Investigating young adults with chronic diarrhoea in primary careBMJ 2015; 350 doi: https://doi.org/10.1136/bmj.h573 (Published 25 February 2015) Cite this as: BMJ 2015;350:h573
- Thomas P Chapman, gastroenterology clinical research fellow1,
- Lucia Y Chen, foundation year 1 doctor2,
- Laurence Leaver, general practitioner3
- 1Translational Gastroenterology Unit, University of Oxford, John Radcliffe Hospital, Oxford OX3 9DU, UK
- 2Maidstone Hospital, Maidstone and Tunbridge Wells NHS Trust, Maidstone, UK
- 3Department of Primary Care Health Sciences and Green Templeton College, University of Oxford, Oxford, UK
- Correspondence to: T P Chapman
- Accepted 12 December 2014
The bottom line
A focused history is key, with organic disease suggested by continuous or nocturnal diarrhoea, significant weight loss, and symptoms for less than three months
Refer urgently to secondary care if there are any red flags (such as weight loss, rectal bleeding, or anaemia)
Irritable bowel syndrome is the most common cause of chronic diarrhoea and can be diagnosed clinically with minimal testing (normal full blood count and C reactive protein with negative coeliac serology)
Consider checking faecal calprotectin if there is diagnostic uncertainty between irritable bowel syndrome and inflammatory bowel disease—a concentration below 50 µg/g makes inflammatory bowel disease unlikely. Do not perform the test if there are red flag indicators or a high suspicion of inflammatory bowel disease, or in older patients, as these patients require referral
A previously well 25 year old man presents to his general practitioner with a six month history of crampy abdominal pain, bloating, and diarrhoea, with passage of loose stools up to six times a day. There is urgency but no incontinence. On clinical examination he has mild lower abdominal tenderness, but digital rectal examination is normal.
What are the next investigations?
Chronic diarrhoea can be defined pragmatically as the passage of loose stools more than three times a day for at least four weeks.1 A patient’s own perception of diarrhoea may be markedly different, and it is essential to clarify—for example, faecal incontinence is commonly misconstrued as diarrhoea.1
Irritable bowel syndrome (IBS), a functional disorder, is the most prevalent cause of chronic diarrhoea.2 Other important causes include inflammation (inflammatory bowel disease (IBD), microscopic colitis, and more rarely infection), bile acid diarrhoea, malabsorption (coeliac disease, lactose intolerance, and pancreatic insufficiency), drugs, food additives, and endocrine conditions (thyrotoxicosis and diabetes). Clinicians should always be mindful of an underlying cancer, particularly in patients over 45 years, and the presence of any red flags, primarily suggestive of cancer or IBD, should prompt urgent referral to secondary care (box 1).3
Box 1: Red flags prompting urgent referral to secondary care3
Unintentional weight loss
Family history of bowel or ovarian cancer
Looser or more frequent stools persisting longer than six weeks in patients over 60 years
Abdominal or rectal masses
A careful history provides important clues to the cause and guides further investigation (box 2). Questioning should seek to differentiate functional symptoms, typically caused by IBS, from organic symptoms. Key features suggestive of an organic cause include nocturnal or continuous diarrhoea, serious weight loss (>5 kg), and shorter duration of symptoms (less than three months).1 Conversely, IBS is suggested by the presence of variable bowel habit, bloating, and abdominal pain or discomfort for at least six months. The prevalence of IBS peaks at 20-40 years of age, and it is twice as common in women as in men. Associated somatic symptoms such as headache and back pain, and psychosocial distress often coexist.4 Specific IBS diagnostic criteria such as Manning and Rome III have been devised, but sensitivity and specificity are only about 70%.5
Box 2: Key points in clinical assessment
Are there red flag indicators?
If present refer immediately to secondary care
Differentiating organic from functional disorders
Suggestive of organic disorder
Nocturnal or continuous diarrhoea (or both)
Shorter duration of symptoms (<3 months)
Onset after 45 years
Serious weight loss (>5 kg)
Suggestive of functional disorder
Variable bowel habit
Abdominal pain or discomfort relieved by defecation or associated with change in stool frequency or consistency, as well as bloating
Screening for other common or important causes
Drugs and food additives
For example, antacids, antibiotics, chemotherapy, metformin, sorbitol, and caffeine
Dairy products can exacerbate diarrhoea in patients with lactose intolerance
Excess alcohol can cause rapid gut transit or pancreatic insufficiency
Family history or autoimmune disease
Cholecystectomy, ileal disease or resection, or steatorrhoea
Bile acid diarrhoea
Although physical examination rarely provides a specific diagnosis, it contributes valuable information, including assessment of severity of diarrhoea through appraisal of fluid and nutritional status. A digital rectal examination should always be performed, including careful inspection of the perianal area for fistulae or abscesses, which may be seen in IBD.
