Premium
Faecal calprotectin testing for identifying patients with organic gastrointestinal disease: systematic review and meta‐analysis
Author(s) -
An YoonKyo,
Prince David,
Gardiner Fergus,
Neeman Teresa,
Linedale Ecushla C,
Andrews Jane M,
Connor Susan,
Begun Jakob
Publication year - 2019
Publication title -
medical journal of australia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.904
H-Index - 131
eISSN - 1326-5377
pISSN - 0025-729X
DOI - 10.5694/mja2.50384
Subject(s) - medicine , calprotectin , feces , population , meta analysis , gastroenterology , organic disease , area under the curve , predictive value of tests , predictive value , inflammatory bowel disease , disease , biology , paleontology , environmental health
Objectives To assess the clinical effectiveness of faecal calprotectin ( FC ) testing for distinguishing between organic gastrointestinal diseases (organic GID ), such as inflammatory bowel disease ( IBD ), and functional gastrointestinal disorders (functional GID s). Study design Studies that assessed the accuracy of FC testing for differentiating between IBD or organic GID and functional GID s were reviewed. Articles published in English during January 1998 – June 2018 that compared diagnostic FC testing in primary care and outpatient hospital settings with a reference test and employed the standard enzyme‐linked immunosorbent FC assay method with a cut‐off of 50 or 100 μg/g faeces were included. Study quality was assessed with QUADAS ‐2, an evidence‐based quality assessment tool for diagnostic accuracy studies. Data sources MEDLINE and EMBASE ; reference lists of screened articles. Data synthesis Eighteen relevant studies were identified. For distinguishing patients with organic GID (including IBD ) from those with functional GID s (16 studies), the estimated sensitivity of FC testing was 81% (95% CI , 74–86%), the specificity 81% (95% CI , 71–88%); area under the curve ( AUC ) was 0.87. For distinguishing IBD from functional GID s (ten studies), sensitivity was 88% (95% CI , 80–93%), specificity 72% (95% CI , 59–82%), and AUC 0.89. Assuming a population prevalence of organic GID of 1%, the positive predictive value was 4.2%, the negative predictive value 100%. The difference in sensitivity and specificity between FC testing cut‐offs of 50 μg/g and 100 μg/g faeces was not statistically significant ( P = 0.77). Conclusions FC testing is clinically useful for distinguishing organic GID (including IBD ) from functional GID s, and its incorporation into clinical practice for evaluating patients with lower gastrointestinal symptoms could lead to fewer patients with functional GID s undergoing colonoscopy, reducing costs for both patients and the health system. PROSPERO registration CRD 4201810507.