This article was last reviewed on
This article waslast modified on 10 July 2017.
The major diagnostic task in patients who have persistent or recurrent bouts of abdominal pain and diarrhoea is to decide whether they have one of the inflammatory bowel diseases (Crohn’s disease or ulcerative colitis) or irritable bowel syndrome (IBS). The standard test is to carry out colonoscopy. This time-consuming procedure requires several days’ preparation to ensure that the patient’s large bowel is empty. The patient is then sedated and a flexible tube is passed to allow inspection the lining of the large bowel and sometimes the last part of the small bowel. Tissue biopsy samples are taken to be examined under a microscope for features of Crohn’s disease or ulcerative colitis. Biopsies from patients with irritable bowel syndrome show no abnormalities. A laboratory test to rule out inflammatory bowel disease would reduce the need for colonoscopy.

Calprotectin is a protein found in cells involved in inflammation. It is present in abundance in white blood cells called neutrophils and is released when they die. The concentration of calprotectin in faeces is increased in inflammatory bowel disease and has been shown to be a better marker of the disease activity seen on tissue biopsy than blood test markers of inflammation (white blood cell count, ESR and CRP). Its measurement is relatively cheap and it is stable in faeces at room temperature for several days.

Dr Patrick van Rheenen and colleagues from Groningen University in the Netherlands reviewed published studies in which patients with persistent or recurrent abdominal pain and diarrhoea had been referred to hospital for investigation for possible inflammatory bowel disease, and who had had a measurement of faecal calprotectin. They included only studies in which biopsies were taken from the right side of the large bowel or the last part of the small bowel at colonoscopy to make sure microscopic inflammation had not been missed. This was taken as the reference standard by which the faecal calprotectin results were judged. The statistical analysis of the combined data – a diagnostic meta-analysis – was published on BMJ.com on 15 July 2010 and in summary form in the BMJ on 24 July 2010.

The authors’ meta-analysis included 13 published studies, six in adults with a total of 670 patients and seven in children and teenagers with 371 patients. Colonoscopy confirmed inflammatory bowel disease in 214 (32%) of the adults and 226 (61%) of the younger group.

A raised calprotectin value was found in 215 adults, of whom 200 (93%) had inflammatory bowel disease on colonoscopy. Of the 455 adults who had a normal calprotectin value, colonoscopy was negative in 437 (96%). The test was less efficient in the children and teenagers. A raised calprotectin value was found in 243 of whom 208 (86%) had the disease, and a normal value was found in 128 of whom 110 (86%) had a negative colonoscopy.

Dr van Rheenen and colleagues discussed a number of factors that might account for the false positive and false negative test results. Raised faecal calprotectin values have been reported in bowel infections, cancer of the bowel and other gastrointestinal diseases, and false negative results might have been the result of the long time interval, up to several weeks, between colonoscopy and calprotectin sampling in some studies.

They discussed whether the number of false positives and negatives would be acceptable if only those patients with suspected inflammatory bowel disease and an abnormal faecal calprotectin result were sent for urgent colonoscopy. They considered a hypothetical population of 100 adults with suspected inflammatory bowel disease and with an actual disease prevalence of 32%. Three patients without the disease would go on to have endoscopy. However, this might uncover another cause for their symptoms such as bowel cancer. The downside would be that two patients with the disease would have their diagnosis delayed. A major advantage is that faecal calprotectin screening would reduce the number of adults requiring endoscopy by 67%.

The authors emphasise that their conclusions are drawn from studies in hospital practice, where a screening test is used to ‘rule in’ a disease to be diagnosed by a more complicated and expensive procedure. They have reservations about their application in general practice where the true prevalence is much lower and a test is more often needed to ‘rule out’ disease. They would certainly discourage the use of faecal calprotectin to screen asymptomatic patients.