Association of inflammatory markers based on routine blood with prognosis in patients underwent percutaneous coronary intervention

Inflammation contributes to the pathophysiological processes of coronary artery disease. We evaluated the association between inflammatory biomarkers, neutrophil-to-lymphocyte ratio (NLR), red cell distribution width (RDW), systemic inflammatory index, platelet-lymphocyte ratio, and 1-year all-cause mortality in patients underwent percutaneous coronary intervention (PCI). In this retrospective cohort, we consecutively enrolled 4651 patients who underwent PCI. Baseline demographic details, clinical data, and laboratory parameters on admission were analyzed. The primary outcome was 1-year all-cause mortality after PCI. We performed Cox regression and restricted cubic spline analysis to assessed the association between the inflammatory biomarkers and the clinical outcome. The area under the curve from receiver operating characteristic analysis was determined for the ability to classify mortality outcomes. A total of 4651 patients were included. Of these, 198 (4.26%) died on follow-up. Univariate Cox regression showed that NLR (heart rate [HR]: 1.070, 95% confidence interval [CI]: 1.060-1.082, P < .001), RDW (HR: 1.441, 95% CI 1.368-1.518, P < .001), systemic inflammatory index (HR: 1.000, 95% CI 1.000-3.180, P < .001), platelet-lymphocyte ratio (HR: 3.812, 95% CI 1.901-3.364, P < .001) were significant predictors of 1-year all-cause mortality. After adjusting for other confounders in multivariate analysis, NLR (HR: 01.038, 95% CI 1.022-1.054, P < .001) and RDW (HR: 1.437, 95% CI 1.346-1.535, P < .001) remained significant predictors. Restricted cubic spline analysis showed the relationship between RDW, NLR, and 1-year all-cause mortality was linear after adjusting for the covariables (P for non-linearity < 0.001). The multivariable adjusted model led to improvement in the area under the curve to 0.83 (P < .05). Nomogram was created to predict the probability of 1 year mortality. Among the laboratory indices, RDW and NLR showed the best performance for mortality risk prediction. Multivariate predictive models significantly improved risk stratification.