Authors: Marcos, A (1), Loureda, M.C. (1) and Diez-Cuervo, A. (2)
(1) Centro Mixto Instituto de Nutrición y Bromatologia. (2) Medical Consultant Associations "Nuevo Horizonte" and "Pauta".
Address: Centro Mixto Instituto de Nutrición y Bromatologia. Facultad de Farmacia. UCM.
Ciudad Universitaria. 28040 Madrid. SPAIN
SUMMARY
Autistic children have been described
as having bizarre eating habits, which may lead to situations
of clinical or subclinical malnutrition. It is well recognized
that dietary factors play an important role in maintaining immune
defences. Thus, immunocompetence has been recently shown to be
a sensitive and functional measure of the nutritional status.
Since, autistic children have been reported to be exceptionally
free from infectious diseases, and because of the literature is
scarce about the relationships between nutritional status and
immunocompetence in this syndrome, the purpose of this work was
to find out the nutritional assessment of autistic children by
evaluating their immunocompetence. The results were compared
to those obtained from a control group. The study involved 20
autistic children ranging in ages from 4 to 12 years, who were
diagnosed according to DSM IV (American Psychiatric Association,
1994). The patients were divided into two groups: 1) with eating
disorders (EDA) (n=9) and 2) without eating disorders (NEDA) (n=11).
Control subjects included 11 healthy children (brothers of the
patients) matched by age and sex, who were free of medical, psychiatric,
and neurological conditions. Leukocyte and lymphocyte counts
were tested. Lymphocyte subsets: CD2, CD3, CD4, CD8, CD19 and
CD57 were determined by flow cytometry. No modifications were
found in lymphocyte subsets between all the autistic children
(n=20) and the control group, but both total number and percentage
of CD19 cells were higher in the autistic children. However,
when the three groups were compared each other, surprisingly the
highest values for CD2, NK and CD19 cells were found in the EDA
group. The results suggest that contrary to what was expected,
neither EDA nor NEDA show signs of malnutrition, having the highest
values of lymphocyte subsets the autistic group with eating disorders.
Therefore, there might be some defence mechanisms involved where
neurotransmitters could play an important role.
INTRODUCTION
From a "symptomatic" nutritional perspective, it is important to stress the fact that autism, as a behavioural syndrome, is characterized, very frequently, by abnormal behaviour towards food (1,2). This behavioural abnormality can be expressed in different forms, such as anorexic and bulimic behaviour, or an "extreme selectivity" of intake behaviour that could be related, perhaps, to the trait of "lack of flexiblilty" and "insistence on sameness", characteristic of autism (3).
In school years, and in relation to "educational treatments" given to autistic children, abnormal behaviour affecting nutrition is less frequent and, usually, less intense. Other factors, such as sex, seem to be also related to frequency and intensity of abnormal food behaviour: although autism is more frequent in males, this bizarre disorder is more frequent and intense in autistic females (Riviere, unpublished data).
The potential for behaviour affecting nutrition is at least as great as the potential for nutrition to affect behaviour. When behaviour involves eating disorders, it may lead to situations of clinical or subclinical malnutrition. Thus, it is probable the development of a "feedback system", between nutrition and behavioural abnormalities or deficiencies in many cases of autism (4).
There is no evidence of an increased requirement for specific nutrients in autistic children. "Abnormal food behaviour" of autistic patients is not well known, and there is not a clear interpretation of their significance. It is not known, for example, whether autistic patients tend to "select" specific foods or nutrients, in their selective nutritional behaviours or tend to avoid some others. Incidence in autism of bizarre nutritional behaviours is unknown. Moreover, little attention has been paid to nutrition in the broader sense, as an integral component in the health of the autistic child (5).
Although there has been considerable debate, -partially as a consequence of the influence of hypotheses and research - about the potencial role of nutritional factors in autism (6), there is little concrete evidence to support either an etiological link or nutritionally based treatment modality in autism (4). Furthermore, it is not clear if autistic children suffer from the hidden
symptoms of malnutrition (5).
