Review Chocolate-divine food, fattening junk or nutritious supplementation?
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
European Journal of Clinical Nutrition (1997) 51, 341±345 ß 1997 Stockton Press. All rights reserved 0954±3007/97 $12.00 Review ChocolateÐdivine food, fattening junk or nutritious supplementation? S RoÈssner Obesity Unit, Karolinska Hospital S-171 76 Stockholm, Sweden Descriptors: body weight; chocolate; cocoa bean; craving; sugar Introduction designated to plant these beans were kept away from their women several days before the work. Their opportunity to Chocolate is an interesting food for numerous reasons. Its meet again with their partners immediately after the beans historical background and mystical nimbus, its hedonic had been planted would ensure that their sexual activity potential and its nutritious properties have all contributed was also transferred into the growth of the beans. to make chocolate a fascinating and controversial food component. This review will describe brie¯y the historical background and the development of modern chocolate products, the sensory and hedonic properties of chocolate, Cocoa beans to Europe the medical risks and bene®ts associated with chocolate The beverage which Christobal Colon and later Hernando consumption and some of the modern myths surrounding Cortez encountered in America was something quite dif- chocolate. ferent from what we today regard as chocolate. `Xocoatl', meaning `noisy water' (Mueller, 1957), was made of roasted ground beans resembling cocoa mass. Spices were Historical background (Nielsen, 1995) added which made the drink either red or black. It was said The cocoa bean has a complex historical background. that Montezume consumed more than 50 golden cups of Besides being an important ingredient in a drink only this drink daily. Christobal Colon returned in the beginning available for the highest in the Aztec society, the cocoa of the 16th century with some cocoa bean samples, but in beans were as close as one could get to a uni®ed monetary Europe nobody showed great interest at that time. Twenty system in the middle Americas of the Aztecs and the years later the interest had grown and it became suddenly a Mayas. Tax was paid in cocoa beans and it is possible to habit, con®ned to the Spanish nobility to consume the reconstruct price lists from that time. A large pumpkin or a chocolate drink. The secret was kept with the help of turkey was 4 beans, a prostitute over night 10 beans. For monks and the harvest from the colonies was strictly 100 beans a healthy slave could be bought. The role of the reserved for the high class of the Habsburg empire. As cocoa bean as a combined nutrient and monetary system with any other food or drink which is available in restricted was obviously unique. In a letter to the pope Clemens VII amounts, chocolate was immediately given special proper- Peter the Martyr writes about the cocoa beans and describes ties. Both Madame DuBarry (later called Madame Pompa- them as `blessed money which prevent the owner from dour) and the venetian Casanova argued that chocolate greed, since they cannot be stored or hidden in the earth'. functioned as an aphrodisiac. In the middle of the 17th The fact that the cocoa tree was the basis for a monitory century Dr Henry Stubbs wrote a thesis about chocolate, system gave it its ®rst Latin name Amygdalae pecuniariae, called `The Indian nectar' stating that `chocolate literally `money almond'. However, Linne later exchanged . . . provokes lust like nothing else'. Statements of this the term and rather underscored the divine and culinaric kind could only have one effect: The interest in chocolate aspects of the tree products, calling it Theobroma cacao. rose dramatically. Plantation of the cocoa tree was a holy ritual and the Since chocolate had become so popular in 16th century farm workers of the Maya people had a god of their own Spain it also came under the scrutiny of the pope. In 1569 who was worshipped to ensure a rich crop. This god Pius V was served a cup of chocolate and found it so Ekchuah was thoroughly celebrated in a happy ceremony disgusting that he was convinced that nobody would make which did not require any human sacri®ce, only the killing drinking chocolate a habit. For this reason it was not even of a dog, provided it had a cocoa coloured patch in its skin. considered necessary to ban the decoction. However, other When the beans were planted this had to take place during pure catholics were unhappy with this decision and numer- optimised circumstances. The beans were exposed for ous dissertations were written to establish whether choco- moonlight for four nights and the farm workers who were late was a food or a drink and whether it would be allowed to consume it on holidays and during fast. In 1662 the Received 10 November 1996; accepted 30 January 1997 cardinal Brancaccio declared that `liquidum non frangit
