[Clinical value of endoscopic hemostasis in acute nonvariceal upper gastrointestinal bleeding].

ABSTRACT

OBJECTIVE: To evaluate the clinical value of endoscopic hemostasis in acute nonvariceal upper gastrointestinal bleeding. METHODS: This was a retrospective study of 223 patients with acute nonvariceal upper gastrointestinal bleeding and receiving endoscopic treatment who were admitted to Peking University Third Hospital between January 1, 2005 and December 31, 2009. Endoscopic diagnosis, lesion location, lesion size and stigmata of recent hemorrhage were recorded. Stigmata of recent hemorrhage was evaluated by Forrest classification. All the patients were scored by Rockall for rehemorrhage and death risk. Endoscopic treatment comprised medicine aspersing, injection, thermocoagulation, clips and combination therapy. RESULTS: Hemorrhagic lesions of Forrest Ia-IIb were selected for endoscopic treatment, in which 214 patients(96.0%,214/223) underwent first endoscopic hemostasis successfully, while rehemorrhage occurred in 34 patients(15.2%,34/223). The first hemostatic achievement rate was 80.7%(180/223). And 17 patients received surgery or died because of endoscopic treatment failure. Total effective rate of endoscopic treatment was 92.4%(206/223). The total effective rates of Rockall high-risk group, moderate-risk group and low-risk group were 80%(40/50),95.7%(156/163) and 100%(10/10) respectively. The effective rates of epinephrine injection and combination therapy were 92.6%(137/148) and 77.6%(38/49) respectively. The rehemorrgagic rates of epinephrine injection and combination therapy were 14.2%(21/148) and 18.4%(9/49) respectively. The proportion of combination therapy in the second attempt at endoscopic therapy was 65.0%(13/20), and the achievement rate was 61.5%(8/13). CONCLUSION: Endoscopic hemostatic therapy is the preferred emergency treatment in acute nonvariceal upper gastrointestinal bleeding. Endoscopic treatment should be used for emorrhagic lesions of Forrest Ia-IIb. Endoscopic therapy could be completely hemostatic in Rockall low-risk group. Rockall score directly influences endoscopic treatment effectiveness.