On the other hand, it is well recognized that dietary factors play an important role in maintaining immune defences, since malnutrition is the most common cause of secondary immunodeficiency all over the world, impairing mainly cell-mediated immunity, which is altered at an early stage in the development of undernutrition (7). Different immunological changes have been seen with different degrees of malnutrition and may or may not be clinically apparent. Some of them are subclinical and only evident by specific testing (8,9). Defence and lymphoid tissues can be affected not only by illness, but also in apparently normal situations. Thus, an immunocompetence study may be a useful tool to evaluate immune function in altered nutritional status (10).
Since the literature is scarce about
the relationships between nutritional status and immunocompetence
in autism, and as immunocompetence has been recently shown to
be a sensitive and functional measure of the nutritional status
(1 1,12), the aim of this work was to assess the nutritional status
of autistic children by evaluating their immunocompetence.
SUBJECTS AND METHODS
The study involved 40 autistic children,
ranging in ages from 4 to 12 years. Diagnosis of autism according
to DSM-IV (13) was assured by clinical assessment of two experts.
Patients with autism were divided into two groups, according
to their food behaviour: 1) with eating disorders (EDA) (n=9,
5 girls and 4 boys) and 2) without eating disorders (NEDA) (n=31,
6 girls and 25 boys).
The control group involved 11 healthy
children (6 girls and 5 boys) were brothers of the autistic patients
and matched by age and sex. They were free of medical, psychiatric,
neurological conditions, and these subjects were also medication
free during the study.
All the parents of all the subjects
tested in the current study (autistic patients and controls) together
with patient caregivers gave informed consent once the purpose
and nature of the study was explained. The procedures followed
were in accord with the Helsinki Declaration as updated in Tokyo
(Japan), 1975 and revised in 1983.
METHODOLOGY
Blood samples were taken after 12-15h of fasting. Leukocyte and lymphocyte counts were assessed by routine analytical methods (Coulter Counter, Hialeah, Florida).
Assessment of the lymphocyte subpopulations: Whole blood samples (100 l) were incubated with 10 l of appropiate titered monoclonal antibodies (Coulter Clone, Coulter Corporation, Hialeah, Florida) at 4ºC for 10 min. in order to evaluate the following lymphocyte subsets: CD2 (pan-T cells), CD3 (mature T cells), CD4 (helper T-cells), CD8 (cytotoxic/supressor Tcells), CD 19 (B-Iymphocytes) and CD57 (natural killer cells) by flow cytometry. Each sample was processed by Immunoprep EPICS lymphocyte preparation system. The Immunoprep reagents include a lysing agent for elimination of erythrocytes, a stabilizer for the leukocytes, and a fixative to maintain sample integrity (14). The fluorescence of the subsets was analyzed with an FACSTAR PLUS DUAL LASER Cytometer (Becton Dickinson, Sunnyvale, CA). Forward light scatter intensity combined with ring-angle scatter was analyzed using the appropiate software.
Statistics: All the results are expressed
as means ± SD. Comparisons between controls and all the
autistic children were performed by the Student's t test. When
the three groups (controls, EDA and NEDA) were compared each other,
the differences among them were subjected to analysis of variance
to detect differences between groups (ANOVA). Statistical analysis
was performed by using the SAS computer program (P<0.05) (15).
