P184 A novel automated calprotectin analysis method increases the diagnostic accuracy of the test
T. Lehto*, P. Hedberg, T. Vaskivuo
University of Oulu and Northern Finland Laboratory Centre Nordlab, Oulu, Finland
Faecal calprotectin is a promising biomarker for inflammatory bowel disease (IBD). The differentiation of IBD from functional gastrointestinal disease can be difficult because of symptoms alone. Calprotectin has been shown to have value both in the diagnosis and in the determination of relapses of Crohn’s disease and ulcerative colitis. We compared new automated calprotectin analysis method (Liaison Calprotectin Kit, DiaSorin) against an established ELISA method (Bühlman Laboratory) and determined its clinical and analytical accuracy.
Stool samples from 100 patients with intestinal symptoms were analysed with both Liaison and Bühlman ELISA methods. Clinical data including colonoscopy findings could be obtained from a sub-group of 62 patients. The thresholds given by the manufacturers were used to determine whether the assays could detect active Crohn’s disease or ulcerative colitis. The analytical performance of the Liaison method was determined and its accuracy was compared with the Bühlman ELISA. Both the imprecision of the Liaison method and the specimen extraction were examined.
The positive predictive value of the methods capability to identify IBD was 68.8% with the Liaison method and 70.4% with the Bühlman ELISA. The negative predictive values of the assays were 93.3% and 85.7% respectively. The correlation between the methods (R2) was 0.12. Within-run precision of a single faecal extract analysed 10 times with the Liaison method was 1.7%. However, when the same faecal sample was analysed from 10 different extraction points the precision of the analysis was 17.7%.
Novel Liaison method for analysing calprotectin increased the negative predictive value of the method from 85.7% to 93.3% when compared with Bühlman ELISA. Negative predictive value is probably the most important aspect of the calprotectin. When compared with each other, the correlation between the Liaison method and Bühlman ELISA was poor, probably indicating the differences in the measurement assays. Further, it should be noted that calprotectin might not be evenly distributed in the faecal sample and the extraction point influences the test results. The Liaison method offers improved diagnostic accuracy and the automated procedure of the assay can help clinical laboratories meet the increasing demand for calprotectin testing.