An Italian Chart for Cardiovascular Risk Estimate Including High-Density Lipoprotein-Cholesterol
Background: Most current clinical guidelines on primary coronary heart disease prevention emphasize the importance of risk stratification. Tools for cardiovascular risk estimation have been produced in many countries, although their use has been limited. The availability of new tools that include additional risk factors might lead to their more widespread use. The objective of our study was to produce an updated version of an existing chart for the estimation of cardiovascular disease risk using Italian population data and including high-density lipoprotein-cholesterol (HDL-C) levels as a predictor. Methods: Data were analyzed from nine population studies run in Italy, which included a total of 8054 men and 3206 women aged 45-74 years. The individuals included in these studies had no history of cardiovascular events or diabetes mellitus. During a mean follow-up of 10 years (range 5-15), incidence data were collected for non-fatal and fatal cases of major cardiovascular diseases (coronary heart disease, cerebrovascular diseases, and peripheral artery diseases). Findings for major cardiovascular risk factors (i.e. sex, age, systolic blood pressure, serum total cholesterol levels, HDL-C levels, smoking habits) at study entry and their relationship with the occurrence of events during the follow-up were used to develop models for the prediction of cardiovascular events. These were multivariate models, based on a log-linear model incorporating the Weibull distribution, and separate models were developed for men and women. Results: In 10 years, 461 new cardiovascular events occurred among men and 147 among women. The models showed good predictive power, with around 30% of events located in the upper decile of the estimated risk, and around 50% in the upper quintile of estimated risk. The area under the receiver operating characteristic curve, calculated based on internal validation only, was 72%, indicating favorable diagnostic performance of the models. The independent predictive power of HDL-C was strong, with 1% increase in HDL-C level being associated with a decrease in the incidence of cardiovascular diseases of almost 1% among men and almost 2% among women. A chart accommodating sex, age, total cholesterol level, HDL-C level, systolic blood pressure, and cigarette consumption was subsequently produced. The inclusion of HDL-C levels in this chart was novel for a risk chart in Italy, as it had not been included in previous editions of the same tool. A special feature of this chart was a new section dealing with the estimate of the `relative risk,' defined by the ratio of absolute risk to the risk expected on the basis of the age, sex, and average age-specific risk factor levels of the involved populations. Conclusions: The cardiovascular risk assessment devised in the current study represents an improved means for physicians to determine cardiovascular risk and discuss the risk with patients. The chart could be used in countries where the background risk is similar to that of the Italian population; however, external validation of the model is required to adequately assess transferability, and until then the chart should be used with caution in non-Italian populations. Compared with earlier tools, it has the advantage of including HDL-C levels as a predictor of cardiovascular risk.
In order to set up a list of libraries that you have access to,
you must first login
or sign up.
Then set up a personal list of libraries from your profile page by
clicking on your user name at the top right of any screen.