5th Congress Autism-Europe
Articulos / Proceeding
Autism-Spain

FEEDING BEHAVIOR IN A GROUP OF AUTISTIC MENTALLY RETARDED CHILDREN

INTRODUCTION

Among the symptoms of autism, feeding behavior disturbances, both in terms of nonspecific (hyperactivity, aggression, stereotypies) and specific (anorexia, hyperphagia, pica) alterations, have often been described (O'Brien and Whitehouse, 1990; Raiten, 1987).

These problems in autistic patients have been thought, in the past, to be related to gastrointestinal changes (such as malabsorption, poor digestion, etc.); however, we tried to look at a totally new approach to these behavioral abnormalities relating them with brain blochemistry.

In fact, dysfunctions of various neurotransmitters and neuropeptides have already been described in autism such as increased serotonin (Anderson et al., 1987) and norepinephrine (Launay et al., 1987). However, in this respect, Coleman and Gillberg (1985) emphasized the variability of most biochemical results and suggested the existente of subgroups of autistic subjects.

Also, an "opioid excess" theory of autism has been suggested by Panksepp (1 979) which was further confirmed by Gillberg (1988) who reported increased levels of endorphin-like substances in the cerebrospinal fluid of some autistic patients.

It is interesting to note that many neurotransmitters and neuropeptides are deeply involved in the modulation of appetite and feeding. In fact, dopamine (Leibowitz and Rossakis, 1979), norepinephrine and serotonin (Leibowitz and Jhanar-Uniyal, 1991), opioids (Morley, 1983), corticotropin-releasing factor (CRF) and glucocorticoids (York, 1991) play an important role in animals and in humans, and also benzodiazepines have been reported to have significant effects on food intake (Cooper, 1989).

Thus, the aim of our investigation was to analyze the features of the feeding behavior of a group of autistic subjects with mental retardation and those of a control group comprised of subjects affected by mental retardation alone in order to better describe their characteristics and to try to understand the importance of the neurochemical involvement of these subjects as revealed by their feeding behavior alterations.

PATIENTS AND METHODS

We examined 9 autistic subjects, diagnosed according to the DSM-III-R (American Psychiatric Association, 1987), aged 10-15 years (mean age 12 years), who were also mentally retarded (IQ ranged between 20 and 50), and an age-matched control group formed by the same number of inpatients with a diagnosis of mental retardation alone (IQ 20-50).

The degree of the autistic disorder was assessed by means of the Childhood Autism Rating Scale (CARS; Schopler et al., 1980).

The observation lasted for one month, at meal time, using a questionnaire filled-out by the children's tutors during their daily activities, investigating their feeding habits.

Data were reported in a form with 16 items comprising: autonomy, mastication, hyperactivity, stereotypies, aggression, play, regurgitation, merycism (rumination), vomiting, pica, food refusal, and request for food/voracity (see Table l). Also quantitative aspects of feeding, qualitative aspects of feeding, mean duration of the meal, and pharmacological therapy at the time of the observation were considered (see Table 11).

For each subject, a general summary was obtained including: feeding anamnesis, sleep behavioral characteristics and nocturnal request for food, and ponderal parameters. The summary contained also the results from laboratory blood tests for anaemia, metabolic diseases and food allergy tests. The Fisher exact probability and the MannWhitney test were used for the statistical evaluation of results (see tables I,II and III).

RESULTS

The mean CARS evaluation was 47.05 (range 37.5-51.5) in the autistic patients, all the control subjects had a CARS evaluation lower than 30.

Table 1 shows the data from the observation form, most of the subjects reached a satisfactory autonomy level for food consumption, with no differences between the two groups, even if some of them liked to use their hands andlor liked to be fed.

Mastication was also adequate in most patients, only two autistic subjects failed to reach an adequate mastication.

As expected, stereotypies were significantly more evident in the autistic subjects and were not affected by the presence of food.

Aggressive behavior was not a severe entity and did not seriously compromise food intake, on the other hand the availability of food did not reduce aggression; however, a slight difference could be observed between the two groups, with a statistical significance for aggression to objects, and for self aggression which were more evident in the autistic group.

Play was more frequently present in autistic subjects, vomiting and regurgitation were not frequent but the latter was significantly more often present in the autistic subjects. Also pica was more often present in the autistic patients.

Food refusal and request for different foods and voracity were all significantly more evident in the autistic children. Overall, it is possible to note that the incidence of feeding disturbances was lower in the control group.

Table 11 shows the quantitative aspects of feeding which are different between the two groups under investigation because some autistic subjects consumed abundant food and others only selected types of food. Also the qualitative aspects of feeding are different between autistic and mentally retarded subjects because most autistic children preferred spicy/tasty or sweet food.

