Endoscopy 2019; 51(04): S217
DOI: 10.1055/s-0039-1681816
ESGE Days 2019 ePosters
Friday, April 5, 2019 09:00 – 17:00: Colon and rectum ePosters
Georg Thieme Verlag KG Stuttgart · New York

ENDOSCOPIC PREDICTION OF INVASION DEPTH BY CONVENTIONAL COLONOSCOPY IN EARLY COLORECTAL CANCER. A PROSPECTIVE STUDY IN PERU 2014 – 2018

L Marin Calderon
1   Hospital Nacional Edgardo Rebagliati Martins, Departamento de Aparato Digestivo, Lima, Peru
,
JB Medina Morales
1   Hospital Nacional Edgardo Rebagliati Martins, Departamento de Aparato Digestivo, Lima, Peru
,
F Palacios Salas
1   Hospital Nacional Edgardo Rebagliati Martins, Departamento de Aparato Digestivo, Lima, Peru
,
P Bardalez Cruz
1   Hospital Nacional Edgardo Rebagliati Martins, Departamento de Aparato Digestivo, Lima, Peru
,
J Vasquez Quiroga
1   Hospital Nacional Edgardo Rebagliati Martins, Departamento de Aparato Digestivo, Lima, Peru
,
E Alva Alva
1   Hospital Nacional Edgardo Rebagliati Martins, Departamento de Aparato Digestivo, Lima, Peru
,
JE Medina Morales
2   Universidad de San Martin de Porres, Lima, Peru
,
M Davalos Moscol
1   Hospital Nacional Edgardo Rebagliati Martins, Departamento de Aparato Digestivo, Lima, Peru
› Author Affiliations
Further Information

Publication History

Publication Date:
18 March 2019 (online)

Also available at
 

Aims:

Endoscopic resection has been established as the treatment of choice for adenomas and cancer colorectal early (ECRC), with minimal risk of lymph node metastasis. Hence, the tumoral invasion prediction might help to determine the best therapeutic approach. The aim of the study is to assess the prediction of tumor invasion by means of conventional colonoscopy in early colorectal neoplastic lesions.

Methods:

A prospective study for the validation of a diagnostic tool was performed. We included those patients with ECRC who were referred for endoscopic management during the period from January 2014 to July 2018. We evaluated the following endoscopic findings in each lesion: loss of lobulation, expansive appearance, depressed, eroded or ulcerated area demarcated, fold convergence, rigidity, and non-lifting sign. We define the depth of invasion in two groups: as intramucosal (M) with the presence of any or 1 endoscopic findings and massive-submucosal (SM-M) with the presence of at least 2 endoscopic findings. We correlated the prediction of depth invasion using the endoscopic findings with the pathologic stage of the lesions and determined its diagnostic performance. Subsequently, the association of each endoscopic predictor and the histological presence of SM-M invasion were determined with univariate and multivariate analysis.

Results:

Global accuracy for prediction of tumor invasion by endoscopic findings was of 93%. Sensitivity, specificity, positive predictive value and negative predictive value for the first group (M) was 91.8%, 95.5%, 97.8%, and 84%, respectively; whereas for the second group (SM-M) was 95.5%, 91.8%, 84%, and 97.8%, respectively. We find that expansive appearance and rigidity were independent risk factors, which predict significantly the submucosal invasion.

Conclusions:

The use of endoscopic predictors by conventional colonoscopy with white light is useful to determine the depth of tumor invasion in early neoplastic colorectal lesions.