Search in the Abstract Database

Search Abstracts 2016

* = Presenting author

P128 Faecal calprotectin has an acceptable sensitivity for detecting small bowel Crohn’s disease: results from real-world clinical practice

J. Wu*, E. Harrison, C. S. Chey, H. Johnson, S. Weaver, S. Mclaughlin

The Royal Bournemouth and Christchurch Hospitals NHS Foundation Trust, Gastroenterology, Bournemouth, United Kingdom

Background

Faecal calprotectin (FC) is a quick, non-invasive, and inexpensive test proven to correlate well with colonic inflammation. It is used in our institution to also exclude active small bowel Crohn’s disease (CD), although its reliability in such cases has recently been questioned. Reliable FC cut-off values remain undefined.

The aim of this study was to assess whether FC compared with gold standard magnetic resonance enterography (MRE) and can reliably be used in small bowel CD as a screening tool to select those needing further investigation.

Methods

Data from all CD patients who had undergone MRE to assess for active small bowel CD between January 2012 and November 2015 were reviewed. Patients with a known or new diagnosis of ileal or ileo-caecal CD (Montreal classification L1 phenotype) or ileo-colonic CD (L3 phenotype) with endoscopic and histologically proven quiescent colonic disease were included. Data were analysed using Mann–Whitney U testing.

Results

In total, 194 MREs were performed for investigation of possible small bowel CD during the study period, and 64 patients were included for analysis: L1 disease = 90.6% (58 of 64 patients) and L3 = 9.4% (6 of 64).

Demographics: 50% were female. The median age was 48 years (range 17–76 years). Median time between FC and MRE was 30 days (range 0–180 days). Out of the 64 patients, there were 35 patients (54.7%) with Montreal disease behaviour B1; 24 patients (37.5%) B2; and 5 patients (7.8%) B3. Four patients (6%) had peri-anal involvement. Twenty-six patients (41%) had previous CD-related resectional surgery.

Further, 39 patients (60.9%) had evidence of active small bowel CD at MRE. The median FC in the active CD group was 246 mg/kg (IQR 556), and 49 mg/kg for the inactive small bowel disease group (IQR 177.5); p < 0.0001 (Figure 1).

Figure 1. Box whisker plot showing median faecal calprotectin concentration (horizontal line), 25th and 75th percentile (box), and range (whiskers) in those with active small bowel disease at MRE (n = 39) and no small bowel inflammatory activity at MRE (n = 25).

Table 1 shows results for sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of different FC cut-off values in detecting any degree of small bowel activity at MRE.

Table 1

Faecal calprotectin cut-off value (mg/kg)Faecal calprotectin cut-off value (mg/kgFaecal calprotectin cut-off value (mg/kg
50100200
Sensitivity (%)907259
Specificity (%)506870
Positive predictive value (%)747874
Negative predictive value (%)776154

Sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of different FC cut-off values in detecting any degree of small bowel inflammatory activity at MRE

Conclusion

In this study, we compared FC results to MRE to exclude or confirm active small bowel CD in real-world clinical practice. The data demonstrate that by using a cut-off value of 50 mg/kg, an acceptable sensitivity for detecting active small bowel CD, is achievable. However, at this cut-off value, specificity is low; nevertheless, using FC will still reduce demand for MRE.