HIV disease: Working Group Report of the First World Congress of Pediatric Gastroenterology, Hepatology, and Nutrition.

Abstract

I. SUMMARY OF THE PROBLEM The picture of human immunodeficiency virus (HIV) infection is one with two faces. Throughout the world, over 90% of children acquire HIV through perinatal transmission, but access to medications and adequate nutrition plays a major role in determining the outcome and natural history of this infection. In children born with congenitally acquired HIV disease in countries in which medications and food are available, the vertical transmission rate has fallen in the past 10 years from 30% to less than 3%. The faces of these HIV-infected children reflect hope for future research that will provide therapy to prevent and treat opportunistic infections as well as improve growth. In countries that lack resources to provide medical and nutritional therapy for HIV-infected children, the faces of the children express despair. The impact of a chronic disease on endemic malnutrition, economic failure, and social instability results in high childhood as well as parental mortality and unmeasurable developmental and intellectual impairment. The acquired immune deficiency syndrome (AIDS) pandemic in Africa compounds the problems of endemic severe malnutrition. Despite declining childhood mortality rates in Africa over recent decades, case-fatality rates for severe malnutrition, particularly for kwashiorkor, in most urban hospitals have remained high (20%–30%). Attributing this observation to poor case management, the World Health Organization (WHO) has focused on improved treatment protocols through Integrated Management of Childhood Illness (1). A reduction in case fatality rates from 17% to 9% in a report from Bangladesh was attributed to implementation of the severe malnutrition protocol, but the median age of the children was 6 months, and HIV infection was presumably unlikely. However, the use of standard treatment with intensive nursing care in the high HIV prevalence setting of Malawi only reduced the case-fatality rate of severe malnutrition from 30% to 25% in children with a median age of 33 months and in whom kwashiorkor cases predominated. Although enteric infections may contribute to malabsorption in many children, studies of gastrointestinal tract function suggest that changes in permeability may occur independently of infection status. Some malnourished children with HIV infection do gain weight with nutritional rehabilitation and specific treatment for Isospora belli. However, most HIV-related wasting is probably cytokine-mediated and unresponsive to feeding alone. The pathogenesis of HIV-related malnutrition remains a challenge for investigators. Because treating symptomatic HIV infection in children is not feasible in poor countries, the highest priority must be given to interrupting mother-to-infant transmission. WHO estimates that every day about 1,600 children are infected with HIV, 90% of them in the developing world. Short course oral zidovudine in seropositive pregnant women (cost $60) reduces vertical transmission by about 35% to 40%, and a two-dose nevirapine regimen (mother intrapartum and infant postpartum, costing $4) reduces transmission by 47% compared with an equivalent zidovudine regimen. Ethical issues have hampered collaborative research on HIV with institutions and sponsors in industrialized countries; however, a recent consensus statement has endorsed five ethical principles, which would allow this vital research to progress (2). The diagnosis of HIV disease is established by laboratory criteria involving culture or identification of virus in the blood or tissues. In regions in which these diagnostic tests are not available, the WHO has created the following clinical algorithm (WHO 1993): WHO Clinical Case Definition of AIDS in Children Major Signs Weight loss or failure to thrive Chronic diarrhea > 1 month Prolonged fever > 1 month Minor Signs Generalized lymphadenopathy > 0.5 cm in 2 or more sites with bilateral lymph nodes counting as one site Oropharyngeal candidiasis Repeated common infections Persistent cough > 1 month Generalized dermatitis Confirmed maternal HIV infection Pediatric symptomatic HIV is suspected in an infant/child with at least 2 major signs plus at least 2 minor signs in the absence of a known cause of immunosuppression. As a scientific community, we have learned much about the pathobiology of retroviral infections through the allocation of funding for HIV research. With great regret, we acknowledge the existence of a great disparity in child health between nations. As pediatricians, we understand the importance to all children of diminishing these inequalities in health care. II. ISSUES IN NEED OF INVESTIGATION OR IMPLEMENTATION The five major areas in need of investigation relating to the gastrointestinal and nutritional problems in HIV-infected children are: prevention of mother-to-child vertical transmission; gastrointestinal, liver, and pancreatic dysfunction; opportunistic infections; growth failure and body composition; and nutrition. Prevention of Mother-To-Child Vertical Transmission Children are infected via vertical transmission from their mothers in more than 90% of cases. This occurs in 22% to 43% of pregnancies in untreated HIV-infected women and accounts for approximately 15% of HIV infections in Africa. However, in regions of developed countries where every HIV-infected pregnant woman receives counseling, medication, and access to prenatal care, the transmission of HIV can be reduced to near zero. The risk of vertical transmission to children is influenced by a number of factors. In an evolving heterosexual epidemic, females outnumber males (male: female ratio of HIV-infected adults is 0.73:1 in South Africa). The infection rate is highest in young females, and therefore a large number of women of childbearing age are at risk of transmitting the disease to their offspring. In all published series, the vertical transmission rate in Africa has exceeded that of the developed world; this difference has been accentuated markedly by the inability of African countries to introduce established strategies shown to reduce mother–infant transmission rates. Identified Risk Factors maternal systemic viral load the local viral load in vaginal, cervical, amniotic fluids, and breast milk lowered maternal immunity (such as that associated with malnutrition or vitamin A deficiency) the presence of associated sexually transmitted diseases and local inflammatory or ulcerative conditions of the genital tract prolonged duration of ruptured membranes and