|

 

Pii: s0031-9384(00)00321-8

Physiology & Behavior 71 (2000) 87±94 Effects of (ÿ)-hydroxycitric acid on appetitive variables Richard D. Mattes*, Leslie Bormann Department of foods and Nutrition, Purdue University, 212 Stone Hall, West Lafayette, IN 47907-1264, USA Received 25 January 2000; received in revised form 18 April 2000; accepted 25 May 2000 ( ÿ )-Hydroxycitric acid (HCA) reportedly promotes weight loss, in part, through suppression of hunger. However, this mechanism has never been evaluated in humans in a controlled study. Eighty-nine mildly overweight females were prescribed 5020-kJ diets for 12 weeks as part of a double-blind, placebo-controlled parallel group study. Forty-two participants ingested 400-mg caplets of Garcinia cambogia 30±60 min prior to meals for a total dose of 2.4 g/day (1.2 g/day HCA). Forty-seven participants ingested matched placebos. Weight and body composition were assessed at baseline and every other week for 12 weeks. Food intake and appetitive variables were assessed at baseline and monthly for 12 weeks. Both groups lost body weight with the active group achieving a significantly greater reduction (3.7 ‹ 3.1 kg versus 2.4 ‹ 2.9 kg). No effects of the HCA were observed on appetitive variables. The active treatment group did not exhibit better dietary compliance or significant correlations between appetitive variables and energy intake or weight change. This study does not support a satiety effect of HCA. D 2000 Elsevier Science Inc. All rights reserved.
Keywords: Garcinia cambogia; Hydroxycitric acid; Hunger; Weight loss; Human; Appetite serotonin-and norepinephrine-reuptake inhibitor, appears promising, but may increase blood pressure and heart rate Dietary approaches for the management of obesity have in some patients [16]. It also has the potential to promote been largely unsuccessful due, in part, to feelings of hunger dependency if abused [16]. Attempts to manipulate satiety that undermine adherence to weight loss regimens. Pharma- hormones such as cholecystokinin or bombesin to achieve cologic agents designed to suppress hunger have promoted sustained weight loss have proven elusive [17].
weight loss, but are often accompanied by unacceptable side The limited success and potential complications of these effects. Amphetamine-based anorexiants are effective in pharmacologic weight loss aids has led to a large and some patients, but leave them feeling anxious and are prone growing market for alternative therapies such as herbal to abuse and chemical dependency [33]. This limits their products. Garcinia cambogia, grown primarily in Southeast long-term use and cessation typically results in prompt regain Asia, is one popular representative. The dried and cured of body weight. Beta-phenethylamine derivatives have lower pericarp of the fruit of this species contains up to 30% by abuse potential but may still cause insomnia, anxiety and weight of ( ÿ )-hydroxycitric acid (HCA) [19]. These rinds irritability [33]. While useful in many patients, their limited are used in regional cooking practices and are reported to efficacy prompted research into a new class of agents, ones make meals more filling [4]. This claim is bolstered by acting on serotonergic neurotransmission. Dexfenfluramine cursory observations from clinical studies [2]. A satiety hydrocholoride and fenfluramine hydrochloride were widely effect has been demonstrated experimentally in rats and effective, but were implicated in the development of cardiac associated with weight reduction [26,36,37]. Because HCA valvulopathy [15] and withdrawn from the market. The most does not appear to enter the brain, it does not elicit CNS side recent introduction in this class of drugs, Sibutramine, a effects that may limit its acceptability.
HCA may promote weight reduction through sup- pressed de novo fatty acid synthesis, increased lipid oxidation and reduced food intake [22]. Enhanced satiety * Corresponding author. Tel.: +1-765-494-0662; fax: +1-765-494- may account for the reported suppression of energy con- E-mail address: [email protected] (R.D. Mattes).
sumption. One potential mechanism accounting for the 0031-9384/00/$ ± see front matter D 2000 Elsevier Science Inc. All rights reserved.
