Nutritional deficiency and susceptibility to infection.

Bulletin of the World Health Organization 57 (2): 167-177 (1979) Nutritional deficiency and susceptibility to infection R. K. CHANDRA' Undernutrition and infection, individually but more often together, are major contributors to morbidity and mortality worldwide, particularly in underprivileged young children. Apart from the common causal factors of poverty, poor personal hygiene, and low standards of sanitation, nutritional deficiency increases the susceptibility to infection. Deficiencies of energy, protein, iron, folic acid, vitamin A, pyridoxine, and zinc impair various immune responses and could result in infection. Delayed hypersensitivity and the number of circulating T lymphocytes are consis- tently depressed. In small-for-gestation, low-birth-weight infants, cell-mediated immunity may remain impaired for several months. The number of intraepithelial lymphocytes is reduced. B lymphocytes, immunoglobulin levels, and antibody responses are generally normal but antibody affinity and the level of secretory IgA are reduced. Opsonization is decreased at lowplasma concentrations and intracellu- lar killing of microorganisms by polymorphonuclear leukocytes is impaired. Most serum complement components are low and the production of lysozyme and interferon is reduced. The level of protective immunity achieved by immunization of malnourished individuals requires further study. It is suggested that an integrated approach be employed to tackle the malnutrition-infection complex; its components should include the promotion of breast feeding, making immunization more effective, prevention of impaired immunocompetence associated with low birth weight, short- term nutritional supplementation, and immunopotentiation. Many underprivileged populations continue to be additionally handicapped by the interlinked problems of nutritional deficiency and infectious disease. However, this common association of malnutrition and infection does not necessarily imply a causal relationship. It must be recognized that the same economic, sociocultural, and ecological factors that contribute to undernutrition may well lead to a high frequency of infection. In this connexion, the significance of poor sanitation and personal hygiene and of early weaning with inappropriate or inadequate replacement foods needs to be highlighted. In this review, an analysis is made of (a) the epidemiological and clinical evidence pointing to an increased susceptibility of malnourished individuals to infection, (b) the effects of nutritional deficiencies and of infection, both individually and collectively, on the immune response, (c) the public health significance of these findings with particular reference to immunization, and (d) intervention strategies proposed to tackle the malnutrition-infection complex. SUSCEPTIBILITY OF UNDERNOURISHED INDIVIDUALS TO INFECTION Clinicians and public health professionals working in developing countries have long believed that infections occur more frequently and are more severe in the malnourished, and evidence has been cited to support this observation. Epidemiological data from field I Professor of Pediatric Research, Memorial University of Newfoundland, St John's, Newfoundland AIB 3V6, Canada. 3789 - 167 - 168 R. K. CHANDRA surveys in Asia and the Americas have confirmed the intimate association of nutritional deficiencies, growth failure, and infectious illness, particularly diarrhoeal disease. In a prospective study in Mexican children followed weekly from birth to 5 years of age, moderate malnutrition correlated significantly with the duration and severity of episodes of infectious disease and to a lesser extent with the incidence of diarrhoeal disease in the age period 8-18 months. The majority of hospitalized children with severe forms of malnutrition have an associated infection at the time of admission; this is more often observed in kwashiorkor than in marasmus. A Pan American Health Organization survey of childhood mortality patterns in the Americasa showed that 57% of children under 5 years of age who died had signs of intrauterine and/or postnatal nutritional growth retardation as either the primary or an associated cause of death. Many infectious diseases run a more severe course in nutritionally deprived children.b, c For example, measles is known to produce fatal giant-cell pneumonia in children with kwashiorkor. It is interesting that there is an unexplained geographical difference in measles-associated morbidity and mortality: the frightening severity of measles in African infants with kwashiorkor has not been observed in undernourished Asian children. Herpesvirus