ChocolateÐdivine food S RoÈssner 342 jejunum', that is drink does not break the fast, but the which results in temperatures above 50 C, the beans lose controversy lasted into the next century. Drinking chocolate their adhering pulp and after 4±6 d are ready to be dried. became so popular that high society ladies even had cups of After one week of drying in good sunshine a stable product the drink brought into church during mass. When Anne of is obtained which can be stored and transported. However, Austria, daughter of the king of Spain, married Louis XIII not even at this point has the distinctive chocolate ¯avour of France in the year of 1615 she brought with her not only yet been developed. In spite of considerable work it has not her own chocolate but also a servant named LaMolina been possible to identify the characteristic chemical com- whose main chore it was to prepare the chocolate. Choco- pounds which constitute the chocolate ¯avour. More than late became immediately very fancy and in several Eur- 400 identi®ed compounds in the end product seem to be opean countries the sovereigns in vain tried to prevent the responsible for the ¯avour and aroma we refer to as consumption. After some years it was realised however that chocolate. chocolate houses might be a better alternative to the coffee During roasting the moisture of the bean is further shops, which in turn had developed as alternatives to the reduced to below 2% which also makes it easier to pubs. The hope of the authorities was that there would be remove the adhering shell. After milling to a particle size less gambling, intrigue making and conspirations in the < 100 m a suspension of solid cocoa particles within a chocolate houses than in the preceeding coffee houses, a matrix of liquid cocoa butter is obtained. This material hope that certainly was not ful®lled. Chocolate spread all named `cocoa mass' or `cocoa liquor' is the starting over Europe and when it could not be prohibited, the material for the further manufacture of chocolate. The governments at least tried to tax the produce exactly as cocoa press invented by van Houten made it possible to had been done with liquor, coffee and tobacco. separate the cocoa butter from the fat free cocoa solids, During these years the product had developed and was which contains far the main part of the characteristic cocoa now served as a sweet drink containing the entire compo- ¯avour. The cocoa butter is unique due to its composition nents from the dried and roasted cocoa bean but with sugar of 35% stearic acid, 35% oleic acid and 25% and vanilla added. Sometimes starch was added as initially palmitic acid, which, along with a considerable amount in Mexico to provide a more homogenous soup. Still the of a-tocopherol, determines its natural stability against preparation of a cup of chocolate was a tidious and hard oxidation. work requiring 30 min of whipping of the hot mixture in a In the production of chocolate products particle size bowl. reduction is necessary to give chocolate the creamy texture and hedonic properties that chocolate consumers expect. With smaller particle sizes of ingredients such as dried Modern chocolate products milk, sugar and cocoa solids higher quality follows. An In 1828 a technical breakthrough came. The Dutch Coen- important property of chocolate is that the melting curve raad van Houten separated the cocoa butter and milled the for cocoa butter shows a transition from a hard material to remains to cocoa powder. Peter Vevey in Switzerland liquid oil at mouth temperature. added milk to the product in 1878 basically creating the Chocolate has a strong and complex ¯avour. This is one modern chocolate of today. reason why chocolate can be used to mask undesirable Chocolate has a ¯avour and a texture unlike any other ¯avours and thus help to convey acceptability to foods and (Morgan, 1994). The tree and the ¯ower from which it beverages which are nutritionally superior or to drugs grows, the pod and its seed and the treatment of the product which might be more easily consumed by children if a are different from those of any food ingredient. Cocoa disagreeable taste is masked by the more attractive choco- beans are a tree crop from small trees, generally planted late (Hoskin, 1994). under larger trees providing shade and protection from the wind. Most cocoa beans are harvested in geographical areas Chocolate consumption patterns close to the equator ( 10 north and south). Cocoa beans were originally cultivated by the Aztecs The per capital chocolate candy consumption shows an and the Mayans from Central and South America. After the interesting distribution pattern which only to a small extent discovery