HOW RELIABLE IS THE CLINICAL HISTORY OF NONSTEROIDAL ANTI-INFLAMMATORY DRUG USE IN PATIENTS WHO PRESENT WITH UPPER GASTROINTESTINAL HEMORRHAGE?

Abstract

There is paucity of literature regarding the reliability of clinical history for assessing the use of nonsteroidal anti-inflammatory drugs (NSAIDs) in patients presenting with upper gastrointestinal hemorrhage. Platelet function testing using the PFA-100 has been used to detect the effects of NSAIDs on platelets. We hypothesized that surreptitious use of NSAIDs occurs only in a small minority of such patients. Study Design Adult hospitalized patients undergoing endoscopy for evaluation of melena, hematemesis, or coffee-ground emesis were included in the study. Exclusion criteria included hematocrit less than 25% or platelet count less than 100,000/mm3 as these factors cause invalid measurements by the platelet function analyzer. Results Sixty-three patients met inclusion criteria. Sixteen patients were excluded due to anemia and thrombocytopenia as above. The average (SD) age of patients was 53.65 (15.55) years. Caucasians formed 40% of the study population, whereas the remaining were African Americans. Patients were predominantly males (64%). Twenty-one percent of the patients tested had positive PFA-100 measurements that suggested NSAID use. However, 60% of those with positive PFA-100 test had denied a history of prior NSAID use. Conclusions Clinical history is not a reliable indicator of prior NSAID use in patients presenting with upper gastrointestinal bleeding. Alternate testing to evaluate for surreptitious use of NSAIDs should be considered in patients with bleeding due to non-HP, non-NSAID ulcers.