infection is often generalized and fatal in kwashiorkor; adrenal, hepatic, and cerebral haemorrhages are seen at autopsy and there is a conspicuous lack of inflammatory response. Septicaemia due to Gram-negative micro- organisms and respiratory infections caused by Pneumocystis carinii frequently compli- cate moderate-to-severe protein-energy malnutrition. Undernourished populations usually have a heavier load of parasites, particularly Ascaris lumbricoides, Necator americanus, Strongyloides stercoralis, Giardia lamblia, and Entamoeba histolytica. This is not to say that nutritional deficiency necessarily leads to heavy parasite infestation-the reverse is probably often true. However, in laboratory animals experimental protein deficiency generally facilitates protozoal and helminthic infection. Also, undernourished persons and small-for-gestation, low-birth-weight infants often show greater rates of infection with enteropathogens, for instance Shigella spp. The incidence and severity of pneumonia and diarrhoeal disease is greatly increased, with a consequently higher death rate, and the detection rate of hepatitis B surface antigen in the serum is several-fold higher. The intimate association between inanition and tuberculosis so well described by Osler in 1907 has been validated by recent observations. Protein-energy malnutrition There are a few isolated clinical and experimental observations that point to enhanced host resistance to some viral infections in undernutrition. However, most data suggest an increased susceptibility of individuals with protein-energy malnutrition (PEM) to infectious diseases. This has been supported by extensive experiments in laboratory animals in whom nutritional deprivation is almost invariably associated with aPUFFER, P. R. & SERRANO, C. V. Patterns of mortality in childhood. Washington, DC, Pan American Health Organization, 1973 (Scientific Publication No. 262). b ScIMHAw, N. S. ET AL. Interactions of nutrition and infection, Geneva, World Health Organization, 1968 (Monograph Series, No. 57). CCHANDRA, R. K. & NEWBERNE, P. M. Nutrition, immunity and infection: mechanisms of interactions. New York, Plenum, 1977. NUTRITIONAL DEFICIENCY AND INFECTION 169 increased morbidity and mortality following challenge with a variety of microorganisms. In recent years, there has been increasing recognition of the frequent presence of PEM in hospital patients. In such subjects there is a good correlation between the presence of nutritional deficiencies on the one hand and infection and mortality on the other. In a study of patients with lymphoreticular malignancies, the occurrence of complicating Pneumocystis carinii infection was related to lower concentrations of serum proteins and albumin. This confirmed controlled observations in rats, whose susceptibil- ity to P. carinii infection was enhanced by protein deficiency. Hospitalized patients in many affluent countries have been observed to have a high frequency of malnutrition as judged by body weight, skinfold thickness, and serum albumin level, and this was correlated with their decreased ability to mount a delayed hypersensitivity reaction to common recall antigens. Postoperative sepsis and mortality were correlated with preoperative albumin and transferrin levels and with anergy (R. K. Chandra, unpublished observations, 1978). Nutritional support may reverse these changes and reduce the incidence of fatal pyogenic infections. These observations suggest that nutritional assessment is useful for selecting those individuals that may be at high risk of developing infection and who would benefit from short-term nutritional supplementation before imminent stress such as surgery. Other nutrient deficiencies Whereas there are considerable data on the mutually augmenting effects of PEM and infection, similar critical information for deficiencies of individual nutrients is not available. This is understandable, since nutritional deficiencies in man are often multiple and each nutrient may exert a variable, possibly even opposing, influence on the occurrence of infection. Some surveys have reported a decrease in the frequency of infection among infants fed iron-fortified milk, whereas others have failed to detect such an association. Iron deficiency has been noted to be a common denominator of chronic mucocutaneous candidiasis and therapy with iron preparations rapidly improved the dermatological lesions and associated immunological changes. An association between a higher frequency of infection and iron-deficiency anaemia has been seen in Indonesian plantation workers; iron supplements decreased the incidence of common gastrointesti- nal and respiratory