of this part of the world the cocoa trees gradually agrees with data for overweight and obesity distribution spread over other tropical parts of the world and today (Seligson et al, 1994). The highest consumption is reported almost 50% of the entire cocoa production comes from from Switzerland (9.9 kg/y). In the United States, which West Africa. A considerable part is also grown in South has the highest mean body mass index in the western world, East Asia (Indonesia and Malaysia). the chocolate consumption is less than half of the Swiss The cocoa tree is peculiar in several respects (Hoskin, ®gures (4.6 kg/y). Interestingly, Denmark, Norway and 1994). In contrast to most fruit or nut trees which have a Sweden are among the top ten states with Norway second distinct time for pollination followed by a well-de®ned in the world (7.9 kg/y). The Finnish annual consumption is harvest time, cocoa trees continuously produce ¯owers and only 3.5 kg/y in a country known for a high prevalence of mature fruits at the same time. In some parts of the world obesity, coronary artery disease and premature athero- cocoa pods can be harvested during nine months of the sclerotic complications. Of the countries in South East year. The pods are cut from the tree by hand and the beans Asia only Japan is found with 1.7 kg/y. are removed from the pod by a manual procedure. The American studies have described the chocolate candy cocoa butter and the cocoa solids are isolated within the eating pattern during the day in the 1987±1988 National cotyledons of the bean and surrounded by water which Food Consumption Survey which demonstrates that only must be removed. The cocoa taste develops during the 34% of all chocolate is consumed at breakfast, lunch and ensuing two processes of fermentation followed by drying. dinner (Table 1). The rest of the chocolate consumption can During fermentation pulp sugars are converted to alcohol be observed as snack foods generally starting in the early and from alcohol to acetic or other organic acids in addition afternoon. During the time period 20.00±24.00, 22% of all with several other biproducts. During this active process chocolate consumption takes place. More chocolate is
ChocolateÐdivine food S RoÈssner 343 Table 1 Chocolate consumption pattern over the day. Data from US National Food Consumption 1990 their patients to avoid chocolate products and in many cases this approach seems to be founded more on moral Chocolate time % than on scienti®c grounds. Some recent studies have demonstrated that chocolate in the diet may not have the Breakfast 2 Lunch 18 adverse effects anticipated. Kris-Etherton et al (1993) Dinner 14 studied the effects of whole food diets high in stearic acids where one of the experiments included cocoa 34% butter. They study demonstrated that this diet elicited a Snacks 00±04 h 2 neutral cholesterolemic response in comparison with a 04±08 h 1 corresponding diet where the fat source included dairy 08±12 h 7 butter. In a later study (Kris-Etherton et al, 1994) it was 12±16 h 21 demonstrated that when a milk chocolate bar was con- 16±20 h 13 sumed instead of a high-carbohydrate snack HDL-concen- 20±24 h 22 66% tration increased and plasma triglycerides fell signi®cantly. 100% The authors speculate that the relative hypocholesterolemic effect of the cocoa butter diet can be attributed to the high proportion of stearic acid. In treatment of diabetic teenagers the need for three consumed in the winter than in any other seasons. Overall main meals and two snacks is generally underscored as a chocolate candy seems to be a minor contributor (0.7± cornerstone of the dietetic treatment. For historical reasons 3.4%) to the dietary intake of total energy, fat, saturated milk chocolate bars and potato chips, which obviously are fatty acids and stearic acids. This of course does not popular snacks among teenagers, have been considered exclude the fact, that for individuals, chocolate might be inappropriate for these patients. Cedermark et al (1993) even the main source of energy intake. decided to evaluate the glycemic response of these popular snack foods compared to a more regular reference diabetes Nutritious properties of chocolate snack (sandwich and apple). The study demonstrated that an occasional exchange of such regular diabetes afternoon The major component of cocoa mass (cocoa liquor) are: snack for an isocaloric amount of a milk chocolate bar or 54% cocoa butter, 11.5% protein, 9% cellulose, 6% tannic potato chips did not have any negative effect on the acids and colour, 5% water, 2.6% mineral salts, 1.2% postprandial blood glucose curve and thus is helpful in theobromin, 1.0% sugars, 0.2% caffeine and 9.5% organic providing these children with a wider and more acceptable acids and aromas (Nielsen, 1995). Depending on the alternative