Endoscopic Activity Prediction in Ulcerative Colitis Using Hemogram-Derived Inflammatory Indices: Impact of Azathioprine Use on Diagnostic Performance

Vollständiger Abstract

Worum geht es in dieser Arbeit?

Objective: In ulcerative colitis (UC), endoscopic assessment remains central to disease monitoring, but repeated colonoscopy is invasive and resource-intensive. This study evaluated the association and discriminatory performance of the systemic immune–inflammation index (SII; platelet × neutrophil/lymphocyte) and the hemoglobin–albumin–lymphocyte–platelet (HALP; hemoglobin × albumin × lymphocyte/platelet) score for endoscopic activity in UC and explored whether azathioprine use modified these associations. Methods: Among 428 retrospectively reviewed patients with UC, 218 met the inclusion criteria. Mayo-defined activity (endoscopic score ≥ 2) and Rachmilewitz EAI-defined activity (score ≥ 4) were analyzed as separate outcomes. Laboratory parameters were compared between active disease and remission groups. Receiver operating characteristic (ROC) analyses, multivariable logistic regression, sensitivity analyses accounting for disease duration and treatment category, and azathioprine interaction analyses were performed. Results: According to the Mayo and Rachmilewitz EAI definitions, 80 (36.7%) and 150 (68.8%) patients, respectively, had active disease. For Mayo-defined activity, CRP showed the highest discrimination (AUC = 0.720), followed by SII (AUC = 0.692), whereas HALP showed limited discrimination (AUC = 0.598). In the primary multivariable Mayo model, CRP, SII, younger age, and male sex were independently associated with activity, whereas HALP and azathioprine use were not. Adding SII to CRP produced only a small numerical increase in AUC (0.720 to 0.752), and adding HALP produced essentially no further change. No significant interaction between azathioprine use and SII or HALP was identified. Conclusion: SII was associated with endoscopic activity but showed only modest discriminatory performance, while HALP had limited discriminatory and independent predictive value. CRP remained the strongest individual biomarker. SII and HALP should therefore be interpreted as adjunctive inflammatory indices rather than substitutes for established biomarkers or endoscopic assessment.