Non-steroidal anti-inflammatory drug gastropathy: causes and treatment

Aspirin and non-aspirin non-steroidal anti-inflammatory drugs (NSAIDs) almost invariably cause acute gastroduodenal injury and probably account for approximately 12,000 ulcer bleeding episodes and 1200 deaths per annum in the United Kingdom. Clinically significant intestinal toxicity is also recognized but less clearly defined. The main risk factors for NSAID-related peptic ulcer complications are age, past history, use of higher risk individual NSAIDs, drug dose, concurrent use of warfarin or corticosteroids. The underlying reason for NSAID use and Helicobacter pylori status is not clearly associated with increased risk. Whether NSAIDs cause drug-induced non-ulcer dyspepsia is also controversial. Acute injury occurs more readily with aspirin than with non-aspirin NSAIDs, and the spectrum of acute injury is of little value in predicting clinically significant end points in comparison with different NSAIDs. However, acute studies of co-prescribed protective agents are highly predictive of performance in clinical practice. Gastric mucosal integrity is maintained by the interplay of three protective networks: prostaglandin synthesis, nitric oxide synthesis and the activity of the enteric nervous system. Aspirin and NSAIDs act by inhibiting prostaglandin synthesis catalysed by two cyclooxygenase enzymes. Most existing NSAIDs are unselective and inhibit the activity of the constitutive cyclooxygenase (COX) 1 enzyme in the stomach as much as the cyclooxygenase (COX) 2 enzyme which is induced at sites of inflammation such as joint disease. There are, however, prospects for selective cyclooxygenase 2 inhibitors. Some NSAIDs, particularly aspirin, have additional topical toxicity, which may in part reflect mucosal trapping. Some data favor an effect of NSAIDs in inhibiting mitochondrial oxidative phosphorylation. The principal physiological mechanisms which are compromised by NSAID use are mucosal blood flow, secretion of mucus and bicarbonate and maintenance of a hydrophobic mucosal surface. Animal data suggest that polymorphonuclear leukocytes (PMNs) are important for acute damage though the relevance to humans has yet to be established. As well as damaging the mucosa, NSAIDs impair ulcer healing and are associated with reduced epithelial proliferation and evidence of diminished angiogenesis. An interaction between prostaglandins and growth factors, as well as the synthesis of antiproliferative products of arachidonic acid metabolism may be involved. Healing of NSAID ulcers by conventional doses of H2 antagonists is slow and H2 antagonists are poor at preventing gastric ulcer development or recurrence. These problems can be overcome by the use of more potent acid suppression. Misoprostol heals NSAID-associated ulcers, though whether healing is as fast as with non-NSAID ulcers has not been formally established. Misoprostol is effective prophylactic treatment but has a significant incidence of adverse events, particularly diarrhoea. Misoprostol has been reported to prevent ulcer complications, and in endoscopic studies has been superior as prophylaxis to standard dose ranitidine. Results of its efficacy compared to proton-pump inhibitors are awaited. For many patients appropriate management is avoidance of NSAIDs or use of less potent/toxic alternatives such as ibuprofen. Reductions in prescribing arising from a prior authorization scheme show that this can be achieved. For high-risk patients requiring continuing NSAID use co-prescription of omeprazole or misoprostol should be considered.