PII: S0031-9384(00)00321- 8 R.D. Mattes, L. Bormann / Physiology & Behavior 71 (2000) 87±94 satiety effect of HCA may involve inhibition of ATP sition determined, completed chemosensory function tests citrate lyase. This would limit the availability of acetyl and received dietary guidance. They were then randomly coenzyme A (acetyle CoA) for lipid synthesis during assigned to receive either caplets of G. cambogia or place- carbohydrate feeding. As a result, carbon is diverted to bo. A log of hunger ratings and activities was kept over the glycogen synthesis. Based primarily on studies with mice next 24 h. During that week, participants were called twice [7,8] and rats [24], it has been argued that glycogen levels and asked to keep 24-h diet records. One week after the serve as a primary signal for energy regulation. However, initial meeting, they began their 12-week diet. Exercise was this has been questioned by findings from human clinical encouraged, but no formal regimen was prescribed. Diet trials [31,34]. Further, the efficiency of carbohydrate con- records and hunger and activity logs were kept and chemo- version to fat under conditions of energy excess in humans sensory function was assessed during weeks 4, 8 and 12. At is extremely low [13] so inhibition of this pathway would the end of weeks 2, 4, 6, 8, 10 and 12, participants reported be expected to hold limited consequence.
to the laboratory for repeat assessments of body weight and A second possible mechanism for an anorectic effect of composition. The protocol was approved by the Human HCA holds that by reducing acetyl CoA, malonyl CoA Subjects Review Committee of Purdue University.
levels are depressed thereby reducing negative feedback on carnitine acyltransferase [21]. This leads to increased lipid transport into the mitochondria and inefficient oxidation with resultant ketone body formation. Ketones are purported Participant eligibility criteria included: 18±65 years old; appetite suppressants, however, several groups have failed 10±50 lb over ideal body weight [23]; interested in losing to observe an association between ketosis and reported 10±20 lb; not adhering to any prescribed diet or taking hunger level [3,32].
medications (except birth control); and self-reported normal Despite an hypothesized prominent role of HCA-induced taste and smell function. A total of 167 individuals were satiety on reduced energy intake and weight loss, there has recruited. An error in coding of pill bottles provided to the been little experimental evaluation of this action in humans.
researchers (detected after the study, but prior to data Given the mechanistic issues raised above, recent evidence analyses) resulted in 28 participants receiving a mixture of that HCA may not promote weight loss [14] and widespread active and placebo pills. Thus, these participants were use of products containing HCA for weight management, excluded from analyses. Based upon pill counts (ingestion the question of whether HCA is an appetite suppressant of at least 80% of the administered caplets) and attendance warrants further consideration. The present study was de- at requisite evaluation sessions, a total of 106 individuals signed to assess the effect of G. cambogia on appetitive were deemed compliant with study procedures. Among the indices and their relationship with weight loss during non-compliant group, 20 had been assigned to active treat- moderate energy restriction.
ment and 13 to placebo. Only 17 of the eligible sub-sample were male. Because the small number of males precluded meaningful gender-specific analyses and there are reports of sex differences in appetitive ratings [24,28,43], including to HCA treatment [2], as well as well known differences in 2.1. General protocol energy intake, analyses were focused on the 89 compliant females. Eighty-seven participants were Caucasian, with Participants were recruited by public advertisement into a one African American and one Asian. Table 1 contains randomized, double-blind, placebo-controlled, parallel- other baseline characteristics of the total sample as well as group design study. During an initial baseline visit, all the active treatment and placebo treatment groups. Only participants completed health, demographic and dietary disinhibition scores differed significantly between groups restraint questionnaires, had their body weight and compo- (t = 2.07, p = 0.042).
Total sample (N = 89) Active treatment (N = 42) Placebo treatment (N = 47) Body mass index (kg/m2) Weight loss goal (kg) Cognitive restraint (Three-Factor Eating Ques- tionnaire (TFEQ)) Disinhibition (TFEQ) R.D. Mattes, L. Bormann / Physiology & Behavior 71 (2000) 87±94 ability, itchiness and distractability (the ancillary appetitive questions) on scales ranging from not at all'' to extre- mely.' During another baseline day and weeks 4, 8 and 12, participants also coded hunger by outlining the place(s) on a gender-appropriate human figure where they felt the sensa- tions they associate with hunger occurred [9] These areas were cut out of the form and weighed. They were coded into three regions Ð head and neck, trunk, limbs.
Dietary restraint was assessed by the TFEQ [35].
2.6. Body weight and composition Body weight was measured on a clinical scale with subjects wearing only a hospital gown. They voided just prior to weighing. Measurements were obtained at approxi- mately the same time of day for each individual. Fat mass, fat-free mass and body water were determined by bioelec- trical impedance analysis (Tanita Body Fat Analyzer, TBF- 105, Tanita, Skokie, IL).
