Eating behavior and adherence in obese adult subjects treated

Carlo
Maurizio

to dietary prescriptions with 5-hydroxytryptophan12

Maria Del Ben, Alessandro Laviano,
Rossi-Fanelli adin obese subjects (14). The present study attempted to confirm

Cangiano,
Muscaritoli,

Fabrizio

Ceci,
Fabiola

Antonia
Antonucci, have (5-HTP) food data adherence 20 obese

Cascino,
and shown intake, over

Filippo that and oral

ABSTRACT
ministration prescriptions of causes

Previous
anorexia,

observations
decreased these

5-hydroxytryptophan To confirm

without a longer

dietary weight period

these data over whether adherence administering

a longer period of observation to dietary restriction could to obese hyperphagic

and to verify be improved by

loss in obese
ofobservation tion could

subjects.

5-HTP

subjects.

and to verify whether be improved by 5-HTP,

to dietary restricpatients were ran-

domly assigned to receive either 5-HTP (900 mg/d) or a placebo. The study was double-blinded and was for two consecutive 6-wk periods. No diet was prescribed during the first period, a 5040kJ/d diet was recommended in 5-HTP-treated for the second. patients Significant during both weight periods.

Subjects
Subjects

and methods
Downloaded from www.ajcn.org by on April 16, 2009

Twenty-eight

obese,

hyperphagic

adult

female

subjects

(mean 30 and of daily

loss was observed

A reduction in carbohydrate intake and a consistent presence of early satiety were also found. These findings together with the good tolerance observed suggest that 5-HTP may be safely used to treat KEY obesity. WORDS

age 43.2 y I .7: SE) with 40 were studied. Hyperphagia

body mass indexes between was defined as the excess

energy introduced with respect according to sex, age, and physical intolerance, from the hyperlipidemia, study. or

to the energy need calculated activity. Subjects with glucose hyperuricemia were excluded

Am J Clin Nutr
Eating obesity, loss behavior, adherence

1992;56:863-7. anorexia, to diet, 5-hydroxytrypappetite for car-

tophan, serotonin, bohydrates, weight

Study The

design research ofthe consent, protocol University subjects was approved by the Faculty Ethics

Introduction
Pharmacological, accumulated rotonin has animals possible genesis this idea and role (5-7). biochemical, and behavioral evidence brain both has sein

Committee informed

of Rome, La Sapienza. were randomly assigned or a placebo composed stearate (both from The drug, which was

After giving to receive

either 5-HTP (900 mg/d) mannitol, and magnesium dustries, capsules per day, subdivided wk periods. tions, during whereas the second Pomezia, Italy).

ofcorn starch, Sigma-Tau Inin the form of

in the last an inhibitory

two decades, suggesting that influence on eating behavior studies system diseases

in humans ( I -4). Reported played by the serotoninergic present Changes in different in the different deprivation metabolism nutritional producing

in favor of a in the pathofurther support serotonin

that do not dissolve until pH 8.6, was taken three times 30 mm before each meal. The 12-wk-study period was for each group of subjects into two consecutive 6During the first period period diet there was were The no dietary 5040 restricsuba 5040-kJ/d recommended to subjects U were

ofanorexia

of brain

in animals experiencing reported (8, 9). Food in serotonin changes stitutes metabolism

states have been a marked increase may reflect in turn conresponses

ofobservation.

in the lateral

hypothalamus

divided as follows: 53% from carbohydrates, and 18% from proteins. No carbohydrate-rich between meals. Subjects were examined every feeding behavior and body measurements were performed weight. Routine for each subject

29% from lipids, foods were allowed 2 wk to evaluate blood chemistry at the beginning

in the availability ofbrain tryptophan; this a potential mechanism by which behavioral

are coordinated with nutritional inputs (10). A variety of different directand indirect-acting agonists at central serotonin synapses such as dexfenfluramine and fluoxetine have been shown to cause weight subjects caused loss as well (1 1- 13). a high as a marked However, of side reduction the effects in food
( 12, 13).

intake of both

in obese drugs study seroany loss Society of typical weight

administration double-blind, of the

incidence

In a previous short-term (5 wk), we observed that oral administration tonin specific precursor dietary 5-hydroxytryptophan prescription, signs,
Nior

crossover physiological without and

(5-HTP), food intake

was followed

by the onset

anorexia-related in,
j

eg, decreased Printed

From the Laboratory of Clinical Nutrition, 3rd Department of Internal Medicine; the Department ofHuman Biopathology; and Servizio di Dietetica, Istituto di Terapia Medica Sistematica, University of Rome, La Sapienza, Rome, Italy. 2 Address reprint requests to F Rossi-Fanelli. Laboratory of Clinical Nutrition, 3rd Department of Internal Medicine, Viale delIUniversit#{224} 37, 00185 Rome, Italy. Received January 22, 1991. Accepted for publication May 25, 1992.

