[A retrospective analysis of clinical characteristics and mortality of gastrointestinal bleeding in 414 elderly patients]

Zhonghua Nei Ke Za Zhi. 2011 Aug;50(8):642-5.
[Article in Chinese]

Abstract

Objective: To analyze clinical characteristics of patients with gastrointestinal bleeding (GIB) and the death-related risk factors.

Methods: A retrospective analysis was conducted in 414 patients hospitalized for GIB during a 16-year period of 1994 to 2009. Logistic regression analysis identified predictors of mortality.

Results: The mean age of the 414 patients is 83.5 years old, ranging from 65 to 96 years old. The main causes of GIB were peptic ulcer (33.1%, 137/414), gastroduodenal mucosal erosion (28.5%, 118/414) and tumor (21.0%, 87/414). The main symptom was melena (71.0%, 294/414). Drugs that induced GIB were mainly non-steroidal anti-inflammatory drugs, including aspirin (11.1%, 46/414), acetaminophen (8.9%, 37/414) and indomethacin (1.9%, 8/414). 14% of patients (58/414) died of GIB in 30 days. The proportion of drug-induced GIB and gastroduodenal mucosal erosion caused GIB had increased significantly during the period of 2004 to 2009 (P < 0.05). Analysis of 30-day mortality risk showed advanced age, low diastolic blood pressure, high heart rate, low hemoglobin levels at presentation and hemorrhage volume in dead GIB elderly patients were significantly different compared with GIB elderly patients alive. Presence of severe comorbidity (heart failure and renal failure) and caused by cirrhosis and portal hypertension in GIB elderly patients were the only independent predictors of 30-day mortality (P < 0.001).

Conclusions: Death of GIB patients occurred predominantly in elderly patients with severe comorbidities and systemic conditions at presentation.

Publication types

  • English Abstract
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Age Factors
  • Aged
  • Aged, 80 and over
  • Female
  • Gastrointestinal Hemorrhage / epidemiology*
  • Gastrointestinal Hemorrhage / etiology
  • Gastrointestinal Hemorrhage / mortality*
  • Humans
  • Logistic Models
  • Male
  • Retrospective Studies
  • Risk Factors