Nonsteroidal anti-inflammatory drugs (NSAIDs) reduce pain and inflammation. Damage to the upper gastrointestinal (GI) tract remains the most common serious adverse event (AE) related to NSAID use; however, small bowel and colonic injury also have clinical consequences. Cardiovascular (CV)-related AEs of these drugs have further complicated an approach to optimize clinical outcomes.1 Clinicians must estimate individual risk for GI- and CV-related AEs, and then consider the impact of various risk modifiers — individual drug(s) and dose(s) — to estimate the overall risk to a patient when prescribing NSAIDs.
NSAID-induced upper GI injury develops due to a relative deficiency of mucosal prostaglandins that weakens the epithelial barrier, allowing acid-related ulceration. Mucosal repair is also prostaglandin dependent. Although the majority of prostaglandins are derived from cyclooxygenase (COX) 1, both COX-1 and COX-2 must be inhibited for ulceration to occur. Serious GI AEs of NSAID use include symptomatic gastric and duodenal ulcers; their complications — perforation and hemorrhage — may occur with acute or long-term therapy. In general, drugs with more prolonged and complete COX-1 inhibition will produce more ulcers and, due to their antiplatelet action, promote bleeding. The annual incidence of NSAID-related clinical upper GI events (complicated and symptomatic ulcers) is estimated to be between 2.5 and 4.5 percent, and the annual incidence rate of serious NSAID-related complications (perforation, hemorrhage, and obstruction) is approximately 1.0 to 1.5 percent.
Risk factors for NSAID-related AEs include the amount used, the type, dose, duration of use, and concomitant drug prescriptions such as antiplatelets (e.g., aspirin), anticoagulants, corticosteroids, or selective serotonin reuptake inhibitors. Non–drug-related risk factors include patient age, prior history of peptic ulcer, dyspepsia — particularly if it persists on antisecretory therapy — Helicobacter pylori infection status, and comorbidities.
The upper GI AEs of NSAIDs can be reduced in several ways — most effectively by discontinuing the drug, by selecting a less-toxic NSAID, or by adding a second drug. Introducing COX-2–selective NSAIDs revolutionized NSAID therapy by providing patients with effective control of inflammation accompanied by fewer ulcers and complication events. Adjunctive therapeutic options include H2-receptor antagonists (H2RAs), proton pump inhibitors (PPIs) and prostaglandin analogs — each of which possess varying effectiveness as protective agents; some cause additional challenges due to their own specific AEs. A recent Cochrane review concluded that standard doses of H2RAs should not be used for prophylaxis against NSAID-related toxicity. Double doses of H2RAs and standard doses of PPIs are effective at preventing duodenal and gastric ulcers (although PPIs are superior), thereby reducing NSAID-related dyspepsia; they are also better tolerated than misoprostol. In patients at high risk for GI events, a COX-2 inhibitor alone or a traditional NSAID in combination with a PPI offer similar but potentially insufficient protection from recurrent bleeding, and use of a COX-2 in combination with a PPI was recommended.
Use of NSAIDs can induce small intestinal and colonic injury with a wide spectrum of manifestations, from clinically silent mucosal injury and anemia to significant ulceration with overt bleeding, intestinal obstruction or perforation. Discontinuation of NSAIDs is the most effective management option. When choosing selective or nonselective NSAID formulations for the patient who requires anti-inflammatory therapy, a key practical consideration is whether NSAID-associated small intestinal injury is clinically relevant. CONDOR, a large, randomized trial, evaluated sustained-release diclofenac in combination with omeprazole (to reduce gastric and duodenal ulcers) compared with celecoxib.2 The results confirmed a low rate of symptomatic and complicated ulcers in both study groups; however, the presumed, small rate of intestinal blood loss was significantly less with study participants receiving the COX-2 inhibitor. Adjunctive misoprostol has also been shown to reduce damage in short-term studies in both the NSAID and aspirin settings.
A personalized approach to NSAID therapy is driven by assessment of an individual’s risk of CV and GI diseases.
Both nonselective and COX-2–selective NSAIDs have been associated with an increased incidence of serious CV events, including myocardial infarction and death. Although theoretical reasons exist as to why COX-2 selective agents may be prothrombotic, the mechanisms remain incompletely understood. A recent meta-analysis has confirmed that CV risk is not unique to celecoxib,3 but comparable risks were also noted with diclofenac and possibly high-dose ibuprofen. Naproxen was not associated with increased CV adverse outcomes but had similar GI effects as other nonselective NSAIDs. Recently, the results of PRECISION, a trial of 24,081 patients randomly assigned celecoxib, naproxen or ibuprofen, were reported.4 During the trial, 68.8 percent of participants stopped taking the study drug, and 27.4 percent of them withdrew from the trial.
The study authors concluded that no difference could be observed in the CV hazard ratio from the three NSAIDs; no evidence suggested that naproxen was safer than the other two drugs; and fewer serious GI events and renal-related AEs were observed in the participants assigned to celecoxib treatment. Many issues preclude these results in significantly altering our current understanding of the risk of NSAID use. The study population did not consist of individuals with a particularly high CV risk, and PPIs were given to everyone — but, similar to aspirin use, this therapy was not monitored.5 Pharmacologically equivalent doses of the NSAID were likely not used, biasing the results toward the safety outcomes observed in the celecoxib group.
A personalized approach to NSAID therapy is driven by the assessment of an individual’s CV and GI risk. CV risk can be calculated by methods such as the Framingham score. Patients at high risk of CV events are those with established CV disease or, in those without established CV disease, an estimated 10-year CV risk of greater than 20 percent. For each patient, clinicians should consider the GI risk associated with treatment as well as the CV risk of the individual. The choice between GI sparing and increased CV risk must be considered, because it will influence the type of therapy, dose, and duration of treatment. If small bowel blood loss is an additional consideration, then celecoxib, in the absence of aspirin use, may offer a specific treatment advantage. One size cannot fit all, and balancing the competing risk remains essential and challenging!
Dr. Scheiman is a consultant to Aralez.
1. Scheiman, J.M. NSAID-induced gastrointestinal injury: a focused update for clinicians. J Clin Gastroenterol. 2016;50:5-10.
2. Chan, F.K., Lanas, A., Scheiman, J., Berger, M.F., Nguyen, H., Goldstein, J.L. Celecoxib versus omeprazole and diclofenac in patients with osteoarthritis and rheumatoid arthritis (CONDOR): a randomized trial. Lancet. 2010;376:173-179.
3. MacDonald, T.M., Hawkey, C.J., Ford, I., et al, Randomized trial of switching from prescribed non-selective nonsteroidal anti-inflammatory drugs to prescribed celecoxib: the Standard care vs. Celecoxib Outcome Trial (SCOT). Eur Heart J. 2017;38:1843-1850.
4. Nissen, S.E., Yeomans, N.D., Solomon, D.H., et al, Cardiovascular safety of celecoxib, naproxen, or ibuprofen for arthritis. N Engl J Med. 2016;375:2519-2529.
5. FitzGerald, G.A. ImPRECISION: limitations to interpretation of a large randomized clinical trial. Circulation. 2017;135:113-115.