Sweeteners: Issues and Uncertainties (1975)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

THE QUESTIONS OF BENEFITS AND RISKS Herman F. Kraybill A great part of this appraisal of the benefits and risks related to the use of sugar will consist of data selected from a wide range of sources, around which I hope to weave a narrative that will present a balanced account. It is my opinion that as biomedical scientists we have become far too engrossed with the risk side of our observations and considera- tions. Therefore, I would like to begin with the reward features from the ingestion of sucrose, as outlined in Table l. Table 2 details the various species that respond to sweet sensation, pointing to exceptions and responders. As to the economics of sweets (Table 3), poorer families apparently use twice as much sucrose per person as higher income groups. The higher income groups supposedly get their sucrose more in the form of desserts and prepared types of foods, rather than sucrose per se. In the Middle Ages, sugar was a very costly item, and it took about a week's salary to purchase a pound of sugar; and that is why it appar- ently was not used. Then the technology of the l800s advanced so much that sucrose presumably became cheaper and more widespread. One could debate that pattern as of l974 and l975. But it is clear that, as the standard of living increased, more sucrose became available. The next data document world and U.S. consumption and production of sugar. Table 4 lists production in tons of both cane sugar and beet sugar and shows the gradual rise that occurred from l958 to l973. Table 5 translates sugar consumption of the United States into pounds per person per year for a 25-year period. As members of the Forum said this morning, one has to keep in mind all types of sugar, so this would modify these figures perhaps a little. Table 6 shows the relative pro- duction of various countries. The United States moves from fifth to 59

60 fourth rank, but the production of Hawaii is included in its total ton production. Relative price listings of sugar (Table 7) go only up to l972. How- ever, it will come as a surprise to no one that there has been a mete- oric rise in the price of sugar since that time. Table 8 is a review of the general benefits of absorbable sugars as I would see it. We do increase the gross national product one way or the other. The sugar industry is an important one, particularly in Hawaii and sections of the South. It provides employment. There is a reward feature. Although it does provide quick energy, that can be either good or bad, depending on what situation you are describing. Sucrose ranks rather high in caloric density compared to the other types of sugars. It used to offer low-cost energy, but I think that current prices might invalidate that. There is some question as to whether it is indeed an appetite stimulator. It aids the preparation of certain foods in certain ways. It is well to take a look at some of the alternatives to sucrose (Table 9) and the possible benefits of sucrose substitutes (Table l0). In regard to appetite stimulation, it has been shown in animal experi- ments that saccharin is as good a stimulator as sucrose or glucose. In weight control, the total caloric value of the diet is important. As it was pointed out earlier, the fact that you take saccharin does not necessarily mean you are going to lose weight or control obesity. The person who thinks he is getting around the problem of obesity by using saccharin or cyclamate and continues to eat pie, cake, ice cream, and everything else, will still put on weight. It is total calories, I think, that is the important problem. I would like to detail some of the effects as to the role of sucrose in the diet that I have gleaned from the literature over the last couple of years. There was an interesting study done at General Foods labora- tories in which sucrose and starch were fed in combination with chromium or alone with chromium. This impinges on the work that Schroeder did earlier. Chromium is a trace metal, and the work that Dr. Mertz is doing at Beltsville has shown that chromium perhaps is tied up with insulin as a glucose tolerance factor. In reviewing Table ll, it is quite marked what effect the chromium has. Indeed, sucrose increases the serum cholesterol levels as compared to starch or chromium alone. There are also some studies that were done by Roberts with soldiers in Antarctica in which the level of sucrose was controlled. Indeed, it was brought almost to zero. An analysis of triglyceride levels (Table l2) shows apparently that for those who had a high level, the effect was marked. This threads through many of the observations that where you have high cholesterol or high triglyceride levels, when the sucrose is lowered markedly in the diet, you can affect the level of the serum triglyceride or the serum cholesterol. Table l3 shows the mean change in the cholesterol levels during the sucrose-free period of the same l8 men in Antarctica, where the high cholesterol group showed a diminution, and with the low cholesterol group it did not have much effect. In other words, for those of us who

6l have normal levels of serum cholesterol or serum triglyceride, appar- ently the effect will not be marked. But if you have the high level, then you may see an effect of lowering the sucrose in the diet. In Table l4 is some data from an epidemiological study by Kessler showing that there was a significantly increased risk of death from pancreatic cancer among diabetics. One may be tempted to conclude that sucrose has a role in pancreatic carcinomas. Indeed, the incidence of pancreatic carcinomas is on the rise, along with others such as colo- rectal and lung cancers. To attribute this all to sucrose as the pre- cursor that is involved in hyperglycemia or diabetes, and to say that this is necessarily associated with pancreatic cancer, I do not know. That could be a moot question. Nevertheless, the Kessler study would indicate that diabetics are a population at higher risk insofar as pan- creatic carcinomas are concerned. Tables l5 and l6 present data on the influence of the dietary carbo- hydrate on the age at death and cause of death in BHE and Wistar rats. We see that longevity was different for those on sucrose than those that were maintained on a corn starch diet or on a glucose. There seemed to be a difference in the disease incidence among these particu- lar rats. The same study was carried out on Wistar rats, another strain. In Table l7 the effect of sucrose on glucose tolerance in the rat is shown. One could debate diet percentages and say that these values are very high, and some of these other studies may reflect this in the lesions, that is, the effect on the kidney and other organs. They did administer a rather severe insult. But the interesting thing to note here is that although levels were high for starch and for sucrose, the effect on the glucose levels in time in milligram percent over l20 minutes was quite different. It leads me to believe this data, with other data, of course, that starch behaves entirely differently. It is a polysaccharide; it is not an absorbable sugar; it takes longer to hydrolyze, and therefore it takes longer to exert its effect metaboli- cally after it is absorbed. Data on the effect of dietary carbohydrates on enzyme activities is presented in Table l8. This gives you a picture of what sucrose, glu- cose, and corn starch will do in terms of certain enzyme systems, show- ing that sucrose apparently magnifies the effect, that it has an accelerating effect on certain liver and blood enzymes when it is administered to the rats. Not all enzyme systems, of course, respond, but some do more than others. Table l9 reveals an old story in l948 data by Sognnaes where he showed that a stock diet plus a purified diet gave you a certain cari- ogenic score. I am just merely amplifying the fact that a purified diet (a semisynthetic diet) with sucrose in it given during the preg- nancy and lactation stage or post-eruptive stage will produce a caries score of 48. In many laboratories today, they elect not to use sucrose or at least a very small percent in the diet, substituting in its place glu- cose or corn starch to get away from the sucrose effect.

62 Here is where we get into a controversial area on the subject of hypoglycemia. The medical community is divided, and so our nutrition- ists and many other scientists. Rachmiel Levine, who is an eminent authority in the area of carbohydrate metabolism and diabetes, said in a recent publication of J.A.M.A. that organic hyperlysenemia is rare, and I think most people agree to that; but he also says that reactive or functional hypoglycemia is rare. There are some clinicians who will take issue with that. Indeed, if hypoglycemia is a reality, as many people believe, then what we are dealing with is a lot of people who have aberrant carbohydrate metabolism, and they cannot handle the load or the insult from absorbable sugars such as sucrose. There is a whole array of symptoms published by Phillips and by Salzer. One could name l5 or 20 different clinical reports in this area. Phillips was one reference that I chose. These are just some of the types of symptoms that a clinician reports for people who presumably have hypoglycemia (Table 20). Some clinicians indicate that the levels that we normally consider as minimum at 50 milligram percent of serum glucose is the value below which you start seeing symptoms. When you get down to about 30 or 25 milligram percent, then one could suspect organic hyperinsulinemia, and one might think about looking for a pancreatic adenoma. Now they say that healthy non-obese males may have a level below 50 milligram percent, and indeed a non-obese female may have a level at 30 milligram percent without symptoms. The main point about this, to go along with Dr. Levine, is that one not only needs to do careful observations on the glucose tolerance test for that data, but also to do an assay for insulin and other hormones. The one thing to bring out is that you have to observe the symptoms after running a glucose tolerance test. If they exhibit the symptoms and are characteristic of this low level of glucose in the blood, which is called hyperinsulinemia or hypoglycemia, then one may be character- ized as a hypoglycemic. This is a question for which I do not know the answer: How many people in the United States or in the world have aberrant carbohydrate metabolism? Some investigators claim that beyond the age of 50 its incidence rises. I have heard figures quoted that it is as high as 40 percent among our adult population. If that is so, then I think we have a problem here for serious consideration. I think then, as a toxicologist, one should emphasize metabolic over- loading. There is a threshold for all chemicals. Salt threshold is 3x, l5x for Vitamin A. Thus, we can agree that one could produce a toxicosis. It may be, as one of the panelists said this morning, that we are only seeing the tip of the iceberg. It has been reported that a first generation of diabetics among the Eskimo occurred when they become ac- quainted with the Canadian and American diet. Because of his gene pool, his genetic makeup, an individual may lack some coenzyme, maybe an enzyme system like glucose 6-phosphatase, or an insulin chromium cofactor.

63 One may have an intrinsic deficiency that shows up when exposed to a high insult of absorbable sugars. How much is that quantitatively? Is it 2 percent sucrose, 5 percent, l0 percent? How much can certain people take who have this deficient metabolic machinery? In Table 2l I have summarized the risk factors of absorbable sugars as I view them. The next two figures illustrate the association of absorbable sugars with allergic manifestations and a hypoglycemic state. This is just an ancillary bit of information that I gleaned back in the l940s when I was quite interested in allergy studies. Arthur F. Coca was considered the allergy authority at that time, and that is not on the next slide. Some of his findings are shown in Figure l. Figure 2 reports observations by Phillips that when serum glucose drops, and this is down below around 50 or lower, there occurs the appearance of severe nasal and GI allergy, nose block, and cramps. Of course, as the serum glucose rises, the symptoms disappear. This refers back to the time Coca was using accelerated pulse read- ings to pick out certain food allergens. He characterized sucrose as a food allergen. I think he was observing, in Figure l, someone who had aberrant metabolism as far as carbohydrates were concerned, and this may fit in now with what Phillips reported. How much sucrose or absorbable sugar can we tolerate? How many of us have abnormal carbohydrate metabolism? Should there be systematic studies done to elicit frank hypoglycemia, which they say is rare? How rare is it? How many people cannot tolerate the high load of absorba- ble sugar, in this case sucrose? There are a lot of people who perhaps can handle a high sucrose load and get away with it. After hearing a talk that I recently gave in Boston, a woman came up to say that she has been eating candy all her 100 r Diagnosis for food allergens using elevation of pulse rate PATlENT C.T. 4567 DAY OBSERVED BEET SUGAR .EATlNG CANE SUGAR. ! LlBERALLY . LESS . SUGARl NO CANE SUGAR niTTiwccc- jD1ZZ1NESS, LEG • HEAD- T1RED-I FEELS NO i incu-' TCCLO : \»/ci i . "w NESS i BETTER: wtLL JSYMPTOMS .1 i » FIGURE l Allergic manifes- tations -- sucrose. From Coca, A. F., Familial Non- reaginic Food Allergy, p. 45, C. C. Thomas Pub- lisher, Springfield, Ill., l945.

64 FASTlNG PATlENT X APPEARANCE OF SEVERE NASAL AND G.1. ALLERGY NOSE BLOCK AND CRAMPS 0.5 1 2 3 4 5 6 HOURS ON GLUCOSE TOLERANCE TEST FIGURE 2 Hypoglycemia in allergy. From Phillips, Am. Pract. Dig. Treat., l0: 97l-77. life, and lots of it. I congratulated her on having inherited a good liver, a good pancreas, a good pituitary, and a good adrenal gland. DISCUSSION MICHAEL SVEDA: We are looking at things at the present time and as we are now. I would like to think back to how we may have been when we were developing either 200,000 years, or a half a million years ago, depending upon which anthropologist you believe or what religion you believe. We became, in part, dependent on the food we had avail- able at that time. I cannot believe that we had a hundred pounds of sugar available then. Can this tip of the iceberg be covering up an awful lot of evolutionary happenings that have taken place? That is my first point. For the sake of all the known scientists in this audience, I would like to make a comment on the objection I have on a lot of things involving cyclamates: depending on who picks the panelists, what their biases are, what the biases of the selector are, we can get a yes or no answer on almost any scientific question. Plus the fact that with all of these things apparently wrong with sucrose, nobody has ever proved anything against cyclamates, yet they are off the market. KRAYBILL: There is a statement made that if you want to prove a point, you pick a committee that is going to side with you; if you want to go to a laboratory and get a certain result, you simply describe what kind of result you want. SVEDA: I thank you, Dr. Kraybill. KRAYBILL: Yes, well that is a little facetious. But getting back to the first point. From what I have read in the literature, obviously