infections: unfortunately, concurrent observations were not made in any control groups. A Tanzanian report suggesting a lower frequency of infection in iron- deficient subjects than in those with other types of anaemia has been difficult to interpret because of lack of data on any healthy controls in the same community. In laboratory animals, controlled deprivation of specific nutrients generally increases host susceptibility to pathogens and infection-related mortality. The extensive literature on the subject has been summarized.d IMMUNE RESPONSE IN NUTRITIONAL DEFICIENCY SYNDROMES The recent increase in immunological knowledge and advances in laboratory methods have stimulated a large number of investigations on the effect of malnutrition on the immunological responses of the host. These efforts have received encouragement, d CHANDRA, R. K. ET AL. In: Ciba Foundation Symposium on Iron Metabolism. Amsterdam, Elsevier, 1977. 170 R. K. CHANDRA coordination, and support from the World Health Organization and other international and governmental agencies. Many studies have employed a uniform, preplanned protocol and in some instances samples have been analysed or rechecked in a central reference laboratory. Ceil-mediated immunity The consistent demonstration of impaired cell-mediated immunity in moderate-to- severe undernutrition has been extensively reported.e f, g Several types of evidence have been collected: cutaneous delayed hypersensitivity to a battery of ubiquitous recall antigens or after deliberate sensitization with a chemical agent; the proportion and number of circulating T lymphocytes; lymphocyte DNA synthesis induced by mitogens and antigens; production of soluble mediators of immunological reactivity in response to mitogens and antigens; mixed lymphocyte response; and rejection of skin homografts. Several methodological principles need to be understood for the correct interpret- ation and comparison of results. The elicitation of a positive delayed hypersensitivity skin response requires the participation of several cell types and their mediators, and also depends on non-immunological dermal characteristics. Three separate components of cutaneous delayed hypersensitivity reactions are recognized: the afferent link, which involves the sensitization of antigen-reactive T lymphocytes and may require prior "processing" of antigen by macrophages; the efferent link, which consists of antigen recognition and the proliferative response of T lymphocytes, with consequent release of various soluble chemical mediators (lymphokines); and the inflammatory response, which is amplified by chemotactic factors. Impairment of one or more of these sequential processes will depress the skin response. Lymphocyte stimulation with plant lectin mitogens and antigens is an in vitro correlate of cell-mediated immunity. Lymphocyte proliferation can be assessed by morphological evaluation of blast transformation and by estimation of DNA synthesis by measuring the incorporation of radiolabelled thymidine. The concentration of lymphocytes, the presence and number of other cells (for example, polymorphonuclear leukocytes, mononculear phagocytes), and the amount of mitogen are critical determinants of the technique. A dose-response curve should be constructed, since the optimal concentration of phytohaemagglutinin required to produce maximum stimulation of lymphocytes may vary from one cell preparation to another. Most studies have shown a reduction in the proportion and absolute number of circulating T lymphocytes, identified by their capacity to form rosettes with sheep red blood cells., i It is possible that a change in the surface characteristics of the lymphocytes impairs rosetting. T cells are rich in membrane-bound acetylcholinesterase and the activity of this enzyme is reduced in malnutrition. Some investigators have found a decrease only in the number of lymphocytes forming " active " rosettes. Alteration in the number of circulating T cells appears to be an early indicator of nutritional deficiency and the abnormality is promptly reversed within 1-6 weeks of the provision of supplementary 'SMYTHE, P. M. ETAL. Lancet, 2: 939 (1971). f CHANDRA, R. K. Journal of pediatrics, 81: 1184 (1972). g NEUMANN, C. G. ET AL. Immunological responses in malnourished children. American journal of clinical nutrition, 28: 89 (1975). h CHANDRA, R. K. & NEWBERNE, P. M. Nutrition, immunity and infection: mechanisms of interactions. New York, Plenum, 1977. i SusKIND, R. M., ED. Malnutrition and the immune response. New York, Raven Press, 1977. NUTRITIONAL DEFICIENCY AND INFECTION 171 nutrients. There is a relative increase in the number of null cells without the