2.7. Dietary assessment Fig. 1. Self-reported hourly hunger ratings obtained over a 24-h period on a nine-point category scale during the pre-treatment week and week 12 of Energy and nutrient intake were determined with version treatment by participants receiving active treatment or placebo.
7.2 of The Food Processor nutrient database (ESHA Re- search, Salem, OR).
Participants were counseled to adhere to a 1200-kcal exchange diet [1] that contained about 30% of energy from fat. They were provided reference materials, recipes and trained to estimate portion sizes with true-size portion charts. Active treatment participants were required to ingest two 400-mg caplets of G. cambogia or matched placebo three times per day (30±60 min before each meal). The source of HCA used in the study was G.
cambogia extract (Citrin1 standardized for a minimum of 50% HCA). Its purity was determined by HPLC. Thus, the total dose of HCA in the active treatment group was 1.2 g/ day. Placebo treatment participants took identical caplets at the same schedule.
2.4. Appetitive questionnaires Hunger, desire to eat, prospective consumption (how much food do you think you could eat right now?) and fullness (the primary appetitive questions) were evaluated by having participants indicate the intensity of the sensation they ascribed to each on a nine-point category scale each waking hour for 1 day at baseline and during weeks 4, 8 and 12. End anchor descriptors are listed in Figs. 1 and 2. In Fig. 2. Mean (‹ SE) peak and nadir self-reported hunger ratings obtained addition, participants indicated how intensely they experi- over a 24-h period prior to treatment (B) or at weeks 4, 8 and 12 of enced feelings of stomach growling, headache, thirst, irrit- treatment with active compound or placebo.
R.D. Mattes, L. Bormann / Physiology & Behavior 71 (2000) 87±94 2.8. Energy expenditure 2.10. Statistical analysis Energy expended in physical activity was determined Body weight, energy and macronutrient intake, appetitive by questionnaire [30] completed at baseline and weeks 4, ratings and sensory function were explored by repeated measures analysis of variance with treatment as a between group factor. Where appropriate, paired t-tests were used for 2.9. Sensory function post hoc comparisons. For the appetitive variables, the primary metric used was the mean self-reported rating Participants rated an array of commercially available during the time each individual was awake on a recording foods for sensation intensity using nine-point category day. Associations between the appetitive variables and both scales with end anchors of no (sweetness, saltiness, fat) dietary intake indices and weight loss outcome were as- at all'' and extremely (sweet, salty, high fat).'' Pleasantness sessed by Pearson correlation coefficients. The criterion for was also rated on a nine-point category scale with end statistical significance was set at p < 0.05, but where multi- anchors of extremely pleasant'' and extremely unplea- ple comparisons were conducted, the Bonferroni correction sant.'' Single bite-sized samples of foods were presented in was applied.
random order and consumed. A water rinse was interspersed between samplings. Ratings were obtained at baseline and weeks 4, 8 and 12. Seventeen foods were selected to be representative of eight overlapping general categories.
A statistically significant loss of weight was observed over the 12-week study period in both the active (t = 7.80, Low sweet±low fat Peaches Lite (Del Monte Foods, p < 0.001) and placebo (t = 5.65, p < 0.001) treatment San Francisco, CA), Golden Loaf, groups. The mean loss with active treatment was 3.7 ‹ 3.1 fat-free and cholestrol-free kg whereas the value was 2.4 ‹ 2.9 kg for the placebo group.
(Entemanns Foods, Totowa, NJ) The difference in weight loss between groups was also High sweet±low fat Glazed Donuts Light (Entemanns statistically significant (t = 2.26, p = 0.026). The decrease Foods), Peaches in Heavy Syrup in fat mass was not significantly different between groups (Del Monte Foods), (active = ÿ 4.1% and placebo = ÿ 3.0%), but the reduction in waist circumference was significant (active = ÿ 3.96 cm, Ice Cream (Prarie Farms Dairy, placebo = ÿ 2.22 cm (t = 2.72, p = 0.008)). Relative to base- line, both groups reported significant reductions in energy Low sweet±high fat consumption during the diet period ( ÿ 1756 ‹ 409 kJ/day (Entemanns Foods) Ð active, ÿ 1574 ‹ 322 kJ/day Ð placebo). Mean daily High sweet±high fat Glazed Buttermilk Donuts intake tended to be lower during active treatment compared (Entemanns Foods), Vanilla Ice to placebo (5534 ‹ 315 versus 6191 ‹ 239 kJ/day), but the Cream (Prarie Farms Dairy), difference was not significant (t = 1.68, p < 0.1). There was Honey-Roasted Peanuts no significant group difference in energy expenditure at any time point or a change over time.