C/in

1992:56:863-7.

in USA.

1992 American

for Clinical

Nutrition

863

864
TABLE Baseline I characteristics of subjects5 Body weight kg Placebo 5-HTP
S

CANGIANO

ET

AL

Total

energy k/Id

Proteins g/d

Carbohydrates g/d
1 1.3 331 29.1 349 47.8

Lipids g/d
120 12.3

94.3 99.7

5.3

I I 398

5.9

13 528

1 020 I 033

104 1 17

8.8

144 12.6

SE: n = 10 for both

groups.

and

at the

end

of each acid

period 24-h

of observation. urinary was also excretion determined method

To

test

subjects 2 wk by seasin-

compliance 3-indoleacetic by the

to treatment, (5-HIAA)

of 5-hydroxyevery described

and unpaired group were compared prevalence effects The icance minimum was

data; modifications compared or placebo

observed within each treatment and the 5-HTP groups were was used to compare symptoms and the treatments. for statistical the side (17). signif-

chromatographic-colorimetnc

( 17). The Chi-square test of both the anorexia-related observed during probability
<

Udenfriend et al (1 5). Daily total energy intake as well as single lection, which may define a subjects eating sessed cluding working every 2 wk by using a 3-d diet diary. all beverages, days. Each were compiled subject was

placebo level

and

5-HTP

macronutrient behavior, were Food diaries,

considered

0.05.

by subjects for 3 consecutive instructed to carefully weigh leftovers. All reports To avoid reported

Results
Twenty ofthe 28 subjects included completed and the the study. The

food
were

before meals and then reweigh any validated by a next of kins signature.

Downloaded from www.ajcn.org by on April 16, 2009

interference due to premenstrual depression, food-intake measurements were not assigned to this time of the month ( I 6). To investigate anorexia, patients were invited to report the presence of a series of symptoms, namely meat aversion, taste and smell alteration, nausea and/or vomiting, All of these symptoms interfere with eating related system invited to a pathological (CNS) regulation to report including stipsis. and and early satiety. are possibly

eight dropouts (three in the study group group) did not complete the study for

five in the control following reasons:

three had family problems, two self-prescribed a low-energy diet during the first study period, two self-administered other anorectic drugs, and one did not complete the follow-up. Two groups of 10 subjects each therefore were were in body expressed studied. weight, selection. As shown for the two in Table groups

1, baseline

characteristics

comparable

modification of the central-nervousoffeeding behavior (5, 6). Subjects were ofeach period myalgia, ofstudy drowsiness, the presence vertigo, weakness,

of subjects who did not differ intake, and specific-macronutrient Changes in body weight,

daily-total-energy ofbasal val-

at the end

as percentage

ofside effects, diarrhea, and

Statistical
All data ing mean,

analysis
were subjected standard error, to standard statistical and Students t tests analysis for both includpaired

ues, are shown in Figure 1 . subjects receiving the placebo did not show any significant change in their body weight during either study period (94.3 5.6 vs 93.2 5.3 kg, I SE, basal value vs end-of-study value). In contrast, subjects receiving 5HTP showed a significant weight loss at the end ofthe first period

of

basal

weight

100

--f-v

i(-___

_f1-____
-

NS
-

98 NO DIET

--

DIET

96

GA

10

12

weeks
FIG
periods P
<

I . Mean body-weight modifications, expressed as percentage of the basal value, during both no-diet ofobservation. D, placebo: #{149}, 5-hydroxytryptophan (S-HTP). 5Significantly different from beginning 0.03. 55Significantly different from value at 6 wk for S-HTP group. P < 0.02.

and diet of study,

5-HYDROXYTRYPTOPHAN (99.7 reduction ofobservation final weight Analysis 5.9 vs 98.0 of their 5.0 body kg, P
<

TREATMENT the quantities treated patients prescribed diet The evaluation two groups questionnaire, subjects nods, cantly group. 80% receiving respectively. higher Reports of subjects that differently occurrence

IN

OBESITY during always the second higher showed the reported the first satiety 5-HTP was the was during receiving which during may The groups such effects than period those

