barrett’s Esophagus After Radiofrequency Ablation

• Published in: Clinical gastroenterology and hepatology Vol. 17; no. 1; pp. 65 - 72.e5
• Main Authors: Tan, Mimi C., Kanthasamy, Kavin A., Yeh, Allison G., Kil, Daniel, Pompeii, Lisa, Yu, Xiaoying, El-Serag, Hashem B., Thrift, Aaron P.
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.01.2019
• Subjects: Aged; Aged, 80 and over; Barrett Esophagus - epidemiology; Barrett Esophagus - surgery; Dysplasia; Esophageal Cancer; Female; Gastroenterology and Hepatology; Humans; Incidence; Intestinal Metaplasia; Male; Middle Aged; Radiofrequency Ablation - methods; Recurrence; Retrospective Studies; Risk Factors; Treatment Outcome; Intestinal Metaplasia; Q1; Dysplasia; quartile 1; BE; EGD; OR; CI; CPT; EAC; PPI; HR; EMR; VA; RFA; Incidence; SD; 3; H2RA; Esophageal Cancer; CEIM; NDBE; IMC; BMI; electronic medical record; hazards ratio; esophagogastroduodenoscopy; odds ratio; proton pump inhibitor; Barrett’s esophagus; body mass index; complete eradication of intestinal metaplasia; current procedure terminology; nondysplastic BE; 2; esophageal adenocarcinoma; 4; Veterans Affairs; confidence interval; intramucosal carcinoma; radiofrequency ablation; H 2 receptor antagonist; standard deviation
• ISSN: 1542-3565; 1542-7714; 1542-7714
• DOI: 10.1016/j.cgh.2018.05.042

Radiofrequency ablation (RFA) is effective treatment for Barrett’s esophagus (BE). However, some patients have recurrence after complete eradication of intestinal metaplasia (CEIM). We investigated the incidence of and factors associated with BE recurrence, with or without neoplasia, after RFA and CEIM using data from the national Veterans Affairs (VA) healthcare system. We performed a retrospective cohort study of Veterans with BE treated by RFA from 2005 through 2016 with follow-up endoscopy. Subjects were followed until BE recurrence, neoplasia, death until October 2016. CEIM, BE recurrence, and factors associated with recurrence were identified by review of medical records. We calculated incidence rates of BE recurrence, with and without neoplasia, after CEIM and identified predictors using Cox proportional hazards models. We identified 430 Veterans with BE who were treated with RFA; of these 337 achieved CEIM (78.4%). Most were men (98.6%), White (83.7%), and 66.0% had baseline dysplasia. Of those with CEIM, 98 patients (29.1%) had recurrence of BE during a total 906.0 patient-years of follow-up (median 1.9 years) after CEIM (incidence, 10.8%/patient-year). Dysplasia developed in 20 patients (2.2%/patient-year) and cancer in 3 patients (0.3%/patient-year). Baseline dysplasia (hazard ratio [HR], 1.71; 95% CI, 1.03–2.84) and long-segment BE (HR, 1.59; 95% CI, 1.01–2.51) increased risk of BE recurrence whereas treatment at high-volume RFA facilities reduced risk of BE recurrence (for quartile 4 vs quartile 1: HR, 0.19; 95% CI, 0.05–0.68). In a nationwide VA system study of outcomes of RFA for BE, we associated baseline dysplasia, long-segment BE, and treatment at low-volume RFA centers with recurrence of BE after CEIM. The findings call for performing these procedures in high-volume centers.