Web edition: September 24, 2007
Nutritionists call them carbohydrates. To most of us, they're simply sugars and starches. And although the fructose in soft drinks and the refined flour in white bread taste quite different, "nutritionally and metabolically they're the same as table sugar," explains endocrinologist David S. Ludwig. That's because the body digests all carbohydrate-rich foods into glucose, or blood sugar.
However, all carbs don't break down at the same rate. The body digests those in many whole-grain products quite slowly. Others become converted to glucose almost immediately.
Rapidly digested carbs aren't healthy for people with diabetes and others watching their blood sugar. A new study by Ludwig and his colleagues at Children's Hospital Boston suggests that such carbs are also problematic for people looking to shed body fat. Indeed, the findings indicate that consumption of the wrong carbs can spur the development of body fat, even with no gain in weight.
In the study, mice that chowed down on a type of rapidly digestible starch didn't gain any more weight than did animals eating a starch that digests slowly. But the first group did accumulate lots of excess fat. The data indicate that something about rapidly digesting carbs signaled the body to convert more of a meal's energy into body fat, into fatty lipids that circulate in blood, and into deposits of fat throughout the liver.
Ludwig considers the observed effect on the animals' livers the most troubling one. Fatty-liver disease has traditionally been regarded as the first stage of damage from alcoholism that can progress to hepatitis, cirrhosis, and death. But researchers in recent years have discerned the beginnings of an epidemic of fatty-liver disease unrelated to alcoholism but correlated strongly with being overweight. Recent data suggest that as much as one-third of children and even a higher proportion of adults have the condition. Ludwig told Science News Online that he suspects that "up to half of the [U.S.] population" now has fatty-liver disease.
The question has been what's fueling this epidemic. Because the disease so often accompanies obesity, many researchers have suspected that high-fat diets and junk foods are responsible. Ludwig's group had another idea.
In recent years, the mushrooming incidence of obesity in the United States has led to a push to get people to lower their intakes of fat. However, reducing fat consumption almost always translates into increasing the intake of carbs (see Counting Carbs). Moreover, the carbs most people reach for first are the refinedeasy to digesttypes found in white flour, white rice, pasta, and potatoes.
Ludwig's team decided to see whether a diet rich in a similar carb promotes fat buildup. They used a proportion of carbs that people on a low-fat diet might eat and compared its effects with that of a diet equal in all respects except that its carbs were mainly a slowly digested starch.
In the September Obesity, the researchers show that animals eating rapidly digested carbs accumulated more fat throughout their bodiesincluding their liversthan did animals eating primarily the slow-to-digest starch.
Says Ludwig, "This is the first study in which a single dietary factorvaried within normal rangesaffected whether the liver remained normal or accumulated seriously elevated levels of fat."
Recipe for pudge
In the new study, Ludwig's team fed 18 recently weaned mice food pellets containing 13 percent fat, 19 percent protein, and 68 percent carbohydrates from corn starch. Half the animals got pellets containing the starch called amylopectin, which is made up of a string of glucose molecules that the gut easily degrades into sugar. The remaining mice ate pellets containing some amylopectin but mostly the starch called amylose. That type of corn starch resists breakdown in the gut.
All the animals ate and drank as much as they wanted for 25 weeks. Throughout the study, the researchers charted weight gain, body fat, fecal excretion of starch, and blood concentrations of glucose and insulin. At the end, the researchers killed the animals and measured their livers' fat contents.
Weight gain didn't differ between the two groups of animals, suggesting that the mice found the diets comparably palatable. However, the animals' bodies responded differently to the two food-pellet recipes. Mice dining on amylopectin-enriched chow became twice as fat as those eating the slower-digested amylose recipe. Mice eating this starch grew a little longer in body, so they looked leaner that the "roly-poly" mice eating easily digested starch, Ludwig says. The latter mice "felt squishy," whereas the slow-digested-starch eaters felt firm, he adds.
