Association between FSP, CVHI, inflammatory cytokines and the incidence of primary stroke.

This case-control study was designed to establish a new risk-prediction model for primary stroke using Framingham stroke profile (FSP), cerebral vascular hemodynamic indexes (CVHI) and plasma inflammatory cytokines including hs-CRP, IL-6, TNF-α and Lp-PLA2. A total of 101 primary stroke patients admitted to Dongguan Houjie Hospital between August 2014 and June 2015 were assigned into the case group, and 156 age- and gender-matched healthy subjects from the Houjie Community were allocated into the control group. The prognostic values of FSP, CVHI and inflammatory cytokines including high sensitive C-reactive protein (hs-CRP), interleukin-6 (IL-6), tumor necrosis factor-α (TNF-α) and lipoprotein-associated phospholipase A2 (Lp-PLA2) were assessed by multivariate logistic regression analysis. Seven risk-prediction models (FSP, CVHI, inflammatory cytokine, FSP+CVHI, FSP+inflammatory cytokine, CVHI+inflammatory cytokine, CVHI+FSP+inflammatory cytokine) were successfully established and the prognostic values were statistically compared by ROC curve and Z test. For FSP, the stroke risk was significantly elevated by 2.85 times when the FSP score was increased by 1 level (P=0.043), increased by 3.25 times for CVHI (P=0.036), 6.53 times for IL-6 (P=0.003), and 7.75 times for Lp-PLA2 (P=0.000). The sensitivity of FSP+CVHI+inflammatory cytokine and CVHI+inflammatory cytokine models was higher than 90%. For model specificity, the specificity of FSP+CVHI+inflammatory cytokine model alone exceeded 90%. FSP, CVHI, IL-6 and Lp-PLA2 are independent risk factors of stroke. Integrating IL-6 and Lp-PLA2 into the models can significantly enhance the risk prediction accuracy of primary stroke. Combined application of FSP+CVHI+inflammatory cytokine is of potential for risk prediction of primary stroke.