65 our ancestors did not have sucrose accessible. They lived largely by foraging on fish, meat, and berries. They may have gotten carbo- hydrates and sweetness from the berries. It was not until the eighteenth century, in George Washington's time, that the intake per day was roughly around l5 grams per person, which is not large. Now we have gone up to about l00 grams as an average, and those who are gluttons may go to 250 or 300 grams a day. The question is, did our ancestors get along well physiologically and nutritionally? Chances are they did. They probably died of other diseases, but certainly as far as sucrose is concerned for energy, they had plenty of energy sources. KASHA: I would like to ask Dr. Kraybill a question. Your Table 20 cited a long list of maladies that many people may feel. I do not know whether suicidal tendencies are common, but headaches, dizzi- ness, other things are. Is there a specific test in the medical profession for hypoglycemia for people who do not suspect an aber- rant carbohydrate metabolism? KRAYBILL: The test is one for glucose tolerance. But as Levine pointed out, you have to run insulin levels and other hormone levels, and then you have to look at the symptoms of the patient after he handles this heavy load of sucrose or glucose. That list reviews the observations by various clinicians about their patients. In regard to the one on suicidal intent, there is an anecdote about a patient who came into Walter Reed because of a suicidal tendency. After a very extensive examination, it was de- cided to check her carbohydrate metabolism. They did this and found that her neuropsychopathic illness was attributed or associated with her handling of carbohydrates, in this case absorbable sugar. When that was corrected, then the other situations fell right in line. She was okay. That sounds like a success story, I know, but these are important little bits of information. ALFRED E. HARPER, University of Wisconsin: I wonder if Dr. Kraybill would like to comment on the problems involved in extrapolating from animal experiments in which some 70 to 80 percent of the diet was sucrose compared to the situation for the average human population with something like l0 to l5 percent of the calories from sucrose. KRAYBILL: I think your remark has a typical resemblance to some we are getting in carcinogenesis work. It has relevance to extrapolation from the animal to man. When that comes about you say, well, what is the ideal model or species? They will say the nonhuman primate. Then one will say, well, there is nothing like man himself. When you come to that, you say what man? There is no such thing as the average man, we all react differently. I think these levels, as I stated, are very high, very extreme and are exaggerated. But the point I wanted to bring out was that

66 although sucrose and starch were high, the impact on the glucose clearance was different. I think this is fundamental. HARPER: In interpreting the results of enzyme studies, where one ob- serves an increase in an enzyme activity upon altering the composi- tion of the diet, one has to look at the problem of adaptation. If there is one thing that is important in human survival, it is the adaptability of the human body and its metabolic systems. To imply that these alterations in enzyme activity are somehow an adverse effect, I think, is very misleading. Often they are the compensa- tory responses of the body to take care of a new substance. KRAYBILL: I hope I did not leave that impression. I just wanted to show you the effect on several of the biochemical parameters, and then one takes it from there, because you can do this with most any type of chemical as to alteration of the microsomal enzymes. That does not necessarily mean that that leads to an abnormal physiologi- cal state. WILLIAM J. DARBY, The Nutrition Foundation: I would like to comment relative to this question of hypoglycemia. This is undoubtedly the most loosely used, almost fraudulent term that a group of would-be physicians and sometimes misled clinicians are bandying about. It is a sort of "entertainer's disease." The American Medical Associ- ation and the American Diabetic Association, as well as a number of other groups, within quite recent times have published analyses of the claims of widespread hypoglycemia. Indeed, Dr. Frank Allan, who described the first case of hypo- glycemia, commented -- and you will find a reference to this in an article entitled "Americans Love Hogwash" by Dr. Edward Rynearson, published in a supplement to Nutrition Reviews, July l974 -- that he had seen very, very rare cases of functional hypoglycemia through- out a lifetime of specialized work in carbohydrate metabolism. I think we have to be quite careful about implying that hypogly- cemia is a widespread disease, or that it has anything much to do with sugar consumption. KRAYBILL: That is one view. Other clinicians can argue the point, and I think it needs to be looked at seriously. I think it is brushed under the rug. Clinical studies and good biochemical studies need to be done. It is true -- and I hope I prefaced my remarks when I said that Dr. Levine said it is a cult - - that too many things have been put into this wastebasket. I might as well admit that right now. I have been examined, and I have been diagnosed as a relative hypo- glycemic. This is not hogwash; this is real to me. I take exception to statements that tend to belittle the fact that there is a real incidence of hypoglycemia in the United States. It stands to reason that there should be, because a lot of people have different genetic

67 pools; they have different metabolic pathways and machinery; they have different biochemistries. With an advance in years, you may have an aberrant metabolic system with enzymes or coenzymes that are lacking. When a person gets relieved, the old saying applies that the proof of the pudding is in the tasting. If by mere dietary management of just reducing your sucrose markedly in the diet and upping the pro- tein by eating a cheese snack at l0:30 in the morning and 2:30 in the afternoon, you start feeling great, this is more than psycho- somatic medicine; I think it is very convincing. RALPH NELSON, Mayo Clinic: It is commonplace today to use functional hypoglycemia as a diagnosis to explain symptoms not related to hypo- glycemia in persons who otherwise are healthy. But in my experience, healthy people who embrace this diagnosis have been unable to cor- relate their symptoms with the decrease of blood sugar during the glucose tolerance test. High-protein diets have been prescribed as therapy by clinicians who believe this entity occurs as a result of increased release of insulin after eating carbohydrate. However, it was shown years ago that some amino acids (the substances absorbed as a result of pro- tein digestion in the intestinal tract) are more potent liberators of insulin than is glucose. So this therapy is not based on physio- logic grounds. We do know four stimuli that will produce hypoglycemia in other- wise normal people: alcohol, excessive exercise, high-protein diets, and -- in lactating women -- calorie restriction. These factors can have effect in combinations. For instance, a lactating woman who takes an alcoholic drink while on a self-imposed diet may show signs and symptoms of hypoglycemia during exercise. So it does occur, but we have some physiologic basis for under- standing it.

68 TABLE l Reward Features -- Satiety Value of Sugars Human oral gratification Tongue is sensitive to sweet taste Oral cavity impulse -- transmission via nerves and spinal cord* Stimulation in brain center (gnostic brain region) Increase in saliva flow SOURCE: Nordsiek, F. W., Am. Sci. , 4l-45, Jan.-Feb., l972. *Chorda tympani traverses middle ear to brain. TABLE 2 Species Responding to Sweet Sensation (via Chorda Tympani) Exceptions Responders Cat Man Pigeon Chicken Dog Rabbit Horse Hamster Cattle Rat* SOURCE: Milner, Physiological Psychology, Holt and Rinehart Publishers, New York, l970. *Rat will respond to saccharin solution equally well. TABLE 3 Economics of sweets Poorer families use twice as much sucrose/person as higher income groups. The reverse situation is true for total sugars (processed, prepared foods, i.e., jams, jellies, cakes, frozen desserts, etc.). Middle Ages -- sugar a costly item (l4th C. -- l Ib of sugar a week's wages). Technology of l800s reduced cost of sucrose. Standard of living increase -- more sucrose became available. SOURCE: Nordsiek, F. W., Am. Sci., 4l-45, Jan.-Feb., l972.

69 TABLE 4 World Annual Sugar Production (Listed on a 3-Year Basis) Year Production (tons x l0 ) Cane Beet Total l958-59 2.9 2.l 5.0 l96l-62 3.4 2.7 6.l l964-65 3.5 2.6 6.l l967-68 4.l 3.l 7.2 l970-7l 4.6 3.2 7.8 l972-73 5.0 3.3 8.3 SOURCE: U.S. Department of Agriculture. TABLE 5 U.S. Per Capita Sugar Consumption (Pounds/Person/Year) l946 74.3l l95l 93.68 l956 98.96 l96l 98.ll l966 98.27 l97l l02.27 SOURCE: Hawaiian Sugar Manual, l974. TABLE 6 Nations Leading in Sugar Production (Tons Produced) Nation Production Nation Production 8.7 x l06 l06 U.S.S.R. Mexico 3.l x 6.9 x l06 1O6 Brazil Australia 3.0 x 6.l x l06 l06 Cuba W. Germany 2.4 x 4.9 x l06 1O6 India Philippines 2.4 x 4.9 x l06 lO6 U.S. S. Africa 2.l x 3.4 x l06 lO6 China Poland 2.0 x 3.2 x l06 1O6 France Italy l.4 x Hawaii (counted separately) l.l x l06 (U.S. would then rank fourth) SOURCE: U.S. Department of Agriculture.

70 TABLE 7 Relative Prices of Sugar (l964-l972, World Market Price) Year Average Year (<f/lb.) l964 5.86 l965 2.l2 l966 l.86 l967 l.99 l968 l.98 l969 3.37 l970 3.75 l97l 4.52 l972 7.4l SOURCE: Licht's World Sugar Statistics, 72/73 TABLE 8 General Benefits of Absorbable Sugars Increases GNP of country Provides employment (farm, factory, transport) Reward feature - satiety value Provides quick energy Calories for low income groups Low cost energy Appetite stimulator Provides consistency and texture to certain foods ^Probably not true in l974-l975. TABLE 9 Alternatives to Sucrose Cheaper material (corn syrup, corn sugar) Sorgo syrups Maple syrup Honey Nonnutritive sweeteners SOURCE: Nordsiek, F. W., Am. Sci., 4l-45, Jan.-Feb., l972.

7l TABLE l0 Benefits (?) of Sucrose Substitute Some nonnutritive sweeteners found to stimulate appetite and food intake High levels of intake of saccharin may have an aversive taste effect Rats preferred 0.25% saccharin solution to 3% glucose solution (male rat preference < female rats') In weight control, total caloric value of diet is important SOURCE: Vadenstern, E. S., et al. , Science, l56:942-43, l967. TABLE ll Effect of Chromium (III) on Hypercholesterolemia in Rats Serum Cholesterol Diet Sucrose Starch Mean Mg (%) Mg (%) -- Control 240!JJ 209^8 237 9nB+l3 iti+l6 Chromium /US 1 T l6l 1 7 ~ X J ~ J. J l82 Mean 236 l83 Chromium in drinking water = 5 ppm. Sucrose diet Cr III = 0.3l ppm. Starch diet Cr III = 0.38 ppm. SOURCE: Data of Staub, H. W., Reussner, G., and Reinhardt, T., Science, l66:746-47, l969. TABLE l2 Mean Changes in Triglyceride Levels During a Sucrose-Free Period for l8 Men in Antarctica All men (l8) -l.3 ± 2.6 Not significant (Mean ± SEM): Change (Mg Percent) High triglyceride Group (5) -l5.l * 5.3 P < 0.0l Low triglyceride Group (l3) +3.9 ± 2.7 Not significant Pre-dietary level: Low triglyceride group = 92 mg percent (l3 men) High triglyceride group = l54 mg percent (5 men) SOURCE: Roberts, A. M., Lancet, l20l-4, June 2, l973.

72 TABLE l3 Mean Changes in Cholesterol Levels During a Sucrose-Free Period for l8 Men in Antarctica (Mean * SEM): Change (Mg Percent) All men (l8) High Cholesterol Group (5) +5.5 ± l.7 -9.l t 3.5 P < 0.005 P < 0.025 Ranges in values: Low cholesterol group = l95 to 225 High cholesterol group = 225 to 250 SOURCE: Roberts, A. M., Lancet, l20l-4, June 2, l973. Low Cholesterol Group (l3) + ll.l t l.6 P < 0.00l TABLE l4 Deaths From Pancreatic Cancers by Time Period (Study of Boston Population: l930-l956) Observed Expected Males Females Males Females l93l-l935 l 4 0.4 0.7 l936-l940 3 3 l.l l.6 l94l-l945 4 6 2.2 2.9 l946-l950 7 9 4.0 4.7 l95l-l955 l0 l4 6.7 6.4 l956-l959 5 l2 6.0 6.2 Total l93l-l959 30 48 20.4 22.5 TOTAL 78 42.9 SOURCE: Kessler, I. I., J. Nat. Cancer Inst., 44(3):673-86, l970.

73 TABLE l5 Influence of Dietary Carbohydrate on Age at Death and Cause of Death (BHE Rats) Diet Longevity Disease (%) Days Kidney Respiratory Sucrose 444 ± 24 7l 70 Cornstarch 595 t 34 77* l5 Glucose 543 t 48 60 20 SOURCE: Durand et al., Arch. Pathol., 85:3l8-24, l968. *Onset at 620 days compared to 47l for sucrose. TABLE l6 Influence of Carbohydrate Source on Longevity and Cause of Death (Wistar Rats) Diet Longevity Disease (%) Days Kidney Respiratory Sucrose 583 * 40 6 62 Cornstarch 636 ± 43 l4 7l Glucose 565 t 48 0 62 SOURCE: Durand et al., Arch. Pathol., 85:3l8-24, l968. TABLE l7 Effect of Sucrose on Glucose Tolerance in Rats Glucose Levels with Time (ing %) Diet Time: 0 30 60 90 l20 72% Starch 60.l*2.2 62.9 ± 4.l 55.3*2.5 44.7*4.2 (no sucrose) 72% Sucrose 82.9*l.3 86.l*2.l 86.3-0.8 63.l*l.3 79% Sucrose 99.l ± l.5 l04.6 ± 3.7 92.7 ± 3.4 72.l ± 3.4 SOURCE: Cohen and Teitelbaum, Metabolism, l5:l034-38, l966.

74 TABLE l8 Dietary Carbohydrates and Enzyme Activities (Wistar Rats) Age (mo. Liver Kidney G6PD Aldolase Alk Pase Serum Alk Pase Aldolase ) Diet G6 Pase 3 Sucrose 70 27 49 36 9 l0.l Glucose 64 l6 30 33 8 l3.l Cornstarch 54 ll 34 27 7.9 l2.0 9 Sucrose 49 l8 40 l3 6 9.0 Glucose 40 l6 35 ll 6 l0.0 Cornstarch 35 l8 34 l3 5 9.0 Standard errors not cited, but are in original paper. SOURCE: Chang et al., J. Nutr. , l0l:323-30, l97l. TABLE l9 Cariogenic Effect of Sugar-Rich Diets (Experimental Animals) Species Diet Caries Score (Teeth Development § Maturation) Hamsters Stock + Purified 6.l Rats Stockl + Purified2 0.0 Mice Stockl + Purified2 0.0 Hamsters Purifiedl + Purified 48.0 Rats Purified + Purified 2.7 Mice Purified + Purified 0.5 SOURCE: Sognnaes, R. F., J. Am. Dental Assoc., 37:676, l948. Ipregnancy and lactation stage. 2Posteruptive stage.