| AGE (years) | |||
| HEIGHT (cm) | |||
| WEIGHT (kg) | |||
| BMI (kg/m2) * | |||
| IBW (%)* | |||
* ANOVA (significant differences, p<0,05)
Different superscripts mean significant differences (Student's
t test, p<0,05)
| HEIGHT (cm) | ||||||
| WEIGHT (kg) | ||||||
| BMI (kg/m2) | ||||||
| LEUKOCYTES (1) | |||
| LYMPHOCYTES (1)* | |||
| CD2 (1)* | |||
| CD3 (1) | |||
| CD4 (1) | |||
| CD8 (1) | |||
| CD19 (1)* | |||
| CD57 (1) * | |||
(1) cells / mm3
* ANOVA ( significant differences, p<0,05)
Different superscripts mean significant differences (Student's
t test, p<0,05)
| LYMPHOCYTES (1) | |||
| CD2 (1) | |||
| CD3 (1) | |||
| CD4 (1) | |||
| CD8 (1)* | |||
| CD19 (1)* | |||
| CD57 (1) | |||
| CD4/CD8 | |||
| CD2/CD19 | |||
(1) percentage
* ANOVA ( significant differences, p<0,05)
Different superscripts mean significant differences (Student's
t test, p<0,05)
RESULTS AND DISCUSSION
Regarding anthropometric parameters, no significant modifications were shown (Table l), except for BMI value and IBW percentage, which were lower in EDA group than in NEDA group, although all of them (controls and autistic children) were within a percentil range between 3 and 97 (16).
It is important to stress the fact
that percentil values of the anthropometric parameters were different
depending on the group tested (Table 2). Thus, regarding height
percentil, NEDA group reached higher values (P87-90) than EDA
group (P80-82). In relation to body weight, percentil values
for controls reached the highest level (P90-93) in comparison
with NEDA group (P63-85) and especially with EDA group (P50-60).
It is noticeable that the most significant difference was shown
in BMI of EDA group which was much lower (P30-35) than the other
two groups (P60-65). These results might reveal a possible deficient
nutritional status for those patients with eating disorders, independently
of sex (16).
In relation to white blood cells
(Tables 3 and 4), all the values were within the normal range
established by Vives (17). Surprisingly, total lymphocyte counting
was higher in EDA group than in controls. In addition EDA group
reached higher CD2 cell number than NEDA group (Table 3).
In the current study EDA group showed
the highest lymphocyte values in comparison with the other two
groups. This outcome would be in disagreement with that observed
by other authors who have pointed out lymphocyte deficits.
Stubbs et al. (1 8) and found that
autistic children ranging in ages from 2 to 12 years exhibited
a depressed lymphocyte transformation response to PHA when compared
to the control subjects. The type of defect postulated would
be of the thymic-derived lymphocytes (T cells) and would be expressed
either as a) weak manifestation of cell-mediated immunity or b)
deficient antibody responses to T-dependent antigens. A defect
in T cell function may make the fetus more susceptible to viral
attack and subsequent damage of dysfunction.
Likewise, Warren et al. (19) found decreased number of T lymphocytes and an altered ratio
of helper to suppressor T cells in 31 autistic patients aged from 3 to 28. However, regarding CD4/CD8 ratio in the current study, no modification among groups was shown (Table 4). This result could mean a correct nutritional status, since CD4/CD8 ratio is well-known to be capable to detect any situation of malnutrition, even at a subcllnical level (10, 11).
A common finding in immune system of patients suffering from an autoimmune pathology is the deficit of certain type of T lymphocytes (suppressor T cells) (20), which mantain immunological homeostasis by preventing immune responses against tissue cells.
Under protein-energy malnutrition, both children and adults show a diminished lymphocyte number (7). Also, there might be a failure of serum thymic factor or thymopoyetin, which is essential for lymphocyte differentiation in malnourished children (21,22). The reduction of lymphocytes is usually at the expense of a decreased T-Ivmphocyte number, since B lymphocytes are not affected (21). The most affected T cell subset has been shown to be CD4, which collaborates with other cells in order to destroy foreign agents from the organism, while CD8 (cytotoxic/suppressor) cells might be even increased (23,24).
Lymphocyte blastogenesis response to certain mitogens is decreased in autism, leading to a higher susceptibility to infection. Immune system alterations might be directly related to underlying biologic processes of autism or these changes may be an indirect reflexion of actual pathological mechanism (20).
Ferrari et al. (25) found in 16 autistic patients between 5 and 16 years old a significant correlation between lymphocyte activity against mitogens and the hyperactivity of the patients. There also are interrelationships between central nervous system and immune system, so that certain brain lesions might alter immune response, parasympathic estimulations increase the antibody production and lymphocyte cytotoxicity, meanwhile sympathic estimulations lead to a reduced response.
