Role of the Metabolic Syndrome in Risk Assessment for Coronary Heart Disease

Commentary

In the mid-1970s, the question was reasonably posed as to whether to treat risk factors for coronary heart disease (CHD) at all.¹ In the last decade, however, with dramatic changes in the knowledge base regarding prevention and treatment of atherosclerosis and CHD, pressing questions have shifted to when to treat, in whom to treat, and how to treat patients to minimize the risks associated with development of atherosclerosis and related CHD. Evidence now conclusively shows that risk factor prevention as well as treatment for risk factors reduce the risk of future CHD^{2 - 4} but questions of cost-effectiveness of risk factor reduction still remain. Accordingly, if those at highest risk for future CHD can be identified and classified correctly, they may be able to benefit most from risk reduction therapies such as diet modification, weight loss, exercise, and/or selected medications. Identifying those at highest risk has therefore been the subject of a number of different risk assessment strategies.^{5 - 8} There is a fervent hope for improved assessment methods in the future era of personalized medicine.⁹

Assessing risk of future CHD by considering risk factors individually is a common method of risk assessment. Individual risk factors, such as low-density lipoprotein cholesterol, tobacco use, or blood pressure levels, and their associated risks for CHD have been extensively investigated. There are currently clinical practice guidelines on treating all of the major risk factors with the goal of decreasing CHD risk overall.^{7 ,10 - 11} The National Cholesterol Program Adult Treatment Panel III⁷ and the Joint National Committee¹¹ on Prevention, Evaluation, and Treatment of High Blood Pressure are good examples of these approaches focused mainly on single risk factors.

Global risk assessment is a clinical approach that recognizes that multiple risk factors considered simultaneously can discriminate risk better than individual risk factors alone.⁶ Various approaches to multiple risk factor assessment or global risk have been reported,^{12 - 15} such as the Framingham risk score, the Joint British Societies Risk Chart, the Prospective Cardiovascular Munster score, and the European System for Cardiac Operative Risk Evaluation. The Framingham risk score is the approach that is frequently recommended, especially in the United States.⁵ Global risk assessment tools, such as the Framingham risk score, take into account multiple risk factors to assess the overall risk an individual has for future CHD. The purpose of a global risk assessment tool is to stratify patients with increased risk of CHD to identify those patients who are at even higher risk who might benefit most from the most intensive risk reduction therapies. In addition, multiple risk factor assessments might identify a set of risk factors that are synergistic (ie, individual components considered together confer a greater risk than the individual components taken alone).

The metabolic syndrome has been proposed as a possible global risk assessment tool and was initially described as a cluster of risk factors that might interact to increase risk excessively.¹⁶ In a recent article in the Archives of Internal Medicine, Wannamethee et al¹⁷ evaluated the metabolic syndrome as a global risk assessment tool and attempted to answer the following question: Is metabolic syndrome a better predictor of diabetes and future CHD than the Framingham Risk Score? Using data from the British Regional Heart Study, a prospective study of 5128 men without a history of cardiovascular disease or diabetes, the authors found that men with the metabolic syndrome had a significantly higher risk of developing CHD, stroke, and diabetes than men without the metabolic syndrome. Increased risk persisted after adjusting for other common CHD risk factors such as age and smoking. Risk also increased with the number of metabolic abnormalities. These results confirm similar findings of previous studies.^{18 - 20}

Wannamethee et al also found that the Framingham risk score was a significantly better predictor of CHD than the metabolic syndrome, while the metabolic syndrome was a significantly better predictor of future incidence of diabetes. The Framingham risk score had a more predictive value for CHD than patients with 4 or more metabolic syndrome abnormalities at both 10-year and 20-year durations of follow-up. In addition, the metabolic syndrome did not confer additional predictive value when added to the Framingham risk score in a multivariable model.

Two additional studies have demonstrated similar findings.^{20 - 21} In the San Antonio Heart Study,²⁰ the Framingham risk score had a significantly higher sensitivity for predicting CHD than the metabolic syndrome. In addition, the metabolic syndrome had a higher false-positive rate when used to predict future CHD. In the Atherosclerosis Risk in Communities study,²¹ the metabolic syndrome did not predict CHD better than the Framingham risk score. In both of these studies, the metabolic syndrome was present in about a quarter of the population studied and those with the metabolic syndrome were more likely to have CHD than those without the metabolic syndrome.

In addition to being inferior to the Framingham risk score for risk prediction for CHD events, the metabolic syndrome appears to confer no greater CHD risk as a whole than does the sum of its individual parts. Several studies have demonstrated that even though the metabolic syndrome was prevalent in the population and it predicted an increased risk of CHD events once the individual components of the metabolic syndrome were accounted for, the predictive value of the metabolic syndrome disappeared.^{21 - 25} Specifically, the metabolic syndrome was a significant predictor of CHD in data from the Third National Health and Nutrition Examination Survey by univariate analysis but the metabolic syndrome was no longer additionally predictive of CHD when blood pressure, high-density lipoprotein cholesterol, and diabetes were included in the multivariable analysis.²⁵ These data are consistent with the conclusion that no additional information is contained in considering the combination of factors in the metabolic syndrome compared with considering the component risk factors individually in a typical logistic regression (Framingham-like) model.

Using the metabolic syndrome as a CHD risk assessment tool in the clinical setting poses several other dilemmas. First, diagnosing patients with the metabolic syndrome does not identify more patients with increased risk of CHD than using the Framingham risk score or the individual risk factors included in the metabolic syndrome. Thus, it could be argued that including more than 1 type of risk assessment beyond the Framingham score is unnecessary. Second, the treatment for patients identified using the metabolic syndrome, individual risk factors, or the Framingham risk score is the same.²⁶ The metabolic syndrome does not yet have a defined etiology that can be targeted by a specific treatment. Treatments recommended by the National Cholesterol Program Adult Treatment Panel III and treatments for the metabolic syndrome are the same as those recommended for the individual risk factors—ie, cholesterol-lowering therapy, weight loss, increased physical activity, and diet modification. Third, there are no studies examining patient motivation to adhere to risk reduction therapies once diagnosed with the metabolic syndrome.

Based on the evidence to date, including the study by Wannamethee et al,¹⁷ the metabolic syndrome as a clinical entity appears to add little or nothing to the prediction of future CHD or to the primary prevention of CHD compared with the Framingham risk score or other methods. Data such as these have led the American Diabetes Association²⁷ and other clinical groups to raise questions about the utility of the metabolic syndrome. One of the initial goals of the metabolic syndrome, improved risk prediction of CHD, has proved disappointing. Further research may be warranted to determine if modifications to the current definition of the metabolic syndrome with the addition of CHD risk parameters will improve its predictive value. The second goal of the metabolic syndrome is to identify a cluster of CHD risk factors that confer increased risk when taken together now seems proven untrue.