The mean duration of the meal is considerably shorter in autistic subjects, 15 minutes versus 25 minutes in mentally retarded patients. Table 11 shows information about therapy in both groups and, even if autistic patients were more often under drug therapy, none of the drugs was administered in a large number of subjects.

From the anamnestic forms (Table III) we can observe a lower frequency of breast feeding in autistic patients who also had presented more frequently mastication difficulties during their eariy childhood.

History of anorexia, malabsorption, anaemia, and ponderal parameters were not statistically different in the two groups and did not show any important incidence. Food allergy was reported in 2 subjects with autism.

It is interesting to note that the simple behavioral analysis of sleep shows differences between autistic subjects and mentally retarded children; however, this is not of crucial importance for this study and more information can be found in the papers by Elia et al. (1991) and Segawa et al. (1992) about this topic.

DISCUSSION

Several experimental studies have demonstrated that various monoamines, amino acids and neuropeptides, correlated with the hypothalamic function, are able to modulate the feeding behavior.

(1) Dopamine agonists and, in particular, D2 agonists, like lisuride, injected into the lateral perifornical hypothalamus, reduce food intake (Leibowitz and Rossakis, 1979). On the other hand, Dl agonists do not influence food consumption. D2 antagonists, however, like sultopride and (-)sulpiride antagonize the effects of lisuride and induce an increase in food intake (Nisoli et al., 1991).

(2) Norepinephrine induces an increase in carbohydrate intake (Leibowitz and Jhanwar-Uniyal, 1991), while serotonin has an inhibitory effect on feeding behavior and .reduces the consumption of carbohydrates probably through an activation of satiety mechanisms (Leibowitz and Jhanwar-Uniyal, 1991).

(3) Opioids increase and opioid antagonists decrease food intake in the animal and in humans (Morley, 1983). In particular, opioid administration seems to increase fat and protein consumption (Bhakthavatsalam and Leibowitz, 1986).

(4) CRF and glucocorticoids show important effects on food intake and energy expenditure. CRF, in particular, shows an inhibitory effect on food consumption and enhances termogenesis through a central stimulation of the sympathetic system (York, 1991).

(5) Benzodiazepine receptors play an important role in taste preference for sweet (Cooper, 1989) and treatment with benzodiazepines can eficit a hyperphagic response (Cooper, 1980). Also, new benzodiazepine partial agonists have been described which only increase food intake without sedative effects (Yerbury and Cooper, 1987) and bezodiazepine receptor inverse agonists show anorectic effects (Cooper, 1985; Cooper et al., 1985).

Regarding our autistic subjects we identified three feeding patterns definitely different from those of the mentally retarded subjects.

The first pattern is that of food refusal and regurgitation. The three subjects with this pattern usually consumed their meal in a very short time (10-15 minutes) and in two of them mastication was almost absent. Autonomy was very low and there was pica.

As pointed-out above, decreased food intake is connected with the action of dopamine agonists and/or serotonin (Leibowitz, 1979; Nisoli et al., 1991) and with CRF (York, 1991) and opioid antagonists (Morley, 1983) effects, in animals.

The second pattern is that of request for food, voracity, abundant food and fast eating, observed in other three autistic subjects. These subjects usually showed good autonomy and mastication, and two had sporadic pica.

Following the above indications, hyperphagy and appetite are correlated with the action of D2 antagonists (Nisoli et al., 1991), norepinephrine (Leibowitz and JhanwarUniyal, 1991), opioids (Morley, 1983), and benzodiazepines (Cooper, 1980), in experimental studies.

The third feeding behavior pattern identified in our subjects was that of the three children who refused food but also requested it, at the same time. This is only apparently contradictory because we usually observed, in these patients, refusal for some types of food and request for spicyltasty or sweet foods.

It is important to note that opioids seem to increase fat and protein intake (Bhakthavatsalam and Leibowitz, 1986) while the GABAergic system could be involved in preference for sweet food (Cooper 1989), in animals.

Finally, we do not think that our results were significantly influenced by therapy because subjects under centrally-acting drugs were equally present in all behavioral subgroups.

CONCLUSIONS

Our behavioral study can only provide thorough indications on neurotransmitters and neuropeptides involved in autism; however, a clear involvement of different systems is evident and it is difficult to ascribe to only one of them a primary role. From our point of view it is possible to indicate that an heterogeneity of feeding behavior can be observed in autistic mentally retarded patients causing the three different patterns described above and probably connected with different neurochemical mechanisms.

The future advances of this research will be in the direction of increasing the number of subjects under investigation and correlating their behavioral characteristics to seric and urinary indicators of different neurochemical functions. We do not believe that cerebrospinal fluid analyses can be carried out as extensively as suggested by Gillberg (1990) only for research purposes.

ACKOWLEDGEMENTS

We would like to thank Dr. Mary Coleman and Dr. Paul Shattock for their helpful suggestions.

