Skip navigation
Link to Back issues listing | Back Issue Listing with content Index | Subject Index

NSAIDs and bowel injury

Prevalence study with NSAID [2]
NSAID plus omeprazole [3]
Randomised trial [4]

The trouble with NSAIDs is that, despite their effectiveness for many types of acute and chronic pain, we continue to discover that they are injurious, with many different adverse effects. As well as being involved in renal failure, congestive heart failure, and bleeding in stomach and duodenum, there is now an unfolding story about additional bowel injury.

In the 1990s the use of radioactive indium-labelled white cells showed that faecal excretion of white cells, a marker of intestinal inflammation, was elevated with oral NSAID. Calprotectin, a calcium binding protein found in neutrophilic granulocytes, monocytes, and macrophages, which resists faecal degradation can also be used as a marker. Use of the test in 312 patients taking NSAIDs showed that 44% had raised faecal calprotectin concentrations, much the same as estimates with indium studies [1].

Capsule endoscopy is a technique that allows direct visualisation of the bowel, with the pictures captured on video. Injury to the bowel can be seen directly. Three new studies have used the technique to study bowel injury with NSAIDs.

Prevalence study with NSAID [2]

Healthy men and women aged 18-70 years in general good health who were taking NSAIDs for at least three months because of arthritis were compared with non-NSAID using controls. Regular use of histamine antagonists or proton pump inhibitors was permitted, as was low dose aspirin.

After an eight-hour fast, the capsule was swallowed. Two independent blinded observers separately reviewed captured data for injury, from the point at which the capsule had passed the pylorus. Categories of injury were normal, red spots, small erosions, large erosions, or ulcers, using prior definitions.


There were 21 NSAID users and 20 controls, predominantly men, and with an average age of 50 years. Two control subjects had signs of mild injury (Table 1). Fifteen NSAID users had signs of injury (71%) and in five (24%) the injury was severe.

Table 1: Small bowel lesions seen by video capsule endoscopy in 21 NSAID users and 20 controls not using NSAIDs. Use of H2A and PPI permitted

NSAID users
Red spots
Large erosion/ulcer
Total (%)
15 (71%)
2 (10%)

NSAID plus omeprazole [3]

Here 40 healthy volunteers not taking aspirin or NSAIDs initially underwent a faecal calprotectin test and video endoscopy. They then took 150 mg slow-release diclofenac and 40 mg omeprazole daily for two weeks, when calprotectin test and video endoscopy were repeated. Images were independently reviewed by three gastroenterologists blinded to identity and treatment. In this case the damage was scored as reddened folds with erythema, denuded area with loss of villous architecture, red spots, mucosal breaks, blood without visualised lesion, other lesions.


Volunteers were 23 men and 17 women aged 21-60 years. They had no pathology initially, but 27 (68%) had some pathology after taking diclofenac 150 mg and omeprazole, with 15 (38%) having more than one category of lesion. Mucosal breaks were seen in 16, and blood in three. Thirty (75%) had faecal calprotectin levels above the upper limit of normal when taking diclofenac plus omeprazole compared with five (13%) beforehand; 36 showed an increase in faecal calprotectin with diclofenac.

Randomised trial [4]

A properly randomised and blinded (triple dummy) trial examined the effects of placebo, celecoxib 400 mg daily, and naproxen 1000 mg daily plus omeprazole 20 mg daily in healthy volunteers. Volunteers aged 18-70 years and not taking NSAIDs or aspirin were given an initial capsule endoscopy, and those without mucosal breaks randomised. After two weeks of treatment the endoscopy was repeated. Review of the images locally and centrally was blinded, with images judged using preset categories, eight of which were relevant.


The average age of volunteers was about 33 years, and about 40% were men. The numbers randomised with usable images were 113 on placebo, 115 on celecoxib, and 111 on naproxen/omeprazole.

Significantly more volunteers had a mucosal break with naproxen (55%) than celecoxib (16%), and both were significantly greater than placebo (7%; Figure 1). Not only were there more patients with small bowel breaks, but there were more breaks or lesions of any sort with naproxen/omeprazole than celecoxib, and again both were significantly greater than with placebo (Table 2). The number of occasions when blood was found without a lesion was the same for celecoxib and naproxen/omeprazole.

Figure 1: Average number of mucosal breaks seen per patient by video-capsule endoscopy

Table 2: Analysis by element and overall, shaded areas showing statistical improvement

Small bowel mucosal preaks per patient
Any small bowel lesion per patient
Blood in small bowel, without visualised lesion (%)


Capsule endoscopy studies are confirming what earlier studies have suggested, that NSAIDs are associated with a high prevalence of injury to the bowel. The clinical implication is that patients may bleed from their small intestine, and lose protein, and become anaemic. Anaemia is also associated with NSAID use, though studies are sparse so far.

Though the implication of each of the various lesions seen by gastroenterologists is less than clear (to Bandolier that is, not necessarily the gastroenterologists), the fact is that prevalence is high. It is also clear that proton pump inhibitors do not reduce this prevalence, and probably have little effect beyond the stomach. Celecoxib, and probably other coxibs as well, reduce the problem very substantially, perhaps by 80-90%, but do not eliminate it altogether.

Bandolier worries about anaemia as another problem with NSAIDs. A recent randomised trial comparing topical with oral diclofenac [5] showed that 10% of patients with knee arthritis on oral diclofenac developed anaemia in 12 weeks, much more than with topical diclofenac. There may be unresolved problems with anaemia in older people (Bandolier 137), who are much more likely to have arthritis and take NSAIDs. Much food for thought here.


  1. JA Tibble et al. High prevalence of NSAID enteropathy as shown by a simple faecal test. Gut 1999 45: 363-366.
  2. DY Graham et al. Visible small-intestinal mucosal injury in chronic NSAID users. Clinical Gastroenterology and Hepatology 2005 3: 55-59.
  3. L Maiden et al. A quantitative analysis of NSAID-induced small bowel pathology by capsule endoscopy. Gastroenterology 2005 128: 1172-1178.
  4. JL Goldstein et al. Video capsule endoscopy to prospectively assess small bowel injury with celecoxib, naproxen plus omeprazole, and placebo. Clinical Gastroenterology and Hepatology 2005 3: 133-141.
  5. PS Tugwell et al. Equivalence study of a topical diclofenac solution (PENNSAID) compared with oral diclofenac in the symptomatic treatment of osteoarthritis of the knee: a randomized controlled study. Journal of Rheumatology 2004 31: 2002-2012.

previous or next story