[Analysis of etiological and related factors responsible for acute gastrointestinal hemorrhage]

To investigate the etiology, related factors and endoscopic characteristics of acute gastrointestinal hemorrhage.

The data including age, gender, medical and medication history, and endoscopic characteristics of patients receiving emergency treatment for acute gastrointestinal hemorrhage between February 2006 and February 2010 were collected to analyze the etiological profiles of this disorder.

(1) A total of 1415 patients with a 2: 1 male-to-female ratio visited our hospital for acute gastrointestinal hemorrhage in the past 4 years. There was a higher mean age of disease onset in men than in women [(51 +/- 20) years old vs (61 +/- 17) years old, P = 0.000]. The numbers of patients were 399, 361, 242 and 413 for 4 respective quarters in order of sequence. (2) And 1030 patients received endoscopy. Among them, there were 897 (87.1%) with upper gastrointestinal hemorrhage and 133 (12.9%) with lower gastrointestinal hemorrhage. Significant differences existed in the mean age of two groups [(51 +/- 20) years old vs (57 +/- 18) years old, P = 0.000]. The male-to-female ratio was 656: 241 and 65:68 for these 2 groups respectively (P = 0.000). The percentage of patient with a history of NSAID (non-steroidal anti-inflammatory drug) treatment was 22.1% (n = 198) and 12.0% (n = 16) for these 2 groups respectively (P < 0.01). (3) The most common causative diseases of upper gastrointestinal hemorrhage were peptic ulcer (n = 546, 60.8%), esophageal & gastric varices hemorrhage (n = 130, 14.5%) and gastric cancer (n = 40, 4.6%). When the patients were divided into 5 groups of < 12 h, 12-24 h, 24-48 h, 48-72 h and > or = 72 h per time window of gastroscopy, their percentages with endoscopically active hemorrhage were 24.1% (20/83), 14.9% (24/161), 9.6% (16/166), 7.5% (8/106) and 7.6% (29/381) for these groups respectively with statistically significant differences. When peptic ulcer was examined by the Forrest classification, the ratio of grade I a- II c decreased gradually while the ratio of grade III increased gradually among 5 groups (chi2 = 80.414, P = 0.040). (4) The most common causative diseases of lower gastrointestinal hemorrhage were ischemic colitis (n = 44, 33.1%), small intestinal hemorrhage (n = 26, 19.5%) and colonic polyps (n = 18, 13.5%). (5) When the patients were divided into > 65 years old group (n = 277) and < or = 65 years old group (n = 620), the ratio of gastric ulcer and cancer in upper gastrointestinal hemorrhage was higher in the former than in the latter [23.5% (n = 65) vs 8.9% (n = 55) & 9.7% (n = 27) vs 2.1% (n = 13), P < 0.01)]. While the ratio of duodenal ulcer was lower in the former than in the latter [22.4% (n = 62) vs 49.7% (n = 308), P < 0.01]. The ratio of small intestinal hemorrhage in lower gastrointestinal hemorrhage was higher in the former than in the latter (all P < 0.01).

At the lowest in the third quarter, the incidence rate of acute gastrointestinal hemorrhage is higher in males than that in females at a lower age of onset. More common than lower gastrointestinal hemorrhage, upper gastrointestinal hemorrhage has a lower mean age of onset. Peptic ulcer is the most common disorder in upper gastrointestinal hemorrhage. Ischemic colitis is the most common disorder in lower gastrointestinal hemorrhage. The rate of gastric ulcer and gastric cancer in the old age group is higher than that in the young group. Emergency gastroscopy is recommended.