Why cutting fats may harm the heart

 Since dietary fats can play a role in boosting blood-cholesterol concentrations and heart-disease risk, does it make sense to short-circuit the process by dramatically paring fats from the diet?

 For most men, at least, the answer may prove an emphatic no, according to new research conducted at Lawrence Berkeley (Calif.) National Laboratory (LBNL).

 In most people, genetic programming directs the body to respond to drops in total dietary fat by repackaging fatty particles called lipids, the study shows. In general, these changes tend to increase risk of cardiovascular disease, notes Ronald M. Krauss, who led the research.

 In a pair of earlier studies, his team directed 238 healthy men to spend periods of 3 to 6 weeks eating each of two diets: one deriving at least 40 percent of its calories from fats, and the other with just 20 to 24 percent fat.

 While on the high-fat diet, one-quarter of the volunteers exhibited a worrisome blood-lipid profile. Their cholesterol-shuttling, low-density lipoproteins (LDLs)-the so-called bad lipoproteins-were unusually small and dense. Such small LDLs have been linked to an especially high risk of heart disease.

 Even more troubling, when the men with normal-size LDLs ate the low-fat diet, a third of them began making the small, dense LDLs. With this change came an increase in blood concentrations of triglycerides, another heart-disease high-risk factor, and a drop in the cholesterol exiting their blood via high-density lipoproteins (HDLs), the good lipoproteins.

 In the most recent work, the LBNL researchers selected 38 men who had proven resistant to lipid changes in the previous studies and put them on an even lower-fat diet. For 10 days, they ate menus deriving just 10 percent of their calories from fat. Once again, a third began making the especially small LDLs, Krauss's group reports in the March American Journal of Clinical Nutrition.

 These men also increased their triglycerides and decreased HDLs-especially HDL2B, which is considered especially beneficial.

 In contrast to the earlier two studies, the men whose lipid profiles changed in this test also exhibited an increase in blood concentrations of intermediate-density lipoproteins (IDLs). Though most physicians don't measure IDLs, studies that assayed them generally found high levels "to be the indicator most strongly connected to heart-disease risk," Krauss notes. "It trumps all other standard [lipid] measurements in predicting risk."

 Overall, the LBNL data indicate that at least two-thirds of men carry genes for these lipid disturbances, though in most cases the effects only show up on a low-fat diet. Krauss suspects that the benefits of exercise and low calorie intake may compensate for the increased risk imparted by the adverse lipid profile.

 The data may underestimate the genetic predisposition for lipid change, Krauss contends, since all the volunteers were of normal weight. With obesity, he says, "there's an even higher prevalence of these lipid disturbances to start with."

 In fact, he worries that the overweight couch potato-precisely the person physicians often target for a low-fat diet-may face the greatest risk of expressing these lipid disturbances. Certainly, Krauss says, his data argue strongly against a one-size-fits-all approach to advice on fat consumption.

 These data "support what a number of us in the nutrition community have been saying-that low-fat diets are not the way to go [to prevent heart disease]," says Meir J. Stampfer of the Harvard School of Public Health in Boston.

"The nutrition community has pushed this notion that fat equals bad," Stampfer says, "without making distinctions between types of fat." In fact, he explains, because it's saturated fats and trans fats that are bad for the heart, "our message to the public should be to replace them with polyunsaturated and monounsaturated fats"-such as the types in nuts.