Single Tool Predicts Global and Specific Cardiovascular Risks

By Charles Bankhead
BOSTON, Jan. 23 -- A patient's global cardiovascular disease risk can be predicted accurately by an office-friendly algorithm, according to investigators here.
What's more, simple adjustments to the general algorithm allow accurate prediction of a patient's risk for each cardiovascular disease component -- coronary disease, cerebrovascular disease, peripheral arterial disease, and heart failure.
So said Ralph B. D'Agostino, Sr., Ph.D., of Boston University, and colleagues online in Circulation: Journal of the American Heart Association, in a Framingham study-based report slated for the Feb. 12 print issue.
Traditional cardiovascular disease risk factors proved to be highly significant predictors of cardiovascular disease risk (P<0.0001), and the general algorithm demonstrated good discrimination for men and women, they found.
"Individuals with a high overall cardiovascular disease risk require more aggressive risk factor modification," said Dr. D'Agostino and colleagues. "The goal of therapy for cholesterol disorders, diabetes, and hypertension should be linked to the global cardiovascular disease risk."
Use of validated cardiovascular disease risk prediction algorithms has lagged in primary care, in part because of the sheer number of algorithms, each targeted toward predicting a patient's risk for an individual cardiovascular disease event, the authors said.
So they set out to develop a single multivariable risk assessment tool that would allow physicians to identify patients with a high risk for any type of cardiovascular disease event, using measurements that are readily available in an office or clinic.
The resulting algorithm was derived from data on 8,491 Framingham study participants followed for 12 years. The patients were 30 to 74 years old at their initial examination and were free of cardiovascular disease.
Investigators developed sex-specific multivariable risk functions that incorporated several readily assessable cardiovascular disease risk factors:
Patient age
Total cholesterol and HDL
Systolic blood pressure
Treatment for hypertension
Smoking status
Diabetes status
During follow-up, 1,174 participants developed a first cardiovascular disease event, which translated into an occurrence rate of 10.08% for women and 18.09% for men. For predicting global cardiovascular disease risk, the algorithm demonstrated good accuracy for men and women, as reflected by c-statistics of 0.762 and 0.793, respectively. In contrast, an earlier Framingham risk algorithm yielded lower c-statistics of 0.756 for men and 0.778 for women.
Comparison of the two algorithms resulted in a "net reclassification improvement from using the new model that was statistically significant for men [P<0.001] and women [P=0.003]," the authors said.
To determine a patient's risk for individual cardiovascular disease events, the global risk score was calculated and then multiplied by the proportion of total cardiovascular disease events represented by the specific event. For example, a woman's risk for coronary heart disease events was determined by multiplying her global risk score by 0.61, the proportion of first cardiovascular disease events that were coronary heart disease events.
Discussing the clinical implications and potential utility of the algorithm, the authors noted that "a multivariable risk formulation for global cardiovascular disease made up of standard risk factors is particularly relevant for primary prevention of atherosclerotic cardiovascular disease because it is intuitive that measures taken to prevent any one cardiovascular disease outcome can be expected to also prevent risk of the other cardiovascular disease outcomes."
"Therefore, use of a general cardiovascular disease risk score is an attractive option in office-based primary care practices," they continued. "Serial assessment of global cardiovascular disease risk could be used to monitor progress of patients on treatment and improvement in their multivariable risk scores."
They also pointed out that "other risk factors not included in the general risk profile must be taken into account in evaluating risk and selecting the most efficacious treatment. These include abdominal obesity, ECG evidence of left ventricular hypertrophy, indications of insulin resistance, triglycerides, and a strong family history of premature cardiovascular disease. Obesity is not included because its influence is largely attributable to its promotion of insulin resistance and its attendant cardiovascular disease risk factors."
In acknowledging limitations of the study, the authors noted that the algorithm did not include all cardiovascular risk factors. They also acknowledged that the algorithm requires validation in other populations.
Dr. D'Agostino has served as a consultant or adviser for Sanofi and Pfizer.
Primary source: CirculationSource reference:D'Agostino RB, et al "General cardiovascular risk profile for use in primary care: the Framingham heart study" Circulation 2008; 117: DOI: 10.1161/CIRCULATIONAHA.107.699579.