The predictive model of carotid angiopathy set from randomly sampled community data
10.3760/cma.j.issn.1671-0282.2014.07.019
- VernacularTitle:基于社区抽样调查数据的颈动脉疾病预测模型
- Author:
Lusha TONG
;
Wenhong JIANG
;
Shenqiang YAN
;
Min LOU
;
Jianmin ZHANG
- Publication Type:Journal Article
- Keywords:
Carotid plaque/thickened intima;
Cervical vascular ultrasound;
Primary stroke prevention;
Predictive model
- From:
Chinese Journal of Emergency Medicine
2014;23(7):801-805
- CountryChina
- Language:Chinese
-
Abstract:
Objective To establish a practical and effective predictive model for predicting carotid plaque/thickened intima,and to provide a novel perspective for stroke screening.Methods A retrospective analysis of the community screening data collected from September 2012 to March 2013 was performed.The data of blood pressure,serum lipid profile and fasting blood glucose and ultrasonographic checks of carotid artery were documented.One thousand subjects were randomly sampled from totally 14 226 individuals and divided into 2 groups according to the presence of carotid plaque/thickened intima.Analysis of relevant data of subjects with Logistic regression were performed.Predictive model was established according to the odds ratio for each risk factor and evaluated by C statistics.Results In the randomly sampled 1000 subjects,164 ones were found to be with carotid plaque/thickened intima.The mean age was (60.23 ± 10.09) years old,110 (67.1％) were male.Smoker,hypertension,angina pectoris,diabetes and hyperlipidemia were seen more often in the subjects with carotid plaque/thickened intima.Logistic regression analysis showed the independent risk factors for carotid plaque/thickened intima included advanced age (OR =1.15,95％ CI:1.12-1.84,P＜0.01),male (OR=3.97,95％CI:2.49-6.33,P＜0.01),smoking (OR=1.80,95％CI:1.24-2.56,P =0.002),diabetes (OR =1.80,95％ CI:1.15-2.98,P =0.012),hypertension (OR =2.82,95％ CI:1.15-2.98,P ＜ 0.01),high total cholesterol level (OR =1.63,95％ CI:1.31-2.02,P ＜ 0.01) and high LDL-C (low density lipoprotein-cholesterol) level (OR =1.70,95 ％ CI:1.35-2.14,P ＜ 0.01).The advanced age was found to be the strongest predictive factor in this model.The C statistics of this model is 0.804,which overweighs other models previously reported.The optimal cutoff point for prediction was 3.5,with a sensitivity of 0.713 and a specificity of 0.757.Conclusion An practical and effective model for predicting carotid plaque/thickened intima established from sampling data of community residents is favorable for both stroke screening and primary stroke prevention.