75 TABLE 20 Some Symptoms Recorded in Hypoglycemic Syndrome (42 Symptoms for 600 Cases Observed) Symptoms Percent in Cases Symptoms Percent in Cases Nervousness 94 Gastrointestinal 68 Irritability 89 Insomnia 62 Fatigue 87 Internal trembling 57 Weakness, cold sweats 86 Tachycardia 5l Depression 77 Allergies 43 Vertigo 73 Blurred vision 40 Headaches 7l Suicidal intent 20 SOURCE: Phillips, K., Am. Pract. Dig. Treat., l0:97l-77, l959, J. B. Lippincott Publishers. TABLE 2l General Adverse Effects of Absorbable Sugars (Sucrose) High intakes replace calories from other macronutrients (empty calories). In aberrant metabolism contributes to obesity. Contributor to dental caries induction. In aberrant metabolism causes and aggravates: Hyperglycemia (hypoinsulinemia -- diabetes) Hypoglycemia (hyperinsulinemia) Implicated in hypertriglyceridemia. Implicated in hypercholesterolemia. Associated with diabetes and increased pancreatic cancer risk. In experimental animals, reduces longevity. In experimental animals, accelerates kidney disease. In experimental animals, makes increased demands on enzymes of liver, kidney, and serum. In experimental animals, impairs glucose tolerance. Associated with allergic manifestations in hypoglycemic states. Alleged contributor to atherosclerotic processes. Potential effects on growth and maturation from overconsumption.

NUTRITION: SUMMARY OF EVIDENCE Paul M. Newberne In conducting nutritional experimental work for over 20 years, I must admit, in company with most other nutritionists, that carbohydrates have usually been the last thing considered in designing the diet. We con- sidered protein, fat, vitamins, minerals, and finally, carbohydrates to complete dietary proportions. More recently, however, we have had to come to grips with this on many different fronts in considering various sources, the quality and quantity of carbohydrate in providing an opti- mum diet for man and animals. One of the questions central to this Forum is, what is sugar? Many people have their own concept of what sugars may be. Sugar is a carbo- hydrate, but there are many different kinds. Sugars may be derived from animal or vegetable sources. The ones that we are most familiar with are the disaccharides and the monosaccharides. All of them, according to definition, are white, crystallizable, soluble in water and in dilute alcohol. Before one can consider intelligently a subject such as this, we have to consider what our nutritional needs may be under varying physiological states. In addition to water, there are those that we do need. We need sources of protein, fat, minerals, carbohydrates, vitamins, and fiber. It is this area of carbohydrates that is of concern to this Forum, and it is in this category that sugars fall. However, even when considering carbohydrates, and more specifically the sugars that make up a bulk of that dietary category, one cannot forget any of the other major ingre- dients since a balance of all of them is required for optimum nutrition. Sugars, first of all, are foods. Sucrose, as such, is a palatable food. It is readily available. It is easily packaged. It stores well, and generally it is relatively inexpensive on a calorie basis. Sucrose is broken down into glucose, and it is in this form that the body 76

77 utilizes it for energy. You get about 20 calories from an average tea- spoon of sucrose. Now, let us assume that the average individual needs anywhere from l,800 to 3,000 calories per day, depending on the degree of physical activity. Table l indicates the average caloric needs and the consump- tion of sugar with that portion of calories derived from sugar. This may be a bit high, but let us assume that we do take in one-third of a pound of sucrose daily on the average. That gives us about 600 calories, and that is just about 25 percent of the average need for calories in our generally inactive society. Sugars have become much more important in recent years, because we do have a distinct craving for sweetness (Table 2). It is a human craving, but it is not limited to those of us in the people family. Animals like it as well. When we consider the amount of sugar consumed by various population groups around the world, we find it to be signifi- cant, constituting about l8 percent of the total calories; and the con- sumption is increasing, particularly in developing countries. In any consideration of dietary constituents or the diet as a whole, the question arises, why do we eat? We eat primarily to satisfy energy needs; in addition to enjoying food we have to have energy to survive. The predominant motivation for food consumption is to gain energy. About three-fourths of the ingredients in a normal, well-balanced diet is used specifically for calories. Our energy requirements depend on a number of factors, for example, activity; this is probably the most important one. Another is body size and composition; the larger you are, the more you have to move about, the more energy it requires. With increasing age, we need fewer calo- ries. The climate, whether it is cold or warm, and the kinds of clothes we wear are additional factors and important determinants of our caloric needs. All these factors are highly important. In the final analysis, TABLE l Range of Caloric Needs of Individuals and Average Calories Derived from Sugar in U.S. Require l,800 -- 3,000 Kilocalories Daily Total = 450 Grams -- 750 Grams (One -- One and Two-Thirds Pounds) Average Consumption Sucrose ll0 Pounds Per Year in U.S. or About l/3 Pound Daily l50 Grams = 600 Kcalories 25% Average Caloric Needs

78 TABLE 2 Reasons for High Sugar Consumption United Kingdom Australia Switzerland United States Human Craving for Sweetness 50 to 60 kg/Capita Sugar l8% of Calories 50 kg/Capita Cheap Source of Calories Favored by City Dwellers Consumption Rising Steeply in Developing Countries if caloric intake is less than our needs for heat, work, and other essential body functions, we lose weight because body stores of fat, and later protein, are drawn upon for energy. If caloric intake exceeds requirements we store it as fat. Figure l illustrates the concept of energy balance (l). Life styles and their accompanying energy needs have changed over the last century. In l850, the average person required about 2,400 calories if engaged in a light work load; l8 percent of the population was engaged in work considered to be light. Table 3 shows that over the last century the proportion of people doing light work has shifted to the point that in l966 about two-thirds of the U.S. population is engaged in light work. Likewise, the percentage of very heavy working individuals has dropped from l6 percent to l percent as a result of mechanization and for other reasons. In consequence, we as a society do not need the same quantity calories as was needed in the past. It has been pointed out elsewhere in this Forum that world sugar production has gone up. There has been a steady increase in production since the turn of the century, and that has paralleled, in an inverse way, the need for calories by U.S. populations. HEAT ENERGY 1NTAKE EXCRETA FIGURE l Illustration of balance of energy. Intake should equal usage of calories (heat, work, excreta) in order to maintain body weight gain. If intake exceeds usage, there will be weight gain. Conversely, if usage exceeds intake, there will be weight loss.

79 TABLE 3 Shift in Caloric Requirements from l850 to l966 Percent of Cal. Required Work Per Day l850 l966 Light 2,400 l8 62 Heavy 3,300 26 l0 Very Heavy 4,000 l6 l SOURCE: Schettler and Schlierf, reference l. In speaking to the role of sugars in nutrition, what are some of the questions that one should ask? What are desirable levels in the diet? What are the benefits of particular levels? What may be excessive in- take? What are the risks associated with excessive intake? The controversial aspects of sucrose have been pointed out by many different investigators, and there is an extensive literature on the subject. John Yudkin has written a book titled Sweet and Dangerous (2) in which he contends that sucrose is harmful. He calls it the quiet killer, and alleges that it causes heart disease. On the other hand, as one example, Fewkes (3) has written a very good review pointing out that in his estimation the direct evidence for what Yudkin says is lacking. The four areas where sugar and human disease are associated by many include cardiovascular disease, obesity, diabetes, and dental caries. Each of these can be taken up separately for brief discussions. Sugar and heart disease relationships have usually been associated with the levels of lipids in the blood (cholesterol and triglyceride) and suggestions have been made that triglycerides may be as important as cholesterol. Szanto and Yudkin (4) conducted a study in human volunteers in which high-sucrose diets caused no significant changes in blood cholesterol but increased the serum triglycerides and serum insulin concentration (Table 4). Since there was no significant change, the authors concluded that probably heart disease was being mediated via hyperinsulinism. Haldi and Wynn (5) studied 44 medical students given a test load of l.5 grains of sucrose or glucose per kilo of body weight (Table 5). Blood concentrations were about the same for both groups, and blood sugar levels in both groups returned to a normal level within about 4 hours after giving the test load. Results of this study clearly indicate that glucose and sucrose had about the same effect on blood sugar levels. There is a disease in man referred to as primary endogenous hyper- triglyceridemia, a condition where patients have a much higher level of triglycerides in their blood than is normal. A group of five patients was chosen because they had 200 milligrams percent or more triglycerides but they had a normal fasting blood sugar. They were examined following a test loading of either starch, fructose, or sucrose (Table 6) and

80 TABLE 4 Dietary Sucrose -- Male Human Subjects Prelim Sucrose Rest Starch Rest Kcal/day 3390 3350 3370 — Wt/kg 73.0 74.0 73.4 73.6 73.5 Cholest (mg°6) 249 248 250 243 248 Trigly (mg%) l43 l50 l42 l40 l42 Blood Glucose (mg%) 0 Min 69 99 68 69 70 l20 Min 68 68 68 67 68 Serum insulin (uU/ml) 0 Min 22 23 23 22 2l 30 Min 63 86 67 64 64 SOURCE: Szanto and Yudkin , Reference 4 . TABLE 5 Blood Sugar After Consuming Glucose or Sucrose Minutes Post-Ingestion Sugar Basal (mg%) 25 l5 30 Glucose 95 99 l08 l34 l55 Sucrose 95 l00 l07 l34 l48 SOURCE: Haldi and Wynn, reference 5. triglyceride concentrations were different with a resting level of tri- glycerides increased in those given sucrose compared to those with starch or fructose. In addition, the turnover rate was higher. It was concluded that sucrose increased the triglyceride concentration of the blood, and further, the turnover rate, but did not have any influence on the excretion or the disposition of it. McDonald reported some interesting studies in humans and in animals (7). He gave human patients either corn starch or sucrose and found that neither increased serum lipid levels significantly in these adult patients. It must be borne in mind, however, that he was giving

8l enormous doses for a relatively short period of time and measuring the effects without giving much consideration for the long-term effects. McDonald and coworkers also conducted studies in rabbits. Since they were interested in kwashiorkor -- a protein-calorie malnutrition disease in children with very low dietary protein but high carbohydrate intake -- they attempted to reproduce this condition and were able to reproduce fatty livers and other derangements associated with the childhood dis- ease. These investigators were trying to define the difference between a child with kwashiorkor and a child that was simply in effect starved; in the latter you do not find lipid accumulating in the liver. He found in rabbits that sucrose produced more liver lipid than an equal weight of dietary starch, but again these results were from a relatively short- term experiment. Anderson did a study in twelve normal men, examining the serum choles- terol levels that were induced by glucose or sucrose, or lactose and glucose (Table 7). This was a study in which 3l percent of the total caloric intake was changed within time periods of about two weeks each. TABLE 6 Triglyceridemia Concentration in Primary Endogenous Hypertriglyceridemiaa TG TG TG Cone. Turnover (h-l) Turnover (mg/h/kg) Diet (mg/l00 ml.) Starch 220 ± 45 0.l80 ± 0.04 l4.l ± 2.l Fructose 229 ± 36 0.l72 ± 0.02 l4.4 ± 2.3 Sucrose 270 ± 95 0.l79 ± 0.04 l6.8 ± 4.5 Five subjects with primary endogenous hypertriglyceridemia, 3 different diets. Each patient had 5-l0 measurements, each period. From Nikkila, reference 6. TABLE 7 Serum Cholesterol Levels (mg/l00 ml) of Twelve Men Diet Mean Glucose Sucrose Lactose fj Glucose Difference l80 ± 6.5 l85 ± 7 -- 5 ± 3.5 l77 ± 9.7 -- l78 ± ll.2 l ± 3.4 _- l79 ± ll.l l74 + l0.l 5 ± 4.5

82 In examining the data, there was no difference in switching from glucose to sucrose or from sucrose to lactose and glucose in terms of serum cholesterol levels. Some have associated sugar intake in coronary patients with that disease, and Table 8 lists the data from one study (8). Although there was considerable variation in sugar intake among coronary patients, there was no convincing evidence that sugar intake and coronary heart disease were causally related. Some of the differences were associated with smoking and with consumption of sweet drinks such as cocoa and coffee. In l970, Masironi published in the WHO bulletin (9) data relative to dietary factors and coronary heart disease from 37 countries. Mortality data from degenerative heart disease and per capita consumption of fat, sucrose, complex carbohydrates, protein and total caloric intake were considered. On the basis of a large volume of information he concluded that total unsaturated fats positively were correlated with the death rates, but complex carbohydrates were not. An interesting discussion pointed out that diet and heart disease were still controversial. Table 9 lists examples of untreated adult onset diabetes and the effect of starch, fructose, and sucrose on blood sugar in untreated adult-onset type diabetes. As these data point to individual differences among patients with no significant differences associated with any one of the individual sugars, one cannot generalize in such cases. TABLE 8 Sugar Consumption by Male Coronary Patients with Control Subjects Number of Subjects Sugar Intake (g/day) Author Patients Controls Patients Controls p Little et al. (l965) 86 84 47 65 <0.0l Papp et al. (l965) 20 20 l2l ll7 >0.05 Begg et al. (l967) 63 33 39 55 <0.05 Paul et al. (l968) 66 85 ll6 96 — Finegan et al. (l968) l00 50 66 69 >0.05 Burns-Cox et al. (l969) 80 l60 l00 97 >0.05 Howell and Wilson (l969) l70 ll58 67 79 >0.05 Working Party Medical Research Council (l970) l50 275 l22 ll3 >0.05 Gatti (l970 47 3l 57 45 >0.05 From Grande, reference 8.

