Endoscopy 2020; 52(S 01): S157
DOI: 10.1055/s-0040-1704484
ESGE Days 2020 ePoster Podium presentations
Friday, April 24, 2020 09:00 – 09:30 Upper GI: Endoscopic diagnosis 1 ePoster Podium 6
© Georg Thieme Verlag KG Stuttgart · New York

IMAGE ENHANCED ENDOSCOPY FOR PRENEOPLASTIC CONDITIONS AND NEOPLASTIC GASTRIC LESIONS

MR Carrasco
1   Portuguese Oncology Institute of Porto, Gastroenterology Department, Porto, Portugal
,
G Esposito
2   Sant’Andrea Hospital, Department of Medical-Surgical Sciences and Translational Medicine, Rome, Italy
,
D Libânio
1   Portuguese Oncology Institute of Porto, Gastroenterology Department, Porto, Portugal
,
P Pimentel-Nunes
1   Portuguese Oncology Institute of Porto, Gastroenterology Department, Porto, Portugal
,
M Dinis-Ribeiro
1   Portuguese Oncology Institute of Porto, Gastroenterology Department, Porto, Portugal
› Author Affiliations
Further Information

Publication History

Publication Date:
23 April 2020 (online)

Also available at
 

Aims Image Enhanced Endoscopy (IEE) improves the accuracy of endoscopic diagnosis. The majority of studies evaluated NBI technology with favourable results. However, there is a lack of evidence concerning the diagnostic accuracy of other technologies, and standardisation of mucosal patterns for some preneoplastic conditions. We aimed to evaluate the diagnostic accuracy of available IEE technologies for gastric preneoplastic conditions and neoplastic lesions.

Methods PubMed and Embase were searched until December 2018. Studies allowing calculation of diagnostic measures were included. Meta-regression and subgroup analysis were performed to explore sources of heterogeneity. When possible, pooled measures were estimated using fixed or random-effect models. For each technology and outcome, analysis was performed based on per-patient and per-biopsy analysis.

Results Among 1338 studies, 44 met the inclusion criteria (29 studies with NBI, 8 with AFI, 1 with FICE, 3 with BLI, 2 with LCI and 1 with i-SCAN). Pooled analysis was only possible for NBI technology and for GIM and dysplasia outcomes. For GIM diagnosis, the presence of tubulo-villous mucosal pattern improved substantially pooled sensitivity and specificity, and decreased significantly heterogeneity. For this outcome, on per-patient analysis, NBI demonstrated a pooled sensitivity of 0,86 (95%CI:0,76-0,92), specificity 0,96 (95%CI:0,93-0,97) and DOR 125,7(95%CI:55,94-282,5); on per-biopsy analysis it showed a pooled sensitivity of 0,87(95%CI:0,84-0,89), specificity of 0,97 (95%CI:0,96-0,97), and DOR 198,4 (95%CI:137,6-286,1). For dysplasia, pooled sensitivity and specificity with NBI was 0,85 (95%CI:0,83-0,88) and 0,97 (95%CI:0,97-0,97) respectively. Subgroup analysis assessing the use of WLE before NBI and the use of magnification using irregular microvascular/microsurface with a demarcation line, did not significantly influence pooled measures.

Conclusions This systematic review confirms the high accuracy of NBI for GIM and dysplasia. It is necessary to standardize endoscopic criteria for some of the other technologies, namely those that emerged in recent years and suggest results similar to NBI.