Faecal calprotectin in primary care: is its current use supported by the evidence?

Introduction

Faecal calprotectin (FC) is a calcium-binding cytosolic protein that is secreted by neutrophils.¹ Its use is well established as a screening test for gastrointestinal inflammation, with multiple studies showing elevated levels in patients with inflammatory bowel disease (IBD) compared with control groups, such as patients with irritable bowel syndrome (IBS). It also has a role in the management of IBD because of its correlation with markers of disease activity.² Its low cost and potential ability to reduce unnecessary secondary care referrals have led to its widespread adoption in primary care.³

However, concerns have previously been raised that patients with a borderline elevated calprotectin may be referred for unnecessary investigation.⁴ In addition, many of the initial FC research studies were performed in a secondary care setting, despite its widespread use in primary care. In this article the guidelines for FC use will be reviewed, as well as its evidence in primary care. This article will make suggestions for best practice and highlight areas in which outstanding questions and uncertainties remain.

Guidelines and supporting evidence

NICE guidelines on faecal calprotectin

The National Institute for Health and Care Excellence (NICE) guidelines currently recommend the use of FC in patients with new-onset lower gastrointestinal symptoms, where a differential diagnosis includes IBD or IBS, and specialist referral is being considered.⁵ Lower gastrointestinal symptoms are defined as abdominal pain or discomfort, bloating, or a change in bowel habit over a minimum of 6 weeks. It is also stated that FC should only be used if cancer is not suspected, defined by symptoms described in the NICE guideline on suspected cancer, because of the inadequate sensitivity of FC for detecting colorectal neoplasia.⁶ Patients meeting criteria for suspected colorectal cancer (CRC) should undergo faecal immunochemical testing (FIT), which has been shown to outperform FC for diagnosing CRC.⁷ While not specifically discussed in the NICE guidelines, other published guidelines, including the NICE-endorsed consensus document, advise avoiding FC use in patients with bloody diarrhoea who will require endoscopic assessment irrespective of FC result, and in patients with recent non-steroidal anti-inflammatory drug use because of their association with higher FC levels.^8–10

Diagnostic performance of faecal calprotectin

The NICE FC guidelines were based on a systematic review and economic analysis performed in 2013.¹¹ For the performance of FC in differentiating between IBS and IBD, seven studies were analysed giving FC a pooled sensitivity of 93% and specificity of 94% (using a cut-off of 50 µg/g). The economic analysis suggested that, at these anticipated levels of performance, FC cost would be offset by a reduction in subsequent endoscopic investigations performed. Combined scoring systems using clinical data coupled with FC testing may further improve diagnostic performance.¹² While the NICE guidelines acknowledge uncertainty surrounding the effect of pre-analytical factors on FC levels, one of the advantages of FC is its stability of up to 3 days at room temperature, making its use suitable in healthcare settings such as primary care where rapid sample processing is not possible.¹³

Faecal calprotectin in primary care and re-testing strategies to improve performance

The diagnostic performance of FC, particularly its sensitivity, and its potential to reduce costs have meant that its use has become widespread in primary care. A review of The Healthcare Network Improvement (THIN) database in 2016 (covering about 6% of the UK population) found that 17 072 patients had had at least one FC test performed and the use of FC has continued to increase subsequently.³ An increase in the proportion of remote consultations during the COVID-19 pandemic may also have increased the use of tests, such as FC, which can screen for more serious conditions.

However, the meta-analysis by Waugh et al, on which the NICE guidelines were based, is limited by the fact that all studies were secondary care based with low numbers of participants. Subsequent larger studies in secondary care have tended to support the finding of this meta-analysis with regards to the sensitivity of FC.¹⁴ However, in a primary care setting where the prevalence of IBD is lower, use of FC at a cut-off of 50 µg/g is less specific, potentially resulting in unnecessary referrals and invasive investigations.⁴ The findings of Freeman et al’s analysis of the THIN database confirm this, reporting a sensitivity of 92.9% and specificity of 61.5% (at a threshold of 50 µg/g), compared with 93% and 94% in the Waugh meta-analysis.³

