Limited incremental value of three complete-blood-count inflammatory indices for cardiovascular mortality: a development and external validation study in 30,533 US adults

Abstract Routine blood tests can be used to calculate several inflammation-related ratios, including the neutrophil-to-lymphocyte ratio (NLR), systemic immune-inflammation index (SII), and platelet-to-lymphocyte ratio (PLR). These indices are widely reported as markers of cardiovascular risk, but it is unclear whether they provide comparable information or whether SII, which is derived from platelet count and NLR, adds anything beyond NLR. We analyzed two independent samples from the National Health and Nutrition Examination Survey (NHANES): 15,141 adults from 2005-2010 (development cohort) and 15,392 adults from 2011-2016 (validation cohort), each with complete blood count data and linked mortality follow-up. Survey-weighted Cox models, restricted cubic splines, multiple imputation, and model comparison were used, and the development model was applied to the validation cohort. During a median follow-up of 11.4 years, 795 cardiovascular deaths occurred in the development cohort; the validation cohort recorded 364 cardiovascular deaths over 5.75 years. In the development cohort, hazard ratios per 1-SD increase were 1.32 (95% CI 1.21-1.44) for NLR, 1.21 (1.12-1.32) for SII, and 1.13 (1.04-1.22) for PLR. When NLR and SII were entered together, only NLR remained associated (HR 1.49, 1.23-1.81). The validation cohort showed the same pattern: NLR was strongest (HR 1.35, 1.22-1.49), and neither SII nor PLR remained significant after mutual adjustment. The frozen model showed good discrimination (C-index 0.8991 to 0.9023) and calibration (slope 0.96). Adding NLR produced only a small improvement in discrimination, and reclassification results were inconsistent between cohorts.