P624 Chromoendoscopy and narrow band imaging versus conventional white light endoscopy for detection of neoplasia in ulcerative colitis – a systematic review and meta-analysis
Har-Noy O.1, Katz L.1, Avni T.2, Bessissow T.3, Young D.4, Koulaouzidis A.4, Engel T.1, Eliakim R.1, Ben-Horin S.1, Kopylov U.*5
1Chaim Sheba Medical Center, Gastroenterology, Ramat Gan, Israel 2Rabin Medical Center, Internal Medicine, Petach Tikva, Israel 3McGill University Health Center, Gastroenterology, Montreal, Canada 4Royal Infirmary of Edinburgh, Endoscopy Unit, Centre for Liver and Digestive Disorders, Edinburgh, United Kingdom 5Sheba Medical Center, Gastroenterology, Tel Hashomer, Israel
an increased risk for development of colorectal cancer (CRC) in patients with ulcerative colitis (UC) was confirmed in multiple studies. In light of this, professional authorities recommend surveillance of CRC in patients with long-standing UC, favoring chromoendoscopy (CE) with targeted biopsies of identified lesions over white light endoscopy (WLE). Narrow-band Imaging (NBI) has been evaluated in several studies, however it is still unclear whether this modality can be considered an equally accurate alternative.
The aim of the study was to compare the diagnostic yield of WLE, CE and NBI for detection of neoplasia in UC patients.
We performed a meta-analysis of prospective trials comparing the accuracy of CE, WLE or NBI for detection of neoplasia in patients with IBD MEDLINE and EMBASE search were performed using the search terms “ulcerative colitis”, “chromoendoscopy”, “narrow band imaging”. We compared the diagnostic accuracy for detection of any neoplasia individual patient examined, as well as per number of neoplastic lesions per patient.
The search yeilded eleven studies eligible for analysis. Five studies compared CE to WLE. CE (n=361 patients) was found to be superior to WLE (n=358 patients): per-patient analysis odds ratio (OR)-2.05 (95% CI 1.26,3.35; p=0.004; I2=0%); per lesion analysis OR-2.79 (95% CI 2.08,3.73; p<0.00001; I2=22%). In four studies Comparing NBI (n=305 patients) to WLE (n=305 patients), no difference was found: Per-patient analysis OR-0.97 (95% CI 0.62,1.53; p=0.91; i2=0%); per lesion analysis OR-0.94 (95% CI 0.63,1.4; p=0.68; i2=0%). Two studies compared CE (n=104 patients) to NBI (n=104 patients) and were not statistically significant different: per-patient analysis OR-1.0 (95% CI 0.51,1.95; p=0.73; i2=0%); per-lesion analysis OR-1.29 (95% CI 0.69,2.41; p=0.93; i2=0%).
Our results suggest that CE has a superior DY for detection of neoplasia in patients with UC. NBI was not significantly different to either WLE or CE, but due to the low number of studies, further evaluation is needed.