865 in the 5-HTPincluded behavior included 100% second always for the period, reported were than and 90% study significontrol by were to 20%, a transitory were (Fig 3). myalgia, All of not in the in the in a of pe-

0.03) end

followed of the

by a greater second period

consumed were (Fig

weight

at the

(98.0 5.0 vs 94.7 loss was 5% of the of the mean total

5. 1, P < 0.02). In the group basal body weight. intake, as calculated from changes in the placebo spontaneous reduction to 9778 was not decrease 907 to kJ/d 8644 statistically

2). of anorexia Among satiety 5-HTP Reports

different symptoms by and

energy

of subjects. early

subjects reports, showed no significant group during either period (Fig 2). The ofdaily observed significant, energy nor intake was from of the the 1 1399 first study further at the end

1021 period negligible

of early

for subjects of nausea, on this 5-HTP during reduced

episodically

1 3 19 ki/d observed during the diet period. There was a significant reduction of total energy intake in subjects receiving 5-

first study

significantly suggesting effect reported The

the last 6 wk ofobservation be considered other as symptoms weakness, of subjects

HTP
total 647 The

from
energy kJ/d

13528 intake in the

1033 showed second

to 7892 a further study

773 period.

kJ/d.

In these reduction

subjects, to 5326

symptom by the two of side

significant

of 5-HTP

administration.

A specific behavior was observed total protein intake remained during was were not the study period. reduced modified progressively further

in macronutrient unmodified In the during during 5-HTP the

selections. in the control group, half protein to values of the

group intake that

drowsiness, vertigo, of these symptoms patients vation 50-fold

diarrhea, and stipsis were investigated. were equally distributed in both groups between the urinary 5-HIAA the two with groups treatment (Fig two periods excretion and provided

the first period second

with no differences (Fig 4). The total difference between

of obsershowed a evidence

study (Fig statistically (Fig cant 50%

2). The carbohydrate different from the subjects of their reduction

intake, though reduced, was not basal value for the control group receiving intake 5-HTP showed a signifiat the end of carbohydrates

for subject-compliance

5).

Downloaded from www.ajcn.org by on April 16, 2009

2). Conversely,

Discussion
The role of amino acids in the regulation data of food intake has in by acid been supported plasma amino affecting the by experimental acid concentrations brain availability suggesting may modify that changes food intake amino

ofthe first study period. This was followed by a further reduction in the second study period during which

significant the total included showed although

carbohydrate
in the changes

intake was even lower than the quantities dietary prescription. Finally, the lipid intake similar to those observed for carbohydrates

of neurotransmitter

ENERGY

INTAKE

PROTEIN

INTAKE

12

12

we.ks

12

l2ws.ks

PLACEBO

S - HIP

PLACEBO

HIP

CARBOHYDRATE

INTAKE

LIPID

INTAKE

350

*0

C PLACEBO

12

B S
-

12w..ks

12

12

w..ks

HIP

PLACEBO

HIP

()

FIG 2. Mean SE modifications of total energy and single periods of observation in the two groups of subjects. B ()

macronutrient basal records;

intakes during both no-diet PD., prescribed diet.

(0) and diet

866
NO
20 40

CANGIANO
DIET
60 80 100 20 40

ET

AL
NO DIET
50 t 50 I 100 I 20 40

DIET
00 50 100 20 I

DIET
60 50 100

Early

satisty

____________ _____________
I I I I I

40 I

I p<.0I

IIIII1

____

p<.005

I.
I

W#{149}skn#{149}ss

Nsus./Vomttng

pc04

Dtowsissss

Tssts

slt#{149}rstlon

Sm.tt

sttstlon

Dtsrthoss

I I
of no 0,

Msst

avirsion

FIG 3. Prevalence of the symptoms used to evaluate anorexia in the two groups of subjects during no-diet and diet periods of observation. 5Significantly different from no diet for 5-hydroxytryptophan (5-HTP) group, P < 0.01 . 0, placebo: 0, S-HTP.