surface characteristics of T or B lymphocytes; these null cells suppress the functions of other lymphocytes and are cytotoxic for a variety of xenogeneic cells in culture.JThe lymphoid organs show depletion of lymphocytes, particularly the thymus, the paracortical areas of the lymph nodes, and the periarteriolar regions of the spleen. Mitogen-induced lymphocyte DNA synthesis is generally reduced in moderate-to- severe PEM,k, 1, m but in one report the response of marasmic infants was normal. The production of a leukocyte inhibitory factor is normal. Sera of malnourished infected subjects may inhibit lymphocyte responses in vitro. Delayed hypersensitivity skin tests employing a variety of ubiquitous recall antigens or chemical sensitizing agents generally show a reduction in the number of positive responses and the size of induration. Immunoglobulins Serum levels of immunoglobulins are generally elevated in undernourished persons. Rarely, the IgG concentration may be low, particularly in low-birth-weight marasmic infants. There is limited data on the turnover of immunoglobulin; the plasma half-life is increased in most malnourished children, probably as a consequence of associated infection. The marasmic infant with a low serum IgG level may show a significant prolongation of the plasma half-life of radiolabelled IgG. In most instances adequate antigenic challenge elicits a normal serum antibody response, although antigens requiring the help of T lymphocytes may fare less well. For instance, antibody production after immunization with heterologous red cells is reduced in experimental animals deprived of energy, proteins, lipotropic factors, pyridoxine, iron, vitamin A, or zinc. Secretory IgA and mucosal antibody responses to viral antigens are also significantly reduced. This may be the result of a decrease in the numbers of IgA-producing plasma cells (Table 1) or reduced synthesis of secretory component. Impaired mucosal immunity may facilitate the development of septicaemia and of food antibodies. Data on antibody affinity in mal- nourished human subjects is conspicu- ously lacking. Table 1. Number of plasma cells (mean ± S. D.) in the jejunal mucosa, by immunoglobulin class a IgA 1gM IgG Malnourished 52 ± 15 31±10 9 ± 3 persons (57±16) (33±11) (10±3) Well nourished 91±17 19±5 4 ± 2 persons (80±15) (18±4) (4±2) ' No. of cells per tissue segment of 6 ,n x 500 lim. Figures in parentheses are percentages of all plasma cells. Opsonic activity Opsonic activity is related to the concentration of plasma employed in in vitro techniques. At a concentration of 5 % or higher, the ability of the serum to opsonize particles is comparable for both well nourished and undernourished individuals; however, at greater dilutions the serum of malnourished children has a lower opsonic activity. i CHANDRA, R. K. & NEWBERNE, P. M. Nutrition, immunity and infection: mechanisms of interactions. New York, Plenum, 1977. kSMYTHE, P. M. ET AL. Lancet, 2: 939 (1971). 'CHANDRA, R. K. Journal ofpediatrics, 81: 1184 (1972). m NEUMANN, C. G. ET AL. Immunological responses in malnourished children. American journal of clinical nutrition, 28: 89 (1975). 172 R. K. CHANDRA Phagocytosis In malnourished individuals phagocytosis is normal but intracellular bacterial killing is reduced., 0 The importance of using a discriminatory ratio of bacteria to phagocytes has recently been stressed. Intraepithelial lymphocytes The number and location of intraepithelial lymphocytes have recently been investi- gated; these cells have many of the surface characteristics and functions of T lymphocytes. In nutritionally deprived children there was a significant reduction in the number of intraepithelial lympho- cytes compared with healthy controls. This is likely to be the direct result of thymic atrophy associated with starva- tion, since neonatal thymectomy in rats produces similar effects and the addi- tional stress of nutritional deprivation in thymectomized animals does not increase the extent of alterations in the number of intraepithelial lymphocytes (Table 2). Leakage of these lymphocytes into dilated lymphatics of the villus has also been postulated. Table 2. Effects of neonatal thymectomy and nutri- tional deprivation on intraepithelial lymphocytes in 6-week-old rats "Stress" lntraepithelial lymphocytes (number per 100 mucosal epithelial Thymectomy Deprivation' cells) b + + 4.0±1.1 + 0 4.1±0.9 0 + 5.7 ± 1.4 0 0 15.6 ± 1.9 a Partial starvation was induced by changing the litter size. b Mean ± standard deviation. Other factors Many non-specific host protective factors are impaired in nutritional deficiency states. These include the complement system, interferon, and