Background:Non-steroidal anti-inflammatory drugs(NSA-Ds), including low dose aspirin, are widely used. However, they can cause gastrointestinal injury, in particular gastric orduodenal ulcers.In Japan, there have been few studies of NSAID-induced bleeding ulcers. Aim:We conducted a prospective study to determine the characteristics of NSAID-induced bleeding ulcers. Materials and Methods:Between January 2000 and December 2005, 0ne hundred ninty two consecutive patients with bleeding peptic ulcers were admitted to our hospital.On history alIpatients were asked about their use of NSAIDs, including concomitant anti-ulcer agents, andanticoagulant therapy, as well as H. pylori infection. Results:Of the 192 bleeding ulcers, 89 were NSAID-induced ulcers(46.4%). Patients with NSAID-induced bleeding ulcers, developed bleeding unrelated to the duration of tinge that they took NSAIDs. Compared to patients with non-NSAID-induced ulcers among NSAID-inducedb/eeding ulcer patients there were statistically significantly more females and elderly(p<0.01, p<0.001, respectively), smokers and drinkers(p<0.01, p%lt;0.001, respectively). Bleeding ulcers were most frequently located in the middle of the stomach. In addition, multiple plural lesions, including non-active bleeding ulcers were statistically significantly observed more often in NSAID-induced ulcer patients than in non-NSAID-induced ulcer patients(P<0.001). Therewere no significant differences in:H. pylori infection, history of peptic ulcer disease, hemoglo-bin level on admission, presence of exposed vessels, and success of endoscopic hemostasisbetween the two groups. Most of the patients were successfully treated with endoscopic hemostasis:88/89 or 98.9% of the NSAIDs-induced ulcer group 102/103 or 99.0% of the and non-NSAIDs-induced ulcer group. However, the NSAID-induced ulcer group had a significantly higher mortality(11/89 or12.4%)compared to the non-NSAID-induced ulcer group(3/103 or 2.9%)(p<0.05). The 11 patients with NSAID-induced bleeding ulcers died due to the aggravation of anunderlying disease or serious complications. Conclusions:The occurence of an NSAID-induced bleeing ulcer is unrelated to the dura-tion of time that the drug was taken. Compared to the non-NSAID-induced bleeding ulcergroup, the NSAID-induced bleeding ulcer group had a higher proportion of:elderly, females and multipled plural ulcer. Endoscopic hemostasis was very successful in both groups. The NSAID-induced bleeding ulcer group had a 12.4% mortality rate due to the aggravation of anunderlying disease or serious complications.
View full abstract