To Establish And Predict The Nomogram Model Of Carotid Plaque And The Estimation Equation Of Plaque Score

Objective:To investigate the risk factors of carotid artery plaque and establish and predict the nomogram model of carotid plaque and the estimation equation of plaque score,so as to explore a new method for preliminary screening of carotid plaque formation.Methods:The study collected 1400 subjects who underwent carotid artery color doppler ultrasound and were screened by the exclusion criteria at the Department of Ultrasound,Maanshan People’s Hospital from June 2016 to June 2017.The information collection of the subjects mainly included the questionnaire data,physical examination,clinical biochemical tests and carotid ultrasonic plaque examination.The collection of questionnaire data mainly included general demographic characteristics:age,gender,behavior characteristics such as smoking and drinking status and other past medical history such as the history of hypertension and diabetes.Physical examination mainly included height,weight and blood pressure.The clinical blood biochemical tests included blood lipid indicators(total cholesterol,triglycerides,high density lipoprotein,low density lipoprotein,lipoprotein a)and renal function indicators(uric acid,urea nitrogen,creatinine,cystatin C).Carotid color doppler ultrasound included measuring the carotid intima-media thickness,carotid plaque thickness,location,echo and the carotid artery(common carotid artery,carotid bulb,internal carotid artery)blood flow velocity as well as resistance index.The differences of indicators between Carotid plaque group and normal group were compared with t test or chi-square test.Binary Logistic regression model was used to analyze the risk factors of plaque formation and establish the nomogram model of carotid plaque.Bootstrap self-sampling method was used for internal validation.Carotid plaque score estimation equation was established by multiple linear regression analysis and fitted a quadratic curve mod,and the model was evaluated by ROC curve.Results:1.The single factor analysis of general demographic characteristics:univariate analysis showed that the age(t=-12.46,P<0.001),gender(χ~2=12.66,P<0.001),hypertension(χ~2=107.56,P<0.001),smoking(χ~2=5.00,P=0.025),drinking(χ~2=6.53,P=0.011),pulse pressure(t=-10.76,P<0.001)and BMI(t=-2.67,P=0.008)were significantly different between the plaque formation group and control group.The difference of diabetes between the two groups was not statistically significant(χ~2=1.90,P=0.168).2.Univariate analysis of blood biochemical data on plaque forming and control groups:Lipoprotein a(t=-4.06,P<0.001),Uric acid(t=-5.01,P<0.001),Creatinine(t=-5.05,P<0.001)and cystatin C(t=-10.18,P<0.001)of plaque group were significantly higher than those in control group.High-density lipoprotein concentration in the plaque group(t=3.84,P<0.001)was significantly lower than that in control group.There was no significant difference of other biochemical indexes between two groups.3.Comparison of carotid artery ultrasound data:PSV(t=4.64,P<0.001)and EDV(t=7.30,P<0.001)of the right common carotid artery in plaque group were lower than those in control group,while RI(t=-4.87,P<0.001)was higher than that in control group.The PSV(t=0.01,P=0.994)of the right internal carotid artery had no statistical significance between groups.EDV(t=2.78,P=0.006)in plaque group was lower than that in control group,while RI(t=-5.01,P<0.001)was higher than that in control group.Compared with the control group,the PSV(t=3.98,P<0.001)and EDV(t=7.74,P<0.001)of the left common carotid artery in plaque group were all lower,while RI(t=-7.36,P<0.001)was higher.There was no significant difference of PSV(t=0.73,P=0.465)between the left internal carotid artery and the control group,while EDV(t=3.40,P<0.001)was lower and RI(t=-5.99,P<0.001)was higher in plaque group compared with control group.4.Analysis of the risk factors for the formation of carotid artery plaque:the logistic regression analysis showed that age(OR=1.056,P<0.001),gender(OR=1.603,P<0.001),hypertension(OR=2.108,P<0.001),LDL(OR=1.308,P=0.003),LPa(OR=1.002,P=0.002),Cys C(OR=3.921,P<0.001),right internal carotid artery EDV(OR=1.029,P=0.010)and PP(OR=1.032,P<0.001)were risk factors of carotid artery plaque formation.High density lipoprotein(OR=0.477,P=0.002)and left carotid artery EDV(OR=0.962,P<0.001)were the protective factors.The nomogram model was established based on logistic regression,bootstrap self-sampling internal verification C-index was 0.780.5.Linear regression analysis of clinical blood biochemical parameters,carotid ultrasound data and carotid plaque score:age(t=12.17,P<0.001),PP(t=10.85,P<0.001),LPa(t=2.66,P=0.008),HDL(t=-3.03,P=0.003),UA(t=4.02,P<0.001),BUN(t=2.46,P=0.014),Cr(t=6.43,P<0.001),Cysc(t=10.59,P<0.001),right common carotid artery EDV(t=-5.95,P<0.001)and RI(t=7.21,P<0.001),right internal carotid artery PSV(t=2.87,P=0.004)and RI(t=5.10,P<0.001),the left common carotid artery EDV(t=-6.15,P<0.001)and RI(t=7.82,P<0.001),the left internal carotid artery PSV(t=3.43,P=0.001)and RI(t=6.52,P<0.001)had a linear relationship with the carotid plaque score.6.ROC curve analysis of prediction equation of carotid plaque score:Seven carotid artery plaque integral prediction equations were constructed.Among them,the highestsensitivityofequationY=(-5.063+0.051Age+0.811Sex+0.030PP-0.314HDL+1.338CysC+0.011RICA PSV-0.043LCCA EDV+0.019LICA PSV)was 0.760,the highest specificity of equation Y=(9.005-0.276Age+0.018PP-0.187LCCA EDV+0.001Age PP+0.002Age LCCA EDV+0.002Age~2-0.001PP LCCA EDV+0.002LCCA EDV~2-0.0001PP~2)was 0.690.Conclusion:In this study,a nomogram model of plaque formation and linear model of carotid plaque score in different application ranges are constructed by using age,pulse pressure blood lipid,renal function,and hemodynamic indexes,which has good prediction effect on carotid plaque formation.And it can accurately estimate plaque score,and can be neck Provide a basis for the initial screening of arterial plaque.