UNLABELLED: To establish the influence of nonsteroidal anti-inflammatory drugs (NSAIDs) on the clinical course of upper gastrointestinal bleeding (UGIB), we designed a longitudinal cohort study to assess the effect of prior use of these drugs on several outcome variables in patients with UGIB. RESULTS: 164 (46.6%) of 352 patients with UGIB consumed NSAIDs in the week previous to the onset of bleeding. This group had significantly fewer previous episodes of peptic ulcer and UGIB (p < 0.01), more associated comorbidity, and increased use of steroids (p < 0.05). On endoscopic examination, gastric ulcer was the most frequent lesion in patients with NSAIDs as compared to duodenal ulcer in patients without NSAID use (p < 0.0001). There were no differences in the multivariate analysis between groups regarding lowest hemoglobin level, need for transfusion, stability of bleeding, surgical or endoscopic management, depth of lesions and hospital stay. CONCLUSIONS: a relevant proportion of patients admitted for UGIB had used NSAIDs during the previous week. The most frequent lesion in this group was gastric ulcer. In our study the clinical course of NSAID-associated UGIB was similar to that in other nonvariceal types of bleeding.
UNLABELLED: To establish the influence of nonsteroidal anti-inflammatory drugs (NSAIDs) on the clinical course of upper gastrointestinal bleeding (UGIB), we designed a longitudinal cohort study to assess the effect of prior use of these drugs on several outcome variables in patients with UGIB. RESULTS: 164 (46.6%) of 352 patients with UGIB consumed NSAIDs in the week previous to the onset of bleeding. This group had significantly fewer previous episodes of peptic ulcer and UGIB (p < 0.01), more associated comorbidity, and increased use of steroids (p < 0.05). On endoscopic examination, gastric ulcer was the most frequent lesion in patients with NSAIDs as compared to duodenal ulcer in patients without NSAID use (p < 0.0001). There were no differences in the multivariate analysis between groups regarding lowest hemoglobin level, need for transfusion, stability of bleeding, surgical or endoscopic management, depth of lesions and hospital stay. CONCLUSIONS: a relevant proportion of patients admitted for UGIB had used NSAIDs during the previous week. The most frequent lesion in this group was gastric ulcer. In our study the clinical course of NSAID-associated UGIB was similar to that in other nonvariceal types of bleeding.
Authors: William J Blot; Thomas Fischer; Gunnar Lauge Nielsen; Søren Friis; Michael Mumma; Loren Lipworth; Raymond DuBois; Joseph K McLaughlin; Henrik T Sørensen Journal: Pharm World Sci Date: 2004-12