Skip navigation

GI complications and NSAIDs

Bandolier 25 described a randomised controlled trial (RCT) involving nearly 9,000 patient with rheumatoid arthritis taking NSAIDs for at least a six month period [1]. The study examined GI complications associated with NSAIDs collected in a RCT. Half the patients were randomised to co-administration of misoprostol, which significantly reduced the incidence of serious and probable GI complications by 44%.


These data translated into a NNT of 83 - that is, overall 83 patients had to be treated for one year with misoprostol rather than placebo to prevent one GI complication. The NNT figure used here is annualised. Though the study was of six months duration presenting the result in this standard fashion is justified because GI events occurred consistently throughout the six months, as has been shown in long-term studies [2].

Risk factors

Certain groups of people are more likely to develop GI complications while on NSAIDs. The size of the RCT enabled some of these "at risk" groups to be defined. Analysis showed that an age of 75 years or more, history of peptic ulcer, history of GI bleeding and history of cardiovascular disease were all significant indicators of increased risk. As examples, patients with GI bleeding had an odds ratio of 2.5 (95% CI 1.5 - 4.1) and with CVD 1.8 (1.1 - 3.2).

With any one risk factor the risk of a GI event was 2% in one year. Combinations of these risk factors were additive: with combinations of three of the factors the one-year risk increased to about 9% and with all four it was 18%.

Risk factors and NNT

Misoprostol reduced the risk significantly within each of these groups, but, as Bandolier 25 pointed out, the information in the paper did not allow calculations of NNTs for high risk groups. The NNTs are now available for individual risk factors, and are shown in the Table. NNTs for the prophylactic action of misoprostol against GI complications (all events 1-11 in the original paper) with NSAIDs can be as low as 7 for patients over 75 years with prior GI bleed. NNT values below 20 were obtained for younger patients who had prior history of peptic ulcer disease or GI bleed. NNTs for other classifications (1-8) are to be found in a BMJ letter [3].

Number of patients who would need to be treated for one year for one serious GI complication to be prevented
Patients All ages Age 65+ Age 75+
All 83 76 105
Prior cardiovascular event 68 53 58
Prior peptic ulcer disease 23 16 11
Prior GI bleed 17 14 7

Clinical implications

As the earlier Bandolier article suggested, and as this now shows, patients taking NSAIDs at high risk of GI complications can be identified, and low NNTs ascribed to misoprostol treatment. This is useful as one step in determining whether there is a bias towards prophylaxis. Groups where prophylactic misoprostol should be considered for preventing NSAID-associated GI complications are:-

  • Increasing age (especially over 75 years)
  • Cardiovascular disease
  • Peptic ulcer history
  • GI bleed history


  1. FE Silverstein, DY Graham, JR Senior et al. Misoprostol reduces serious gastrointestinal complications in patients with rheumatoid arthritis receiving nonsteroidal anti-inflammatory drugs. Annals of Internal Medicine 1995 123:241-9.
  2. JH Kurata, DE Abbey. The effect of chronic aspirin use of duodenal and gastric ulcer hospitalisations. Journal of Clinical Gastroenterology 1990 12:260-6.
  3. MJ Shield, SV Morant. Misoprostol in patients taking NSAIDs. British Medical Journal 1996 312: 846.

previous or next story in this issue