Framingham, MA - The Framingham CVD risk-prediction tool is inadequate for predicting CV events in people with chronic kidney disease (CKD) or end-stage renal disease (ESRD), two studies suggest [1,2]. Researchers for the studies say there is an urgent need for clinical trials enrolling patients with kidney disease to better understand what risk factors can accurately predict future CVD in this group of patients.
The studies were first presented during the American Society of Nephrology Renal Week 2006 in November but have been rereleased to coincide with World Kidney Day, March 8, 2007. In one, Dr Daniel E Weiner (Tufts-New England Medical Center, Boston, MA) and colleagues looked at 934 patients participating in the Atherosclerosis Risk in Communities Study and the Cardiovascular Health Study over a 10-year period. All patients had no preexisting coronary disease but had an estimated glomerular filtration rate (GFR) of 15 to 60 mL/min per 1.73 m2. Baseline Framingham risk factors were used to calculate their 10-year CVD risk.
Over the 10-year period, almost 10% of women and 20% of men had an initial cardiac event, yet the Framingham equations underestimated patient risk by as much as 50%.
"The most common cause of death among patients with chronic kidney disease is cardiovascular disease, and the number of Americans with chronic kidney disease is expected to increase within the next decade," senior author on the study, Dr Mark Sarnak, commented in a press release. "We need to move quickly to identify more effective heart-disease screening and treatment methods in order to monitor and preserve the long-term health of this susceptible population."

Prediction tool falls short in ESRD
In a second study, Dr Darshan S Dalal (Johns Hopkins University School of Medicine, Baltimore, MD) and colleagues examined the utility of the Framingham risk equation in 4229 patients with ESRD but no prior cardiovascular disease. Data was drawn from the Dialysis Morbidity and Mortality Study (DMMS) and the United States Renal Data System (USRDS). On the basis of baseline Framingham risk factors, the predicted risk of CVD was around 3% for women and around 2% for men and increased to roughly 6% for men and 12% for women at four years. However, over a mean follow-up of 24 months, the observed risk of CVD was considerably higher, perhaps due nontraditional risk factors in ESRD patients.

"Further studies are needed to develop risk equations that better estimate the risk for CVD in ESRD and can be used in clinical settings," the authors conclude.

Findings no surprise
Commenting on the studies for heartwire, a medical officer for the Framingham Heart Study, Dr Caroline S Fox (National Heart, Lung, and Blood Institute, Framingham, MA), said it is "not surprising" that the Framingham risk score underestimates coronary heart disease risk in patients with chronic kidney disease.
"The Framingham risk score was not developed for this patient population, which has a high burden of traditional and nontraditional CVD risk factors, as well as a higher overall CHD event rate," she said. "In addition, a substantial proportion of patients with CKD have diabetes, and the Framingham risk score has been shown to underestimate CHD risk in patients with diabetes, which may be a potential explanation for these findings."
In ESRD patients in particular, she added, "traditional risk-factor relationships with CVD outcomes can be obscured due in part to the high degree of comorbidity, concomitant inflammation, and malnutrition."

Sources
Weiner DE, Tighiouart H, Griffith JL, et al. The Framingham predictive instrument in chronic kidney disease. Renal Week 2006: ASN Annual Meeting; November 14-19, 2006; San Diego, CA. Abstract TH-PO393.
Dalal DS, Guallar E, Zhang L, et al. The Framingham risk score is not useful for cardiovascular risk prediction in end-stage renal disease. Renal Week 2006: ASN Annual Meeting; November 14-19, 2006; San Diego, CA. Abstract F-FC045.