lysozyme. Structural changes in various tissues may also contribute to increased susceptibility to infection. These aspects of the immune response are reviewed elsewhere. q IMMUNIZATION OF MALNOURISHED INDIVIDUALS Since nutritional deficiency is associated with impaired immunological responses, the effectiveness of prophylactic vaccination must be carefully examined. The extensive literature on antibody production in response to a variety of bacterial, viral, rickettsial and other antigens has been analysed and summarized. It was concluded that the serum antibody level following adequate immunization is generally normal and that suboptimal response in undernourished subjects can be improved by giving a larger dose of the antigen, choosing appropriate adjuvants, and excluding serious concomitant infection. There have been reports of geographical variations in the response of young children to live attenuated poliovirus vaccine. In two studies in India, the frequency of seroconversion and titre of antibody were lower, but this was not correlated with nutritional status. Virus "take" indicated by the presence of vaccine virus in the stools nSETH, V. & CHANDRA, R. K. Archives of disease in childhood, 47: 282 (1972). 0 SELVARAJ, R. J. & BHAT, K. S. American journal of clinical nutrition, 25: 166 (1972). P CHANDRA, R. K. & NEWBERNE, P. M. Nutrition, immunity and infection: mechanisms of interactions. New York, Plenum, 1977. q SusiCND, R. M., ED. Malnutrition and the immune response. New York, Raven Press, 1977. NUTRITIONAL DEFICIENCY AND INFECTION 173 was generally associated with positive seroconversion. Antibody response improved and was comparable with titres achieved in trials in industrialized countries when poliovirus vaccine was administered five times. Occasionally, some malnourished children show a secondary booster response following the apparently primary administration of an antigen, thereby suggesting previous exposure and memory even though antibody was not detected prior to immunization. There are few available data on the local mucosal antibody response in malnutrition. This has been examined in only one study,' in which the presence and levels of secretory IgA antibody to live attenuated measles and poliovirus vaccines were significantly reduced. Clearly, there is a need for further information in this important area. There is limited information on specific cell-mediated immunity following prophylac- tic immunization. Following BCG vaccination, tuberculin conversion is observed less often in malnourished individuals. This impairment can only partly be overcome by testing with a higher concentration of purified protein derivative. Low-birth-weight infants who are small-for-gestation also show reduced tuberculin conversion and decreased lymphokine production. The critical factor in this area is the correlation between nutritional status and protective immunity to infection rechallenge, irrespective of the magnitude of the immune response produced. The protective titres of specific antibody in man are not well defined for most infections. It is conceivable that the lower levels of immune response encountered in nutritional deficiency syndromes may still be effective in warding off natural disease. Furthermore, careful studies are required of the incidence, nature, and severity of complications in immunized malnourished children following exposure to natural infection in the community setting. Moreover, most studies of immunological- nutritional interactions have demonstrated impaired responses in severe forms of malnutrition, such as kwashiorkor and marasmus, whereas these patients constitute only 1-2% of the malnourished population. The responses to immunization of individuals with mild-to-moderate PEM and of those with deficiencies of individual nutrients should be evaluated. Based on current information, there is full justification for continuing comprehensive immunization programmes in all populations, perhaps with some additional precautions to ensure an adequate response. INTERVENTIONS AFFECTING THE MALNUTRITION-INFECTION COMPLEX Practical intervention strategies to tackle the widespread problems of mal- nutrition and infection have recently been discussed.s A multifaceted attack on the twin problems of nutritional deficien- cy and infection must be mounted. Some of the important elements of this strategy are shown in Fig. 1. Fig. 1. Various types of intervention in the malnu- trition-infection complex. SANITATION / | MALNUTRITION- INFECTIO AGRICULTURAL v/ ~~CO^MPLEX PRODUC IONF & _'> % ~~~~~~ECONtY SELECTIVE SNORT-TERM NUTRITIONAL SUPPLEE WREVENTION OF LOW BIRTH WEIGHT IMMUNOPOTEN ION CHANDRA, R. K. British medical