All patients will require a set of screening blood tests as follows:
If IBS is considered the most likely diagnosis after history and examination, the National Institute for Health and Care Excellence (NICE) recommends full blood count, C reactive protein, erythrocyte sedimentation rate or plasma viscosity, and antibody testing for coeliac disease.3 However, there is little evidence that erythrocyte sedimentation rate and plasma viscosity have added benefit.6
If clinical features do not clearly differentiate functional from organic disease, also request thyroid function tests (to exclude hyperthyroidism), liver function tests (low albumin suggests inflammation or malabsorption and inflammatory bowel disease is associated with liver disorders), tests for urea and electrolytes (organic diarrhoea may cause raised urea and electrolyte derangements), calcium (both raised and depressed serum calcium levels are associated with lower gastrointestinal symptoms), vitamin B12, serum folate, and ferritin.1 Measure serum Ca-125 in women (particularly if ≥50 years) with symptoms that suggest ovarian cancer in line with the NICE guideline on ovarian cancer.7
Interpreting blood results
The presence of anaemia, hypoalbuminaemia, hypokalaemia, or raised inflammatory markers has high specificity for organic disease, but normal blood results do not exclude it.8 For example, although inflammatory markers are mainly requested as a screen for IBD, they are normal in up to half of patients with the condition, most commonly those with distal ulcerative colitis.9
A raised platelet count is another useful marker of inflammation. Anaemia is considered a red flag indicator. Although iron deficiency is common in IBD and coeliac disease, cancer must be excluded.10 However, young women may have iron deficiency anaemia as a result of menstrual blood loss, with a microcytic picture indicating chronicity. Conversely, vitamin B12 or folate deficiency may cause macrocytosis. Low ferritin, vitamin B12, or folate can be seen with small bowel enteropathy, notably coeliac disease.
Coeliac disease is underdiagnosed in primary care and serology should be requested in all patients.3 11 IgA anti-tissue transglutaminase antibody (TTG) is now the preferred first line investigation,12 although IgA anti-endomysial antibody (EMA) is also widely used. Both have sensitivities greater than 90% and specificities greater than 98% in patients with abdominal symptoms.13 However, there is a risk of false negative results in patients with IgA deficiency. This is important in coeliac disease, in which the prevalence of IgA deficiency is 2%—10 times higher than in the general population. Consequently, many laboratories also provide a total IgA level; if low, request IgG based tests for these antibodies. A raised total IgA level with a negative EMA or TTG result does not indicate coeliac disease and does not require referral to gastroenterology. Finally, antibodies typically become negative on a gluten-free diet, so diagnostic testing should be performed when the patient is on a gluten-containing diet.
Stool microscopy, culture, and antigen testing
Infectious agents seldom cause chronic diarrhoea in Western populations unless there is associated immunodeficiency. Giardiasis, amoebiasis, and cryptosporidiosis are the most common chronic infections, and three fresh stool samples for culture, ova, cysts, and parasites should be sent if there is a history of travel to high risk areas. Up to 40% of protozoal infections may still be missed with this approach, so if suspicion remains, stool enzyme linked immunosorbent assay (ELISA) is a more sensitive assay.14
Request this if there is any question of undiagnosed immunodeficiency.
Clostridium difficile toxin
Clostridium difficile may cause recurrent episodes of diarrhoea. When suspected, particularly after recent hospital admission or antibiotic use, request faecal toxin testing.
Faecal calprotectin is a non-invasive stool test that helps distinguish functional from organic disease in patients with lower gastrointestinal symptoms, with raised levels suggestive of intestinal inflammation.15 New NICE guidance supports its use as an additional method of differentiating IBS from IBD in primary care.16
Limited primary care data—A 2010 meta-analysis determined that a positive faecal calprotectin test has a sensitivity of 93% and specificity of 96% for the diagnosis of IBD in adults, but this did not include primary care data and the positive cut-off value varied between studies.17 Using the most common cut-off value of 50 μg/g, most specialist care studies suggest a sensitivity and specificity of over 80%, and positive and negative predictive values (PPV, NPV) of 70-90% for organic bowel disease.16 Caution is needed in applying these findings to primary care, because the lower prevalence of organic disease will affect predictive values. A recent large retrospective primary care study found a PPV of only 28% with a cut-off value of 50 μg/g, although NPV remained high at 98%.18 Thus, a patient with faecal calprotectin less than 50 μg/g has a 2% probability of having organic disease, underlining the value in primary care of a negative test to support a diagnosis of IBS. Although more research is needed to determine optimal cut-off values in primary care (lower thresholds improve sensitivity and NPV at the expense of specificity and PPV), for ELISA based tests, 50 μg/g may be the most appropriate to reduce the risk of missed or delayed diagnoses of IBD.19
Age limit—There are currently no firm recommendations for an upper age cut off for faecal calprotectin testing. A primary concern in older patients is that a negative test result may falsely reassure in a case of colorectal cancer and lead to a missed diagnosis. Although more research is needed, 45 years may be an appropriate age threshold18 19 at which the risk can be considered sufficiently low.
