Dr H Robert Superko

The results may explain inconsistencies in previous studies, which have produced mixed results as to whether coffee might increase cholesterol levels. Superko believes inconsistent reports in earlier research, including analyses from the Atherosclerosis Risk in Communities, Harvard Physicians Health Study, and Framingham Heart Study, did not distinguish between regular and decaffeinated coffee.

"From our study, we can unequivocally say, caffeinated coffee does not raise cholesterol. We can also unequivocally say that decaffeinated coffee, at least the one we used in this study, does raise cholesterol," Superko told members of the press.

For their cholesterol-effects study, Superko et al chose coffees produced by "the second-largest coffee maker in the US," giving 187 study participants (all of whom were regular coffee drinkers) coffee grounds in blinded bags with specific brewing instructions. After a washout period, study participants were randomized to a decaf, caffeinated, or no-coffee arm for an eight-week period, during which time the coffee-drinkers drank three to five cups of black coffee, without sugar or sweeteners.

Over the eight-week period, the investigators saw no significant changes in triglycerides, HDL, glucose, or insulin, but they did see significant increases in LDL in the decaf drinkers compared with both the caffeinated coffee and no-coffee groups. Levels of ApoB, a protein attached to LDL and believed to be a more accurate predictor of LDL levels, also increased significantly in the decaffeinated-coffee drinkers. Of note, nonesterfied fatty acids (NEFA) also rose in the decaf drinkers compared with the caffeinated and no-coffee groups, although the increase just missed statistical significance.

In an interview with heartwire, Superko speculated that the NEFA increases reflect the higher diterpene content of decaffeinated coffee beans. Most decaffeinated coffee beans, he said, are part of the Robusta species of coffee bean, a bitter bean made palatable during the decaffeination process, he explained. Caffeinated coffee beans, by contrast, are normally from the milder Arabica species, which contain fewer diterpenes.

Superko was careful to emphasize that the findings from his group may or may not extend to all types of coffee and that different decaffeination processes may affect decaf coffee's effects on LDL.

"I think what you can accurately say is that the bean and decaffeination process used by the second-largest coffee maker in the US causes this effect. What I can't say is that if you get Starbuck's decaffeinated Somalia brand you get the same thing. We did that on purpose because there are so many types of coffee; we wanted to pick one that's used by a lot of people."

For physicians encouraging patients to adopt a "lifestyle approach" to either cholesterol or blood-pressure control, "these results could be relevant," Superko said. "They are also going to tell their patients to exercise, lose weight, eat fiber, etc. Each of these things has about a 5% effect, but if you do three, four, or five of them together and suddenly you're down 20% to 30%, that can be really significant."

Still, he stressed, the findings must be kept in perspective. "I don't think there's a health threat to coffee—that's where somebody might overinterpret this," Superko cautioned. "The interesting thing is that there is clearly a caffeinated/decaffeinated difference. One related to cholesterol, one to blood pressure, and it appears to be the fatty acids that drive the lipid effects."

Those who enjoy drinking coffee, he adds, "[should continue to] enjoy drinking coffee. However, if you're concerned about your cholesterol, then look to the decaffeinated coffee as a possible contributor to your high cholesterol. If you are concerned about your blood pressure, then look to your caffeinated coffee as a possible contributor to your high blood pressure."

And, Superko quipped, for those concerned about both: "There's always green tea."