Gastroenterology specialists Dr Michael Colwill, Dr Kevin Jacob and Dr Andrew Poullis advise on how to manage a patient with ongoing intermittent diarrhoea after a positive FIT but negative colonoscopy
A couple of months ago you saw an otherwise well 74-year-old male patient with a four-month history of intermittent diarrhoea. His FIT was positive so he underwent straight-to-test colonoscopy under the urgent cancer pathway. This showed no obvious abnormality other than a few scattered diverticulae. He is now back to see you. ‘It was good to know the test was OK,’ he says. ‘But what can we do about this diarrhoea – it’s becoming a real problem?’
This is a very common clinical scenario that affects patients across all specialties. Urgent suspected cancer pathways are under significant pressure, like rest of the NHS, and consequently the teams managing these patients focus solely on diagnosing malignancy. If malignancy is ruled out, the patients will be sent back to the referring team often without management advice or an ongoing plan.
In the case of this patient, he has had the gold-standard investigation – ie, a colonoscopy – to rule out colorectal cancer (CRC) and the incidental finding of a few scattered diverticulae is a common and likely insignificant finding. However, his symptoms are clearly still bothering him, so what is the best way forward with diagnosing and helping him?
What other causes should we consider in this situation?
Chronic and intermittent diarrhoea is a common pathology and has a wide differential that includes:
• Inflammatory Bowel Disease (IBD) – while in this case ulcerative colitis has effectively been ruled out by the colonoscopy, Crohn’s disease remains a differential particularly if the procedure did not involve ileal intubation.
• Diverticulosis.
• Chronic infection – eg, Clostridium difficile, Campylobacter, Giardia or Entamoeba.
• Disorders of the gut-brain-axis such as diarrhoea predominant irritable bowel syndrome (IBS-D).
• Microscopic colitis.
• Dietary intolerances – eg, lactose or gluten intolerance.
• Iatrogenic secondary to medications.
• Systemic causes such as hyperthyroidism or diabetes.
• Malabsorption diarrhoea – caused by pancreatic exocrine insufficiency, coeliac disease or small intestinal bacterial overgrowth (SIBO).
• Bile acid diarrhoea – most commonly occurs after cholecystectomy or bowel resection particularly after a right hemi-colectomy or terminal ileal resection.
How can we rule these out? Is the extent to which he should be investigated influenced by his positive FIT test, even though his colonoscopy was normal?
Several studies have shown that a negative FIT is an excellent rule-out test for CRC, with one large study that assessed its accuracy in a symptomatic primary care population showing it to have a negative predictive value of 99.8%. On the other hand, the positive predictive value of a FIT for CRC is variable in the literature, but appears to be around 7% when used by GPs. There are multiple conditions, many of which are listed above, that can result in a positive FIT including inflammatory bowel disease, infectious colitis, diverticulitis, benign polyps, haemorrhoids, fissures and fistulas. Medications that predispose to gastrointestinal tract bleeding, such as anticoagulants or anti-platelets, may also give a positive result. Therefore, with a colonoscopy which did not identify a CRC and may not have given any further indication about the cause for this positive FIT, the primary care team may need to investigate further based upon symptoms and clinical history.
It is standard procedure to obtain colonic biopsies when investigating diarrhoea to assess for possible microscopic colitis. It is essential to check that this has been carried out, as there is a danger that the colonoscopy was focused on the positive FIT and assessing for polyps and cancer rather than investigating diarrhoea. If these were not performed but symptoms persist then routine referral to Gastroenterology or an Advice and Guidance request can be made to clarify the reasons for this and whether a repeat procedure with biopsies is indicated.
Many initial tests, along with a history and examination, can be done in primary care. Clues in the history such as a lack of nocturnal symptoms or weight loss, symptoms worsened by stress and abdominal pain related to defecation point towards IBS-D. The history may also identify risk factors for SIBO such as gut dysmotility due to connective tissue disease, previous bowel surgery, immunodeficiency or a history of chronic liver disease or diabetes. A medication review will often also identify the culprit, particularly when comparing the timing of onset of symptoms and medication changes.
Initial blood tests should include a full blood count, thyroid function, coeliac serology, inflammatory markers, electrolytes, haematinics and albumin to investigate for malabsorption. Stool tests such as a culture, provided with any foreign travel details, faecal calprotectin and faecal elastase are also non-invasive tests that could provide the diagnosis. It should be noted that the assay used to measure faecal elastase can be impacted by a dilutional effect, resulting in a falsely low result, so ideally the patient should provide the sample when their stool is as formed as possible or a repeat carried out if a low level is found, to confirm a true result.
There is controversy over whether repeating a FIT has any value. It was recently highlighted in the latest NICE guidance DG56 that there is insufficient evidence to support the routine use of repeated FIT and questions remain over the timing of such a test and the most appropriate thresholds. Previous work has shown that while repeated FIT can improve sensitivity for CRC it can also reduce the specificity and therefore the value of this strategy is unclear, and not currently recommended by NICE. However, some parts of the UK, such as Scotland and Northern Ireland, do advocate for a repeat FIT if there are persisting symptoms of concern. NICE also noted in their guidance and committee discussions when developing DG56 that it is likely that in the context of persisting symptoms and a desire to exclude a sinister pathology, a safety netting strategy in primary care will likely include a second FIT. Given this inconsistency, and while the evidence base for repeated testing remains uncertain, a repeat FIT is a reasonable strategy to adopt in primary care if felt to be clinically appropriate by the requesting practitioner.
When might the patient require further referral?
Despite its strong credentials as an effective rule-out test, a FIT is not perfect and will miss a very small percentage of CRC (0.6% or 1 in 168) and so if there is ongoing clinical concern with new or worsening ‘red-flag’ symptoms outlined in the NICE guidance then an urgent suspected cancer referral is indicated.
Further referral to Gastroenterology may also be indicated if the symptoms are persisting and there is still no diagnosis despite the investigations discussed earlier. This can happen because some of the tests required to make the above diagnoses, such as an MRI of the small bowel to look for ileal Crohn’s disease, or a SEHCAT scan to investigate bile acid diarrhoea, are not available to the vast majority of GPs, and so if there is strong clinical suspicion of these diagnoses referral to secondary care should be made.
If symptoms persist and remain problematic, assuming infection has been ruled out and there is no contraindication, then agents such as loperamide when required can provide symptomatic benefits to patients in the interim until they are reviewed. Avoiding any triggering foods or medication is also advisable.
If investigations reveal a chronic gastroenterological diagnosis such as IBD or pancreatic exocrine insufficiency then these should also be referred to secondary care.
Dr Michael Colwill is Gastroenterology Research Fellow, Dr Kevin Jacob is Gastroenterology Clinical Fellow and Dr Andrew Poullis is Consultant Gastroenterologist at St George’s University Hospitals NHS Foundation Trust, London
