Association of CBC‑derived inflammation indices with prevalent CKD: a cross‑sectional analysis of a Chinese data set and NHANES

Background Given the established association between chronic kidney disease (CKD) and systemic inflammation, this study examined the relationships between complete blood count (CBC)-derived inflammatory markers and prevalent CKD.Methods Two data sets (Chinese clinical data, n = 4,518; NHANES 2017–2020, n = 7,724) are analyzed. Seven CBC‑derived inflammatory indices were calculated: SII, SIRI, NLR, dNLR, NMLR, MLR, and PLR. Each index was categorized into quartiles (Q1–Q4). Weighted logistic regression (for NHANES) and unweighted logistic regression (for the Chinese clinical data) were applied to estimate associations with CKD, with sequential adjustment for demographics, body mass index, and comorbidities. To control for multiple comparisons, false discovery rate (FDR) correction was applied. Furthermore, subgroup analysis, restricted cubic spline analysis and sensitivity analysis were also performed in this study.Results In the two fully adjusted models, the highest quartile (Q4) of SIRI, NLR, and NMLR was significantly associated with higher odds of prevalent CKD compared with Q1. After FDR correction, these associations remained significant in both cohorts: in the Chinese cohort, all padj < 0.001; in NHANES, padj = 0.028 for SIRI, 0.040 for NLR, and 0.041 for NMLR. Sensitivity analyses consistently supported the primary findings: excluding outliers, log2 transformation, and the alternative CKD definition yielded similar effect directions and significance levels. Heterogeneity was observed across subgroups, and restricted cubic splines revealed nonlinear dose‑response relationships (p for nonlinearity <0.05).Conclusions Elevated SIRI, NLR, and NMLR are associated with prevalent CKD, suggesting systemic inflammation may play a role, yet prospective studies are required to establish temporality and causality.