83 Dr. Kraybill has referred to chromium as an additional factor that has been associated with blood sugar levels. The chromium content of different types of sugars is listed in Table l0. The data in Table l0 were supplied by Walter Mertz of the USDA Human Nutrition Laboratory where he has been working with this for many years, and the indication is that there is some relationship between chromium and the level of blood glu- cose. In Table l0 it is revealed that different kinds of analyses yield different results; nevertheless the trend is in the same direction. The more refined the sugar, the less the chromium content. Finally, TABLE 9 Effect of Dietary Fructose and Sucrose on Blood Glucose in Untreated Adult-Onset Type Diabetesa Case Starch Fructose^5 Sucrose* K. H. l4 .l + l. l l3. 6 ± l .0 l4.2 ± l.3 T. V. l0 .7 ± 0. 8 l2. 0 ± 0 .7 — V. N. 6 .9 ± 0. 4 6. 2 ± 0 .3 5.7 ± 0.4 Kl .H. l4 .4 ± l. 5 l4. 7 ± 0 .7 l4.l ± l.l J. G. l3 .4 ± 0. 7 l4. 5 ± 0 .6 l5.2 ± 0.7 V. P. 9 .5 + 0. 4 ll. 8 ± 0 .9 MEAN ll .3 l2. l l2.3 Mean ± S.D. of 6-l0 values for each period, mM. Daily dose 80-l00 g substituting for starch. TABLE l0 Chromium Content in Different Types of Sugars (ng/g Sugar ± SEM) Sugar Number Samples Direct Analysis (l,000° Ash) Muffle Ash (450°) Low Temp . Ash (l50°) Molasses 3 29 ± 5 l29 ± 54 266 ± 58 Unrefined 8 37 ± l3 88 ± 20 l62 ± 36 Brown 5 3l ± 2 53 ± 8 64 ± 5 Refined 7 l0 25 ± 3 20 ± 3 Supplied by Dr. Walter Mertz.

84 when low temperature ashing method was used, more accurate data were ob- tained, indicating a need for sensitive, accurate methods. The question of the relation between sugar and obesity is a real one. Some diseases are associated with obesity in a very positive way (Figure 2). I do not think anyone would argue that obesity is unrelated to longevity, diabetes, liver cirrhosis in the male, appendicitis, gall- stones, and perhaps cardiovascular disease, some more strongly related than others. In any case, there are convincing associations of disease with obesity. In regard to sugar and obesity, can we say there is a causal relationship? Based on what we know today, I do not think so. Obesity is simply a result of taking in more calories than are used up. DEATHS-Percent actual of expected Cardiovascular- Renal Disease Diabetes mellitus Liver Cirrhosis Appendicitis Gallstones Cancer (All) Tuberculosis Peptic Ulcer Suicide Accidents—Total Sex {Death rate of persons accepted for standard insurance) 100 200 300 400 100 200 300 400 FIGURE 2 Effects of obesity on susceptibility to various diseases. Black bars represent increased susceptibility in overweight individuals. Reproduced from Marks, H. H., Influ- ence of obesity on morbidity and mortality, Bull. N.Y. Acad. Med., 36:296-3l2, l960.

85 Now, let us turn to the area of dental caries where the evidence is much more convincing. Sugars do indeed contribute to the development of dental caries, sucrose perhaps more than others. There are many studies in animals that support this concept, and an earlier one by Haldi and colleagues (l0) illustrates the point. Rats were given sucrose by three different ways as shown in Table ll. If the entire ration was given by stomach tube, thereby bypassing the teeth, there were no caries. If given in an oral solution there was a signifi- cant incidence of caries, but the highest rate was associated with ad- ministration of granular sucrose in the diet. This permitted longer, continuous contact with the teeth. The source of energy provided the microflora of the mouth creates an environment for acid production and enamel erosion. In closing, I do want to point out that the data we have in hand indi- cates that it is the quantity of the calories, rather than the source, that contributes to health problems. We must bear in mind, however, that some sources are much easier to get, they are much more palatable, and as a consequence we may take more of them in. TABLE ll Dental Caries Incidence in Rats Fed Sucrose by Various Routes Average No. No. Wt. Increase Caries- Average Caries Score Group Treatment Rats M F Free Rats I Entire Ration 8 23l l96 8 0 Stomach Tube II Granular Sucrose l3 232 l82 2 7.5 Orally III Sucrose Sol'n l3 252 208 7 l.2 Orally From Haldi, reference l0.

HEALTH: SUMMARY OF EVIDENCE D. Mark Hegsted It is difficult to decide what I can add at this stage of the program. This is a summary of a summary of a summary after Dr. Warren, Dr. Kraybill, and Dr. Newberne. The agenda suggests that one can sepa- rate nutritional effects from those on health. I believe that nutrition is concerned with the effects of foods or constituents of foods upon health and that no such separation can be made. Nutritionists are also concerned with taste, cultural habits, and various things that may affect food intake, but primarily because of their relation to health. We know that good nutrition can be achieved with many different kinds of diets. From a strictly nutritional point of view, we, as nutrition- ists, do not care what foods people eat as long as the mixture provides for good nutrition. Obviously, this differs from the commercial inter- ests. We are aware, for all practical purposes, that every food in our diet competes with every other food. There is practically no elasticity in total food demand in the United States. If we eat more of something, we will eat less of something else. This concerns many of you, but it may or may not be of any particular concern to nutritionists. I wish to emphasize that we are entering or have entered into a new era of nutrition. For most of this century we have been concerned with the identification of the essential nutrients -- amino acids, vitamins, and minerals. The underlying thesis has been that if we could identify all of the essential nutrients, obtain reasonable estimates of the di- etary requirements for each, and understand their biochemical function, then it should be easy to develop biochemical tests for nutritional status or to evaluate dietary data. Adequate nutrition would be defined as enough of all essential nutrients. It is clear that this is an oversimplification. We are now concerned with a variety of problems that are food related but have little, if 86

87 anything, to do with essential nutrients. The prime examples probably are obesity and heart disease. The modern Western diets, the type of diets that universally characterize affluence around the world, are associated with many diseases. Most of these have been mentioned -- heart disease, obesity, hypertension, cancer of the colon, diverticulo- sis, diabetes, et cetera. There is good reason to believe that these are causally related to diet; there is little reason to believe that they are related to essential nutrients. As shown in Figure l the changes in diet that accompany "development" are usually so uniform that the epidemiologic data do not permit identi- fication of causal factors. Since many other things will show similar correlations -- tin cans, television sets, automobiles, pollution, processed foods, et cetera -- these data alone do not identify saturated fat and sugar, either or both, as causal factors. They simply indicate things that may be worthy of further consideration. There may, of course, be "natural experiments" that permit some in- sight into causality. High consumption of sugar in populations like those in Jamaica or Cuba, where fat consumption may not be correspond- ingly high, may be examples. If data are available from such popula- tions, however, I do not know of them. Even so, one must be careful in interpreting such data since a high fat-high sugar diet may not yield the results that would be expected from studies of high-fat diets and high-sugar diets alone. There are, in fact, data indicating effects upon serum lipids that are not observed when either is fed alone. The primary problems possibly associated with sugar consumption have been discussed -- heart disease, diabetes, obesity, and dental caries. Table l summarizes the data obtained by the British Medical Council to check the Yudkin hypothesis. Yudkin concluded from retrospective di- etary studies that people who had had heart attacks generally consumed much larger amounts of sugar than comparable controls. The studies of the Medical Research Council simply do not confirm Yudkin's data nor do a variety of other studies, many of which were summarized by Dr. Newberne. It must be emphasized that it has been shown repeatedly and almost universally that modification of dietary fat will result in a modifica- tion of circulating cholesterol levels and that such cholesterol levels are associated with severity of atherosclerosis and risk of heart dis- ease. Thus, to this degree, the effect of dietary fat is explained. It is this fact, not the epidemiologic data, which implicates dietary fat and cholesterol as causal agents in heart disease. Such mechanisms are not available to explain a supposed role for sugar. I take exception to many of the experimental studies presented by Drs. Kraybill and Newberne. There have been many studies in which rats or other experimental animals or human subjects have been fed diets in which all of the carbohydrate was either sugar or starch or other carbo- hydrates. One can often demonstrate metabolic differences, but the relevance is unclear. In the United States a low-sugar diet may provide l0 percent of the calories as sugar, and a high-sugar diet 25-30 percent of the calories. The question is whether changes of this magnitude have

88 any significant effect on health. Such data as are available indicate that they are minimal with regard to lipid metabolism or risk of heart disease. 800 r- USA 700 O Sugar Australia • Saturated Fat • ,, Canada 600 - Finland N. Zealand 40 1 * O • N.Z. § 500 X 35 - Un. King. '= - 0 OCan. QAuit. OU.S. UK. o 8 0 • 5 ^ Q "^ Fin. Swed | 400 — "o O O "^ a? 25 •~ W. G«r. °Q c"|,z ~ Denmark o ^f 0 {? 300 •0 u- 20 ~ O 1 Norway • o * Nor. O • ° S»iu. S 15 200 0 / ° 1 W. Germany (/> ._ 1taly ""V r, . — O 100 — O* Japan MIH.», g — • O Japan 1 1 1 1 1 1 | J P1 1I 11! 1 20 40 60 80 100 120 Sugar Consump., 1b/yr. 20 40 60 80 100 120 140 Sugar Consump. 1 1 1 1 1 1 1 1 5 10 15 20 25 30 Saturated Fat Consump., % Calories FIGURE l Correlations between fat and sugar consumption and mortality from coronary heart disease. Taken from McGandy et aZ., N. Engl. J. Med,, 277:4l7, 469, l967. TABLE l Summary of Results of the Medical Research Council Working Party CHD Patients Controls No. Sugar Intake (g/day) No. Sugar Intake (g/day) Middlesex hospital 80 l00 l60 97 Hammersmith hospital 2l l03 2l l00 Scottish study 49 l22 94 ll3 SOURCE: Lancet, 2:l265, l970

89 The epidemiology of diabetes is similar to that of heart disease. Given our preoccupation with blood sugar levels in this disease, it is not surprising that sugar seems a logical causal factor. The experi- mental studies of Cohen have shown the development of a mild diabetes in rats fed diets containing very large amounts of sugar. However, many people with elevated blood lipids demonstrate abnormalities in glucose tolerance, and diabetics are at high risk of atherosclerosis and heart disease. These are presumably common dietary factors influencing these diseases. The story of dental caries seems to be the clearest of the four con- ditions under discussion. Sugar consumption has a relation to the development of dental caries, but it is also clear that the way the sugar is consumed is probably more important than the amount of sugar consumed. With regard to obesity, there is no evidence that the calories in sugar are any better or worse than other forms of calories. It does seem likely that highly acceptable diets -- diets high in sugar and fat -- are more likely to induce an overconsumption of calories. Diets lower in sugar and fat may be useful in controlling obesity. One of our problems is that we are looking for a cause of a disease when this is probably unreasonable. We are to some degree misled by experience with infectious disease where we ordinarily attribute the disease to the infectious agent. We forget that most people who are exposed to the tuberculosis organism, for example, do not get tubercu- losis. Obviously, exposure is a necessary but not sufficient condition to produce the disease, and one can make a better argument that other factors, either of the host or the environment, are more important than the bacillus itself. All diseases have a complex etiology, and this is particularly true of the so-called degenerative diseases, the major health problems of the United States. We cannot expect to find single causal agents, and we must find ways to deal with the variety of factors that contribute to the development of disease. I feel that much of the data that have been presented about sugar con- sumption in the United States is somewhat misleading. It has been stressed that consumption has remained at approximately l00 pounds per year for a considerable period. This ignores the fact that the total calorie consumption has been falling. Many people now consume diets that provide only l,500 calories or so per day. The proportion of the diet provided by sugar is rising. When total calorie intake falls this low and 40 percent of the calories are supplied by fat and 25 percent by sugar, only 35 percent of the calories -- some 500 calories or so -- are available to carry the essential nutrients. This is not the most desirable situation and is associated with some risk of nutritional inadequacy. Although we cannot identify sugar specifically as a causal factor for any specific disease, we do know a good deal about the risk of con- suming the type of diet we usually consume. Half of us will die of heart disease, for example. It seems to me the choice should and must

90 be made on the basis of the risk of doing nothing versus the risk of doing something that seems most sensible at this time. The sensible action is to move toward the so-called "prudent diet," a diet that is lower in fat, in sugar, in meat, in cholesterol with less food in gen- eral, while increasing our consumption of fruits, vegetables, and cereals -- especially whole grain cereals. I see no risk from the nu- tritional point of view and the possibility of substantial advantage. I personally believe that we all have a tendency to rely too heavily on the regulatory agencies. We tend to encourage them to take all sorts of actions that happen to be favorable for specific interests and to swear at them when they do things not in our specific interest. They obviously should be involved when there are clear-cut dangers. Other- wise, we should remember that we all have the right, within limits, to make fools of ourselves, and we want to preserve that right. All dietary recommendations have advantages and disadvantages for some part of the food industry or other. These effects, whatever they may be, cannot be of much concern in developing dietary recommendations. Some products will be harmed and may fail, but opportunities are pro- vided for other products. The food industry will survive. DISCUSSION KASHA: May I raise a factual question to begin with? We were shown by Sidney Cantor that the sucrose level of consumption had remained essentially constant or slightly decreased, but I thought he also definitively showed that the total sugars had gone up by as much as 30 percent. HEGSTED: I think we are discussing consumption and disappearance, and I do not know how much of the disappearance figures represent con- sumption. I do think there is good reasons to believe that the pro- portion of the calories in the diet provided by sugar has probably gone up substantially in the last 25, 30, or 40 years. KASHA: Does the highly desirable placement of France and Italy on the chart you showed include wine as a source of sugar? HEGSTED: I am sure it did not. I do not know how much sugar is left in wine; if it is dry, I assume there is very little. JOHN NEWTON, Clinton Corn Processing Company: I am a carbohydrates, starch, and enzyme chemist. I have been very concerned that in most of your feeding trials, you are using raw native starches that are indigestible by human beings, and that you compare them at a 70 to 75 percent level with a water insoluble, completely available sugar. I would like at least to see some of the feeding trials done on a

