PREVALENCE AND CHARACTERISTICS OF MISSED GASTRIC CANCER

Aims To analyze 1) The missed gastric cancer (MGC) rate in our area, 2) Baseline characteristics of the gastric cancer (GC) diagnosed and 3) Factors related to the MGC.

Methods Retrospective study of GC diagnosed in our area (secondary hospital with specialties) between October/2003-December/2018. Patients were identified through Pathology database. MGC was defined when a normal prior EGD was performed from 3 to 36 months before the diagnosis. Clinical and demographic characteristics of both non-MGC and MGC group were evaluated. The comparison of variables between the two groups was performed with the λ² test and multivariate analysis.

Results A total of 349 patients with GC were diagnosed being 6% (95% CI 3.97-9.01) MGC. In the MGC group, the localization was less frequent in body-antrum (52.4% vs. 75%; p = 0.04), and more frequent in the anastomosis of previous gastric surgery (14.3% vs. 0.9%; p = 0.0002). And this group also had less alarm symptoms (28.5% vs. 74%; p = 0.0001). There were no differences in the age, sex, smokers, presence of Helicobacter Pylori infection and histological type of tumor. Regarding the features of the EGD, performing the procedure without sedation was the only factor related to the presence of MGC (19% vs. 45.7%; p = 0.03). No differences were found regarding the type of endoscope (Olympus-GIF-Q145 / 165 vs. Olympus-GIF-Q190). In multivariate analysis, the only factors associated with MGC, was the sedation with OR 3.6 (95% CI 1.1-11.3) and anastomosis localization with OR 18.3 (95% CI 3.2-104.5).

Conclusions The MGC rate in our area was 6%. Performing EGD under sedation is associated with less possibility of MGC probably because it allows a more accurate visualization of the whole stomach, especially fundus, subcardial and anastomosis areas, where the GC is most often non-diagnosed