Surveillance intervals and biopsy protocols after complete remission of intestinal metaplasia (CRIM) post radiofrequency ablation (RFA) in Barrett’s oesophagus (BO) are intensive and not based on substantial evidence. We aimed to assess the timeline, location, and histology of recurrence following CRIM with the goal of assessing the appropriateness of current recommendations.Methods
Data on patients undergoing RFA for BO-related neoplasia were obtained from prospectively maintained databases of five (3 USA and 2 UK) tertiary referral centres with expertise in management of BO-related neoplasia. Patients underwent RFA following endoscopic mucosal resection (EMR) of visible lesions. RFA was performed every three months till CRIM was confirmed endoscopically and histologically on two consecutive endoscopies. Subsequent surveillance was performed at 3, 6, 9, and 12 months thereafter. Recurrence incidence was estimated using Kaplan-Meier method and Cox Proportional Hazards models were used to assess predictors of recurrence.Results
594 patients achieved CRIM as of April 1 st 2017 and were included in the analysis. Mean (standard deviation (SD)) age was 67 (10) years and 86% were males. Median (interquartile range (IQR)) BO segment length was 4 (2–6) cm. 90% of patients were treated for dysplasia or carcinoma. 151 subjects developed recurrent BO over a median (IQR) follow up of 2.8 (1.4–4.4) years. BO recurred at the gastroesophageal junction (GOJ) in 67% of subjects and in the tubular oesophagus in 33%. 84% of BO recurrences in the tubular oesophagus occurred within 5 cm of the GOJ. Histology of recurrences included cancer (9%), high grade dysplasia (HGD) (8%), low grade dysplasia (LGD) (12%), indefinite for dysplasia (2%) and non-dysplastic BO (69%). Annual incidence of any recurrence was 9.6%, dysplastic (LGD/HGD/cancer) recurrence was 2.8% and HGD/Cancer recurrence was 1.6%. The recurrence hazard rate did not vary over the follow-up (p=0.74) with 19% risk within 2 years and an additional 49% risk over the next 8.6 years. Recurrence hazard rate of any dysplasia and HGD/Cancer while lower, also did not vary over the duration of follow up (p=0.94 and p=0.88, respectively) (Figure 1). In a multivariable model, baseline HGD/cancer predicted recurrence (hazard ratio 1.9, 95% CI 1.2–3.1, p=0.004).Conclusions
In this large multicentre and international cohort study, BO recurrence risk (at least in the first 5 years following CRIM) did not appear to vary over time suggesting that continued surveillance remains important. Most recurrences appear to occur at the GOJ or distal 5 cm of the oesophagus. Sampling the GOJ and the distal 5 cm of the oesophagus in the absence of visible lesions may be adequate for surveillance.