TY - JOUR T1 - A Novel Composite Endpoint to Evaluate the Gastrointestinal Effects of NSAID Through the Entire GI Tract: Introducing CSULGIE JF - The Journal of Rheumatology JO - J Rheumatol SP - 6 LP - 8 DO - 10.3899/jrheum.091207 VL - 37 IS - 1 AU - CHRIS J. HAWKEY Y1 - 2010/01/01 UR - http://www.jrheum.org/content/37/1/6.abstract N2 - The notion that nonsteroidal antiinflammatory drugs (NSAID) cause gastric and duodenal ulcers, which can bleed or perforate, has certainly had its fair share of attention over the past 40 years. Yet, in rodent models, the burden of injury is in the terminal ileum, not the stomach or duodenum1. A small but growing constituency has argued that small bowel injury may be as important in humans. This proposition, which is commercially attractive to manufacturers of selective cyclooxygenase (COX)-2 inhibitors, which may lack this property, is increasingly discussed. In this issue of The Journal Francis Chan and his colleagues introduce the notion of Clinically Significant Upper and Lower GastroIntestinal Events2 (the acronym CSULGIE proudly maintaining the dysphonia originally established with the acronym NSAID!). They suggest that upper and lower GI events should be aggregated in assessing patients with drug-induced GI damage. In assessing this proposition, it is worthwhile reviewing lessons from the upper GI (UGI) tract. The first evidence that aspirin was gastrotoxic was conveyed by watercolor paintings of acute lesions detected by rigid endoscopy3, but the first systematic investigation used microbleeding as a surrogate for injury4 before volunteer endoscopy took over. Such studies also showed that non-aspirin NSAID caused injury, and numerous epidemiological studies associated aspirin and non-aspirin NSAID with GI bleeding5,6. While endoscopy and a couple of secondary prevention studies were regarded as adequate to establish the validity of Helicobacter pylori eradication, different standards were applied to NSAID and COX-2 inhibitors with the advent of the outcomes study7. As with H. pylori, endoscopy studies are broadly predictive of outcomes. Where there are differences, this may well be because of imprecisions in outcome studies. Nevertheless, the vogue for outcomes shows no signs of abating and accidentally generates interesting … Address correspondence to Prof. Hawkey; E-mail: cj.hawkey{at}nottingham.ac.uk ER -