Abstract
BACKGROUND AND AIM: Inflammatory indices are promising indicators that can be used to evaluate inflammation in inflammatory bowel diseases (IBDs). The present study aimed to investigate the test accuracy of several inflammatory indices to identify endoscopic, and histological activity in a cohort of IBD patients. STUDY: All IBD patients who underwent colonoscopy and blood examination (within 4 weeks and without therapeutic change) were included. For these patients, 10 different inflammatory biomarkers were collected. Our primary outcome was the assessment of accuracy [evaluated with a receiver operating characteristics (ROC) analysis] of each inflammatory biomarker and indices. Furthermore, we tried to establish the optimal cutoff to identify patients with endoscopic and histologic activity among the inflammatory biomarkers and indices with higher performance. RESULTS: Regarding endoscopic activity, at the ROC analysis, the systemic inflammation response index (SIRI) showed the best accuracy [area under the curve (AUC), 0.627; confidence interval (CI), 0.552-0.698]. Whereas the ROC analysis showed a suboptimal AUC for the neutrophil-to-lymphocytes ratio (NLR) and platelets-to-lymphocytes ratio; (AUC, 0.620; CI, 0.545-0.691 and AUC, 0.607; CI, 0.532-0.679, respectively). Concerning histological activity, the C-reactive protein albumin ratio (CAR) presented a higher accuracy among the calculated inflammatory biomarkers (AUC, 0.682; CI, 0.569-0.781) while SIRI and NLR presented a subdued diagnostic performance. CONCLUSION: SIRI and CAR presented the best test accuracy in an IBD outpatient setting to identify endoscopic and histological activity. However, the test accuracy of all the evaluated Inflammatory indices appeared suboptimal. Fecal calprotectin has still the highest accuracy in predicting endoscopic and histological activity in patients with IBD.