P277 Assessment of fatigue as a patient-reported outcome: Correlation with baseline disease activity and therapy response in Inflammatory Bowel Disease

Tran, F.(1,2);Schrinner, F.(2);Nikolaus, S.(1);Kümpers, J.(1);Sievers, L.K.(1,2);Lessing, A.(1);Lintner, J.(1);Lessing, M.(1);Rosenstiel, P.(2);Aden, K.(1,2);Franke, A.(2);Schreiber, S.(1,2);

(1)University Medical Center- Schleswig-Holstein- Kiel Campus, Department for Internal Medicine I, Kiel, Germany;(2)University Medical Center- Schleswig-Holstein- Kiel Campus, Institute of Clinical Molecular Biology, Kiel, Germany;


Currently available disease activity indices for IBD (e.g., CDAI and Mayo score) focus on symptoms directly related to intestinal function, while other relevant health consequences of disease activity including quality of life, neuropsychiatric symptoms like depression, and social functionality, are not captured. With better therapies more stringent definitions of patient health have come into focus. One of the most common patient reported outcomes (PROs) in IBD is fatigue, which can be unrelated to endoscopic disease activity. We have shown earlier a strong degradation of tryptophan in active IBD (Nikolaus et al, 2017, Gastroenterology) and low tryptophan levels have been linked in various chronic disease to high fatigue levels. Here, we aim to study the relationship of fatigue with disease activity in the context of therapy response to a biological treatment.


Patients with Crohn’s disease and ulcerative colitis with high disease activity (Mayo score ≥ 6, CDAI ≥ 220) and planned new therapy with a biologic agent were included into a prospective observational protocol. After informed consent, clinical disease activity, inflammatory biomarkers, tryptophan levels in serum and fatigue levels were assessed longitudinally at week 0, 2, 6 and 14 after start of induction therapy. Complete assessments including week 14 were available for 38 patients for this interim analysis of the ingoing study. For fatigue assessment we used the validated German version of the fatigue subscale of the FACIT-F questionnaire. Patients were stratified into therapy responder (clinical remission (CR), reduction of partal Mayo score (pMayo) > 50%, CDAI < 150), non-responders (reduction of Mayo <30%, CDAI reduction <70) and partial responders at week 14.


In the cross-sectional analysis of all patients (n = 76), the fatigue score strongly (inversely) correlated with pMayo and CDAI and correlated positively with tryptophan levels in serum. In receiver operating characteristic (ROC) analysis, FACIT-F score were able distinguish active vs. non-active disease (p < 0.0001, AUC = 0.7619) with specificity of 0.7103 and sensitivity of 0.7160. We did not observe significant differences in fatigue scores between males and females. In the longitudinal analysis, fatigue score only improved in the CR group (n = 25, median FACIT score 30 at baseline vs. 44 at week 14, p = 0.0221), and was significantly higher in CR vs. non-responder (n = 16, median FACIT score 44 vs. 31.5, p = 0.007).



Fatigue is strongly related but not completely overlapping with disease activity. Fatigue will be evaluated as a a patient-oriented indicator for therapeutic response. This observation will be extended by increasing the cohort size and independent validation.