9l precooked starch of the type that humans use. I think that you would see some differences. HEGSTED: I would say that on many of our animal trials we used dextrin, which I suppose is more digestible, depending on how you prepare it. Anyway, I do not think those studies can be interpreted in terms of human nutrition. A diet in which all the carbohydrate is sugar or all starch is such a foreign situation that it does not tell us anything about what we really want to know. KASHA: Dr. Hegsted, there are candies and cereals that are as much an assault on carbohydrate metabolism as the rat studies in which feed- ings are 40 to 60 percent sugar. HEGSTED: I doubt that. I do not deny that there are products that are all sugar, but I do not know that anybody lives on candy bars. CHOATE: Dr. Hegsted has said that if you take in calories from sugar, you are in effect displacing other categories of food. I wonder if this is always true. Are there not, particularly among the young, some situations where the constant advocacy of sugared foods means that they in effect take in more calories than they otherwise would? This leads me to another question: Are fat cells laid down in the very young any indication of the tendency toward obesity as one gets older? What is the state of knowledge in this? Is heavy advocacy of foods to the very young, therefore, sort of doubly detrimental, not only making people obese while they are young, but obese when they are older? Can somebody speak to that point? HEGSTED: Well, I would comment on the first one. I assume that people get fat because they eat too much food. I do not think anybody denies that. I assume that advocacy of food on television may pro- mote consumption of more food, but the evidence on it is minimal. Most people have had a very difficult time demonstrating that obese children eat any more food than the non-obese. Although I do not deny that there may be something here, I think that my statement still holds, by and large, that if you consume more sugar, you consume less of something else. My understanding is that the development of adipose tissue is still a controversial area. HENRY SEBRELL: I would like to make a brief comment on fat cells, based on the work primarily of Jules Hirsch at Rockefeller and Jerry Knittle at Mount Sinai. In the general handling of obesity, a distinction can be made between childhood-onset obesity and adult-onset obesity. Childhood-onset obesity is much more difficult to handle in control- ling long-term weight loss and the maintenance of normal body weight over a long period of time. Drs. Hirsch and Knittle have shown that most of our fat cells are laid down very early in life, and that if a child or a young infant

92 is made obese by an unwise mother, this child lays down many more fat cells than a normal child. However, if a person becomes obese later in life, he does not make so many new fat cells; he primarily puts more fat into the cells that are there. HEGSTED: Well, there is a little argument over even that, Dr. Sebrell, because the question has already been raised whether you can count a fat cell that does not have any fat in it, and whether the counts are necessarily right. SEBRELL: Along that line, Knittle has recently reported that he has succeeded in cultivating fat cells in tissue culture. This will give us a new tool to work with on this problem. Fat cells without fat in them are a little difficult to identify. FREDERICK STARE, Harvard University: As Jules Hirsch and a few others have pointed out, there may be a second period in development where fat cells are laid down, and that is early in adolescence. SEBRELL: Yes, but I think not as many and so fast as those emerging in a young child. ALFRED HARPER: I do not want to get into the midst of the fat cell controversy, but I would like to emphasize the tremendous capacity for enlargement of any fat cells that exist, regardless of the num- ber that may be there, and suggest that this may be more important than the initial number. The second point I would like to emphasize is that we should look at the behavioral implications of the feeding pattern more than we look at the nature of the diet itself. Behaviorists point out that in the shift from breast feeding to bottle feeding, from the day an infant is taken away from the breast, usually one or two days after birth, it is forced to finish up everything in the bottle because the mother has been told by the physician to provide it with four ounces or six ounces or eight ounces of formula. The mother is un- happy until it is finished and so convinces the child that he or she has to finish if he or she wants any peace and quiet. I would sug- gest that this behavioral pattern contributes to the development of juvenile obesity. The final point has to do with the concern about the energy in- takes of the adolescent and the young child. Energy requirements are highest at the youngest age; as the child grows and matures, energy requirements fall. If we keep this in perspective we should recognize that the active young child, who needs a good deal of energy, may well be able to consume enough food to provide an ade- quate diet, and still consume a considerable quantity of relatively purified, high-caloric foods without serious problem.

COMMENTARY AND DISCUSSION Joan D. Gussow Richard A. Ahrens Aaron M. Altschul Frederick J. Stare Ralph A. Nelson Robert L. Glass Richard L. Veech JOAN D. GUSSOW I feel that some female voice should be heard in this Forum. I would like to speak to that last remark because of a piece of information I happen to have as a result of looking at television commercials. There was a television commercial, no longer on the air, that was aimed at children and that showed a cartoon child, whom I judged to be about two, consuming a certain kind of snack cake and gobbling them out of the box. Obviously the assumption was that you would have more than one. When I checked the caloric value of those snack cakes, I dis- covered that two of them represented 50 percent of that child's total caloric requirement. I find this particularly appalling because that snack cake gets 50 percent of its calories from fat. The calorie re- quirement of a two-year-old child is about l,000. These cakes had 250 or 300 calories apiece. In effect, then, they constituted about 50 percent of that child's caloric requirement for the day if the child consumed as many as were demonstrated on that commercial. I have kept waiting to speak today until information came up to which I could address a question. Since it has not, I am going to make a few points anyway. I am a nutrition educator. I think that there is an unfortunate separation between food and nutrition in this country. There may be an artificial separation between nutrition and health, but there certainly is an even wider separation between those who are interested in food and those who are interested in nutrition. I am interested in food, food consumption, and food consumption patterns, and there are two points that I would like to make very strongly. 93

94 First, we need percentage data on added sugar in foods. We do not have it, and we cannot seem to get it put on packages. I have been offered the data by somebody here who said he would give it to me. But I do not want it just for myself. I want to see a statement of added sugars on packages. I do not think we have to get into details about whether we ought to label oranges because they also have a lot of sugar. I just want to know about the amounts of added sugars in pro- cessed foods. I want the public to know so that they at least have a choice. At this time we are guessing. A recent guess I heard was that one of the new cereals that is being brought out may have up to 60 per- cent by weight of sugar. I think that is appalling. Dr. Beidler said there was a natural drive for sweetness and asked whether the culture tended to overdevelop that drive. Then we heard that we have very little option, that in fact it is in the foods, and it is not a question of whether one wants to add sugar or does not want to add sugar. It is there in the baked beans, in almost everything, as people who are on sugar-restricted diets know. If they have looked at food labels, they know sugar is there, but they do not know how much. The fact that one has no choice about how much sugar one takes in under certain circumstances is a very important issue. Sugar also affects the characteristics of the foods that you can market. Sugar sells foods. Therefore, you can put together a tasty confection out of very little that has any nutritional value. Fill it full of sugar, throw in a few cheap vitamins and minerals, and you can call it nutritious. I think there are some very major questions as to whether it is in fact nutritious. The snack cakes that I was referring to earlier get 50 percent of their calories from fats. Added vitamins would not compensate for their high sugar and fat content. The other thing we need is sugar consumption information. I am really tired of living on l965 household consumption data, when a gen- eration of eaters, as anyone who knows teenagers knows, lasts about three years, if that long. Teenagers are a new breed about every two or three years. I would like to know what they are eating now. I have just looked at some teenager diets, and they surprised me. The staples of the diets of this particular group are not pizzas, hamburgers, french fries -- I might feel better about it if they were -- they are cake, cookies, and milk. Don't take the milk away, incidentally, because then teenagers will be in a-lot of trouble. Somebody told me today that a lot of the sugar in those disappearance studies is going into dog food. As a cat owner I was relieved to know that cats do not have a taste for sugar, but I would still like to know which people are eating that sugar. The very fact that we do not have enough data to talk about the possible hazards of heavy sugar consump- tion is almost irrelevant if we do not know who is eating it. We do not know where it is going, and we are not even looking. I know there is a household menu census, but I am not an industry person, and I cannot get hold of that. I understand the FDA is going to pay to get it. I hope under those circumstances nutritionists will have a chance to see it.

95 I looked at that l965 curve of sugar consumption, and I said to my- self speculatively, there it goes way up there with those teenagers, and then it drops off. What happens to those people's sweet tooth? Do they keep their sweet tooth and then go on to synthetic sweeteners? Or maybe they are the ones who eat so much salt because they cannot taste anything anymore? They are used to such a high sugar intake. I do not know what happens, but I would like to know. I do not see how we can teach if we do not know what happens. Last year I was quoted in the paper as saying that I thought that it was a good thing that the price of sugar had gone up because high sugar prices would do more to improve the American diet than anything else. I received a letter from a lady, who said to me, you do not understand, some people really like to eat sugar. Then she told me that she had lost her husband during World War II and that she liked to drink tea; because it was wartime she could not get sugar; because she was dehy- drated from crying so much she had to drink lots of tea. Now, every time she has to drink tea without sugar, it reminds her of that sad time. You never really know whom you are going to offend. It is always surprising. In any case, I know that people like sugar. I just wonder whether we should encourage, to such an unregulated extent, the consumption of sugar. DISCUSSION ROSS HALL, McMaster University: I was struck by the fact that our speakers have addressed themselves to four health problems as they relate to sugar -- diabetes, dental caries, obesity, and heart dis- ease. Because these health topics have come up several times and quite a bit of data presented, I assume that one of the concerns of this Forum is the health aspect of sugar and sugar consumption. These four health questions can be attacked either by epidemio- logical means or by laboratory means. Of course some of the data are not conclusive, but my concern is that they are the kind of studies that are done with sugar'out of context, in other words, sugar per se. We have just heard from Joan Gussow and also from Mark Hegsted that sugar is not eaten by itself in the North American diet. This may be true of a very tiny portion, but the bulk of sugar is eaten in food. It is these food products that people are eating, includ- ing sugar and everything else that is in them, that concern me. If we are going to address ourselves to health problems, for which there apparently is a concern as far as this conference goes, then we should be looking at the broader question of the kinds of foods in which the sugar is placed, and the kinds of health problems that these foods generate. I think we have to enlarge our horizons.

96 GUSSOW: Let me make just one response to that. I really question whether any nutritionist now living doubts that our health would be better if we ate more complex carbohydrates, more fiber, more fruits, more vegetables, less animal protein, less animal fat, and less sugar. Therefore, in a certain sense, this discussion is moot. We are here today not because somebody many years ago decided that sugar was a good and nutritious food and needed to be introduced into the diet for the health of the populace. We are here because sugar is a very nondestructible, easily portable, cheap, and profit- able item to put into food. Somebody earlier suggested to us that we come up with a question to summarize the whole issue, and this is the question I came up with: Is there any evidence at all that present and/or anticipated levels of consumption of caloric sweeteners in the U.S. are harm- less? I would rather address myself to that question than the ques- tion of whether they are harmful. We are dealing with a food product for which we have no need. So the real question for me as an educator is, if I go out and tell people that I think they are eating too much sugar, if I go out and tell mothers I think they should stop their kids from eating so much sugar because it is bad for them, am I going to get flak from the scientists? Or am I going to be allowed to make that statement without travail, on the grounds that even though we do not have hard evidence to link sugar with a specific disease, we do know that a dietary pattern containing considerably less sugar, in which sugar is replaced by a complex carbohydrate, would be a much healthier diet? RICHARD A. AHRENS In the course of doing our research several years back, we happened to compare in the beginning experimental animals that had been fasted or not fasted before sacrifice. We were surprised to find that when sucrose was the carbohydrate in the diet rather than starch the tight- ness with which fluid was retained during this overnight fast was greatly increased, at least for a period of time. So, we began to search the literature to find out if there was any evidence that sucrose is indeed a hypertensive agent, which would give a mechanism to the relationship that Dr. Hegsted and Dr. Newberne both referred to between sugar and heart disease, diabetes, and a number of other things. Figure l is from a paper by Hall and Hall in the Proceedings of the Society for Experimental Biology and Medicine. This is not really a nutrition study. They were interested in studying hypertension and were trying to make their rats hypertensive. On the left is fluid consumption. This is when the rats were given distilled water to drink, hypertonic saline, and hypertonic saline sweetened with either glucose or sucrose. The blood pressures follow