of snack supplementation. individual chocolate manufacturers recipies the amount of cocoa mass will be present from 7±15% in milk chocolates and from 30 > 70% in dark chocolates. Chocolate craving In addition to these well known ingredients it has A patient coming to an appointment in our obesity unit recently been discovered that chocolate is a major source describes her reason to choose chocolate as her main source of dietary copper (Joo & Betts, 1996). In a study of of nourishment in the following way: `I hate cooking, I live American dietary habits using 3 d dietary records it was on my own, I love chocolate, chocolate makes it quick to demonstrated that dark chocolate and candy made the prepare a meal, I enjoy this meal thoroughly, eating highest contribution to the mean daily copper intake chocolate requires little work in the kitchen, no cleaning, (22% of the estimated safe and adequate daily dietary chocolate eating is satisfying, affordable and gives plea- intake). The other major contributors were various sure, so why should I not live on chocolate?' chocolate-containing foods and drinks. Although craving is a common occurrence the literature Recently Waterhouse et al (1996) quanti®ed the exploring the basis for such cravings is scarce. Chocolate is amounts of antioxidants in chocolate. Flavonoid phenolics the most commonly craved substance in the United States in chocolates prevent the randi®cation of the fat in choco- (Weingarten & Elston, 1991). It is not clear whether the late (a quality which actually made it possible to use craving can be attributed to a pharmacological effect of chocolate as an important source of energy for soldiers chocolate compounds such as phenylethylamine or magne- during heavy combat). Cocoa powder was found to contain sium, or whether sensory properties are more important. 20 mg of total phenol gallic acid equivalent per gram. In Since a pre-menstrual desire for chocolate has been terms of consumption patterns this means that a 40 g demonstrated (Tomelleri & Grunewals, 1987; Rozin et al, chocolate bar contains the same amount total phenol as a 1991; Buffenstein et al, 1995) it seems reasonable to glass of red wine. The authors conclude that if dietary believe that there is a physiological or hormonal basis for phenolics are active also in vivo as antioxidants chocolate this craving. Michener & Rozin (1994) speculate along provides a signi®cant portion of such dietary antioxidants. these lines and have set up a clinical research protocol based on the predictions shown in Table 2. By breaking down chocolate into its components it was possible to Clinical observations isolate the sensory, nutritive and pharmacological compo- There is no doubt that chocolate products are high in nents from each other together with a placebo. This study energy, most of which comes from saturated fats and suggested that there was no role for the pharmacological sugar. However, since a signi®cant proportion of the effects in the satisfaction of chocolate craving but a certain triglycerides in chocolate originates from stearic acid, it role for aroma independent of sweetness, texture and has been questioned whether chocolate consumption has energy. any negative metabolic effects apart from the energy Chocolate craving is a strange phenomenon which is contribution. Most dieticians will tend to recommend still poorly understood. It has previously been confused
ChocolateÐdivine food S RoÈssner 344 Table 2 Theories of chocolate craving and their predictions (Michener and Rozin, 1994) Theory Milk choc. White choc. Cocoa caps. Placebo White choc. Empty cocoa caps. Sensory 0 0 0 Nutritive Calories 0 0 0 Magnesium 0 0 0 Pharmacological 0 0 0 Result 0 0 0 full effect; partial effect; 0 no effect. with the term carbohydrate craving (Wurtman, 1987) but it feelings of guilt or relaxation. These authors conclude that is obvious that when subjects reporting craving are inter- although chocolate provides initial pleasure such pleasure viewed about what they crave for, it is chocolate rather than is short-lived and soon accompanied by feelings of guilt. pure sugar that may cause them to leave home at midnight in search of this desirable food. The association of choco- Other clinical complications late craving has ben compared with that of certain drugs induced psychotic situations, suggesting that pharmacolo- Chocolate has been said to provoke various allergic reac- gically active substances could be involved. Schifano & tions (Fries, 1978). However, in a critical analysis, Fries Magni (1994) have described seven subjects with a history concludes that the evidence is generally weak. Other of drug abuse of MDMA (`ecstasy'). These six young men medical problems attributed to chocolate include pruritus, and one woman describe a distinct pattern: after