Search

Grades of dysplasia can help predict risk of EA. Pathologists cannot distinguish regenerative changes from true dysplasia in up to 20% of patients (Indeterminate for dysplasia -IND). The rate of progression and clinical significance of IND are uncertain. Endoscopic surveillance and treatment patterns are also not standardized. Objectives: To understand the practice patterns and clinical outcomes for patients with BE IND. Methods: We conducted a retrospective study of patients with BE with IND from 2000-2010 at one university referral center. Statistical analysis was performed with SPSS (17.0). Results: 322 patients with BE were identified. 48 patients with IND on index EGD were identified. 24 patients were excluded since they had other degrees of dysplasia prior to or at the time of index EGD. All patients were non-Hispanic white (75% males, median age 53). 14 patients had a documented diagnosis of GERD (58%). 17 patients were on a PPI (71%). 14 patients (58%) had a recommendation for follow up after the first EGD (median 46 weeks, 6-104 weeks). The median actual follow up EGD was 40 weeks (range 12-108). Findings on repeat EGD included IND (23%), no dysplasia (46%), and low grade dysplasia (LGD) (31%). Of the patients who continued to have surveillance EGD, almost 50% had no dysplasia on subsequent biopsies. Endoscopic surveillance intervals ranged from every 2 months to every 16 months. Only 1 patient had high grade dysplasia (HGD) with an incidence of 14.3 patients per 1000 person-year. 11 patients had histologic evidence of esophagitis on the index EGD (46%). Patients without dysplasia on the 2nd EGD were more likely to have had histologic evidence of esophagitis on the 1st EGD. None of the patients with LGD on the 2nd EGD had histologic evidence of esophagitis on the 1st EGD. All patients with persistent IND or documented progression of dysplasia had a hiatal hernia. Patients who progressed to LGD or continued to have IND were more likely to be on a PPI. Immunohistochemical staining for biomarkers was performed on the initial biopsy of 12 out of 14 patients who had a second EGD. p53, -methylacylcoenzyme A racemase (AMACR), and beta-catenin were positive in 9 (75%), 7 (58%), and 6 (50%) patients. All 3 markers were positive in the patient who progressed to HGD. Re-examination of H&E slides by 2 pathologists confirmed indefinite of dysplasia in 12 out of 13 patients. Conclusions: Patients with IND may be at low risk of progression to HGD. Significant variation in endoscopic surveillance patterns for IND was found. Biomarkers may help predict progression of dysplasia. More research is needed to better understand this population and guide resource utilization in patients with BE.