Exploring the Correlation Between the Systemic Immune Inflammation Index (SII), Systemic Inflammatory Response Index (SIRI), and Type 2 Diabetic Retinopathy

Purpose: Systemic immune inflammation index (SII), systemic inflammatory response index (SIRI) are new inflammation indicators calculated after integrating multiple indicators in blood routine. This study aims to investigate the correlation between the SII, SIRI and type 2 diabetic retinopathy (T2DR) and the diagnostic significance of these indices in T2DR. Patients and Methods: A retrospective analysis involved 500 patients diagnosed with type 2 diabetes (T2DM), categorized into two groups based on fundus examination results: the non-diabetic retinopathy group (NDR, n=256) and the diabetic retinopathy group (DR, n=244). We calculated SII and SIRI, and analyzed their associations with T2DR. Results: The DR group exhibited significantly higher SII and SIRI values compared to the NDR group (P<0.001). Mantel-Haenszel's chi-square trend analysis revealed a notable linear trend (P<0.001) between SII, SIRI, and DR. SII and SIRI exhibited moderate positive correlations with DR, (r=0.354, P<0.001; r=0.469, P<0.001), respectively. Binary logistic regression analysis identified SII and SIRI as independent risk factors for DR. Restrictive cubic spline analysis demonstrated a significant linear relationship between SII and DR (P total trend <0.001, P nonlinear = 0.0657). Stratification by gender indicated that SII is more sensitive to the onset of DR in men. A significant nonlinear relationship was observed between SIRI and DR (P total trend <0.001, P nonlinear = 0.0025), with no gender-related differences in SIRI's association with DR. ROC curve analysis demonstrated that the combined use of SII and SIRI achieved the highest diagnostic accuracy for DR, with an AUC of 0.782, 74.6% sensitivity, and 69.9% specificity. Conclusion: Our findings suggested that SII and SIRI as independent risk factors for DR. The high accuracy of SII combined with SIRI in diagnosing DR underscores their potential as early biological indicators for DR diagnosis.