97 *N*C1-5*GlucaM f i 3 WEEK 3 WEEK FIGURE l Fluid intake and blood pressure of rats given various types of fluid to drink. From Hall, C. E., and O. Hall, Comparative effectiveness of glucose and sucrose in enhancement of hypersalimentation and salt hypertension, Proc. Soc. Exp. Biol. Med., l23:370, l966. the pattern shown on the right side of Figure l. The lower line is the blood pressure when distilled water was given. The second line is the blood pressure when hypertonic saline was given. But the investigators were not satisfied with this. They wanted to make the animals have higher blood pressures, so they sweetened the hypertonic saline. The third line is the blood pressure when glucose was the sweetening agent. The top line is the blood pres- sure when sucrose was the sweetening agent. The kidneys were larger in the animals receiving sucrose and they had twice as many kidney lesions. A number of other people have reported that sucrose does lead to kidney lesions. This seems to be associated, at least here, with blood pressure increases. We were concerned. So we formulated a diet, simulating the American diet, with 40 percent of the calories from fat, 45 percent from carbo- hydrate, and l5 from protein. Of the carbohydrate calories, 5 percent came from lactose and in one instance the other 40 percent from starch; in the other instance, l0 percent of the calories came from sucrose and 30 percent from starch. We found that within two weeks this l0 percent of calories from sucrose made a significant difference in blood pressure. Blood pres- sure was higher, and it stayed higher for up to eight weeks. Then if we switched the diet, the blood pressure followed the sucrose, that is, the rats which had higher blood pressures suddenly had lower blood pressure when the sucrose was taken away from them. This caused us to go to a human study, depicted in Table l, from the M.S. thesis of M. L. Mclntyre, University of Maryland, l975. This is data from 26 volunteers, l2 men and l4 women, including my wife and myself, over a five-week period. We took the five different levels of sucrose shown supplementally each day, and half of the sub- jects, that is l3 of them, started at zero, the first week; the second week, 50 grams; third week, l00 grams; fourth week, l50 grams; and fifth week, 200 grams. The other half started at 200 and worked down,

98 TABLE l Mean Diastolic Blood Pressures at Five Different Levels of Supplemental Sucrose Consumed by 26 Volunteers (l2 Men, l4 Women) Over a Five-week Period. Supplemental Sucrose Diastolic Blood Pressure (g/day) (mm Hg) 0 73 .4 + l .3 a, a 50 73 .3 + l .5 b l00 75 .4 + l .3 l50 76 .5 + l .3 200 78 .2 + l .2 Standard error of the mean. Significantly lower during this week than during the week when the subjects consumed 200 g/day of supplemental sucrose (P<0.05). ^Significantly lower during this week than during the week when the subjects consumed 200 g/day of supplemental sucrose (P<0.0l). in order to get rid of any possible time effect that might complicate the experiment. The column on the right shows the diastolic blood pressures at the five different levels of sucrose. You can see at the higher level of sucrose, for the 26 people, there is a difference of about 5 points, which we got in about a week's time. The difference here is signifi- cant, that is, from 0 g to 200 g sucrose/day is significantly different at the l-percent level. However, as was pointed out, there is a biological difference be- tween people. If I had done this study only on myself, I could not confirm it, because I was not a responder. We needed to get this size difference to make it statistically significant. This is really due to more of a response in about seven to eight of the subjects who seemed to be sucrose sensitive. As you give these people the overload of sucrose, which was mentioned earlier, their blood pressure truly does jump, and, therefore, the mean assumes this magnitude. As has been alluded to earlier, I think we are talking largely about sucrose-sensitive people. How large a segment of the population are they? I have seen a number of different estimates. Just from the small sample I had here, it seemed that roughly 25 percent of our subjects responded to this overload of sucrose, and their blood pressure jumped. So we are working on the idea that sucrose is a hypertensive agent, at least in some people. This would provide the association with

99 diabetes, heart disease, and a number of other conditions that hyper- tension has already been linked with. DISCUSSION JOHN PIETZ, Psycho-Biodynamics: This brings up a point that I would like to raise. What studies have been made to show a difference, if any, between refined sugar and such forms of sweeteners as honey, molasses, sorghum, and other such products that are not as highly refined? I have seen a study in which rats were fed saline solutions along with glucose and a number of other sugars, sucrose among them, and the one sugar that they did not seem to be hypertensive on was honey. Honey was the only one of the nonrefined carbohydrates that was used in the study. I have searched the literature, using the computer at the National Library of Medicine, and have not turned up much of anything in the research to determine what difference there is be- tween these other forms of sweeteners and sugar. But this is one case where there was a definite difference. Does anybody know of anything else in this area? BOWEN: I am with the National Caries Program and can answer part of the question. Some years ago there was a study to compare the effects of refined sugar versus nonrefined sugar in caries causation in rats. When the study was first carried out it was found that the animals eating refined sugar got more caries than those that were on the unrefined sugar. The study was later repeated, and the sugars were ground to the same size. Then the differences disappeared. So, in essence, there is no difference as far as caries genicity is concerned between refined and unrefined sugars. SVEDA: I would like to make a comment about youngsters eating sugar because it is available to them. I have been told of a psychologi- cal study to determine why people like sweet things. The psycholo- gist apparently made quite a thorough study and found that sweets are really a substitute for affection and love. He then went a step further and tried to find out something about saccharin and cyclamates, which then, you see, becomes a substitute for a substitute. With the feeling at the present time that young- sters are affected by the rapport or lack of rapport with their parents, this may have bearing on a point that you are making, Dr. Gussow, and I love you for some of the comments that you made.

l00 AARON M. ALTSCHUL Let me raise an issue. What can be done in a public way now about sweeteners? I do not ask this question because I know the answer, or because I am a particularly brave person. But it seems to me that we ought to stop for a second and ask what comes out of all this in a practical way. I have had some experience in this because I was an advisor to a Cabinet officer, and both of us were frustrated. He was frustrated with me because he never got me to move fast enough, and he never got the right kind of answers to suit him. I was frustrated that, despite everything, I was forced to say things that I really did not want to say. So it is a tough problem for a person who has been a scientist to try to get down to what do you do now. There are certain properties of modern diets shown on Figure l, which is a derivation of material prepared by the FAO on the consump- tion pattern of countries as a function of income. I revised it slightly to show several points. First, the amount that you eat: the poorest countries might have l,785 kilocalories and the richest coun- tries may be having up to 3,500 kilocalories per capita per day. So as you get richer, you eat more. Second, it is not that you eat more of everything; you change your food pattern. Particularly, you reduce the carbohydrates from starch, in- crease the carbohydrates from sugar, increase animal fats and separated 3490 Calories per capita per day $us 2600 2000 1500 1 100 800 600 400 300 200 150 & Q *° "T6 20 30 40 50 60 70 80 ~90 100 100 90 80 70 60 50 40 30 20 10 Percent of total calories p—1785 Calories per capita per day—-j FIGURE l Consumption pattern of countries as a function of income.

l0l edible fats. So there is a change both in amounts and in the kind. And the last point is that those of us in the so-called wealthy nations are at the end of the line, right at the top. I would like to think a little bit later about what it means to be at the end of the line. What I have shown is a fact. There might be differences from one society to another, but that is a pretty general pattern with income. Another fact relates to the properties of modern societies. I think that in modern societies deficiency diseases can be controlled. That they are not, here and there, is inexcusable because we have the tech- nology to control them. But there are certain diseases of modern society -- Dr. Hegsted and others have pointed them out -- which appar- ently go with modernization and with higher income. So, we ask our- selves, what is the relationship, if any, between income and these diseases? I think it came out today that the diseases are complex, the etiol- ogy is complex, the elements of life-style that enter into it are com- plex, and the time scale is long. It could be from l0 to 30 years. So it is difficult, and it may even be impossible, to apply a cause and effect relationship between certain aspects of the life-style, which include heredity, smoking, exercise, and control of blood pressure on the one side and coronary heart disease on the other. How does this information help us to deal with sweeteners? The evi- dence on a particular role for sweeteners is difficult to obtain. But I think that most everybody has echoed a concern of what I am going to call "being at the top of the line": being the top calorie consumer, the top fat consumer, the top sweetener consumer, all of these things. There is a sort of general feeling, even in the absence of hard data, which may be impossible to get altogether, that one might be tempted to opt for moderation in all aspects of behavior, including eating behavior. Of course, some genetic groups or high risk groups may wish to in- clude sweeteners as part of their option for moderation. Those who are trying to control weight, if they have to make a choice, might want to choose things that are sweet as things to moderate out of their diet. And people in general may want to consider a little more emphasis on sweeteners as a way of avoiding gluttony. I think that this is a public issue. It is a public issue because the public is frustrated by the rising cost of health care and is look- ing for outs. Sometimes they are looking for easy outs, and sometimes they are looking for far-out outs. But they are looking, and I think they will continue to look and continue to be unhappy with anyone who says that there is nothing that can be done. It is a scientific issue because there are differences of opinion. You heard them today. I think, therefore, that the one thing everybody can agree on, and you have heard some people mention it, is that when there is no hard data, and when it is both a scientific and a public issue, then the public has a right to know what it is eating. The consequences of this are that somehow, by labeling procedures or otherwise, and I do not want to go into details, everybody ought to know the sucrose or

l02 sweetener content of the foods that they are eating. Then they can do what Hegsted suggested -- they can ignore it, or they can decide that "this is for me" and make changes in their diet consistent with what their needs are. There are going to have to be changes in our behavior. We are going to continue to be suspicious of being "at the top of the line." We have a Finnish person working with us to whom I always say how grateful I am to the Finns, because they have a higher heart disease rate than we do, and therefore we are number two instead of being number one. But it might be even nicer to be number three. Maybe the Finns have ideas of their own, and if they work at it, they may become number two, and we would become number one. I think that there is a desire to moderate, and I have a feeling that the first step, at least, that maybe we could agree upon is to in- sist that we know what we are eating. ARTHUR KOCH, Giant Food Company: I would like to pick up on that point and ask a general question about it. We have heard this morning about several suggestions -- warning labels, percentage disclosure, and so forth. The fact is that we do not have for many foods any requirement for even the disclosure that sugar is in that food. I have heard arguments against the warning and even some argu- ments against percentage disclosure. I think maybe we ought to throw open the question: Are there any arguments against the simple disclosure of sugar? If anybody has them, maybe they can state them. If not, maybe we can take the position that those of us here at least agree that there should be some recommendation to the Food and Drug Administration and to the Congress that sugar content be re- quired on all foods -- not percentage, but the mere fact that sugar is there. Then maybe we go further. So I throw it open as a general question. Does anybody have any reasons that we should not have at least that much on the labels? KASHA: I believe I heard in discussions that under the ingredient re- quirement "sugar added" is mentioned. KOCH: There is no requirement now. There is a requirement known as standards of identity. But sugar can be put in anything and does not have to appear on the label. For example, Coca-Cola and all the cola beverages are required to disclose caramel color data. There is no requirement that the sugar be listed on those beverages. And indeed, if you look at many of the cola drinks on the market, the label will not reveal that sugar is in there. That is true of most of your major foods on the market today. HARRY COLE, National Soft Drink Association: What the gentleman says is true as of this moment. But the FDA has propagated a new stan- dard of identity for soft drinks that requires labeling for all in- gredients. It will become effective this spring.

l03 FREDERICK J. STARE The comments that I make will refer very briefly to some of the speakers in the morning as well as the afternoon. Dr. Warren said this morning that sugar is not necessary, but that it tastes good and makes other foods taste better. I would just like to remind the audience that most people eat because eating is one of the pleasures of life. The first disease that he talked about was obesity, and here he emphasized, as Dr. Hegsted did later, that obesity is caused by a calo- ric imbalance regardless of where the calories come from. It is wrong to say that sugar is the cause of obesity. Too many calories going in and not enough used up in activity is the cause of obesity. Dr. Warren also pointed out that there is very little evidence to show that obesity per se is a hazard. But the joker in that argument is that the obese person who is 40, 45, or 50 years of age is most likely also to be an individual who has a little increase in blood pressure, who is a heavy cigarette smoker, who has an increase in cholesterol. You very seldom get to be 50 years of age and have just obesity; when you have some of these other findings, then obesity is a real hazard. I will skip the question of dental caries completely because Dr. Glass is here, and he will probably have some comments about that. He is in a much better position to make them than I am. On diabetes, I would just like to emphasize, as I believe Dr. Warren did, that sugar is not the cause of diabetes. Nobody knows what the cause of diabetes is, and I think that Dr. Leaf, Dr. Nelson, and the other physicians in the audience will agree with me that the greatest hazard to the diabetic is overweight. The best thing that a diabetic can do to help himself is to take off some weight or to keep it off if he is not overweight. On heart disease, I will simply say that I agree with Dr. Warren that those of us who have worked in this area do not have much faith in John Yudkin's arguments. In answer to Robert Choate's earlier question as to whether there is any really good evidence on what a l5-year-old needs in the way of calories, I would say that we have as good evidence as exists on the sugar consumption of adolescents. It is drawn from two studies that we did in boys' boarding schools, one of which was published in the American Journal of Clinical Nutrition, the other in Preventive Medicine. We did not tell these boys what to eat; they ate what they were fed at school. We came out quite clearly with a figure during the wintertime, the cooler months, that approximated 20 percent of their total caloric intake coming from sugars -- I am using that in the plural. During the warmer weather, in the springtime, when more ice cream was consumed, more soft drinks, more iced tea, the percentage of total calories com- ing from sugars increased to 24 percent. I know of two other studies in the literature that deal with adults. These were done in Cleveland and published in the Journal of the American Dietetic Association in l970 and in l97l. One study involved