FIG 4. Prevalence of the side effects investigated in the two groups subjects during no-diet or diet periods of observation. There were statistically significant differences between groups or study periods. placebo; 0, 5-hydroxytryptophan (5-HTP).

precursors volved

( 1 8). Of the in the regulation

two

major

neurotransmitter the serotoninergic

systems

in-

low-energy as past small intake, argues exerted and

diet other

period,

when

the

meal

structure were

was allowed

subverted only in bread proteins, prein

of feeding,

system study by the brain physiological patterns action of is

carbohydrate-rich

food

seems

to play a specific role (19). A pharmacological Leibowitz et al (20), in fact, precisely characterized sites and receptors involved as well as the possible role eating believed patterns medial interact renergic ofendogenous and nutrient serotonin selection. in controlling In particular, energy balance satiety neurons natural serotonins

amounts. which in favor

In addition, the significant reduction is usually eaten in combination with of a specific Thus, the effect of carbohydrates adherence optimal

Downloaded from www.ajcn.org by on April 16, 2009

on appetite, to dietary may have been satiety and low-

by 5-HTP.

to influence both the of eating by activating hypothalamus antagonistically receptors

and the circadian localized in the serotonin seems to and its 2-norad-

scription observed the consequence

in subjects receiving 5-HTP ofa combination ofboth early

(19). In this process, with norepinephrine

(2 1 ). Pharmacologically

induced

variations activity (22, and also

in

carbohydrate intake. The relatively high incidence of nausea during 5-HTP administration may raise the concern that it could, at least in part, have been responsible for the 5-HTP effects observed. However, note that for most subjects, nausea Nausea was significantly less frequent was reported during the in food intake episodically. second 6-wk and weight

brain serotonin concentrations or serotoninergic penmental animals result in fluctuation ofappetite 1981, (24) important Wurtman have shown role et al (4) and that the in macronutrient more recently selection Hill system serotoninergic

in ex23). In Blundell plays an The be in obese

period

when

the

greater

reduction

especially behavior brain

people consuming same authors have counteracted synthesis. concentrations fluramine

large amounts ofcarbohydrate-rich also demonstrated that such enhancing

food. can

loss was observed. Finally, when present, nausea was of low intensity and never caused a subject to drop out of the study. We believe that the effect of nausea on weight loss was negligible. The lack of a significant reduction of body weight observed in the control group during and with the low-energy compliance with dietary reports. both study periods (ie, both without diet) was likely due to subjects nonprescriptions as demonstrated by diet study seem to confirm the system in the regulation

by pharmacologically

serotonin

Pharmacological manipulation with serotoninergic-active and fluoxetine has been reported

of brain serotonin agents like dexfento control food intake

and reduce body weight in obese patients (12, 1 3). Nevertheless, the incidence of side effects reported by subjects receiving both drugs (12, was higher than study that reported that in oral the control administration groups of food (14). 13). In a previous

specific

In conclusion, the results of this role played by the serotoninergic

we observed

mg/24

5-HTP intake The reported.

to adult obese subjects and weight loss without results ofthe In fact, present during study

was followed by decreased affecting the subjects mood seem to confirm 6-wk period those

500

previously

the diet-free,

of observation,

400L
300 lot-

subjects receiving 5-HTP at doses similar to those prescribed in the previous study showed a significant reduction of both body weight and daily carbohydrate intake. Moreover, when subjects were both actenstic second uation duction the diet-free given indices a low-energy was of the at lunch first observed. Italian period diet of the or dinner), diet a further Because (ie, the pasta early study significant of the reduction charand the bread evalrethe to during of meal-structure protein referred altered

first,
satiety may

i1HHH
B 2 4 6 8 10 12 B 2 4 6 5 81012

weeks

have

PLACEBO

HTP

of total carbohydrate in carbohydrate

intake. However, a significant intake was also observed during

FIG 5. Mean total 5-hydroxy-3-indoleacetic during the l2-wk study period in both groups

acid (5-1-IIAA) of subjects.

excretion

5-HYDROXYTRYPTOPHAN of feeding behavior was in fact followed in humans. The administration by a reduction of both daily of 5-HTP total energy of body as well that of

TREATMENT

IN

OBESITY

867

and carbohydrate intakes followed by a significant loss weight. The optimal adherence to dietary prescription as the good this substance obesity. tolerance may to 5-HTP be safely used treatment observed in the long-term

12. 13.

suggest treatment

B
14.

