Fecal calprotectin is reasonably accurate in identifying kids with inflammatory bowel disease in primary care

Référence

Walker GJ, Chanchlani N, Thomas A, et al. Primary care faecal calprotectin testing in children with suspected inflammatory bowel disease: a diagnostic accuracy study. Arch Dis Child 2020;105(10):957-963.

Sommaire

These researchers enrolled 195 children between the ages of 4 years and 18 years with gastrointestinal problems whose primary care clinician decided to perform a fecal calprotectin assay. They excluded children with an established IBD diagnosis, those who were suspected of having cancer, and those who had used nonsteroidal anti-inflammatory drugs. The authors used a quantitative ELISA-based fecal calprotectin assay. Levels of 100 mcg/g and higher were considered positive. The gold standard for IBD diagnosis was based on clinical, radiologic, and histopathologic findings during the 12 months after the index assay. Nearly half the children (46%) were female and slightly more than half (54%) had at least one "red flag" criterion for IBD (eg, rectal bleeding, nocturnal symptoms, and so forth). Approximately half the time, the treating clinician would have referred the child if the fecal calprotectin assay were not available. Thirteen (7%) children were ultimately given a diagnosis of IBD: 8 had Crohn's disease, 2 had ulcerative colitis, and 3 were unclassified. The authors report that fecal calprotectin was 100% sensitive (95% CI 75% - 100%) and 91% specific (85% - 94%), which translates to a positive likelihood ratio of 11.1 (5 - 16.7) and a negative likelihood ratio of 0 (0 - 0.3). In other words, this is a pretty decent test for ruling out IBD. However, in a low-prevalence setting, even a specific test carries a greater than 50% false positive rate. The fecal calprotectin test result modestly influenced clinician behavior: 83% of the patients with positive fecal calprotectin levels were referred compared with 54% of those with negative levels.