Winston-Salem, NC) Fig. 1 depicts the hunger patterns of participants at Low salt±low fat White Corn, air-popped (American baseline and the end of the 12-week study. Because parti- Popcorn, Sioux City, IA), Unsalted cipants awoke and retired at different times of day, data are Original Sourdough Recipe Hard presented only when 10 participants were awake. Between Pretzels (Wege Pretzel, 800 and 2200 h, 30 individuals were awake in each group.
Ratings were coded as missing when participants were High salt±low fat Original Sourdough Pretzels asleep. While hunger did change over the 24-h recording (Wege Pretzel), Low-Fat Original periods (e.g, baseline Ð F(14,728) = 8.44, p < 0.001; week Potato Crisps (Frito-Lay, 12 Ð F(14,560) = 8.93, p < 0.001), no significant group differences were observed at any time point during baseline Low salt±high fat White Corn, air-popped (American or weeks 4, 8 or 12. Mean ratings were also comparable Popcorn), coated with across the study period. Peak and nadir values were similar between the groups at baseline and at the end of weeks 4, 8 (Land O' Lakes, Arden Hills, and 12 and were stable over the study period (Fig. 2). Group MN), Unsalted Cocktail Peanuts variance in reported hunger was significantly greater in the active treatment group at baseline (F test for variance, High salt±high fat Cocktail Peanuts (Nabisco Foods), p < 0.05), but the group variances during treatment were Potato Chips (Frito-Lay) not significantly different.
R.D. Mattes, L. Bormann / Physiology & Behavior 71 (2000) 87±94 Fig. 3. Self-reported hourly desire to eat,'' prospective consumption'' and fullness'' ratings obtained over a 24-h period on a nine-point category scale during the pre-treatment week and week 12 of treatment by participants receiving active treatment or placebo.
The patterns of responses for desire to eat, prospective treatment effect and time by treatment interaction were consumption and fullness at baseline and week 12 are not significant. No other significant treatment or time presented in Fig. 3. Active treatment participants reported effects were observed. Ratings of itchiness, which were a higher desire to eat than the placebo treated participants not expected to vary in relation to the treatment, also did only at 600 h during baseline (t = 3.07, p = 0.004). This not differ between groups or over time.
isolate finding is likely artifactual. Fullness and prospective With a correction for multiple testing, there were no consumption ratings were similar at all time points from significant group differences in the rated sweetness, salti- baseline to the end of the study.
ness, fat level or pleasantness of the test foods at baseline or Table 2 contains self-reported 24-h mean appetite-re- any time during treatment.
lated sensations by active and placebo treated participants Baseline mean appetitive sensations did not correlate at baseline and weeks 4, 8 and 12. Mean thirst ratings significantly with weight change in the full sample (Pearson were higher during baseline relative to all other assess- correlation coefficients ranged from ÿ 0.19 (hunger) to 0.05 ments (all p < 0.05) in the full sample. However, the (fullness), placebo sub-group (r = ÿ 0.24 (hunger) to 0.12 24-h mean (SD) appetite-related sensations (and a malingering check itchiness'' at baseline and weeks 4, 8 and 12 treatment for participants receiving active treatment and placebo. Ratings of 1.0 = not at all, 9.0 = extremely R.D. Mattes, L. Bormann / Physiology & Behavior 71 (2000) 87±94 (fullness)) or active treatment sub-group (r = ÿ 0.08 (hun- individuals [6,20,26,40,41] but pharmacologic enhancement ger) to 0.26 (fullness)). Among the placebo-treated partici- of satiety has proven effective at reducing energy intake and pants, the only appetitive variable significantly correlated weight [12,25]. The present data on appetitive indices are with weight change was mean hunger ratings at week 8 unequivocal. No significant treatment effects were observed (r = ÿ 0.40, p = 0.01). No significant associations were ob- on mean, peak or nadir hunger ratings, mean ratings of served in the active treatment group at any time point.