having rashes, rhinorrea and asthma (Ghosh, 1977; Fries, 1978; given up the MDMA, a craving for chocolate developed Wilson, 1985). Further medical complications include acne, with an intake of up to 2000 calories of chocolate candy per childhood hyperactivity, migraine and irritable bowel eating episode. Sometimes these were described as binges, symptoms (Kaplan et al, 1989; Grant, 1979; Friedman, sometimes not. The typical pattern was an episode of 1989). Since the methylxanthines in cocoa can relax the chocolate craving about once a week with an intake of lower esophageal sphincter re¯ux esophagitis has been about 1000±1500 calories per episode. In most cases a reported (Murphy & Castel, 1988). weight loss was observed of up to 14 kg before the On the other hand chocolate has been described as an development of the craving. The authors conclude that all interesting functional food by Kurihara, who reported that seven MDMA abusers exhibit striking changes in appetite chocolate products, preventing dental decay, have been and food preferences which could be explained by a developed in Japan (Kurihara, 1991). decreased central 5-hydroxytryptamin functioning. They speculate that chocolate which contains phenylethylamine Miscellaneous about chocolate with amphetamine-like properties and the caffeine-like substance teobromin could be useful in counteracting the Deeply entrenched in obstetric and midwifery folklore is effects o MDMA withdrawal. the hypothesis that the weight of candy gifts is proportional These observations are brought forward by the DiMarzo to the birth weight of the newborn. So far this hypothesis et al (1996) and DiTomaso et al (1996), demonstrating a has never been scienti®cally tested until Nordin (1993) novel group of pharmacological constituents of chocolate. conducted a prospective study of postpartum candy gift net Anandamide is a brain lipid which binds to cannabinoid weight and its correlation with birth weight. Thirty-nine receptors with high af®nity and resembles the psychoactive chocolate gifts were recorded during the study period effects of cannabinoid drugs. The authors sampled cocoa during a time period when 1491 infants were delivered. powder or chocolate and demonstrated that these products The box candy net weights ranged from 200±1200 g, the contain three unsaturated N-acylethanolamines that could birth weights from 2260±4926 g. However, the coef®cient act as cannabinoid mimics either directly by activating the of correlation between candy and birth weights was only corresponding receptors or indirectly by increasing the 0.15 (P 0.2). The author assumes that socioeconomic and anandamide levels. Thus it is possible that the euphoria prematurity-related factors were responsible for the poor described by many chocolate cravers could be attributed to candy giving. Possibly parents of high birth weight infants the fact that chocolate products contain substances which are themselves overweight or obese and ®nd it impossible resemble brain cannabinoids in their action. to express their gratitude with a chocolate box of corre- Hetherington & Macdiarmid (1995) have analysed the sponding signi®cant proportions. The design of this study response of chocolate overeaters to different presentations could have been ¯awed by the fact that rumours spread in of chocolate and found that these subjects exhibit repro- the hospital that weight data would be used retrospectively ducible changes in appetite associated variables, when to test the separate hypothesis that midwife waist measure- exposed to chocolate in different forms. These authors ment would be directly proportional to the net weight of the suggest that chocolate overeating individuals can provide boxed candy. This was expected to result in severe under- information also about other aberrant eating behaviours. reporting. In a further study of chocolate addicts Macdiarmid & Rakatansky (1995), disturbed by the report that choco- Hetherington (1995) studied twenty self-identi®ed addicts late craving could be cured by bupropion (Michell et al, together with matched controls. As expected, the addicts 1989), suggests that such a drug might be detrimental to reported a larger amount of chocolate consumption that health and happiness. If chocolate craving is cured, what controls but also that eating chocolate did not improve would the consequences be? Deep depression, a baby boom
