ORLANDO — The various measures of a complete blood count together helped in the development of a risk score that was very effective for predicting a patient's subsequent risk of death or myocardial infarction.
“This relatively inexpensive and easily obtained test has important but unutilized risk information,” Jeffrey L. Anderson, M.D., said at the annual meeting of the American College of Cardiology. “It adds substantially to a predictive model” based on standard risk factors, said Dr. Anderson, associate chief of cardiology at LDS Hospital in Salt Lake City. “Complete blood counts [CBCs] could provide risk stratification and add to treatment decision making at a negligible incremental cost.”
The model was developed using the 19,044 patients who underwent coronary angiography at LDS Hospital during 1993–2003. During an average follow-up of 4 years, these patients had a total of 3,124 deaths and 4,672 nonfatal MIs, an overall event rate of 41%. “This was clearly a high-risk population,” Dr. Anderson said. At baseline, 56% of patients had stable angina, 23% had unstable angina, and 21% had an acute MI. Angiography at baseline showed severe coronary stenosis (at least 70%) in 61% of patients, mild to moderate stenosis in 8%, and no stenosis in 31%.
Dr. Anderson and his associates developed a risk model that used seven measures from the CBC: hematocrit, red-cell distribution width, mean corpuscular-hemoglobin concentration, mean corpuscular volume, platelet count, white cell count, and white-cell differential count. The model also included age, gender, and seven classic coronary disease risk factors such as hyperlipidemia, hypertension, diabetes, and smoking. All of these factors were combined into a scoring formula that rated patient risk on a scale of 0–10. Other measures from the CBC, such as hemoglobin and mean corpuscular hemoglobin, were excluded from the final model because they added no additional risk stratification.
During follow-up, patients with a baseline score of 0 had a 12% incidence of death or nonfatal MI; patients with a baseline score of 10 had a subsequent 65% event rate. These extremes show a “substantial differential in risk” by combining the blood count parameters with standard risk factors, Dr. Anderson said.