OBJECTIVE: To analyze the clinical characteristics, outcomes and prognostic factors in elderly patients (aged 75 years and elder) with acute nonvariceal upper gastrointestinal bleeding (UGIB). METHODS: Consecutive patients admitted with acute nonvariceal UGIB who underwent upper gastrointestinal endoscopy were prospectively recruited and subdivided into two age-based groups, elderly (aged ≥75 years) and younger patients (<75 years). The patients' characteristics and outcomes were recorded. RESULTS: Altogether 1136 patients were included in the study, 276 (24.3%) aged ≥75 years. Peptic ulcers, gastroduodenal erosions and esophagitis represented the three most common endoscopic lesions found in 87.7% of the elderly patients compared with 80.8% in younger patients ( P = 0.008). Overall, the rebleeding rate (4.0% vs 3.3%, P = 0.568), need for blood transfusion (66.3% vs 61.0%, P = 0.122), surgery rate (1.2% vs 1.4%, P = 0.947) and in-hospital mortality (13.0% vs 10.0%, P = 0.157) were not different between the two groups. In elderly patients, serum albumin was the only predictive variable independently associated with mortality in the overall analysis (OR 5.867, 95% CI 2.206-15.604, P < 0.001) and in the subgroup patients with peptic ulcers (OR 5.230, 95% CI 2.099-13.029, P = 0.001). Elderly patients with serum albumin >23.5 g/L at admission presented a low mortality (negative predictive value 97.3%). CONCLUSIONS: Clinical evolution and mortality do not differ between the elderly and younger patients with acute nonvariceal UGIB. Serum albumin level at admission is a prognostic marker for mortality in elder patients.
OBJECTIVE: To analyze the clinical characteristics, outcomes and prognostic factors in elderly patients (aged 75 years and elder) with acute nonvariceal upper gastrointestinal bleeding (UGIB). METHODS: Consecutive patients admitted with acute nonvariceal UGIB who underwent upper gastrointestinal endoscopy were prospectively recruited and subdivided into two age-based groups, elderly (aged ≥75 years) and younger patients (<75 years). The patients' characteristics and outcomes were recorded. RESULTS: Altogether 1136 patients were included in the study, 276 (24.3%) aged ≥75 years. Peptic ulcers, gastroduodenal erosions and esophagitis represented the three most common endoscopic lesions found in 87.7% of the elderly patients compared with 80.8% in younger patients ( P = 0.008). Overall, the rebleeding rate (4.0% vs 3.3%, P = 0.568), need for blood transfusion (66.3% vs 61.0%, P = 0.122), surgery rate (1.2% vs 1.4%, P = 0.947) and in-hospital mortality (13.0% vs 10.0%, P = 0.157) were not different between the two groups. In elderly patients, serum albumin was the only predictive variable independently associated with mortality in the overall analysis (OR 5.867, 95% CI 2.206-15.604, P < 0.001) and in the subgroup patients with peptic ulcers (OR 5.230, 95% CI 2.099-13.029, P = 0.001). Elderly patients with serum albumin >23.5 g/L at admission presented a low mortality (negative predictive value 97.3%). CONCLUSIONS: Clinical evolution and mortality do not differ between the elderly and younger patients with acute nonvariceal UGIB. Serum albumin level at admission is a prognostic marker for mortality in elder patients.