Likewise, a great number of authors have pointed out a defect of lymphocyte activity both in experimental malnutrition (26) and in malnourished subjects (27). Thus, cell-mediated immunity determined "in vivo" by delayed hypersensitivity skin tests has been shown to be decreased under malnutrition (28).
Regarding natural killer cells, no modifications were observed in the current study when autistic children were compared with controls. In additlon, EDA group showed a higher number of CD57 cells than NEDA group (Table 3).
On the opposite, Warren et al. (19) speculated about the possibility of the existance of an underlying autoimune process in autistiv children, since a lower natural killer activity was found, which would be associated to autoimmune processes. Similar observations have been pointed out in acute starvation, showing a depressed natural killer cell activity in proteincalorie malnutrition which could be corrected by proper dietary intake (29).
According to Tables 3 and 4, EDA group showed a higher CD19 lymphocyte number than the other two groups. When CD2/CD19 ratio values, index of nutritional status (12), were compared between controls and autistic children, a decrease was observed in the autistic patients (Table 4).
In this sense, B lymphocytes have been shown not to be significantly different in autistic patients, although T cells are commonly affected (18). In fact, Plioplys et al. (30) evaluating 17 autistic children (aged between 3 and 23 years) did not found alterations in B cell number or function throughout proliferation and "in vitro" IgG and IgM synthesis in response to pokeweed mitogen. Two possible explanations are: 1) that autistic children have a genetic predisposition to a relative T cell deficiency and not B cell deficiency; and 2) that certain viruses have a predilection for interfering with the thymus which differenciates T cells and not a predilection for interfering with the bone marrow which differenciates B cells (18).
Regarding CD19 lymphocyte function, certain type of alterations have been described under malnutrition conditions (31). Thus, serum immunogiobulin rates could be normal or slightly increased due to concomitant infections, suggesting a priority in the antibody production to save B lymphocyte functionality in malnourished subjects.
Despite the fact that the anthropometric parameters could reveal a deficient nutritional status for those autistic patients with eating disorders, immunological parameter results do not show a subclinical malnutrition situation. On the opposite and surprisingly, an increase of immunological parameters is found in the autistic children with eating disorders.
This outcome could be due to an overlap
on the immune system of neurotransmitters, which are known to
be altered. Further research is required to elucidate which mechanisms
could be envolved.
REFERENCES
1.- Wing L. Social, behavioural and cognitive characteristics, En: Rutter M, Schopler E, editores. Autism. A reappraisal of concepts and treatment. New York: Plenum, 1978;27-45.
2.- Wing, L. Children Apart. Washington, D.C.-. National Society for Autistic Children, 1979.
3.- Kanner L. Autistic disturbances of affective contact. Nervous Child 1943;2:217-50.
4.- Ralten DJ, Massaro TF. Nutrition and developmental disabilities: an examination of the
orthomolecular hypothesis. En: Cohen DJ, Donnellan A, Paul R, editores. Handbook of autism and disorders of atypical development. New York: Wiley, 1986.
5.- Ralten DJ. Nutrition and Developmental Disabilities. Clinical Assessment. En: Schopler E, Mesibov G, editores. Diagnosis and assessment in autism. New York: Plenum, 1988;211- 23.
6.- Rimland B. High dossage levels of certain vitamins in the treatment of children with severe mental disorders. En: Hawkins D, Pauling L, editores. Orthomolecular Psychiatry. S. Francisco-. WH Freeman, 1973;513-39.
7.- Miller K. Nutrition and immunity. Nutr Bull 1987-49-.32-40.
8.- Marcos A, Varela P, Santacruz 1, Muñoz-Vélez, Morandé G. Nutritional status and
immunocompetence in eating disorders. A comparative study. Eur J Clin Nutr 1993;47:787-93.
