P254 Clinical implications of low-grade dysplasia detected during surveillance of IBD patients: a closer look at chromoendoscopy and high-definition white light endoscopy
J. Ten Hove*1, E. Mooiweer1, A. Van der Meulen2, E. Dekker3, C. Ponsioen3, B. Oldenburg1
1University Medical Centre Utrecht, Utrecht, Netherlands, 2Leiden University Medical Centre, Leiden, Netherlands, 3Amsterdam Medical Centre, Amsterdam, Netherlands
Current guidelines advocate the use of chromoendoscopy (CE) in the surveillance of patients with IBD, because CE increases the number of neoplastic lesions as compared with other imaging modalities. Whether these lesions found using chromoendoscopy indicates a similar risk of high-grade dysplasia or cancer on follow-up compared with lesions detected using white light endoscopy is currently unknown. Aim: To compare the risk of advanced neoplasia after detection of lesions containing low-grade dysplasia identified with CE or WLE in patients with IBD.
A retrospective multicentre database was used to identify patients undergoing IBD surveillance for UC or colonic Crohn’s disease between 2000 and 2014. Patients were selected if low-grade dysplasia was detected during surveillance, which was not managed by immediate colectomy. Sub-groups were made based on the endoscopic technique used to detect the first lesion with low-grade dysplasia during the observed surveillance period. Visible lesions detected without pancolonic dye spraying were classified as SD-WLE or HD-WLE based on the colonoscope type used. LGD detected in random biopsies was classified as an invisible lesion. Incidence of advanced neoplasia was defined as histopathologically proven high-grade dysplasia (HGD) or colorectal cancer (CRC). Censoring was performed in case of colectomy, death, or the last known surveillance colonoscopy.
Of 1 065 IBD patients undergoing surveillance, 159 patients (71 ulcerative colitis patients [55%]), 51 Crohn’s disease patients (40%), and 7 IBD-unclassified patients (5%) had at least 1 episode of low-grade dysplasia, which was a visible lesion in 133 cases and invisible in 26 cases. Mean age was 56 years (SD 11), and 81 patients (60.9%) were male. On follow-up, 10 advanced neoplasias were detected (5 HGD and 5 CRC). The overall incidence rate of AN was 1.31 per 100 patient-years, with a median follow-up of 4.7 years and a median time to progression of 3.3 years. ANs occurred in the sub-groups for visible index lesions in 7/10 cases, (0 for CE, 1 for HD-WLE, 6 for SD-WLE), although median follow-up time for the CE-group was significantly lower (2.0 years, p < 0.01). Incidence rates of AN (per 100 patient-years) for the sub-groups were 2.29 for invisible lesions, 1.29 for SD-WLE, 1.24 for HD-WLE, and 0.0 for CE. There were no significant differences between incidence rates between the sub-groups (log rank, p = 0.726).
In accordance with the literature, high-grade dysplasia or cancer occurs infrequently after detection of low-grade dysplasia in patients with IBD undergoing endoscopic surveillance. Based on these results, the natural history of visible dysplastic lesions detected with CE or WLE is similar.