ChocolateÐdivine food S RoÈssner 345 when sex replaces chocolate craving or other unwanted dairy butter, and milk chocolate on the plasma lipids of young men. consequences of approaches to compensate for the choco- Metabolism 42, 121±129. Kris-Etherton PM, Derr JA, Mustad VA, Seligson FH & Pearson TA late loss? (1994). Effects of milk chocolate bar per day substituted for a high- Chocolate addiction is a documented problem (Max, carbohydrate snack in young men on an NCEP/AHA Step 1 Diet. Am. J. 1989). A very personal testimony came from Doctor Clin. Nutr. 60 (suppl 6), 1037S±1042S. Patterson (1993) who eloquently has described his life as Kurihara K (1991). In functional foods. Nutek Report 1992 68, 42. Macdiarmid JI & Hetherington MM (1995). Mood modulation by food: An a chocoholic and revealed that chocolate made it possible exploration of affect and cravings in `chocolate addicts'. Br. J. Clin. for him to pass his exams or overcome low spirits. His Psychol. 34, 129±138. dietary combination of a huge brownie plus half a litre of Max B (1989). This and that: chocolate addiction, the dual pharmacoge- chocolate milk eventually served as a form of aversion netics of aspargus eaters, and the arithmetic of freedom. Trends therapy and changing his addiction from chocolate to . . . Pharmacol. Sci. 10, 390±393. Michell GE, Mebane AH & Billings CD (1994). Pharmacological versus hot buttered popcorn. sensory factors in the satiation of chocolate craving. Physiol. Behav. 56, 419±422. Morgan J (1994). Chocolate: a ¯avor and texture unlike any other. Am. J. References Clin. Nutr. 60 (suppl. 6), 1065S±1067S. Mueller W (1957). Seltsame Frucht Kakao. Gordian p. 18. Buffenstein R, Poppitt SD, McDevitt RM & Prentice AM (1995): Food Murphy DW & Castel DO (1988). Chocolate and heartburn: evidence of intake and the menstrual cycle: A retrospective analysis, with implica- increased esophageal acid exposure after chocolate ingestion. Am. J. tions for appetite research. Physiol. Behav. 58, 1067±1077. Gastroenterol. 83, 633±636. Cedermark G, Selenius M & Tullus K (1993). Glycaemic effect and Nielsen N (1995). Chocolate. Stockholm: Trevi. satiating capacity of potato chips and milk chocolate bar as snacks in Nordin AJ (1993). A prospective study of postpartum candy gift net teenagers with diabetes. Eur. J. Pediatr. 152, 635±639. weight: Correlation with birth weight. Obstet. Gynecol. 82, 156± DiMarzo, Fontana A, Cadas H, Schinelli S, Cimino G, Schwartz JC & 158. Piomelli D (1994). Formation and inactivation of endogenous cannabi- Patterson R (1993). Recovery from this addiction was sweet indeed. Can. noid anandamide in central neurons. Nature 372, 686±691. Med. Assoc. J. 148, 1028±1029, 1032. DiTomaso E, Beltramo M & Piomelli D (1996). Brain cannabinoids in Rakatansky H (1995). Chocolate: pleasure of pain? Rhode Island Med. J. chocolate. Nature 382, 677±678. 78, 179. Friedman G (1980). Nutritional therapy of irritable bowel syndrome. Rozin P, Levine E & Stoess C(1991). Chocolate craving and liking. Gastroenterol. Clin. North Am. 18, 513±524. Appetite 17, 199±212. Fries JH (1978). Chocolate: a review of published reports of allergic and Schifano F & Magni G (1994). MDMA (`ecstasy') abuse: psychopatho- other deleterious effects, real or presumed. Ann. Allergy 41, 195± logical features and craving for chocolate: a case series. Biol. Psychiatry 207. 36, 763±767. Gosh JS (1977). Allergic reactions to chocolate. Am. J. Clin. Nutr. 30, Seligson FH, Krummel DA & Apgar JL (1994). Patterns of chocolate 834±835. consumption. Am. J. Clin. Nutr. 60 (suppl. 6), 1060S±1064S. Grant EC (1979. Food allergies and migraine. Lancet 1, 966±969. Tomelleri MS & Grunewald KK (1987). Menstrual cycle and food Hetherington MM & Macdiarmid JI (1995). Pleasure and excess: liking for cravings in young college women. J. Am. Diet. Assoc. 87, and overconsumption of chocolate. Physiol. Behav. 57, 27±35. 311±315. Hoskin JC (1994). Sensory properties of chocolate and their development. Waterhouse AL, Shirley JR & Donovan JL (1996). Antioxidants in Am. J. Clin. Nutr. 60 Suppl. 6, 0168S±1070S. chocolate. Lancet 87, 311±315. Joo S-J & Betts NM (1996). Copper intakes and consumption patterns in Weingarten HP & Elston D (1991). Food cravings in a college population: chocolate foods as sources of copper for individuals in the 1987±88 A questionnaire study. Appetite 15, 167±175. nationwide food consumption survey. Nutr. Res. 16, 41±52. Wilson NM (1985). Food related asthma: a difference between two ethnic Kaplan BJ, McNicol RD, Conte RA, Moghadam HK (1989). Dietary groups. Arch. Dis. Child. 60, 861±865. replacement in preschoolaged hyperactive boys. Pediatrics 83, 7±17. Wurtman RJ (1987). Dietary treatments that affect brain neurotransmitters: Kris-Etherton PM, Derr J, Mitchell DC, Mustad VA, Russell ME, effects on caloric and nutrient intake. In Human Obesity. Annals of the McDonnel ET, Salabsky D & Pearson TA (1993). The role of fatty New York Academy of Sciences, eds RJ Wurtman & JJ Wurtman. acid saturation on plasma lipids, lipoproteins, and apolipoproteins: I. Annals of the New York Academy of Sciences, New York, Vol 499, Effects of whole food diets high in cocoa butter, olive oil, soybean oil, pp 179±189.
You can also read