This concern has led to the NICE-approved York Faecal Calprotectin Care Pathway (YFCCP), which advocates the use of FC testing as a diagnostic tool within a decision tree.¹⁵ This approach was supported in the NICE consensus document, which produced a similar pathway building on the work of the YFCCP.⁸ Both pathways suggest using an initial higher threshold of 100 µg/g, rather than the 50 µg/g suggested in the original NICE guidance, because of data suggesting this increase improves specificity without significantly impacting sensitivity.¹⁶ The pathways then suggest the re-testing of results over this threshold: patients with two consecutive results <250 µg/g are then defined as low risk for IBD and treated for IBS in primary care initially, with routine referral only if symptoms persist despite first-line treatment.¹⁵ The consensus pathway suggests that patients with an initial FC level of >250 µg/g are considered for immediate gastroenterology referral if symptoms are severe or worsening, while the YFCCP suggests all positive tests are repeated. Both pathways suggest that FC should only be used on patients under the age of 60 years, because of increasing risks of colorectal cancer above this age. A study of 1005 patients found a sensitivity for FC of 94% and specificity of 92% when the YFCCP was followed, as well as demonstrating its cost-effectiveness.^15,18 The British Society of Gastroenterology (BSG) IBD guidelines broadly support the approach of the YFCCP, though note the lack of published evidence to confidently recommend re-testing strategies.¹⁰

Areas of ongoing uncertainty

Managing borderline results

One of the key uncertainties surrounding FC is the level above which results should be considered raised, leading to difficulties managing patients with borderline elevated results. The YFCCP attempts to address this, though this approach has not been widely validated beyond the evidence described above. Re-testing strategies also need to consider patient compliance: return rates as low as 35% have been reported for FC in IBD patients, raising questions of adherence to this strategy in real-world clinical practice.¹⁹ Diagnostic delays are associated with worse clinical outcomes for IBD patients. Therefore the potential increased time to secondary care referral, caused by multiple tests being performed in primary care, also needs to be considered.²⁰

Patient selection for faecal calprotectin testing

There is also uncertainty surrounding exactly which patients should be targeted for FC testing in primary care. The NICE guidelines state that all patients with ‘lower gastrointestinal symptoms’ are eligible, as long as cancer is not suspected, but some specialty guidelines (such as the BSG guidelines on IBD and IBS) suggest only patients with chronic diarrhoea as their primary symptom should be targeted.^10,17

In fact, there are no specific data on the performance of FC in patients without diarrhoea, who are passing formed stools.

There is also a lack of clarity as to whether there should be a clearly defined age cut-off for the use of FC. Many studies have excluded older adults (with varying age thresholds) because of the higher risk of cancer in this age group.²¹ The BSG guidelines and YFCCP recommend age cut-offs of 40 and 60 years, respectively, but no clear threshold is recommended in the NICE guidance (analysis of FC usage in clinical practice confirms its frequent use in older adults).³

Technical considerations

In addition, while stability at room temperature for up to 3 days is a strength of FC, there remain outstanding technical questions, with studies showing both intra-individual and inter-assay variability.^13,22 The role of recently developed FC home testing kits in primary care pathways also needs to be clarified; high usability scores suggest they may be a method of improving compliance.^23,24

Conclusions and recommendations

FC’s high sensitivity for detecting IBD, as well as its low cost, makes it a useful diagnostic tool in primary care. However, this review of the evidence suggests more research is needed in the following areas:

• the use of FC in older adults and in patients presenting with symptoms other than diarrhoea;

• the exact role of re-testing strategies — evidence suggests that the YFCCP may improve FC performance but more validation studies are needed to definitively recommend this approach;

• the respective roles of FC and FIT testing, and how best to use each test in patients presenting with lower gastrointestinal symptoms; and

• the optimal way to standardise collection procedures and FC analysis, as well as the role of point-of-care home testing kits.

The current authors would recommend the use of FC in younger patients presenting with diarrhoea who do not meet criteria for more urgent referral, where there is a differential of IBD and IBS (Figure 1). Re-testing strategies should be agreed at a local level, with referral thresholds determined by the FC assay used and capacity of local services.

Figure 1.

Proposed flowchart for FC testing in primary care based on existing guidelines.^{8,10,15,17 a}Exact thresholds to be determined locally and dependent on FC assay used. FC = faecal calprotectin. IBS = irritable bowel syndrome. NICE = National Institute for Health and Care Excellence.

Provenance

Freely submitted; externally peer reviewed.

Competing interests

The authors have declared no competing interests.