Heart disease worries? Watch the decaf.

Several studies have suggested a link between heavy coffee drinking and heart disease. Though most did not differentiate between coffee types or brewing methods, scientists suspected that any adverse effect must trace to caffeine. Now, researchers report that decaffeinated coffee--but no regular--may nudge cholesterol levels in the direction of increased heart risk.

The 16-week study, directed by H. Robert Superko of the University of California's Center for Progressive Atherosclerosis Management in Berkeley, involved 181 healthy, nonsmoking men who routinely drank three to six cups of coffee per day. The researchers provided all volunteers with regular, drip-grind coffee and instructed them on how to brew it. Eight weeks later, they randomly assigned each man to one of three regimens: the same coffee, a switch to decaf, or abstinence from coffee. Participants were asked to avoid other caffeine sources throughout the study.

Those who drank regular coffee and those who abstained showed no changes in blood cholesterol levels during the study, the team reports in the September AMERICAN JOURNAL OF CLINICAL NUTRITION. The decaf group, however, experienced a roughly 6 percent increase in low-density lipoprotein (LDL) cholesterol, the so-called "bad" cholesterol linked to heart attacks.

Using the general rule that a 1 percent rise in total cholesterol boosts the risk of heart disease by 2 percent, the researchers conclude that the LDL changes in their study "may increase coronary artery disease risk by around 10 percent." This could have important implications, says Superko, who notes that decaf represents some 20 percent of the 139 billion cups of coffee downed each year in the United States.

If caffeine isn't responsible for coffee's cholesterol effects, what is? "We think there's one compound in the bean that's causing this effect on a molecular level," Superko told SCIENCE NEWS. The culprit compound may occur only in certain types of beans, he suggests.

Other reports have indicated that caffeinated coffees usually come from arabica beans, whereas most decaffeinated blends rely on robusta beans. Superko is now comparing the chemistry of the two brews used in his study, searching for differences that might explain the LDL increase.