l04 80 adult, professional men. They averaged l2 percent of their total calories coming from sugars. In the other study, dealing with patients who had multiple sclerosis, the subjects averaged ll to ll.5 percent. As I think Joan Gussow implied, it would be nice if we had better data on the sugar consumption of different groups of people and partic- cularly children. I cannot quite believe that there are many children who receive half of their total calories from sugar, but I do not know this to be a fact because I do not know of such data. On the question of the Yemenites that was mentioned this morning, we seem to forget that when they left Yemen and went to Tel Aviv, not only did they consume more sugar, but they consumed more food in general, and they all gained weight. I do not think that one can blame the sugar consumption for the fact that more of these people developed dia- betes. There were many other changes in their diet, and they gained weight. Dr. Stumpf referred to a report that George Mann had told him of some time recently, about the five-year drug study recently completed. This involved two drugs, atromid and niacin. They do lower cholesterol. But in this study the lowering of cholesterol made no difference on the incidence of heart disease. But you did not mention, Dr. Stumpf, that this study dealt with secondary prevention. The subjects had all had a coronary. What is not known yet is whether a lowering of cholesterol is help- ful to those who have not had a coronary, that is, in primary preven- tion. Many of the clinicians in this audience would say that it is highly debatable whether lowering the cholesterol by drugs or dietary means is effective in lessening the chances of a second coronary, that is, in secondary prevention. I think any physician who did not try to lower cholesterol in high-risk patients even after a first coronary might be guilty of malpractice, even though there is not any evidence that I know of to support it. On Dr. Kraybill's comments, I must be frank and say that I cannot agree .with too much of what he said, particularly on hypoglycemia. As far as I know, functional hypoglycemia is a very rare condition. It is my opinion -- and again I defer to Dr. Leaf, Dr. Nelson, or Dr. Warren -- that hypoglycemia is pretty largely a figment of the imagination. The high intakes of sugar, as both Dr. Harper and Dr. Hegsted pointed out, in various rat studies with 60 to 65 to 70 percent of the carbohydrates coming from sugar have absolutely no relevance to man. Some 20 years ago, when Oscar Portman was with us, our laboratory was one of the first to show that in the rat when all of the carbohydrate is starch, the cholesterol level goes down; when all of the carbohydrate is sucrose, the cholesterol level goes up. But we do not eat that way, and I do not know what relevance such studies have to man. I have absolutely no faith in the idea that sucrose is a potentiator of allergic reactions. Also, Dr. Hegsted made an important point when he emphasized a general decrease in total calories -- a decrease in fat, a decrease in sugar, a decrease in meat, decrease in egg. The one thing he did not mention, but I am sure he agrees with me, is a decrease in alcohol- -

l05 HEGSTED: Within limits. STARE: Within limits. This reminds me that I occasionally mention the fact that I seldom have any desserts. I love pumpkin and apple pie with ice cream or cheese. But the reason I seldom have pie is that I do know I have to control my caloric intake, and I would rather have my calories from a couple of martinis than I would from apple pie with cheese. This is just a personal choice. I would say in concluding my comments that in moderation sugars have a useful role to play in the diet. I do not care whether you are talk- ing about sucrose or fructose or honey or what. I would define modera- tion as anywhere from l0 to 25 percent of the total calories, because, after all, that does leave you with 90 percent or 75 percent of the calories to come from meat, rutabagas, milk, cottage cheese, peacjies, and other things that you might enjoy and that will provide the 50-some nutrients we need to be well-nourished. I have emphasized the point that eating is one of the pleasures of life. Why some people or why most of us enjoy sweet taste, I just do not know. As for dental caries, as far as I know the children in Grand Rapids, where they have had fluoridation as long as any other place, eat the same amounts of snacks and sweets as the children in Boston, and yet they have 60 to 70 percent less tooth decay. The reason is they have had fluoridated water, and we in Boston have not. DISCUSSION GUSSOW: I get in enough trouble when I am properly quoted, so I do not want to be misquoted on what I said. I did not say I thought most children were getting 50 percent of their calories from sugar. I said precisely the following: There is, or was, a television commercial advertising snack cakes to children that showed a child of about two eating a number of them. The amount of snack cakes consumed, when I checked the calories, represented 50 percent of that child's calories for the day. Those snack cakes get 50 percent of their calories from fat. So, I was not saying that children are getting 50 percent of their calories from sugar. HOWARD SELTZER, Office of Consumer Affairs, HEW: Dr. Stare, 'since you seem to be less concerned about the possible deleterious effects of sucrose and some of the other sweeteners, but you believe that it has a role to play if used in moderation in the diet, I wonder if you would comment on how you feel about the stating by percentage of weight of added sugar both in labeling and possibly in advertising. I am thinking of the FTC nutritional advertising proposal.

l06 STARE: I would be in favor of labeling relative to sugar. The only question that I cannot answer at this time is whether this would simply be added sugar or total sugar, because as far as I know, the effects on health are no different from the sugar that is in bananas or oranges or the sugar that may be added to orange juice or to sliced bananas when you put them on cornflakes. I do think it would be useful to know -- instead of just saying so much carbohydrates, to have a figure that states so much starch, or so much sugars. SELTZER: What about in advertising as well? STARE: I would say yes, although advertising is not my business. AHRENS: I think we should not be left with the impression that the only time we have been able to demonstrate differences between sucrose and starch is when huge quantities have been given. Dr. Kraybill mentioned a study by Roberts that was done on volun- teers in Antarctica in which no sugars were added. The subjects were asked to reduce the sugar from what I assume was ordinary in- take. Five out of l8 volunteers -- again we are dealing with this biological variability that I referred to -- showed a significant decrease in their serum triglycerides by cutting well below l0 per- cent of the calories. STARE: I do not know the study, so I really cannot comment; but I would be interested to know what period of time was involved and what happened to the weights of these individuals. AHRENS: The study was published in Lancet. HEGSTED: . I would like to comment on that. Removing all the sugar from the diet is not going to get us anywhere either. You may think it would be desirable to get all the sugar out of the diet, but let us be reasonable about this. We are talking about maybe l0 percent at the lowest level, and I think moderation is a lower level than Dr. Stare thinks it is. But we are talking about moderate amounts rather than none, I am sure. KRAYBILL: There are some comments that have been made here that con- cern me. We do many epidemiological studies in many other areas, such as cancer. There is great concern about colorectal cancer and the association with certain dietary components: we have thought of sugar; we have thought of meat; we have thought of salt, in terms of the Japanese in gastric carcinoma, and many other things. What I cannot understand is why there is difficulty in this area when you are talking about sucrose or other absorbable sugars. Why is that an exception? We make approaches in other fields. One way to look at this problem is, as is done in the pesticide area, that you either have a

l07 cause looking for a disease, or a disease looking for a cause. This is done repeatedly in many other fields. Are there population groups around the world where there are higher levels of absorbable sugars intake as compared to other areas where there is minimal level of consumption that we can look at epidemiologically? That is my first question, if someone wants to answer it. I also would like to ask if Pennington's theory as related to mechanisms of obesity is still in vogue? STARE: No. KRAYBILL: That answers that question, because it was proposed some years ago that people who got fat had a deficiency of a coenzyme, I think it was pyruvic acid conjugase. BOWEN, National Caries Program: I cannot let Dr. Stare's comments on fluoridation pass by. Let us briefly look at what is necessary for dental caries to develop. You must have a caries susceptible tooth, a cariogenic flora, and a suitable dietary substrate. Those of us concerned with trying to prevent dental caries believe that each one of those can be attacked together, and it is not a question of either/or. We believe that dietary restriction has a role to play in the control of dental caries even in those areas where water is fluori- dated. The sort of comment that Dr. Stare made -- namely, that there is a 60 to 70 percent reduction in caries in Grand Rapids, which is inaccurate anyway, and that people there eat all the carbo- hydrate they wish without risk -- is a gross disservice to the public and to those of us who are trying to prevent dental caries through all available means. Imagine the effect if dietary restric- tion is practiced in an area where the water is fluoridated. JOHN PIETZ, Psycho-Biodynamics: I was interested in the comment that sugar intake has little to do with obesity. I am wondering if that is really the case. It seems to me that there have been studies showing that a high intake of sugar tends to pervert the taste, and that you can eat a lot more of carbohydrates in the form of refined sugars than you can in some of the more natural forms such as honey or fruits. Isn't it possible that people who have a high amount of their diet from refined carbohydrates or sucrose could become obese much more easily from that source of carbohydrate than from honey, sorghum, molasses, fruit, or any of the other sources of sugar? I would particularly like to hear from Stare and Gussow. STARE: My only comment would be that I think that is the type of infor- mation one would read in Prevention. GUSSOW: I cannot understand how there would be any difference in some- thing like honey, which is really effectively hydrolyzed sugar with some trace nutrients.

l08 PIETZ: Is there not a gag reflex in too much honey that would tend to set a limit? GUSSOW: I would not know anything about that. I think if one eats sources of concentrated sugars, which most of the things you men- tioned are, that it is much easier to overconsume sugar than it is if one had to eat a sugar beet or a piece of sugarcane or some other natural source of sugar like a piece of fruit. It is very hard to consume enough apples or oranges to get the amount of sucrose you get in a candy bar. Whenever you are talking about a concentrated form of refined carbohydrate, I do not think you can make that kind of distinction. I do not have any information, but I just would not see why it would make a difference. SVEDA: May I follow up on her comment? An apple is 90 percent water. Some fruits are 95 percent water. To bear you out, you would have to eat a tremendous volume to get much sugar. MARSHA COHEN: In terms of eating up to 25 percent of your calories in refined sugar, wouldn't that present-- STARE: I said "sugars" including the sugar out of orange juice and other foods. COHEN: But you are not precluding the fact that under your theory people could choose to eat 25 percent of refined sugars, and you would still approve of their diet. STARE: Yes. COHEN: You would approve of that diet? STARE: Well, it would depend on what the other 75 percent is. COHEN: The question I would like to ask deals with that other 75 per- cent. Would it be possible for people generally -- and I am think- ing of my own situation. I am bigger than I should be and I should not eat as many calories -- to get all the necessary nutrients out of the remaining 75 percent? STARE: Yes. COHEN: I am thinking of a normal diet, and I am thinking of especially trace minerals and iron. STARE: Yes, you can. Don't worry. GUSSOW: You can do it with probably 50 percent sugar, and all liver and broccoli.

l09 KASHA: Ladies and gentlemen, I would like you to know that at least by tomorrow afternoon the chairman, either this one or tomorrow's, will welcome anonymous questions from the audience. We have one now: "Is it possible to ask if there is anyone in the audience who would oppose the percent of sugar labeling on processed food?" UNIDENTIFIED: Will you accept anonymous answers? KASHA: Anonymous answers to anonymous questions. MILTON R. WESSEL, Adjunct Professor of Law, New York University: I do not think that procedure would be fair. Somebody wants to quote the group as being unanimous. I did not come here to be forced to speak in order to avoid being quoted as part of a unanimous group on labeling. KASHA: The force of this question, as I understood it, is to make a nonunanimous statement. But I see the other interpretation, too. WESSEL: Why don't you ask what might be the reasons for not labeling? KASHA: We could well discuss that point. I believe, though, that we will leave that open until tomorrow when it is our intention to bring up all examinations of questions. I started out this morning by pointing out that this Forum is not here to make recommendations but to open questions. That is our intention, and I hope we manage to do that. PAUL KHAN, Continental Baking Company: I do not wish to be in opposi- tion to percentage labeling. But I think it is all very premature to say that we must have regulation or legislation that will request stating the percentage of sugar added to a food, when all we have been hearing all day is that none of the broad sweeping allegations of the harmfulness of sugar have really been clearly demonstrated in a cause and effect relationship, with the possible exception of den- tal caries. It would not be a surprise if the next speaker states that the percentage of sugar has little relationship to the cario- genicity or the caries-producing capability of the food in which it is contained. I am concerned about percentage labeling if it is not coupled with an educational effort. We have this experience with nutritional labeling right now. We might as well have put that nutritional in- formation detail in Greek or Arabic or some other language that is rarely spoken. The number of people who understand it is infinites- imal ly small. KASHA: Thank you. Now I would like to hold back further questions, because we have three more panel members to make statements.

ll0 RALPH A. NELSON I would like to confine my remarks to some personal observations and data we have collected at Mayo Clinic concerning sugar in applied clin- ical nutrition, and to present some thoughts about grams of sugar con- sumed per kilogram of body weight in different population groups. Hospitalized patients receiving total parenteral nutrition (some- times referred to as "hyperalimentation") via a central venous catheter near the heart receive glucose as their primary source of calories. A 50-kg patient might be given 2,000 calories per day, and this would mean about 500 g of glucose infused each day -- roughly l0 g/kg body weight. Although it is virtually impossible to eat sugar at this rate, persons whose intake is wholly parenteral for a year and more do very well from a clinical standpoint while receiving these potent solutions. Their health improves and they gain weight. The evidence from this period implies that high-glucose feedings are a safe source of calories. Sucrose is known to have a positive therapeutic effect in patients with chronic renal disease. One of the most outstanding developments in clinical nutrition during the past decade has been the use of con- trolled protein diets in treatment for these disorders. Lowering of protein intake reduces the production of the end-products of protein metabolism that must be excreted by the kidney. Accumulation of these end-products produces the azotemic or toxic state. When their produc- tion is reduced to equal the reduced excretory capacity of the kidneys, the patient feels better and begins to live a more normal life. In these patients, sucrose decreases the amount of end-products formed from protein so that there is even less to be excreted from a diet al- ready limited in protein. About 3 g sucrose per kilogram of body weight in the chronic renal diet has been prescribed. However, Palumbo, Kottke, Briones, Nelson, and Huse (l-3) have studied more than l30 men with proved coronary heart disease who had been eating about l.5 g of sucrose per kilogram. (This information was from diet interviews, and was considered accurate in the 40 to 50 per- cent of cases where the urinary excretion of nitrogen was within 2 g of calculated dietary intake.) For the study, 2 g/kg of sucrose was pre- scribed; and the increase of sucrose intake from l.5 to 2 g/kg lowered both cholesterol and triglycerides in these men with coronary heart disease. The calorie intake was kept constant to maintain weight, be- cause loss of body weight reduces blood fats rather drastically. But when sucrose was fed at 4 g/kg, which is about the limit of what is possible to eat, blood triglycerides increased moderately while choles- terol decreased. My colleagues and I concluded, in regard to people with coronary heart disease, that it is best they eat 2 g/kg body weight of sucrose or less. This is in the range of usual sugar consumption today, amounting to about 40 kg (90 Ib) a year. We further have been involved with the NIH SCOR study (Specialized Center for Research for Atherosclerosis) with Dr. W. H. Weidman, a pediatric cardiologist, as the principal investigator and P. A. Hodgson,

