• M.Michael Wolfe
    Requests for reprints should be addressed to M. Michael Wolfe, MD, Section of Gastroenterology, Boston University School of Medicine, 88 East Newton Street, Room Evans 201, Boston, Massachusetts 02118
    Department of Medicine, Boston University School of Medicine, Boston, Massachusetts, USA
    Search for articles by this author
Open AccessPublished:August 16, 2004DOI:
      As indicated by both prescription and over-the-counter sales, nonsteroidal anti-inflammatory drugs (NSAIDs) are one of the most commonly used classes of drugs.
      • Paulus H.E.
      FDA Arthritis Advisory Committee meeting postmarketing surveillance of nonsteroidal antiinflammatory drugs.
      The success of the NSAID market is probably attributable to the effectiveness of these drugs as anti-inflammatory and analgesic agents. For the most part, NSAIDs are very safe drugs; only a very small percentage of patients experience adverse gastrointestinal (GI) events that result in serious complications. However, this problem has achieved enormous proportions because of simple arithmetic: NSAIDs are consumed by many millions of individuals perhaps hundreds of times annually, and a small percentage of an extremely large number is itself a large number. These drugs are remarkably effective yet carry a significant risk potential; thus, they have sometimes been characterized as a double-edged sword.
      • Lichtenstein D.R.
      • Syngal S.
      • Wolfe M.M.
      Nonsteroidal antiinflammatory drugs and the gastrointestinal tract the double-edged sword.
      Because the potential risk is always present but is only realized in a fraction of the population, it might also be represented, in keeping with the sword analogy, as a sword of Damocles, with the thread by which the sword hangs representing the risk factors and the toxicity dependent upon which strands of the thread break.
      Despite the widespread use of these agents and the extensive database potentially available for study, the strands of the thread holding the sword have not been fully characterized; major gaps remain in our knowledge of how, when, and why these drugs effect toxicity and demonstrate variability in their adverse-events profiles. To address these issues and to attempt to distinguish fact from opinion, a roundtable of gastroenterologists active in clinical research was convened. The goals of this roundtable were to (1) critically analyze the current state of knowledge regarding NSAID toxicity; (2) generate discussion concerning unresolved issues and what may be required to resolve them; and (3) determine what the future may provide for decreasing or eliminating the problem of NSAID-associated GI toxicity. This supplement to The American Journal of Medicine constitutes a summary of that roundtable. From the results, it can be seen that these articles, representing different aspects of, and divergent opinions about, NSAID-associated GI toxicity may have raised more questions than they have provided answers. However, there are 2 important themes that recur throughout.
      One of these themes is especially important in evaluating clinical trials, both past and planned: the need for clear and consistent definitions of toxicity, injury, risk, clinically relevant events, or endpoints in clinical studies of new or approved drugs. These issues are addressed in the articles by Drs. Denis McCarthy, Andrew Soll, Jamie Barkin, and Michael Kimmey. Another theme, considered in the articles by Drs. David Bjorkman, Richard Rothstein, James Scheiman and Jon Isenberg, and Michael Wolfe, is the importance of postmarketing surveillance in determining the true safety profile of an NSAID in the general population, once it has been tested in the selected populations represented in clinical trials.
      The articles are arranged in almost chronologic order, starting with a statement of the problem, based on previous epidemiologic data, and moving toward the future consideration of new or novel strategies that are being used or tested to decrease NSAID-associated GI toxicity.
      The first article, by Dr. McCarthy, uses an epidemiologic approach to state and define the extent of NSAID-associated GI toxicity. His discussion of the risk factors provides the foundation for the articles by Drs. Soll and Bjorkman, which describe the types and manifestations of the injury. The article by Dr. Soll summarizes the pathogenesis of acute and chronic NSAID-associated toxicity of the gastric mucosa, the region most commonly associated with both symptoms and clinically significant adverse events. His major contention is that it should be possible both to identify patients at risk and to better understand the mechanisms underlying NSAID-associated pathogenesis by determining what distinguishes the population that tolerates these drugs from the population in which clinical problems are observed. The less common but still clinically important toxicity of these drugs in the bowel, esophagus, and liver is discussed by Dr. Bjorkman, with correlation to currently used NSAIDs for which data are available.
      The next article, by Dr. Barkin, focuses on one particular and contentious issue: the presence of Helicobacter pylori infection in NSAID users. This organism represents a known risk factor for the development of gastroduodenal ulcers. However, its presence as a potentiating risk factor and the usefulness of its eradication in NSAID users have been debated.
      Nowhere in this supplement is the issue of definitions as critical as it is in Dr. Kimmey’s article, which discusses the role of endoscopy in NSAID clinical trials. The implications of this discussion are integral to placing the results of past and future clinical studies in their appropriate context, and it establishes the tone for the following articles on the assessment of preventive and therapeutic strategies.
      Prophylaxis and therapeutic strategies have usually relied on the concomitant administration of one of a variety of agents that have not necessarily proved beneficial. These agents are the subject of the article by Drs. Scheiman and Isenberg. This article also defines the distinction between efficacy and effectiveness, which is essential to our understanding of clinical outcomes and their relation to the usefulness of a drug.
      The last 2 articles in this supplement focus on recent advances in our knowledge of gastric physiology and the mechanism of action of NSAIDs. The past few years have witnessed the introduction of several new NSAIDs. The pharmacologic attributes of the leading drugs on the US market and the relation of these attributes to the drugs’ observed GI safety, or lack thereof, is the subject of the article by Dr. Rothstein, who reviews current data on the safety of oxaprozin, nabumetone, and etodolac.
      Other novel approaches, such as NSAIDs coupled to nitric oxide or zwitterionic phospholipids, or administered with trefoil peptides, are discussed by Dr. Wolfe. Although some of these approaches reveal insights into mechanisms of gastric physiology and may ultimately provide viable alternatives to standard NSAIDs, their clinical relevance has yet to be established. However, a more immediate approach is represented by NSAIDs that inhibit cyclo-oxygenase-2 (COX-2). Preferential COX-2 inhibitors, such as meloxicam, have been on the market in many countries for several years. Several specific COX-2 inhibitors are currently completing clinical trials in the United States. These drugs have been purported to have an improved GI safety profile, and expectations are high. However, as discussed by Dr. Wolfe, their actions are complex and, at least for the specific COX-2 inhibitors, which have only been tested in clinical trials, several issues remain that must be resolved during postmarketing surveillance.
      It is hoped that this journal supplement will provide a comprehensive overview of the problems involved in assessing NSAID-associated toxicity and will lead to further discussions on the resolution of these important issues.


        • Paulus H.E.
        FDA Arthritis Advisory Committee meeting.
        Arthritis Rheum. 1985; 28: 1168-1169
        • Lichtenstein D.R.
        • Syngal S.
        • Wolfe M.M.
        Nonsteroidal antiinflammatory drugs and the gastrointestinal tract.
        Arthritis Rheum. 1995; 38: 5-18