Kathleen L. Wyne, MD, PhD, is an assistant professor, Division of Endocrinology and Metabolism, University of Texas Southwestern Medical Center, Dallas.

The risk calculator reported in the Lee et al paper was developed to predict the 10-year cardiovascular (CV) disease probability in an American Indian population. The calculator grew out of the Strong Heart Study, an ongoing longitudinal study of CV disease in American Indians that was initiated in 1988. (The calculator is accessible at strongheart.ouhsc.edu/CHDcalculator/calculator.html.)

There has been speculation that the risk calculator derived from the Framingham population underestimates the CV risk in other populations. The work presented in the present paper clearly shows the need for population-specific risk calculators. The equations used to model the risk indicated that the factors listed in the Table each contributed significantly to CV risk in this population.

The surprising result of this study was the strong influence of microalbuminuria and macroalbuminuria, even in the absence of diabetes. The presence of microalbuminuria is an established CV risk factor, although it has not been included in the Framingham Risk calculator or the National Cholesterol Education Program (NCEP), Adult Treatment Panel III definition for the Metabolic Syndrome. Some have proposed checking for microalbuminuria in all patients with hypertension (but without diabetes) as a standard of care. However, such widespread screening has not yet permeated into daily clinical practice. Given the clinical significance of albuminuria, it is not such a surprise that it has emerged as a strong marker of CV risk. The presence of albuminuria, even just microalbuminuria, indicates that the person already has endothelial damage. And if there is endothelial damage in the kidney, then there is likely also endothelial damage in other organs, including the heart. Given that microalbuminuria is a marker of endothelial damage, then perhaps physicians should consider checking for microalbuminuria in patients thought to be at risk for CV disease.

Several concerns pertaining to the broader applicability of this study, however, must be raised. First, these equations were developed from data derived from a population of American Indian men and women age 45 to 74 years. Further studies are needed to determine whether these calculations are valid for anyone who does not meet this specific profile. Second, this risk calculator needs to be disseminated to all practitioners for use in their patients who have American Indian ancestry, not just to those who practice in the American Indian health system. Given that these patients have a risk for CV disease greater than that of the general population, this risk calculator can be used to heighten awareness and intensify therapy, possibly preventing the devastating complications associated with the diabetes and CV disease that has become an epidemic in this population.