TBA (16S107)
Incidence Rate and Predictors of Progression in Patients’ With Barrett’s Esophagus: Experience From a Large Irish Tertiary Centre
Author(s)
Grace Chan, Jun Liong Chin, Marie O’Brien, Cian Muldoon, Ravi Narayanasamy, John Reynolds, Dermot O’Toole preprocess
Department(s)/Institutions
Department of Clinical Medicine and Gastroenterology, Pathology and Surgery, St James's Hospital.
Introduction
Low- (LGD) and high-grade dysplasia (HGD) are known risk precursors for development esophageal adenocarcinoma (EAC) in Barrett’s esophagus (BE). The significance of indefinite for dysplasia (IND) on histology is poorly characterised with limited data on risk of progression.
Aims/Background
To identify the incidence and rate of progression of BE-IND to LGD, HGD and EAC, in a large homogeneous cohort of patients followed in a dedicated BE programme. We also assessed the predictors of progression for patients diagnosed with BE-IND.
Method
Data was extracted from a prospective electronic patient record cohort in our Barrett’s registry. Patients diagnosed with BE-IND at diagnosis or follow-up since January 2006 were included (patients with previous grade > IND excluded). All pathology specimens (Vienna classification) were reviewed by two expert GI pathologists. Following a diagnosis of BE-IND, twice daily proton-pump inhibitors were prescribed with surveillance endoscopy performed between 6 to 12 months by senior endoscopists (using NBI, FICE and quadrantic biopsies every 1cm).
Results
110 of 1383 patients (8%) were diagnosed with BE (49 at initial endoscopy and 61 patients developed IND during surveillance period) with a mean age of 63.1 ±11.4 years (70%, were males with a mean BMI of 28.8±5.9kg/m2). The mean follow-up duration was 70.8 ± 63.4months. The incidence rate for development of BE-IND in patients who initially had non-dysplastic BE was 9 new cases per 1000 persons/year. Over this period, 31(28.2%) patients progressed from BE-IND: 24(21.8%) to LGD; 1(0.9%) progressed to HGD and 6(5.5%) to EAC. The progression rates to LGD, HGD and EAC were 2.5, 0.1 and 0.6 per 100 person/years, respectively. Higher BMI (30.9kg/m2 for progressors vs 27.5 kg/m2 for non-progressors, p=0.022) was associated with significant risk of progression of IND. Age, gender, smoking history, alcohol consumption and length of BE did not significantly increase the risk of IND progression in this cohort.
Conclusions
In our single center homogeneous cohort of BE, the incidence of IND and rates of progression to LGD/HGD/EAC were similar to those reported elsewhere. Higher BMI was found to be a predictor of progression and should be targeted in risk stratification in surveillance and management of BE.