Comparison and validation of systemic immune-inflammation index, C-reactive protein/albumin ratio, and C-reactive...

This study explored the value of the systemic immune-inflammation index (SII), C-reactive protein/albumin ratio (CAR), and C-reactive protein-albumin-lymphocyte index (CALLY) to evaluate rheumatoid arthritis (RA) activity and guide clinical decision-making.

We analyzed the levels of SII, CAR, and CALLY in 280 participants (180 RA patients as the training cohort and 100 healthy controls) and compared the differences between RA groups and their correlations with the Disease Activity Score-28 (DAS28), Erythrocyte Sedimentation Rate (ESR), and C-reactive protein (CRP) levels. Another 83 patients (stratified by disease activity into three subgroups: 21, 37, and 25 patients) formed the time validation cohort, and the area under the curve (AUC) differences were compared between the training and validation cohorts. The model performance was evaluated using calibration curves and decision curve analysis.

SII and CAR levels were significantly higher in the RA group than in the healthy control group, whereas CALLY levels were significantly lower (P < 0.05). CAR was positively correlated with DAS28-ESR (r = 0.605), whereas CALLY was negatively correlated with DAS28-ESR (r =  - 0.575). Correlation analyses revealed near-perfect collinearity of CAR and CALLY with CRP (r = 0.996 and r =  - 0.966, respectively). CALLY exhibited an independent predictive value after controlling for CRP levels. Receiver operating characteristic curve analyses revealed that all models yielded relatively high AUC values without statistically significant differences. Calibration and decision curve analyses revealed that the SII + CAR model provided a more stable calibration performance and net benefit.

Overall, CALLY serves as an independent predictor of RA disease activity and provides a clinically meaningful assessment of patients nutritional status and immune reserves. SII + CAR has more practical application value in terms of calibration performance and net clinical benefit. Multicenter prospective studies with long-term follow-up are required to validate its stability and clinical utility. Key Points • Biological markers for assessing RA disease activity, systematic comparative and validation studies, and guidance for clinical decision-making analyses remain to be identified. • We evaluated the performance of the SII, CAR, and CALLY in assessing RA disease activity using ROC, calibration, and decision curve analyses. • CALLY serves as an independent predictor of RA disease activity, and SII + CAR has greater practical value in terms of discriminatory ability, calibration performance, and net clinical benefit.