journal, 2: 583 (1975). t International Symposium on Nutritional Interventions, Instituteo de Nutrici6n y Tecnologia de los Alimentos, Universidad de Chile, Santiago, 20-24 August 1978. R. K. CHANDRA Breast feeding Human milk contains a number of physicochemical and cellular protective factors, including lactoferrin, antibodies, phagocytes, and macrophages. Breast feeding is associated with a lower fre- quency of diarrhoea and respiratory in- fections, and complications such as dehy- dration and septicaemia also occur less often. These effects are achieved in all communities but are greatest and most useful in communities with heavy environmental contamination and poor standards of personal hygiene. The com- parative data obtained from two studies in India and Canada are shown in Table 3. Table 3. Effect of the type of feeding on growth and the incidence of infection in Canada and India Delhi, India St John's, Canada Breast Milk Breast Milk milk formula milk formula (n = 50) (n = 50) (n = 30) (n = 30) Growth failures 11 34 0 0 Obesity' 1 1 2 9 Respiratory infectionsb 87 156 31 77 Otitis5 37 79 5 41 Diarrhoea b 108 326 3 9 Dehydration b 9 51 0 2 Septicaemia5b 1 9 0 0 ' Number of children affected. b Number of episodes of illness for the group over a 12-month period. Immunization Schedules of vaccination should be improved to make them more effective. This may necessitate modification with regard to the amount of antigen, number of inoculations, timing in respect of nutritional status, and the type of adjuvants employed. Immunopotentiation The demonstration of impaired im- mune responses in malnutrition has led to preliminary attempts at stimulation of the immune system by cell extracts or phar- macological agents. Transfer factor Table 4. Effect of transfer-factor administration on morbidity Transfer factor Control (n = 20) (n = 20) Diarrhoea ' mild 174 193 severe 31 40 Two studies have evaluated the effect Chest infection' of the administration of transfer factor. mild 115 126 Their protocols differed as did the severe 17 29 results. In one, no benefits were seen, whereas in the other there was a reduc- 'Episodesper 1000 child-weeks. tion in the occurrence of diarrhoea but not in that of chest, middle ear, or skin infections. In our controlled study, transfer-factor administration was associated with a decrease in the severity of infection and associated complications but not in the frequency of infectious illnesses (Table 4). 174 NUTRITIONAL DEFICIENCY AND INFECTION 175 Levamisole This drug has been widely used as an anthelmintic and has been demonstrated to stimulate cell-mediated immunity both in vivo and in vitro. Side effects of its use on a short-term basis are rare. The effects of levamisole administration on cellular im- munity in malnutrition are shown in Fig. 2. The role of short-term im- munopotentiation with one ore more agents in severely malnourished children before or during critical periods of stress, such as infection, requires careful evalua- tion. 75 IMWftATION INDI* 250- -20- 25- 0 5 10 15 20 25 30 DAYS Fig. 2. Levamisole in malnutrition. The drug was given in a dose of 150 mg daily for two 5-day periods with a gap of 2 days. Lymphocyte function in vitro improved before a significant change in weight. Manipulation of lymphocyte subpopulations In malnutrition, there are significant changes in the number, proportion, and function of various subsets of lymphocytes and such alterations probably contribute to impaired immunity, especially cell-mediated immune response, in nutritional deficiency. With the recent advances in immunology, it may be possible to stimulate selective populations of immunocompetent cells. Short-term nutritional supplements The provision of nutritional supplements shortly before or during a period of metabolic stress, such as infection, may reduce the impairment of immunity associated with undernutrition. This may be particularly relevant for hospitalized patients. Specific nutrients may be required in individual subjects. Prevention of low birth weight Low birth weight is associated with significant and prolonged impairment of cellular immunity, particularly in those who continue to have physical growth retardation.' Impaired immunocompetence associated with fetal malnutrition can to some extent be prevented by appropriate antenatal and perinatal practices. The strategies for dealing with this problem must necessarily take into account the prevalence of various risk factors in the community and the magnitude of risk associated with each of these factors. CHAiRA, R. K. American journal of diseases of children, 129: 450 (1975). CHAINDRA, R. K. ET AL. Biology of the neonate, 31: 15 (1977). R. K. CHANDRA R:SUMA Carences nutritionnelles et sensibilite 'a l'infection Un certain nombre d'enquetes epidemiologiques dans diverses parties du monde ont confirme qu'il existait une etroite association entre les carences nutritionnelles et les maladies infectieuses. C'est non seulement l'incidence des infections, mais aussi leur gravite, qui sont augmentees et on a ainsi constate chez les nourrissons africains souffrant de kwashiorkor des infections rougeoleuses ou herpetiques evoluant de maniere fatale. L'infestation parasitaire, les maladies diarrheiques, la pneumonie, l'hepatite B et la tuberculose sont aussi generalement plus frequentes et plus graves chez les sujets mal nourris et les enfants de faible poids a la naissance. De nombreuses observations, chez l'animal et chez des malades hospitalises, ont permis de noter une bonne correlation entre la malnutrition proteino-energetique et les taux d'infection et de mortalite; les autres carences alimentaires exercent des effets variables moins bien etablis. Les etudes portant sur la maniere dont la malnutrition affecte la reponse immunitaire se sont recemment multipliees et elles sont soutenues et coordonnees notamment par l'OMS. En ce qui concerne l'immunite a support cellulaire, diverses epreuves, en particulier celle qui porte sur l'hypersensibilite cutanee retardee, permettent d'evaluer le nombre et la proportion des lymphocytes T circulants. La plupart des experiences ont mis en evidence leur reduction en cas de malnutrition. Les cellules T sont en effet riches en acetylcholinesterase liee a la membrane, enzyme dont l'activite est reduite chez les sujets mal nourris. Ces derniers possedent un nombre accru de cellules <<nulles>> ne presentant pas les caracteristiques de surface des lymphocytes T ou B et empechant l'action des autres lymphocytes. Quant aux lymphocytes intraepitheliaux, dont les caracteristiques de surface et le role sont voisins de ceux des lyrnphocvtcs T, ils subissent egalement une reduction significative chez les enfants mal nourris. On trouve d'autre part chez ceux-ci des taux seriques d'immunoglobulines IgG generalement plus eleves et, dans certains cas, c'est la demi-vie plasmatique de ces immunoglobulines qui est prolongee, sans doute en raison de la presence d'une infection concomittante. Les cellules plasmatiques produisant les IgA sont toutefois moins nombreuses, d'oiu une diminution des IgA secretoires. Enfin, d'autres facteurs de protection non specifiques sont alteres, notamment le systeme de complement, l'interferon et la lysozyme, et divers tissus subissent des modifications structurelles qui peuvent aussi contribuera diminuer la resistance a l'infection. Cet ensemble de constatations a amene a s'interroger surl'efficacite de la vaccination prophylactique chez des sujets mal nourris. Selon les donnees disponibles, le taux d'anticorps seriques est generalement normal en ce cas et on peut l'ameliorer si necessaire en augmentant la dose d'antigene, en utilisant des adjuvants appropries ou en eliminant toute infection concomittante. Toutefois, uneetude a montre que le taux des IgA secretoires consecutivesa l'immunisation avec des vaccins vivants attenues contre la rougeole et la poliomyeliteetait reduit de maniere notable en cas de malnutrition. Quoi qu'il en soit, et independamment de la reponse obtenue, on ne sait pas dans quelle mesure l'etat nutritionnel pourra diminuer la protection en cas d'exposition ulterieure naturelle a l'infection et de nouvellesetudes sont requisesa ce propos. La plupart des etudes deja faites ont porte sur des enfants atteints de formes graves de malnutrition, comme le kwashiorkor et le marasme, et ne representant que 1 a 2% de la population 176 NUTRITIONAL DEFICIENCY AND INFECrION 177 souffrant de carences; la poursuite de programmes complets de vaccination est donc parfaitement justifiee. Pour briser le cercle vicieux malnutrition-infection, on peut employer un certain nombre de moyens, au premier rang desquels figure l'allaitement maternel; le lait humain contient en effet un grand nombre d'elements protecteurs physico-chimiques aussi bien que cellulaires. Les vaccinations peuvent etre faites selon un schema qui les rende plus efficaces. On peut aussi stimuler la reponse immunitaire a support cellulaire par l'administration de levamisole, et les progres recents en immunologie permettront sans doute de stimuler plus particulierement les cellules immunocompetentes. Enfin, on peut recourir en cas de besoin a la supplementation nutritionnelle pour les malades hospitalises et eviter, par des mesures prenatales et perinatales, le faible poids a la naissance, important facteur d'alteration de l'immunite cellulaire.