P309 Ferritin or transferrin saturation vs. soluble transferrin receptor for iron deficiency diagnosis in inflammatory bowel disease

S. DAUDE1, T. Remen2, T. Chateau3, S. Danese4, I. Gastin5, C. Baumann6, J.L. Gueant5, L. Peyrin-Biroulet1

1Nancy University Hospital, Department of Hepato-Gastroenterology and Inserm NGERE, Nancy, France, 2Nancy University Hospital, Methodology Data Management and Statistic Unit, MPI Department, Nancy, France, 3University Hospital of Grenoble, Department of Hepato-Gastroenterology, Grenoble, France, 4Humanitas University, Department of Biomedical Sciences, Milan, Italy, 5Nancy University Hospital, Department of Biochemistry Molecular Biology Nutrition, Nancy, France, 6University Hospital of Nancy, Methodology, Data Management and Statistic Unit, MPI Department, Nancy, France


Iron deficiency is common in inflammatory bowel disease (IBD) and can negatively affect the quality of life even in the absence of anaemia. Diagnosis of iron deficiency is based on ferritin and transferrin saturation (TfS) in routine practice, yet guideline thresholds are not evidence-based. Serum levels of soluble transferrin receptor (sTfR) are the best non-invasive test as it is not influenced by inflammation, but the test is costly with low availability. Thus, the aim of this study was to evaluate for the first time the accuracy of ferritin and/or TfS for diagnosing iron deficiency in IBD and identify the optimal thresholds of these parameters using sTfR as the gold standard.


Serum samples were collected from IBD patients (n = 2,072) receiving a biologic in routine practice. Diagnostic accuracy was assessed using receiver operating characteristic curves for ferritin and TfS levels separately or combined.


No ferritin or TfS threshold had good diagnostic performance in CD patients. In UC patients with CRP <10 mg/l, optimal iron deficiency diagnostic performances were observed with ferritin and TfS thresholds of 65 µg/l and 16%, respectively. For UC patients with CRP >10 mg/l, the thresholds with the best diagnostic performance were 80 µg/l for ferritin and 11% for TfS. There was no added value for combined ferritin and TfS.


In conclusion, we found that ferritin and TfS are reliable parameters for iron deficiency diagnosis only in UC patients, at thresholds different from current guidelines. In CD patients, sTfR should be used given the poor diagnostic performance of ferritin and TfS.