BACKGROUND: We evaluated the patients who underwent surgical or nonoperative treatment for acute nonvariceal upper gastrointestinal bleeding. METHODS: The study included 31 patients (21 males, 10 females; mean age 60 years; range 19 to 82 years) with nonvariceal upper gastrointestinal bleeding. Surgical treatment was performed in 14 patients (10 males, 4 females; mean age 64 years), while 17 patients (11 males, 6 females; mean age 55.6 years) were treated conservatively. The two groups were compared with respect to age, sex, associated diseases, arterial blood pressure, pulse rate, and hematocrit values on admission, number of blood transfusions, endoscopic findings, cause of bleeding, and mortality. RESULTS: Upon admission, all the patients underwent endoscopic examination except for five surgically-treated patients. Peptic ulcer was detected in 74.1% of the bleedings. The incidences of duodenal ulcer and stomach ulcer did not differ between the two groups. No significant differences were found with respect to the causes of bleedings. The mean number of blood transfusions was 4.36 units preoperatively, and 2.29 units in those treated conservatively (p=0.013). The mean systolic (p=0.002) and diastolic pressures (p=0.029), pulse rates (p=0.003), and hematocrit values (p=0.011) obtained on admission differed significantly. Mortality occurred in only one patient (7.1%) due to cardiac failure in the postoperative period. CONCLUSION: Our study yielded elucidative data on the referral of patients to surgical treatment for nonvariceal upper gastrointestinal
bleeding.