REFERENCES

*American Psychiatric Association. Diagnostic and statistical manual of mental disorders (3rd ed., rev.). Washington, DC: Author, 1987.

*Anderson GM, Freedman DX, Cohen DJ, Voikmar FR, Hoder L, McPhedran P, Minderaa RB, Hansen CR, Young JG Whole blood serotonin in autistic and normal subjects. J Child Psychol Psychiatry 1987;26:885-900.

*Bhakthavatsalam P, Leibowitz SF. Morphine-elicited feeding: Diurnal rhythm, circulating corticosterone and macronutrient selection. Pharmacol Biochem Behav 1986;24: 911-917.

*Coleman M, Gillberg C. The biology of the autistic syndromes. New York, USA: Praeger, 1985.

*Cooper SJ. Effects of chlordiazepoxide and diazepam on feeding performance in a foodpreference test. Psychopharmacology 1980;69:73-78.

*Cooper SJ. The anorectic effect of FG 7142, a partial inverse agonist at benzodiazepine recognition sites, is reversed by CGS 8216 and cionazepam but not food deprivation. Brain Res 1985;346:190-194.

*Cooper SJ. Benzodiazepine receptor-mediated enhancement and inhibition of taste reoctivity, food choice, and intake. Ann NY Acad Sci 1989;575:321-336.

*Cooper SJ, Barber DJ, Gilbert DB, Moores WR. Benzodiazepine receptor ligands and the consumption of a highly palatable diet in nondeprived male rats. Psychopharmacology 1985;86:348-355.

*Elia M, Ferri R, Musumeci SA, Bergonzi P. Rapid eye movement modulation during night sleep in autistic subjects. Brain Dysfunct 1991;4:348-354.

*Gillberg C. The role of the endogenous opioids in autism and possible relationships to clinical features. In L. Wing (Ed.), Aspects of autism: Biological research. London,

England: Gaskell, 1988:31-37.

*Launay JM, Bursztein C, Ferrari P, Dreux C, Braconnier A, Zarifian E, Lancrenon S, Fermanian J. Catecholamine metabolism in infantile autism: A controlled study of 22 autistic children. J Autism Dev Disord 1987; 1 7:333-347.

*Leibowitz SF, Jhanwar-Uniyal M. Hypothalamic neurotransmitters controlling food intake. In G. Racagni, N. Brunello, T. Fukuda (Eds.), Biological psychiatry, vol. 2 (273-275). Amsterdam, The Netherlands: Elsevier, 1991:273-275.

*Leibowitz SF, Rossakis C. Mapping study of brain dopamine- and epinephrine-sensitive sites which cause feeding suppression in the rat. Brain Res 1979;172:101-113.

*Morley JE. Neuroendocrine effects of endogenous opioid peptides in human subjects: A review. Psychoneuroendocrinology 1983;8:361-379.

*Nisoli E, Memo M, Tonello C, Pizzi M, Spano PF, Carruba MO. Central and peripheral dopamine receptors involved in the regulation of feeding behavior and energy expenditure. In G. Racagni, N. Brunello, T. Fukuda (Eds.), Biological psychiatry, vol. 2. Amsterdam, The Netherlands: Elsevier, 1991:279-281.

*O'Brien G, Whitehouse AM. A psychiatric study of deviant eating behavior among mentally handicapped adults. BrJPsychiatry l990;157:281-284.

*Panksepp J. A neurochemical theory of autism. Trends Neurosci 1979;2; 1 74-177.

*Raiten DJ. Nutrition and developmental disabilities iissues in chronic care. In E. Schopler & G.B. Mesibov (Eds.), Neurobiological issues in autism. New York, USA: Plenum Press, 1987:325-338.

*Schopler E, Reichier RJ, De Vellis RF, Daly K. Toward objective classification of childhood autism: Childhood autism rating scale. J Autism Dev Disord 1980;10:91103.

*Segawa M, Katoh M, Katoh J, Nomura Y. Early modulation of sleep parameters and its importance in later behavior. Brain Dysfunct 1992;5:211-223.

*Yerbury RE, Cooper SJ. The benzodiazepine partial agonists, Rol6-6028 and Rol7-1812, increase palatable food consumption in nondeprived rats. Pharmacol Biochem Behav 1987;28:427-431.

*York DA. CRF and glucocorticoids: Central regulation of appetite and peripheral metabolism. In G. Racagni, N. Brunello, T. Fukuda (Eds.), Biological psychiatry, vol. 2. Amsterdam, The Netherlands: Elsevier 1991:285-287.

Somet: sometimes; N.S.= not significant. ata in fields "never" and "somet" were pooled together and compared with those in "often" and "always" for the computation of the Fisher exact probability, with the exception of "Regurgitation" where "never" was compared with "somet"

Fisher exact probability ** Mann-Withney test

N.S. = not significant