Association Between Primary Sclerosing Cholangitis and Faecal Calprotectin Levels in Patients With Inflammatory Bowel Disease: A Systematic Review and Meta-Analysis.

Primary sclerosing cholangitis (PSC) is strongly associated with inflammatory bowel disease (IBD), and PSC-associated IBD (PSC-IBD) has a distinct phenotype characterised by extensive, often right-sided colitis, frequent endoscopic and histologic activity despite mild symptoms, and an elevated risk of colorectal neoplasia. Faecal calprotectin (FC) is a validated non-invasive marker of colonic inflammation in IBD, but whether FC concentrations differ between patients with PSC-IBD and IBD without PSC, and how faithfully FC reflects colonic disease activity in PSC-IBD, remain uncertain. We conducted a systematic review and meta-analysis of observational studies reporting faecal calprotectin concentrations in patients with PSC-IBD and a contemporaneous IBD-without-PSC comparator. Risk of bias was assessed using a structured quality assessment approach. Standardised mean differences (SMD, Hedges' g) in faecal calprotectin were pooled using a DerSimonian-Laird random-effects model with Hartung-Knapp adjustment; medians were converted to means using an established conversion method. Reports arising from a single overlapping cohort were represented once in the quantitative synthesis. Robustness was examined by leave-one-out, mean-difference and ratio-of-means sensitivity analyses, the last of which is better suited to right-skewed data. Nine reports, arising from four independent cohorts, met the inclusion criteria; three cohorts provided data suitable for meta-analysis. No statistically significant difference in absolute FC between PSC-IBD and IBD without PSC was detected (pooled SMD -0.09, 95% confidence interval -0.41 to 0.22; p=0.56), a finding that was stable across leave-one-out, mean-difference and ratio-of-means sensitivity analyses. Qualitative synthesis indicated that FC may be discordant with colonic disease activity in PSC-IBD: FC was elevated in patients with endoscopically quiescent colitis; its correlation with endoscopic severity relative to IBD without PSC was inconsistent across cohorts; it discriminated mucosal healing with high accuracy, and it correlated closely with biliary calprotectin, while serum (but not faecal) calprotectin was higher in PSC-IBD. No statistically significant difference in absolute FC concentrations was detected between PSC-IBD and IBD without PSC, although clinically relevant differences cannot be excluded given the wide confidence interval, the small number of cohorts and the very low certainty of the evidence. The available data nonetheless suggest that FC may be elevated out of proportion to colonic inflammation in PSC-IBD and may in part reflect biliary inflammation, a hypothesis that requires confirmation. FC should therefore be interpreted with caution as a marker of colitis activity in PSC-IBD, and adequately powered prospective studies pairing FC with endoscopic, histologic and biliary endpoints are required.