desire to eat, prospective consumption, fullness or sensa- Similarly, correlations between appetitive variables and tions of thirst, stomach growling, headache, distraction, energy intake or change of energy intake were of a low irritability or, as a check on malingering, itchiness. Prior order and not-statistically significant. Among active treat- support for an appetitive effect was based on anecdote [4] ment participants, correlation coefficients between hunger and data interpreted without a control treatment or pure ratings at baseline and treatment weeks 4, 8 and 12 ranged HCA formulation [2]. The appetitive indices also were not from ÿ 0.21 to 0.02 for energy intake and from ÿ 0.16 to significantly associated with energy intake or body weight 0.20 for change of energy intake from baseline. Correlations change within the active treatment participants. An associa- between the appetitive variables of desire to eat, prospective tion between satiety effects and weight reduction has been consumption and fullness and the intake variables of energy reported in rats [27,37,38]. However, the effect is transient consumption and change of energy consumption ranged [11]. The association was examined at weeks 4, 8 and 12 of from ÿ 0.30 to 0.27.
this study and was not apparent at any time point. It is possible that it lasted less than 4 weeks. A diminution of appetite suppression over this time frame has been noted [2] yet, interestingly, weight loss reportedly continued in that study. The weak and transient nature of appetitive effects of Consistent with the prescribed and reported reduction of HCA raise questions about its clinical significance. While energy intake, participants in both the active and placebo negative findings are always open to methodological ques- treated groups lost weight over the study period. The active tions, the consistency of our data across appetitive indices, treatment group achieved a significantly, albeit modest in larger sample size and use of more rigorous methodology absolute terms, greater reduction. This finding is consistent lends credence to our findings. Unlike most other published with several early reports [5,38], but not with a recent, larger work, our study also entailed ingestion of the active pills and more vigorously controlled trial [13]. However, such 30±60 min prior to meals when, based on animal studies, comparisons must be made with caution as there were the HCA reaches peak efficacy [36]. The administered dose differences in the formulations and doses administered and was modest and blood samples were not collected to study populations. Earlier reports were typically based on confirm effective plasma levels were achieved, but the combination products (e.g., HCA plus chromium) (e.g., Ref.
weight loss results suggest the dose was adequate to elicit [2,5,10]) so efficacy cannot be ascribed to the HCA alone.
Interestingly, more consistent weight loss is reported with Increased blood ketones and hepatic or muscle glycogen lower doses of HCA (i.e., approximately 750 mg/day levels have been posited as potential mechanisms for the [2,5,10]) compared to higher doses (i.e., 1300±1500 mg/ satiety effect of HCA [21,22]. These indices were not day [14,29,39]). Further, several had small sample sizes measured in the present study but two recently published [5,39] and/or lacked a placebo control [2]. The work by trials [14,18], involving participants on diets with macro- Heymsfield et al. [14], which yielded no effect, involved nutrient compositions similar that used here, have failed to males and females whereas the present report is limited to note shifts associated with HCA use.
females. In fact, the males in our study exhibited more Alteration of the rewarding properties of foods can lead variable weight responses and if included in the sample, the to reduced intake independent of hunger status [42]. How- significant difference from placebo treatment was elimi- ever, the lack of effect of HCA on either taste intensity or nated. Heymsfield et al. [14] reported controlling for gender hedonic ratings for foods suggests this also is unlikely to did not influence their findings, but our data suggest a account for the present findings.
gender-specific weight-loss response remains a possibility.
To the extent that hunger sensations are sufficiently Additionally, studies of rats suggest obese animals are more unpleasant that they compromise dietary compliance, it resistant to the weight reducing effects of HCA than the lean was hypothesized HCA would lead to higher rates of dietary [11]. The study population used by Heymsfield et al. [14] adherence relative to placebo-treated controls. However, included a higher proportion of markedly obese individuals study attrition rates were comparable in the two groups than the present sample.
(20 from active and 13 from placebo), as noted by others The primary focus of our work concerned the effects of [14]. These data suggest the addition of HCA does not HCA on appetitive variables and whether these could promote improved compliance with a reduced energy diet.
account for any noted effects on weight loss. The associa- However, given the lack of effect on hunger, they do not tion between appetitive sensations, food intake and body address the more general question of whether amelioration weight is weak in non-dieting and dieting, free-living of hunger serves this function.
R.D. Mattes, L. Bormann / Physiology & Behavior 71 (2000) 87±94 There are several qualifications that warrant comment [13] Hellerstein MK, Schwartz J-M, Neese RA. Regulation of hepatic de in this study. First, the study of appetitive properties of novo lipogenesis in humans. Annu Rev Nutr 1996; 16:523±57.