Other causes of raised faecal calprotectin levels—Gastrointestinal diseases other than IBD—including infection, diverticulitis, polyps and malignant tumours, untreated coeliac disease, and food allergy—may raise faecal calprotectin levels. False positive results may be caused by non-steroidal anti-inflammatory drugs, proton pump inhibitors, and high alcohol consumption, and if possible these should be stopped at least two weeks before testing.
Organic causes of diarrhoea with a negative faecal calprotectin test—Negative results will not differentiate IBS from other common causes of diarrhoea, such as bile acid diarrhoea, which may also present with normal screening blood test results (box 3).
Box 3: Other important causes of diarrhoea in primary care that may need further investigation or referral for diagnosis
Suspect if symptoms worsened by dairy products
Diagnosis: consider a therapeutic trial of a lactose-free diet. Formal diagnosis requires lactose hydrogen breath test (but false negative rate up to 25%) or lactose tolerance test
Bile acid diarrhoea
Suspect in a patient with a history of cholecystectomy or ileal disease or resection. Also associated with other gastrointestinal diseases such as coeliac disease and chronic pancreatitis. Symptoms exacerbated by fatty foods, with watery nocturnal diarrhoea or steatorrhoea. Importantly, may also be idiopathic and found in 20-30% of patients diagnosed as having functional diarrhoea20
Diagnosis: consider therapeutic trial of bile acid sequestrant, although definitive diagnosis requires a 75SeHCAT radiolabelled scan if available20
Suspect if watery nocturnal diarrhoea. More common in older women and associated with coeliac disease
Suspect if history of chronic pancreatitis or excess alcohol, with watery diarrhoea or steatorrhoea
Diagnosis: faecal elastase, which has a high sensitivity and specificity for moderate and severe insufficiency but may miss mild cases
Possible role of faecal calprotectin test—Despite these limitations, faecal calprotectin testing has been shown to be superior to serum inflammatory markers and primary care referral decisions based on clinical grounds alone.9 Consider this test in patients with lower gastrointestinal symptoms such as chronic diarrhoea if clinical features and blood tests do not clearly differentiate IBS from IBD. Do not perform the test if there are red flag indicators (box 1) or a high suspicion of IBD because these patients require specialist referral irrespective of the test result, or in older patients (we suggest 45 years as a cut-off age). The high NPV in primary care means that a negative result can help to support the clinical diagnosis of IBS, sparing the patient further investigation, including invasive lower gastrointestinal endoscopy. Testing is not necessary in patients with a clear diagnosis of IBS. Refer patients with a positive faecal calprotectin result for further investigation, although it may be reasonable to monitor those with borderline results (50-150 μg/g) because the concentration may fall again in IBS.18 19
When to refer
Refer all patients with red flag indicators, suspected coeliac disease (positive serology requires confirmation with duodenal biopsy), an uncertain diagnosis, or suspected organic disease such as IBD, for further assessment. Clinical judgment is paramount, and normal initial investigation results should not deter referral if there is a high suspicion of organic disease (for example, a patient with a strong family history of IBD).
If bile acid diarrhoea is suspected, consider referral for a formal diagnosis (box 3), given the need for lifelong medication that may be poorly tolerated or expensive.21 Patients with IBS that is difficult to manage may also benefit from review in a specialist IBS clinic, with dietitian input.
A focused history showed that our patient had intermittent non-bloody diarrhoea, with no nocturnal motions or other features suggestive of organic disease. His blood test screen was normal and coeliac serology (EMA) was negative, with a normal total IgA level. Stool microbiology was not requested because he had no infectious risk factors. Given his sex and absence of psychosocial stressors, which are less typical for IBS, faecal calprotectin was requested but was negative (25 μg/g).
A diagnosis of diarrhoea predominant IBS was made on the basis of his clinical presentation and reassuring investigation results. His symptoms improved with dietary and lifestyle measures,3 together with use of loperamide for diarrhoeal episodes and mebeverine for discomfort and bloating.3
Cite this as: BMJ 2015;350:h573
This series of occasional articles provides an update on the best use of key diagnostic tests in the initial investigation of common or important clinical presentations. The series advisers are Steve Atkin, professor of medicine, Weill Cornell Medical College Qatar; and Eric Kilpatrick, honorary professor, department of clinical biochemistry, Hull Royal Infirmary, Hull York Medical School. To suggest a topic for this series, please email us at.
Contributors: All authors helped in the conception and design of the article, TPC produced an initial draft, and LYC and LL revised the article for intellectual content. All authors approved the final version. LL is guarantor. Thanks to Satish Keshav, Kate Williamson, and Robert Bryant (Oxford Translational Gastroenterology Unit) and Richard Stevens (retired, Oxfordshire Primary Care) for helpful comments on the draft manuscript.
Competing interests: We have read and understood BMJ policy on declaration of interests and declare the following interests: none.
Provenance and peer review: Not commissioned; externally peer reviewed.
Patient consent not required (patient anonymised, dead, or hypothetical).