[Is it possible to rate digestive toxicities of NSAIDs?]

Most often case-control studies, based on serious GI complications reported in hospitalization statistics are available to compare the gastro-intestinal (GI) safety and tolerability of the various NSAIDs currently on the market. Based on pooled data regarding 5 medicinal products among the most widely used and best documented (ibuprofen, diclofenac, naproxen, indomethacin and piroxicam), the relative result consistency of these studies is striking, thus these 5 NSAIDs may be legitimately classified from the least gastro-toxic (ibuprofen) to the most gastrotoxic (piroxicam). MISSING DATA: But epidemiological survey results may be questioned. Indeed, the volume of prescriptions written, individual dosages and the route of administration are not well documented. Above all, the disease justifying the prescription of an NSAID is never specified. Yet, it is likely that some major anti-inflammatory agents are prescribed more often, at higher doses in patients more seriously and chronically affected, and thus more likely to develop serious gastro-intestinal adverse effects. Data for recently marketed NSAIDs are less comprehensive, with the number of subjects exposed to these new NSAIDs being low, and the validity of the results reported, limited. Regarding currently marketed specific cyclo-oxygenase 2 inhibitors, recent cohort studies demonstrating that the relative risk of ulcer is about half compared to subjects treated with traditional NSAIDs are available.

The epidemiological survey results can only be extrapolated to individual clinical practice with great caution.