Good Calories, Bad Calories

in 1975 by Duke University pediatrician James Sidbury, Jr., in the treatment of obese children.

Another common explanation for the absence of hunger on carbohydrate-restricted diets is that fat and protein are particularly satiating—“these foods digest slowly, making you feel satisfied longer,” as Brody has explained in the Times. (Even those investigators who published studies supporting Yudkin’s idea that carbohydrate-restricted diets work by restricting calories would invariably comment that high-protein, high-fat diets still induced the least hunger and the greatest feeling of satiation. “There is a good reason to believe that the satiety value of such diets is superior to diets high in carbohydrate and low in fat, and hence, may be associated with better dietary adherence,” the metabolism researcher Laurance Kinsell wrote in an influential 1964 article entitled “Calories Do Count.”) But this is also unsatisfying as an explanation. The statement that fat and protein satisfy us longer is equivalent to the statement that carbohydrates are less satisfying—they either make us experience hunger sooner than fat and protein or perhaps induce hunger, whereas fat and protein suppress it. This leads us back to the now familiar question: what is it about carbohydrates, or about the speed with which we digest them, that accelerates or exacerbates our sensation of hunger and our desire to eat?

Even Yudkin had struggled with the question of why people would willingly semi-starve themselves on a carbohydrate-restricted diet. “For reasons I do not clearly understand,” he wrote, there must be something unique about carbohydrates that either stimulates our appetites or fails to satiate us. “It would seem from this that carbohydrate does not satisfy the appetite,” he noted; “it may even increase it….”

This conclusion is simply hard to avoid, considering the half century of experimental observations on these diets. It leaves us with two seemingly paradoxical observations. The first is that weight loss can be largely independent of calories. The second is that hunger can also be. Even if we could establish that weight loss on these diets is universally attended by a decrease in calories consumed—no bread, no butter—we then have to explain why the subjects of these diets don’t manifest the symptoms of semi-starvation. If they eat less on the diets, why aren’t they hungry? And if they don’t eat less, why do they lose weight?

“It is better to know nothing,” wrote Claude Bernard in An Introduction to the Study of Experimental Medicine, “than to keep in mind fixed ideas based on theories whose confirmation we constantly seek, neglecting meanwhile everything that fails to agree with them.” In the study of human obesity, that fixed idea has been what Yudkin called “the inevitability of calories,” which in turn is based on the ubiquitous misconception of the law of energy conservation. If we believe that conservation of energy—calories in equal calories out—implies cause and effect, then we will refuse to believe that obese patients can lose significant weight without restricting their energy intake beneath some minimal expenditure. Any reports to the contrary will be rejected on the basis that they cannot possibly be true. “Claims that weight loss occurs even with high-caloric intake, but no carbohydrate, are absurd,” as the American Medical Association insisted in 1974. “Although authors of popular diet books frequently say that loss of body fat can occur regardless of high-calorie intake, this is not supported by evidence and, in fact, is refuted by the laws of thermodynamics.”

Because such a possibility is not refuted by the laws of thermodynamics, we should take such claims seriously, as Alfred Pennington did. Although several of Pennington’s articles appeared in journals that were widely read, including The New England Journal of Medicine, they would have little influence on the thinking about obesity. A few practicing physicians took his work seriously—George Thorpe and Herman Taller, a Brooklyn obstetrician who published a 1961 best-seller based on Pennington’s science entitled Calories Don’t Count—but they only lost professional credibility by doing so. The great majority of clinicians and nutritionists would not go against the conventional wisdom.

Nonetheless, Pennington was on to something. He set out to understand why his DuPont patients lost weight on a calorie-unrestricted diet that they enjoyed. He knew it contradicted the conventional wisdom but was determined to pursue the evidence. First he read what he called the “voluminous experimental literature on obesity.” He concluded that only “meager and conflicting” evidence existed to support the popular contention that calorie restriction would induce long-term weight loss, or even that it should induce long-term weight loss. He came to believe that experts who invoked the first law of thermodynamics to defend their beliefs did great damage. “These tended to distract the general attention from examination of the evidence on the real question, whether or not common obesity arises from a metabolic defect,” he wrote.

Pennington based his analysis of the obesity problem on one fundamental premise that he adopted from the research on homeostasis in the 1930s and early 1940s: Because fuel is ultimately used by the cells themselves, the relationship between fuel supply and demand at this cellular level determines both hunger and energy expenditure. The less fuel available to supply the metabolic demands of our cells, the greater the hunger and the less energy we will expend. The greater the fuel available to the cells, the greater the metabolic activity and perhaps physical activity also. This was something Francis Benedict had suggested in the 1920s and Eugene Du Bois believed. Energy expenditure, wrote Pennington, is an “index of calorie nutrition at the cellular level.”

Pennington considered two facts about obesity to be particularly revealing. One was Hugo Rony’s observation that an obese individual will spend much of his life in energy balance—in the “static phase” of obesity, to use Rony’s term—just as the lean do. “His caloric intake, like that of people of normal weight, is dictated by the energy needs of his body,” Pennington wrote. “His appetite, far from being uncontrolled, is precisely and delicately regulated.”

The second fact was that when obese individuals try consciously to eat less—when they go on a low-calorie diet—their metabolism and energy expenditure inevitably decrease, just as they do when lean individuals are semi- starved. Benedict had observed this diet-induced decrease in energy expenditure in his lean subjects in his 1917–18 semi-starvation studies. Frank Evans and Margaret Ohlson had made the same observation of the obese. Pennington believed, as Benedict, the Cornell nutritionist Graham Lusk, and others had suggested, that this was the natural response to a diminished supply of energy. Less energy is available to the cells, and so they expend less. On a calorie-restricted diet, Pennington suggested, the obese and the lean become hungry and lethargic for identical reasons—“their tissues are not receiving enough nutriment.”

This presented a dilemma. That the tissues of the lean are semi-starved by calorie restriction is easy to imagine; they don’t have a lot of excess calories to spare. But why would this happen with the obese, who do? Pennington found his answer in a 1943 article by the Columbia University biochemist DeWitt Stetten, who reported that the rate at which fatty acids were released from the fat deposits of congenitally obese mice was significantly slower than it was in lean mice. Stetten had suggested that obesity in these animals was caused by a suppression of the flow of fat from the adipose tissue back into the circulation and its subsequent use by the tissues for fuel.

Pennington proposed that the same thing causes obesity in humans. The adipose tissue amasses fat calories in a normal manner after meals, but it doesn’t release those calories fast enough, for whatever reason, to satisfy the needs of the cells between meals. This was the metabolic defect that causes obesity, he said, and it could apparently be corrected or minimized by removing carbohydrates from the diet.

By hypothesizing the existence of such a defect, Pennington was able to explain the entire spectrum of observations about obesity in humans and animals simply by applying the same law of energy conservation that other obesity researchers had misinterpreted. The law applies to the fat tissue, Pennington noted, just as it does to the entire human body. If energy goes into the fat tissue faster than it comes out, the energy stored in the fat tissue has to increase. Any metabolic phenomenon that slows down the release of fat from the fat tissue—that retards the “energy out” variable of the equation—will have this effect, as long as the rate at which fat enters the adipose tissue (the energy in) remains unchanged, or at least does not decrease by an equal or a greater amount.