Predicting Stroke Risk By Combing Cerebral Vascular Hemodynamic Indexes, Plasma Inflammatory Factors And Framingham Stroke Profile

Background and purpose:Despite of the high incidence, disability and mortality rates of stroke, there was no effective and simple method to predict the first occurrence of stroke. Therefore, we aimed to construct a new model to predict the risk of stroke, by combining the Framingham stroke profile(FSP),cerebral vascular hemodynamic indexes(CVHI) and plasma inflammatory factors, including Hs-CRP, IL- 6, TNF alpha and Lp-PLA2.Methods:This was a case-control study. In total, 101 patients with stroke and 156 community-dwelling healthy controls were recruited, from August 2014 to June 2015. All subjects underwent FSP and CVHI assessment and plasma inflammatory factors exams. The predictive values of FSP, CVHI and inflammatory factors for stroke were analyzed in multivariate, stepwise logistic regression, to identify independent risk factors for stroke occurrence. Then seven logistic regression models for stroke prevention were constructed as follows, FSP alone, CVHI alone, inflammatory factors alone, FSP combined with CVHI, FSP combined with inflammatory factors, CVHI combined with inflammatory factors, and CVHI plus inflammatory factors combined with FSP. For the inflammatory factors, only those found to be independently related to stroke were included in these models. The diagnostic efficiency of these models was then analyzed by the receiver operating characteristic(ROC) curves and the Z test. The sensitivity and specificity were also calculated.Results: In multivariate logistic regression analysis, FSP, CVHI, IL-6 and Lp-PLA2 were independently related to stroke, with the odd ratios(OR) of 2.85, 3.25, 6.53 and 7.75, respectively. The area under the ROC curve(AUC) of the seven models as detailed above were 0.588, 0.810, 0.850, 0.845, 0.861, 0.979 and 0.983(p for all <0.001), respectively. Among all these models, CVHI plus inflammatory factors combined with FSP was found of the highest sensitivity and specificity, which were 92.1% and 91.2%, respectively.Conclusions: The combination of CVHI, inflammatory factors(IL-6, Lp-PLA2) and FSP may yield a valuable tool in predicting the stroke risk in primary prevention, which needs further validation in large, community-based cohort studies.