An evaluation of lipid profile and pro-inflammatory cytokines as determinants of cardiovascular disease in those with diabetes: a study on a Mexican American cohort

Abstract

Sedentary life styles coupled with high-calorie diets and unhealthy social habits such as smoking, have put an ever-increasing number of people at risk of cardiovascular disorders (CVD), worldwide. A concomitant increase in the prevalence of type 2-diabetes (hyperglycemia), a risk factor for CVD, has further contributed towards escalating CVD-related mortalities. The increase in number of cases of type 2-diabetes underscores the importance of early diagnosis of cardiovascular disease in those with diabetes. In this work, we have evaluated the sensitivity and specificity of dyslipidemia and proinflammatory cytokines to be used as biomarkers for predicting the risk of CVD in those with diabetes. We hypothesize that interplay between dyslipidemia and diabetes-induced low-grade inflammation in those with type 2-diabetes increases the risk of CVD. A total of 215 participants were randomly recruited from the Cameron County Hispanic Cohort (CCHC). Of these, 99% were Mexican Americans living on Texas-Mexico border. Levels of cytokines, adipokines and lipid profile were measured. Cardiovascular disease (CVD) for this study was defined as prior diagnosis of heart attack, angina and stroke, while diabetes was defined by fasting blood glucose (FBG) of > 100 mg/dL and HbA1c of > 6.5, in accordance with American Diabetes Association (ADA) guidelines. Depending on type and distribution of data, various statistical tests were performed. Our results demonstrated higher rates of heart attack (14% vs 11.8%) and stroke (19.8% vs 10%) in those with diabetes as compared to non-diabetes. The odds of having a heart attack were eight times higher in the presence of elevated triglycerides and pro-inflammatory markers (TNFα and IL6) as compared to presence of pro-inflammatory markers only. The odds for heart attack among those with diabetes, increased by 20 fold in presence of high levels of triglycerides, TNFα, and IL6 when coupled with low levels of high-density lipid cholesterol (HDL-C). Lastly, our analysis showed that poorly controlled diabetes, characterized by HbA1c values of > 6.5 increases the odds of stroke by more than three fold. The study quantifies the role of lipid profile and pro-inflammatory markers in combination with standard risk factors towards predicting the risk of CVD in those with type 2-diabetes. The findings from the study can be directly translated for use in early diagnosis of heart disease and guiding interventions leading to a reduction in CVD-associated mortality in those with type 2-diabetes.

Figure 1

Relative Risks (RR) for CVD on exposure of Diabetes. Relative risk ratios (on the horizontal axis) for stroke, angina and heart attack are plotted along with their confidence intervals.

Figure 2

Normalized importance for three independent variables and ROC curve for dependent variable. (A) Showing the importance of each independent variable in predicting the occurrence of heart attack hence proving to be important risk factors in DM. (B) Shows the ROC curve for the predictive model of heart attack for the neural analysis. Area under the curve (AUC) of 0.803 is indicative of the accuracy of sensitivity analysis.

Figure 3

Normalized importance for four independent variables and ROC curve for dependent variable. (A) Showing the importance of each independent variable in predicting the occurrence of heart attack hence proving to be important risk factors among DM. (B) Shows the ROC curve for the predictive model of heart attack for the neural analysis. Area under the curve (AUC) of 0.817 is indicative of the accuracy of sensitivity analysis of a good model.