HCA under conditions of energy restriction could be [14] Heymsfield SB, Allison DB, Vasselli JR, Pietrobelli A. Garcinia cam- bogia (hydroxycitric acid) as a potential antiobesity agent. JAMA viewed as problematic if the diet promoted extreme sensations. However, this did not occur with the mild [15] Khan MA, Herzog CA, St. Peter JV, Hartley GG, Madlon-Kay R, restriction imposed as evidenced by ratings falling in Dick CD, Asinger RW, Vessey JT. The prevalence of cardiac valvular the middle range of the response scales. Second, given insufficiency assessed by transthoracic echocardiography in obese that an energy-restricted diet would prevent the required patients treated with appetite-suppressant drugs. N Engl J Med enzyme alterations (acetyl CoA±malonyl CoA) that lead [16] King DJ, Devaney N. Clinical pharmacology of sibutramine hydro- to altered substrate metabolism and satiety, the concur- chloride (BTS 54 524), a new antidepressant, in healthy volunteers. Br rent dietary restriction could have hampered induction J Pharmacol 1988;26:607±11.
of HCA's satiety effects. However, the prescribed diet [17] Kordik CP, Reitz AB. Pharmacological treatment of obesity: therapeu- was only mildly energy restricted and still contained at tic strategies. J Med Chem 1999;42:181±201.
[18] Kriketos AD, Thompson HR, Greene H, Hill JO. (ÿ)-Hydroxycitric least 30% of energy from fat. Thus, it likely reflected acid does not affect energy expenditure and substrate oxidation in conditions under which HCA would be used by con- adult males in a post-absorptive state. Int J Obes 1999;23:867±73.
sumers. Third, it may be that HCA is more effective at [19] Lewis YS, Neelakantan S. (ÿ)-Hydroxycitric acid Ð the principal acid moderating weight gain [11] than promoting weight in the fruits of Garcinia cambogia. Desr Psytochem 1965; 4:619±25.
loss. This was not tested, but if true, the compound [20] Mattes RD. Hunger ratings are not a valid proxy measure of reported food intake in humans. Appetite 1990;15:103±13.
may be more useful for weight maintenance after an [21] McCarty M, Majeed M. The pharmacology of Citrin. In: Majeed M, initial loss.
Rosen R, McCarty M, Conte A, Patil D, Butrym E, editors. Citrin.1 A revolutionary, herbal approach to weight management. Burlingame, CA: New Editions Publishing, 1994. pp. 34±52.
[22] McCarty MF. Promotion of hepatic lipid oxidation and gluconeogen- esis as a strategy for appetite control. Med Hypotheses 1994; This work was funded by a grant from the SlimFast [23] 1983 Metropolitan height and weight tables, vol. 64. New York: Stat Bull Metrop Insur, 1983. p. 3 (January±June).
[24] Monello LF, Seltzer CC, Mayer J. Hunger and satiety sensations in men, women, boys and girls: a preliminary report. Ann NY Acad Sci [25] Novin D, Robinson K, Culbreth LA, Tordoff MG. Is there a role for the liver in the control of food intake? Am J Clin Nutr 1985; 42: [1] American Dietetic Association. Handbook of clinical dietetics. 2nd ed.
New Haven: Yale Univ. Press, 1992.
[26] Pasquali R, Besteghi L, Casimirri F, Melchionda N, DeFebo G, Zoccoli [2] Badmaev V, Majeed M. Open field, physician controlled, clinical L, Barbara L, Tassoni U. Mechanisms of action of the intragastric bal- evaluation of botanical weight loss formula Citrin1. Nutracon '95: loon in obesity: effects on hunger and satiety. Appetite 1990; 15:3±11.
nutriceuticals, dietary supplements and functional foods, July 11±13, [27] Rao RN, Sakariah KK. Lipid-lowering and antiobesity effect of (ÿ) Las Vegas, NV.
hydroxycitric acid. Nutr Res 1988;8:209±12.
[3] Baird IM, Parsons RL, Howard AN. Clinical and metabolic studies of [28] Rolls BJ, Fedoroff IC, Guthrie JF, Laster LJ. Effects of temperature chemically defined diets in the management of obesity. Metabolism and mode of presentation of juice and hunger, thirst and food intake in humans. Appetite 1990;15:199±208.
