Background: gastric cancer (GC) is a gastrointestinal (GI) neoplasia which often complicates with GI bleeding. It is uncertain if bleeding worsens mortality in this group of patients.
Aims: to compare 30- and 90-day mortality in patients with unresectable GC (uGC) and tumor bleeding versus patients with the same neoplasia without bleeding.
Methods: a retrospective analysis of patients with uGC, with and without tumor bleeding was performed. Survival analysis for 30- and 90-days mortality was performed using Cox regression. Logistic regression was used to identify risk factors associated with mortality and first bleeding episode.
Results: 202 patients were included in the analysis (105 cases). Mortality at 90 days was 37.14 % for cases and 20.62 % for controls (p = 0.04). There was a significant difference in hazard ratio (HR) at 90 days for cases compared to controls (HR 1.95, 95 % CI 1.14-3.34, p = 0.02). Cases without palliative chemotherapy had the highest 90-days mortality (HR 5.43, 95 % CI 2.12-13.87, p < 0.01), compared to controls treated with chemotherapy. Predictors for first tumor bleeding were clinical stage IV (OR 2.93, 95 % CI 1.04-8.26, p = 0.04), Helicobacter pylori infection (OR 2.80, 95 % CI 1.35-5.80, p < 0.01) and histologic intestinal-subtype (OR 2.14, 95 % CI 1.07-4.30, p = 0.03).
Conclusions: tumor bleeding increases 90-days mortality in patients with uGC. Prevention of the first bleeding episode might improve outcome in these patients and the recognition of high-risk patients might help decision-making.