Today's therapy of bleeding peptic ulcers consists of pharmacological hemostasis, endoscopic therapy and surgery in chronological order. The aim of this study was to objectively assess the contribution of the contemporary algorithm of therapy with the use of endoscopic and surgical hemostasis techniques for the therapy of bleeding peptic ulcers. This study is a retrospective analysis and comparison of two randomized groups A/B with 427/388 patients with endoscopically verified bleeding from peptic ulcer lesions Forrest I-IIb. Patients in group A (1990-1993) were treated without endoscopic intervention, compared with group B patients (1998-2001) who were treated with endoscopic intervention. In both groups we have statistically scored and compared: quantitative operative therapy share, time interval from the beginning of therapy until surgery, APACHE II score of patients at the beginning of therapy and on the day of surgery, complications requiring re-operation and mortality. In groups A/B surgical hemostasis was required in 15.0/10.6% cases, from which 90.6/61.0% operations were resections and bionomic operations. Data evaluation of APACHE II scores from both groups at the beginning of treatment showed no significant difference, but at the time of operative therapy the APACHE II scores were significantly higher in group B (11.83 +/- 6.49/15.00 +/- 4.36). The length of unstable intervals of bleeding in group B compared to group A was quantitatively lengthened (A = 55.6 +/- 19.8/B = 68.6 +/- 37.0 h). Significant differences were also noted in the number of re-operations 7.8/9.8% and mortality 15.6/24.3% between groups A/B. The contemporary accepted sequence of hemostatic therapy is accompanied by the risks of limited selection of optimal methods of endoscopic therapy, protracting the interval of bleeding with unfavorable rise in APACHE II score, and hesitancy in indication for surgery in intractable bleeding after non-surgical therapy.