P177 C-reactive protein/albumin ratio: a novel biomarker for monitoring of the disease activity in Ulcerative Colitis patients
Tomasic, V.(1)*;Biscanin, A.(1,2);Ćaćić, P.(1);Palac, L.(2);Dorosulic, Z.(1);Kralj, D.(1);Babic, F.(1);Ogresta, D.(1);Hrabar, D.(1,2);
(1)Sestre milosrdnice University Hospital Center, Department of Gastroenterology and Hepatology, Zagreb, Croatia;(2)University of Zagreb School od Medicine, Internal medicine, Zagreb, Croatia;
Background
Fecal calprotectin (FC) is a reliable laboratory test for non-invasive assessment of disease activity in ulcerative colitis (UC) patients; it's high cost, absence of widespread availability and low patient adherence limits it's use as a potential surrogate for endoscopy. The association of higher C-reactive protein (CRP) and low albumin levels, both obtained from patient's sera, with disease acitivity in UC, is well established. C-reactive protein/albumin ratio (CAR) is suggested as a biomarker for evaluating disease activity in various inflammatory conditions. We aimed to evaluate association between CAR, FC, CRP and serum albumin values with disease activity in UC patients.
Methods
A single center, noninterventional retrospective chart review of adult UC patients treated with biologics at a tertiary center was performed. Routine clinical (stool frequency score – SFS, and rectal bleeding score – RBS), biochemical (CRP, serum albumin, FC), and endoscopic (Mayo Endoscopic Subscore – MES) findings were assessed prior to initiating biologics. Active UC was defined as SFS 2/3 + RBS 2/3 + MES 2/3. Differences between continuous variables were calculated using independent sample t-tests and Mann–Whitney U test. Sperman’s correlation and ROC curve analysis were carried out. The Z statistic was used to compare correlation coefficients and AUC.
Results
Total of 82 patients (52.2% females, mean age 40.3 ± 15.3 years, smokers 16.9%) with dominantly extensive UC (57.3%) were analysed. The CAR (0.175 vs 0.025; p=0.02) and FC (1363.5 vs 65.5; p=0.01) levels were significantly higher in active vs inactive UC patients. Positive correlation was observed between CAR (rs = 0.354, p = 0.01), FC (rs = 0.448, p = 0.001) and CRP (rs = 0.316, p = 0.005) with active UC. Although serum albumin (r's = - 0.093, p = 0.532) negatively correlated with disease activity it didn't reach statistical significance. There was no statistically significant difference between CAR and FC correlation coefficients. The area under the curve (AUC) for the diagnosis of active UC was 0.896 (SE: 0.058; 95% CI 0.73 - 0,983, p=0.015) and 0.994 (SE: 0.009; 95% CI 0.98 - 1, p=0.001) for CAR and FC, respectively. Although AUC for FC was higher compared to CAR it didn’t reach statistical significance (p = 0.1).
The cut-off value associated with the active UC for CAR was 0.06 (sensitivity 77,2%, specificity 100%), and 120 mcg/g for FC (sensitivity 97.6%, specificity 100%).
Conclusion
There is good correlation between CAR and clinical as well as endoscopic disease activity in patients with UC, comparable to FC. CAR could be offered as a cheap and practical biomarker for detection of active UC disease.