[4] Clouatre D, Rosenbaum ME. The diet and health benefits of HCA [29] Rothacker DQ, Waitman BE. Effectiveness of a Garcinia cambogia (hydroxycitric acid). New Canaan, CT: Keats Publishing, 1994.
and natural caffeine combination in weight loss; a double-blind pla- [5] Conte AA. A non-prescription alternative in weight reduction therapy.
cebo-controlled pilot study. Int J Obes 1997;21:53.
[30] Sallis JF, Haskell WL, Wood PD, Fortmann SP, Rogers T, Blair SN, [6] de Graaf C, Jas P, van der Kooy K, Leenen R. Circadian rhythms of Paffenbarger JRRS. Physical activity assessment methodology in the appetite at different stages of a weight loss programme. Int J Obes five-city project. Am J Epidemiol 1985;121:91±106.
[31] Shetty PS, Prentice AM, Goldberg GR, Murgatroyd PR, McKenna [7] Flatt J-P. McCollum Award Lecture, 1995: Diet, lifestyle, and weight RJ, Stubbs RJ, Volschenk PA. Alterations in fuel selection and maintenance. Am J Clin Nutr 1995;62:820±36.
voluntary food intake in response to isoenergetic manipulation of [8] Flatt J-P. Glycogen levels and obesity. Int J Obes 1996;2:S1±S11.
glycogen stores in humans. Am J Clin Nutr 1994;60:534±43.
[9] Friedman MI, Ulrich P, Mattes RD. A figurative measure of subjective [32] Silverston JT, Stark JE, Buckle RM. Hunger during total starvation.
hunger sensations. Appetite 1999;32:395±404.
[10] Girola M, DeBernardi M, Contos S, Tripodi S, Ventura P, Cuarino C, [33] Silverstone T. Appetite suppressants: a review. Drugs 1992; 43:820±36.
Marletta M. Dose effect in lipid-lowering activity of a new dietary [34] Stubbs RJ, Murgatroyd PR, Goldberg GR, Prentice AM. Carbohydrate integrator (chitosan, Garcinia cambogia extract and chrome). Acta balance and the regulation of day-to-day food intake in humans. Am J Toxicol Ther 1996;17:25±40.
Clin Nutr 1993;57:897±903.
[11] Greenwood MRC, Cleary MP, Gruen R, Blase D, Stern JS, Triscari [35] Stunkard AJ, Messick S. The three-factor eating questionnaire to AC, Sullivan AC. Effect of (ÿ)-hydroxycitrate on development of measure dietary restraint, disinhibition, and hunger. J Psychosom obesity in the Zucker obese rat. Am J Physiol 1981; 240: E72±8.
Res 1985;29:71±83.
[12] Hansen DL, Toubro S, Stock MJ, Machdonald IA, Astrup A. Thermo- [36] Sullivan AC, Hamilton JG, Miller ON, Wheatley VR. Inhibition of genic effects of sibutramine in humans. Am J Clin Nutr 1998; lipogenesis in rat liver by (ÿ)-hydroxycitrate. Arch Biochem Biophys R.D. Mattes, L. Bormann / Physiology & Behavior 71 (2000) 87±94 [37] Sullivan AC, Triscari J, Hamilton JG, Miller ON. Effect of (ÿ)-hydro- [41] Wing RR, Marcus MD, Blair EH, Burton LR. Psychological responses xycitrate upon the accumulation of lipid in the rat: II. Appetite. Lipids of obese Type II diabetic subjects to very-low-calorie diet. Diabetes [38] Sullivan AC, Triscari J. Metabolic regulation as a control for lipid [42] Yeomans MR, Wright P, Macleod HA, Critchley JAJH. Effects of disorders: I. Influence of (ÿ)-hydroxycitrate on experimentally in- nalmefene on feeding in humans disassociation of hunger and palat- duced obesity in the rodent. Am J Clin Nutr 1977;30:767±76.
[39] Thom E. Hydroxycitrate (HCA) in the treatment of obesity. Int J Obes [43] Zylann KD. Gender differences in the reasons given for meal termina- tion. Appetite 1996;26:37±44.
[40] Wardle J. Hunger and satiety: a multidimensional assessment of re- sponses to caloric loads. Phys Behav 1987;40:577±82.

Source: http://womenslife.com-health-and-fitness.pw/lifestyles/index_files/1-s2.0-S0031938400003218-main.pdf