Although blood sugar concentrations didn't differ between the two groups, mice on the amylopectin-rich food developed higher insulin values after a meal. The body uses the hormone to shepherd energy into its cells. Higher blood insulin after a meal, Ludwig explains, indicates that an animal needs more insulin to fully use the food it's eaten. Needing more of the hormone can be a first sign of insulin resistance and impending diabetes.
Ludwig notes, "Insulin is a powerful anabolic hormone, meaning it promotes the storage of fat. In fact, that's arguably one of [the hormone's] main roles." One of the first places newly made insulin ends up is in the liver, where it can trigger the localized creation and stockpiling of fat.
Although the rodents' livers weighed the same whether they ate fast- or slow-digested starch, fat made up 12 percent of the liver in mice fed the amylopectin-rich diet. That's double the fat content of livers in animals that had eaten the slow-digested starch. For perspective, Ludwig notes, people whose livers contain 10 percent fat are considered to be suffering from "advanced" nonalcoholic fatty-liver disease.
What about people?
This isn't the first study to indicate that foods that rapidly break down to glucose in the bodycharacterized as having a "high glycemic index" (see The New GI Tracts)can fuel nonalcoholic fatty-liver disease. For instance, last year Silvia Valtueña of the University of Parma in Italy and her colleagues reported findings from a study of 247 apparently healthy men and women. The volunteers' diets were evaluated and given a glycemic-index (GI) rating.
Low GI foods included corn, dairy products, and fruit. High GI fare included bread, pizza, and baked snacks. The volunteers were grouped into four categories based on the ascending GI rankings of their diets.
Participants with the highest-GI diets were twice as likely to have undiagnosed fatty-liver disease as were other study participants. People in the highest group were also far likelier to be insulin resistant, the researcher reported in the July 2006 American Journal of Clinical Nutrition.
In an editorial accompanying the Valtueña report, David J.A. Jenkins and his colleagues at the University of Toronto argued that the "implication of this study is that a low-GI diet, or selection of lower-GI rather than higher-GI foods, may benefit persons with nonalcoholic fatty liver." Indeed, the commenters suggested, it might be possible for doctors to treat nonalcoholic fatty liver by lowering the glycemic index of an individuals' diets.
That's what Ludwig's group is now investigating: "We hope to enroll 46 kids to a diet for 6 months," he says. The 8- to 17-year-olds and their parents will receive dietary counseling. Half of the recruits will be assigned to a low-fat diet. The rest will receive counseling to lower the glycemic index of their diets. The general guidelines for a low-GI diet call for substituting whole-grain foods for ones made from highly processed cereal fibers and reducing refined sugars in favor of sweet fruits.
"Conceptually," Ludwig says, "fatty liver should be reversiblewe've seen it anecdotally in practice many times, such as when someone loses weight or changes the quality of their diet."
If you would like to comment on this Food for Thought, please see the blog version.
David J.A. Jenkins
Clinical Nutrition and Risk Factor Modification Center
St. Michael's Hospital
61 Queen Street, East
Toronto, ON M5C 3E2
David S. Ludwig
Children's Hospital Boston
Department of Medicine
333 Longwood Avenue
Boston, MA 02115
Department of Internal Medicine and Biomedical Sciences
University of Parma
Pawlak, D.B., et al. 2001. High glycemic index starch promotes hypersecretion of insulin and higher body fat in rats without affecting insulin sensitivity. Journal of Nutrition 131(January):99-104. Available at [Go to].
Raloff, J. 2007. Super-size miceFast food hurts rodents. Science News Online (June 9). Available at [Go to].
______. 2004. Counting carbs. Science News 166(July 17):40-42. Available at [Go to].
______. 2004. Coming soonSpud lite. Science News Online (June 19). Available at [Go to].
______. 2000. The new GI tracts. Science News 157(April 8):236-238. Available at [Go to].
Sloth, I., et al. 2004. No difference in body weight decrease between a low-glycemic-index and a high-glycemic-index diet but reduced LDL cholesterol after 10-wk ad libitum intake of the low-glycemic-index diet. American Journal of Clinical Nutrition 80(Aug. 1):337-347. Available at [Go to].