References
1. Blundell JE. Serotonin and appetite. Neuropharmacology l984;23: 1537-51. 2. Pinder RM, Brogden RN, Sawyer PR, Speight TM, Avery GS. Fenfluramine: a review of its pharmacological properties and therapeutic efficacy in obesity. Drugs l975;l0:24l-323. 3. Blundell JE, Leshem MB. The effect of 5-hydroxytryptophan on food intake and on the anorexic action of amphetamine and fenfluramine. i Pharm Pharmacol l975;27:3l-7. 4. Wurtman Ii, Wurtman Ri, Growdon IH, Henry P. Lipscomb MA, Zeisel SH. Carbohydrate craving in obese people: suppression by treatments affecting serotoninergic neurotransmission. Int I Eating Dis 198l;l:2-l5. 5. Rossi-Fanelli F, Cangiano C, Ceci F, et al. Plasma tryptophan and anorexia in human cancer. Eur I Cancer Clin Oncol 1986;22:8995. 6. Cangiano C, Cascino A, Ccci F, et al. Plasma and CSF tryptophan in cancer anorexia. I Neural Transm 1990;81:225-33. 7. Rossi-Fanelli F, Cangiano C. Increased availability oftryptophan in brain as common pathogenic mechanism for anorexia associated with different diseases. Nutrition 199 1;7:364-7. 8. Curzon G, Joseph MH, Knott P1. Effects of immobilization and food deprivation on rat brain tryptophan metabolism. J Neurochem l972;l9: 1967-74. 9. Perez-Cruet I, Tagliamonte A, Tagliamonte P, Gessa GL. Changes in brain serotonin metabolism associated with fasting and satiation
in rats. Life Sci 1972;1 1:31-9.

15.

16.

17.

18. 19. 20.

2 1.

22.

nucleus of the hypothalamus causes hypophagia. Psychol Forsch I 988;95:550-2. Guy-Grand B, Crepaldi 0, Lefebvre P, et al. International trial of long-term dexfenfluramine in obesity. Lancet l989;2: 1 142-5. Marcus MD, Wing RR, Ewing L, et al. A double-blind, placebocontrolled trial offluoxetine plus behavior modification in the treatment of obese binge-eaters and non-binge-eaters. Am I Psychiatry l990;l47:876-8 1. Ceci F, Cangiano C, Cairella M, et al. The effects oforal 5-hydroxytryptophan administration on feeding behaviour in obese adult female subjects. I Neural Transm 1989;76:109-l7. Udenfriend S, Titus E, Weissabach M. The identification of 5-hydroxy-3-indoleacetic acid in normal urine and a method for its assay. I Biol Chem l955;2l6:449-505. Wurtman II, Brzezinski AA, Wurtman RI, Laferrere B. Effect of nutrient intake on premenstrual depression. Am I Obstet Gynecol 1989;5 1: 122-34. Dixon WI, Massey Fl, Jr. Introduction to statistical analysis. New York: Mcgraw-Hill, 1983. Fernstrom ID. Role ofprecursor availability in control of monoamine biosynthesis in brain. Physiol Rev 1983;63:484-546. Leibowitz SF, et al. The role ofserotonin in eating disorders. Drugs l990;39(suppl 3):33-48. Leibowitz SF, Weiss OF, Walsh UA, et al. Medial hypothalamic serotonin: role in circadian patterns of feeding and macronutrient selection. Brain Res (Amsterdam) l989;503:132-40. Leibowitz SF. Hypothalamic paraventricular nucleus: interaction between 2-noradrenergic system and circulating hormones and nutrients in relation to energy balance. Neurosci Biobehav Rev l988;l2: 10 1-9. Wurtman II, Wurtman Ri. Fenfluramine and fluoxetine spare protein consumption while suppressing caloric intake by rats. Science
l977;I98:l 178-80.

Downloaded from www.ajcn.org by on April 16, 2009

23. Luo S, Ransom

10. Wurtman RI, Wurtman II. Nutrients, neurotransmitter synthesis, and the control offood intake. In: Stunkart Ai, Stellar E, eds. Eating and its disorders. New York: Raven, 1984:77-86. 1 1. Hutson PH, Donohoe TP, Curzon G. Infusion of the 5-hydroxytryptamine agonists RU24969 and TFMPP into the paraventricular

T, Li ETS. Selective increase in carbohydrate intake in rats treated with 8-hydroxy-2-(DI-N-Propylamino)-Tetraline or Buspirone. Life Sci l990;46: 1643-8. 24. Hill Ai, Blundell JE. Model system for investigating the actions of anorectic drugs: effects of d-fenfluramine on food intake, nutrient selection, food preferences, meal patterns, hunger and satiety in healthy human subjects. In: Ferrari E, Brambilla F, eds. Disorders of eating behaviour: a psychoneuroendocrine approach. Oxford: Pergamon Press, I 986:377-89.