9.- Marcos A, Varela P, Santacruz I, Muñoz-Vélez A. Evaluation of immunocompetence and nutritional status in patients with bulimia nervosa. Am J Clin Nutr 1993;57:65-9.
10.- Tojo R, Regueiro BJ. Evaluation of immunological parameters in malnutrition. En: Fidanza F, editor. Nutritional status assessment methodology for individual and populations groups. Perugia: Perugia University, 1986;349-54.
1 1.- Chandra RK. Immunocompetence is a sensitive and functional barometer of nutritional status. Acta Scand Suppl 1991;374:129-32.
12.- Marcos A, Varela P, Santacruz I. Nutritional status in non-infected and human immunodeficiency virus-infected asymptomatic drug addicts. Immunological assessment. Immunol Infect Dis 1993;3-.245-8.
13.- American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders (DSM IV). (1st. ed.). Washington D.C. 1994.
14.- Baker JW. An innovate lymphocyte preparation system for flow cytometry. Am Clin Lab 1988-1120:320-4.
15.- SAS Institute INC. SAS/STAT User's guide version 6. 4th ed. vol 1,2. Cary, NC: USA:SAS Institute Inc., 1989.
16.- Hernández M, Castellet J, Navaíza JL, Rincón JM, Ruiz I, Sánchez E, et al. Curvas y tablas de crecimiento. Instituto de investigación sobre crecimiento y desarrollo. Fundación F. Orbegozo. Bilbao:Garsí, 1988.
17.- Vives JL. En: Hematología clínica. Barcelona: Sans-Sabrafen, 1988.
18.- Stubbs EG, Crawford ML, Burger DR, Vanderbark AA. Depressed lymphocyte
responsiveness in autistic children. J Autism Child Schizo 1977-17:49-55.
19.- Warren RP, Margaretten NC, Pace NC, Foster A. Immune abnormalities in patients with autism. 1986:189-97.
20.- Strelkauskas AJ, Calleri RT, McDowell J, Borel Y, Schlossman SF. Direct evidence for loss of human suppressor cells during active autoimmune disease. Proc Nat Acad Scl 1978;75:5150-4.
21.- Chandra RK. Nutritional regulation of immunity: an introduction. Nutrition and immunology. New York: Alan R Liss Inc, 1988:1-8.
22.- Wade S, Bleiberg FK, Moose A et al. Thymulin (Zn-Facteur thymlque serique) activity in anorexia nervosa patients. Am J Clin Nutr 1985;41:275-80.
23.- Dowd PS, Heatley RV. The influence of undernutrition on immunity. Clin Scl 1984;66:241-8.
24.- Tomkins AM. Protein-energy malnutrition and risk of infection. Proc Nutr Soc 1986,45:289-304.
25.- Ferrarl P, Marescot MR, Moullas R, Bursztejn C, Deville-Chabrolle A, Thiollet M. et al. État immunitaire dans l'autisme infantile. Correlations entre état immunitaire, symptômes autistiques et taux de sérotonine. Encephale 1988-14:339-44.
26.- Lamont AG, Gordon M, Ferguson A. T lymphocyte function in protein deprived mice.
Clin Exp Immunol 1988;72:113-7.
27.- Varela, P., Marcos, A., Ripoll, S., Requejo, A., Herrera, P. y Casas, A. Nutritional status
assessement of positive HIV drug addicts. Eur J Clin Nutr 1990;44:415-8.
28.- Suskind RM. Malnourished child. Nestlé Nutrition Editor. Workshop Series. Vol. 19. New York: Raven Press, 1990.
29.- Salimonu LS, Johnson AOK, Williams AIO, Adeleye GI, Osunkoya BO. Lymphocyte subpopulations and antibody levels in immunized mainourished children. Brit J Nutr 1982;48:7-14.
30.- Plioplys AV, Greaves A, Kazemi K, Silverman E. Autism anti-21OK neurofilament immunogiobulin reactivity. Neurology 1989;39(Suppll):187.
31.- Steihm ER. Humoral immunity in malnutrition. Fed Proc 1980;39:3093-7.