Ill M.S., as nutritionist. More than 2,000 Rochester school children were surveyed and their blood lipids determined (4). From this group, l04 children were chosen, some from the 95th percen- tile for blood cholesterol (264 mg/dl), some from the 90th percentile (l97 mg/dl), some from the 50th percentile (l6l mg/dl), and some from the fifth percentile (ll6 mg/dl). The consumption of sugar, about l00 to l20 g/day, and of total calories, about l,600/day, was similar in all groups. Sugar contained about 25 percent of the total calories. (Sugar was cal- culated from all sources, not separated as "refined" versus "natural.") Therefore, we could not conclude that dietary sugar was related to the level of lipids in this young population group. That is not to say that diet cannot be manipulated to treat people with excess blood lipids. Dietary manipulations do work. From the standpoint of nutri- ent intake, however, it was obvious that sugar (at least in the diet) from all sources did not seem to be related to the level of blood fats. Before summarizing, let me discuss obesity briefly. We studied thin, muscular, and obese children and found -- as others have -- that obese children ate about the same amount of calories as did thin chil- dren. There were fat children eating a lot of calories, but so were there thin children. We also noted that there were thin children eat- ing only 800 calories, but these too could be matched with fat children eating 800 calories. Obesity, unfortunately, has become a moral prob- lem ("fat kids choose to eat too much"); but this attitude is not fair. If calories are calories, less exercise by obese children than by thin ones still can lead to an increase in body fat. Most of the time and money for physical education programs are now spent on the few gifted athletes. Help is needed to provide exercise programs for obese children along with nongifted athletes. This and nutrition counseling are important aids not now given as formal treat- ment. In conclusion, I believe there are two general nutrition factors worth considering in lieu of focusing on one nutrient such as sugar in our diet. We should consider both protein consumption and total calo- rie consumption in the diet. When protein consumption is moderated, fat consumption becomes moderate too. And moderation of total calorie consumption (reducing intake as aging occurs) tends to moderate carbo- hydrate intake likewise. ROBERT L. GLASS In spite of the mixed emotions and motivations in the group here, there seems to be a trend of agreement on the association between sugars in general and dental caries. Several of the panelists have pointed this out, and Dr. Bowen has reminded us that we have known this for some time. However, there is a tendency for people outside of the research area of dental caries to oversimplify the issue. The fact is that dental caries is an extremely complex process, and instead of thinking in

ll2 terms of the old-fashioned chain of causation concept, one has to deal with the concept of a web of causation. This web is most complicated. It is very easy to oversimplify and it is very easy to equate on a one- to-one basis the ingestion of sugar and dental caries incidence. This is not the case. We should mention also the fact that certain sugars are associated with plaque formation. Dental plaque forms very rapidly in the mouth even on clean teeth. Perhaps, a better expression for it is just plain gunk, and microorganisms live in this gunk or dental plaque. Dental plaque also stores the substrate on which the microorganisms can pro- liferate and form acids that are intimately associated with the dental caries process. Again, I am oversimplifying for the sake of brevity. Dental caries is an intermittent disease. In order to carry out any sort of prospective study, one needs a lot of time and a large amount of money. One of the classic studies in this particular field, that really should be updated (if one could get it through one's committee on human studies) is the Vipeholm dental caries study carried out in Sweden some 20 years ago. There were a number of experimental groups in this study. In one group they fed something in the order of 300 grams of sucrose in solu- tion at mealtime. The other group that I will speak of, and skip the others, is a group that got something in the order of 70 or 80 grams of sugars in sticky toffees and caramels that were fed in between meals. There was approximately a tenfold difference in caries incidence, with, of course, the group that had the sticky sweets in between meals get- ting the tremendous increase in tooth decay, whereas the group that had the 300 grams of sucrose in solution at mealtime had a barely detect- able or measurable increase in tooth decay when compared to the control group. It is for this reason that it is not very scientific to attack all sucrose-containing foods as being highly cariogenic. In fact, some of the research that should be done is an identification of the hierarchy of cariogenicity of foods so that we may concentrate on eliminating these or somehow or other altering these in our programs of prevention. Again it is the sticky, in-between-meal sweets that we should be focus- ing on, and not necessarily all sucrose or all foods containing sugars. I was a little bit disappointed that only one of the speakers, most of whom are familiar with the problem associated with dental caries and sugar, mentioned fluorides. One particular method of viewing dental caries is that it is in part a fluoride-deficiency disease. It is also an infectious disease, and investigators at the National Institute of Dental Research, the Forsyth Dental Center, and several other places have identified an organism, namely streptococcus mutans, which is intimately associated with plaque formation and dental caries. Dr. Bowen spoke of a multiple attack on the caries process in order to eliminate it. Many of us believe that dental caries could virtually be eliminated through a combination of, for example, the following: Fluoridation is absolutely basic and something that all people in the health field should be working for.

ll3 We would like to work with agents that can have some sort of anti- bacterial effect to help supplement the preventive effect of fluorida- tion. We also would like to think in terms of being able to improve the oral hygiene of people. But this is extremely difficult and has been successful only in certain dedicated groups, and it is not a very prac- tical public health measure. Nor is it very practical to tell young- sters to avoid eating candy. You can tell them that, but it doesn't work. It did not work with my own children. It did not work in the preventive dentistry programs that I have tried to set up in dental schools in the past, except with a very small, highly motivated group of people. As a practical measure, in this regard, it seems to me, we need to have the development of food additives, some of which are al- ready being researched, that will in fact neutralize the effect of sugar in those foods that are identified as highly cariogenic. Along these lines, of course, we need the cooperation of the Food and Drug Administration in the approval of INDs and ultimately the granting of NDAs, assuming that the material in question is safe and effective. This constitutes a major difficulty. Somebody earlier today asked how the Food and Drug Administration might help in this re- gard. I would urge that it help some of us clinical researchers in this particular area. As I pointed out before, there are a number of people here with dif- ferent backgrounds, different ideas, different fields of research, dif- ferent motivations, and different emotions. If some of the do-gooders, perhaps I should say advocates, for example, of controlling Saturday morning television for children, and advocates of food labeling and so forth would help those of us who are working for fluoridation, and really put the pressure on their state legislatures and on people who are spokesmen for the community, they could do a fine job in reducing the incidence of dental caries and the family dental bill as well. RICHARD L. VEECH I would like to take slight issue with some of the speakers. First, at the present time, the state of biochemical knowledge is such that we could define pretty clearly what in fact controls the rate of fat syn- thesis. I will discuss just this aspect -- the control of fat synthe- sis -- and leave aside the question of the control of cholesterol synthesis. It is generally accepted, although there seems to be some confusion here, that sucrose is about the most lipogenic substrate one can use if one is intending to increase the rate of fat synthesis. I will quote only from animal studies, as such studies are not done in humans. If we feed a rat, let us say, an NIH diet, we can measure most of the parameters that control the rate of fat synthesis. These are really only two -- the Vmax of fatty acid synthase and the concentra- tion of malonyl CoA. We will leave aside the question of the control

ll4 of acetyl CoA carboxylase. I'f we know the malonyl CoA content, we know what acetyl CoA carboxylase is doing; we can then measure the rate of fat synthesis. The rat on an NIH diet (which is largely starch and about 4 percent fat) will synthesize fat at the rate of 0.45 micromoles of C2 units per minute per gram of liver. I might just add that as far as we know, in both the rat and in man, the liver is the main site of fat synthesis. This is de novo fatty acid synthesis. On this normal diet, the rat's fatty acid synthase will have an activity of about 0.7 micromoles per minute per gram. That is the maximum he could synthesize, or the vmax of fatty acid synthase. His malonyl CoA, after meal feeding, will be about 0.025 micromoles per gram. We can switch this animal's diet to include 50 percent sucrose. That is twice the normal sucrose intake said to be eaten by young Americans. We can feed him starch or glucose or sucrose or fructose or a mixture of the two for three days, and then we can measure all these parameters. What comes out, leaving aside all the numbers, is that sucrose increases the activity of fatty acid synthase from 0.68 to 2.6 micromoles/min/g in three days. This will really quadruple the capaci- ty of the liver to synthesize fat, and it will in fact quadruple the rate of fat synthesis as measured by 3H uptake from 0.45 ymoles C2 units/min/g liver to 2.6 pmoles/min/g. Under these conditions, sucrose is more lipogenic than starch, glucose, fructose, or any combination. Now a quadrupling of the rate of synthesis of fatty acid after sucrose feeding need not necessarily be reflected in an increase in the steady state level of the blood triglyceride. We all realize that without an increase in oxidative phosphorylation or some burning of this fat, something is going to have to be done with this excess fat pro- duced. So, I think the conclusion that eating a diet composed of 25 percent sucrose is benign may be a bit premature. In fact, more param- eters really need to be studied. I think we need to know what the rate of synthesis is, what the turnover rate is. A mere study of the steady state levels of blood lipids is not necessarily going to give us the answer, if we are asking Dr. Yudkin's question. The relationship of sucrose to atherosclerosis in general is the question that seems perti- nent here. Second, I tend to disagree with Dr. Hegsted and Dr. Stare when they say all calories are equal in the rate of fat synthesis. As they well know, acetate, butyrate, or any ketogenic substrate will not form fat. Sucrose really forms fat very much better than glucose, or glucose and fructose, in fact any mixture; and I would think that could be substan- tiated. Third, in regard to Dr. Newberne's experiments, many of the clinical studies are terribly confusing, and as Cecil and Loeb's textbook would have said, you must have the prior dietary history. Since one impor- tant control point is the Vmax of fatty acid synthase, which is subject to dietary induction, you must know the prior diet of the animal before you can draw any conclusion. If you just grab some people off the street and shove sucrose down them and measure their blood lipids, you may or may not find any changes since the Vmax changes in the enzymes

ll5 take eight hours or so. These figures that I am giving you are after three days. If you test this after one day, of course, you find no effect of sucrose. And, of course, one can feed sucrose for 28 or 30 days; and then, the question comes up about other dietary deficiencies that might change the results. There are many questions that can be asked on this subject, and we have the tools to answer them. I think that we are going over a lot of older studies that perhaps were not done adequately, when we now have the methods to answer these questions using proper tools. My recommendation to the FDA would perhaps be to convene fewer expert panels and do more research. (I have heard an expert defined as a "drip under pressure"; since I am brought here as an expert I suppose I qualify under this definition.) But I think we have the tools, and these are questions one can answer. Obviously, the relationship be- tween the rate of fatty acid synthesis and the process of atherosclero- sis is a complex one. I do not think anyone knows exactly the relation- ship between blood lipid levels and atherosclerosis, and there is no point in arguing about it. These questions about what controls the rate of synthesis are very well known, and I think an understanding of the control of cholesterol synthesis is not very far off at all. So, my recommendation to the FDA would be to get some good new studies done and you won't need to con- vene panels of experts -- the answers will be obvious. JOAN D. GUSSOW I think that what troubles me as a nutritionist and as a nutrition edu- cator is the basically defensive posture that we are put in relative to the food supply. Sugar is just a very special case of a more general situation. Anyone who wants to put anything at all in the food supply may do so whether or not that item has any role at all in a food supply that al- ready has l0,000 items. Incidentally, if you want to know what is com- ing in, read Advertising Age. There is a new cereal called Norman coming up, and a number of other interesting additions. Anyone who wants to put anything in the food supply may do so, as long as it is made with FDA certified chemicals. There is no require- ment that a need be demonstrated for such a product, that the product will do anyone any nutritional good, or even that it will not do them harm. It is simply free to be introduced into the food supply. If those of us who are educators wish to make some kind of move, as some of us have, to say, "enough already, 62 cereals is enough, do not bring out another one," we are told -- not in effect, but literally -- to prove that it will do harm. What we have heard here all day long is that we cannot even prove clearly that something that is in the diet at the level of 25 percent or 30 percent of the total diet does harm, because we do not have the epidemiologic tools to do so. So there is no way of knowing whether

ll6 that last little plastic can of pudding, or that last little synthetic bread stick or whatever it is that we are being introduced to, is the straw that broke the nutritional camel's back. We are simply on the defensive, and we have very few weapons. I came here today hoping very much to hear scientific information that would back me up if I went out on the barricades and fought against the influx of junk into the marketplace; that would give me the confi- dence of knowing that I wouldn't be clobbered from behind by my sup- posed allies. I hoped I would learn that I had scientific support in trying to change diets. I do not feel that I do, and I am sort of de- pressed about that. I mean all I really have is people saying, "Yes, we agree with you that if you make certain kinds of changes in the diet that you are talking about, it would not do anybody any harm." I guess that is what I am going to have to go out and tell people. KASHA: Thank you. I am sure we now are past saturation on the absorp- tion, not of sugars, but of new facts and new ideas. We will resume our search for them tomorrow morning. I thank all of you for your attention and participation.

DAY II