| Cardiovascular disease (CVD) is the leading cause of death in the world. Therefore, it has become one of the major public health problems the mankind are confronted with, and needs urgent attention of the medical science society to curb its hazards. The whole developing process of CVD, from occurrence, development, to an ultimate outcome including death, is the result of multiple risk factors accumulating damaging the vascular system. Early conduction of cardiovascular risk assessment and appropriate intervention to high risk individuals are helpful to reduce the burden of CVD and would play a significant role in public health. Currently, many countries adopt the "10years CVD accumulated risk assessment model", which is represented by the "Framingham model", to evaluate the’overall risk’of CVD. However, these models usually choose a very limited number of predictors, resulting to relatively insufficient predictability.Former studies have found that a range of CVD risk factors usually gather and appear in the same individual, which was defined by the World Health Organization (WHO) as metabolic syndrome (MetS), meaning the pathological state of aggregation of various abnormal metabolic components. The concept of MetS adds to new insights into studies of prediction of CVD risk. Numerous studies indicate that every component of MetS is an independent risk factor for CVD, and the fact that they interact with each other may increase the risk to develop CVD. In recent years, health management industry witnessed tremendous development in China. The data from health check-up systems has a lot of advantages, including its large sample size, information validity, comprehensiveness and abundance, the feasibility to acquire follow-up database for many years, and its appropriate reflection for the health status of urban residents. Therefore, this kind of data has become a hot data source for epidemiological studies.Our study relied on the Shandong Province multicenter health management platform system, based on its large number of samples, took full advantage of the metabolic cardiovascular risk indicators to build the Coronary Heart Disease (CHD) risk assessment model for Urban Han Chinese.Materials and Methods:The study population includes a cohort of all participants who received routine health check-ups from2005to2010at the Center for Health Management of Shandong Provincial QianFoShan Hospital, and the Health Examination Center of Shandong Provincial Hospital. All participants received a general health questionnaire survey, physical examination, Laboratory tests, and other auxiliary examinations.Exploratory factor analysis (EFA) was performed in5311MetS subjects to extract CVD-related factors with specific clinical significance from16biomarkers tested in routine health check-up. Logistic regression model, based on an extreme case-control design with445CHD patients and890controls, was performed to evaluate the extracted factors used to identify CHD. Then, Cox model, based on a cohort design with1923subjects followed up for5years, was conducted to validate their predictive effects. Finally, a synthetic predictor (SP) was created by weighting each factor with their risks for CHD for developing a risk matrix to predict CHD in the practice of routine health check-up.Results:1. The distributions of metabolic cardiovascular risk factors, the prevalence and combination of these components were significantly different between male and female MetS populations.2. Eight independent factors with their specific clinical significances were retained and named for two groups respectively. For the male group, the first factor was named erythrocyte viscosity factor (EVF) and contributed by Hb&HCT, following lipid viscosity factor (LVF) by TC&LDL-C, lipid metabolism factor (LMF) by TG&HDL-C, blood pressure factor (BPF) by SBP&DBP, hepatic enzyme factor (HEF) by ALT&GGT, glucose metabolism factor (GMF) by FBG&UA&CREA, fat accumulation factor (FAF) by NAFLD&BMI, and inflammation response factor (IRF) by WBC&CREA. For the female group, the first was named LVF and contributed by TC&HDL-C&LDL-C, following EVF by Hb&HCT, HEF by ALT&GGT, BPF by SBP&DBP&BMI, IRF by UA&CREA&WBC count, LMF by TG&HDL-C, GMF by FBG&WBC, and FAF by NAFLD&BMI. These suggested that the patterns of factors were similar between males and females, though their ranks with descending order of explained variance were slightly different.3. Logistic regression discrimination model was conducted in extreme case-control study to classify CHD situations. In male populations, the area under the receiver operating characteristic (ROC) curve was0.994(95%CI0.984-0.998), with0.5617as the cut-off point, and its sensitivity and specificity being95.8%,98.5%respectively. In female populations, the area under the ROC curve was0.998(95%CI0.982-1.000), as0.4158being the cut-off point, with the sensitivity and specificity of99.3%and97.0%.4. Cox proportion hazard prediction model was performed in cohort study to predict the CHD incidence. In male populations, the area under the ROC curve was0.871(95%CI0.851-0.889),0.0701as the cut-off point, with the sensitivity of81.1%and the specificity of79.9%. In female populations, the area under the ROC curve was0.899(95%CI0.873-0.921), with0.0739as the cut-off point, its sensitivity and specificity being86.4%and85.2%, respectively.5. Based on the Cox proportion hazard prediction model, we drew the5-year CHD risk assessment matrix. It provided a simple tool for conducting CHD prediction in health management and clinical practice.6. The proportion of identified high CHD risk individuals among the28200observed individuals shaped an S curve as age growing. People before the age of45years were at relatively very low risk for CHD over the next5years. For the age group from45to65, the curve showed a high-speed increasing phase, indicating the risk of CHD might increase rapidly during this age period. After65, people were at the highest risk for CHD, and almost all individuals over the age of65were supposed to develop into CHD within the next5years.Conclusions: 1. Results of our study suggest that the pathogenesis, progression and symptoms of MetS may be different between male and female populations. This may be caused by the difference of genetic, endocrine, predisposition factors, exposure to pathogenic factors, or other factors affected by gender.2. The development process of MetS is quite complicated:a single pathogenesis is not enough to explain all the metabolic abnormalities. MetS may be caused by mixed effects of high blood viscosity, dyslipidemia, hypertension, dysfunction of liver, obesity, insulin resistance, inflammation and many other pathophysiologic mechanisms.3. A weighted SP index was created by the extracted8factors in our study. The SP-based CHD risk assessment model can well discriminate and predict the incidence of CHD among Chinese urban residents. The5-year risk matrices we drew provided a simple and convenient tool for conducting CHD prediction in health management and clinical practice.4. In Chinese populations, the CHD risk rises with age. The CHD risk increases rapidly after middle age and is extremely high among the old age groups, thus indicating that prevention and intervention should especially be focused on the the people before the age of45years to reduce the increase of incidence risk for developing CHD later in life. |
PDF Full Text Request