ABSTRACT
AIM: To evaluate the incidence and etiology of acute non-malignant upper gastrointestinal bleeding (ANMUGIB) in northern Greece due to increased use of non-steroidal anti-inflammatory drugs (NSAIDs), including low-dose aspirin (L-A), exposure and geographical variability of Helicobacter pylori (Hp) seroprevalence. METHODS: A retrospective study of 110 patients admitted for hematemesis or melena during a 6-month period. All patients had undergone a gastrointestinal (GI) endoscopy during hospitalization. The presence of Hp was identified by biopsies and a (13)C-urea breath test in the case of Hp(-) biopsy bleeding peptic ulcer (BPU). The activity of ANMUGIB was assessed according to Forrest's classification. Statistical analysis was made by the chi(2)-test and Yates' correction. RESULTS: Most patients were in the two medium age groups with no significant difference between them (P < 0.001). NSAID or L-A (100 mg/day) use was reported in 42.73% of patients in a ratio 1:1 (P > 0.1) and Hp infection was found in 29.09% of patients. BPU, with approximately two-thirds in the bulb, erosions and varices were the most frequent sources. Hp infection was found in 60.65% of BPU, 65.57% were related to NSAIDs or L-A and 8.19% were non-Hp non-NSAID/L-A BPU. Flat spots were most commonly found with a significant difference (P < 0.001) to other stigmata of recent bleeding, except for clean base. CONCLUSIONS: In northern Greece, persons aged over 40 years are prone to ANMUGIB with a non-significant relationship to males. Hp infection and medication use, such as NSAIDS and L-A, are deeply involved in its etiology. Non-Hp non-NSAID/L-A BPU are a small proportion. ANMUGIB seems to have a generally good prognosis.