instrumentation during delivery prematurity breast feeding Maternal Systemic Viral Load Maternal transmission to her fetus and baby depends on her own viral load. The viremia of recent infection, advanced disease, or nutritionally derived immunosuppression is associated with a higher risk of transmission. Short-term antiretroviral treatment during the last weeks of pregnancy, labor, and in the first week of life has been shown to reduce the infection rate. Even though recent protocols (e.g., PETRA study of AZT/3TC) with shortened duration of combination therapy, or single drug schemes (e.g., HIVNET 012 Nevirapine) have become less expensive, they still require the availability of antenatal testing, drug procurement, administration and monitoring (3). Economic support for communities lacking the resources to provide these services to women and children should be identified and provided. An early rebound plasma viremia has been observed after withdrawing maternal antiretroviral therapy. In women who choose to breast feed, this may reduce the benefit of anti-retroviral prophylaxis and needs to be evaluated fully. Local Risk Factors Transmission is increased in the setting of inflammation of the placental membranes or the presence of ulcerating or inflammatory conditions of the genital tract. Because most infection occurs during birth, the management of labor and mode of delivery may influence this risk. This includes the duration of ruptured membranes and the extent and duration of instrumentation during delivery. Vaginal antiseptic douching during labor has not been shown to be effective. Delivery by elective caesarean section decreases the vertical infection rate, but this is not likely to be a management option in the developing world. Maternal Malnutrition Maternal malnutrition has a number of consequences for her own viral load and therefore for the risk of transmission to her offspring. A poor maternal nutritional state leads to a higher proportion of premature and low-birth-weight babies. In most large African hospitals, the low birth weight rate exceeds 15%. A number of studies have addressed the question of the relationship between the maternal vitamin A status and supplementation during pregnancy and the risk of infant infection with HIV. These have shown so far that neither maternal vitamin A deficiency, nor maternal vitamin A supplementation during pregnancy, can be correlated directly with mother-to-infant transmission rates (4). Conversely, underlying vitamin A deficiency is common, and one Durban study showed a reduced risk of preterm delivery in mothers supplemented during pregnancy. In these preterm babies, mothers supplemented with vitamin A had a decreased infection risk compared with babies whose mothers had received placebo. Accordingly, adequate and intensive maternal–infant nutritional support should be provided in developing countries as an adjunctive measure. Breast Feeding Breast feeding is associated with significant risk of HIV transmission. When mothers are seropositive before the pregnancy, the increased attributable risk of infection from breast feeding has been estimated as 14%, but when mothers have seroconverted during lactation, this risk may be much higher. Breast milk has been shown to contain infectious HIV virus. Some evidence suggests that virus enters breast milk more readily in the presence of mastitis and other local factors, including cracked nipples. Even though the virus load may be relatively small (less than 1,000 copies/mL), the volume of milk consumed by babies means that the total exposure to virus may be considerable. The amount of virus entering breast milk also depends on the mother's plasma viral load. Accordingly, the risk of vertical transmission through breast milk is increased when the mother has seroconverted during lactation or has late disease. Similarly, an increasing duration of breast feeding has been associated with an increased cumulative risk of transmission. This risk tends to be maximal in the first few months of life and then tapers off, suggesting that both maternal viremia and also virus entry into breast milk stabilize. Higher maternal parity and maternal age appear also to be associated with a decreased risk of infection through breast feeding. Breast milk contains specific and nonspecific protective factors, including local1y active antibodies. A suggestion has been that exclusive breast feeding for 3 months may actual1y be protective against vertical transmission (5). The hypothesis that the intestinal barrier function is compromised by exposure to nonhuman milk or other feeds, due to either the presence of injurious factors in those feeds, lack of protective breast milk factors, or a combination thereof, requires further study. Studies are needed on the impact of feeding advice and policy on community-wide patterns of infant feeding. In countries with very high infant mortality rates and the need to prevent morbidity and mortality from malnutrition and enteral infection, when the availability, safety, and nutritional adequacy of substitute feeding cannot be guaranteed, nothing should be done to influence a move away from near universal breast feeding. In communities with relatively low infant mortality rates up to about 40/1,000, a selective feeding policy of breast or artificial feeding, free antenatal testing, and informed choice by HIV-positive mothers is possible. When breast feeding is chosen, this should be exclusive, up to the age of 3 months. Calculations in Kenya suggest that the risk of HIV transmission by breast feeding outweighs its cost-effective protective value by age 3 to 7 months. HIV-infected women in more developed, educated, and affluent communities in developing countries should certainly not be denied the information on which to base their choices. The Gut as Portal of Entry for HIV Infection Breaks of mucosal integrity have been assumed to be the entry route for HIV infection in the gut. Mouth ulcers, small mucosal breaks as occur during the period of teething, or surface inflammation such as with oral thrush, may all play a part in permitting entry of virus from breast milk. These processes may apply to the mouth, esophagus, stomach, small intestine, or in the perianal skin, rectum, or descending colon. Particular care should be exercised in managing thrush and stomatitis in breast feeding infants of HIV-positive mothers. Virus particles can also be detected in oropharyngeal and gastric aspirates of newborn infants. Premature and low-birth weight infants have a higher risk of mother–child HIV transmission; the role of reduced gastric acid output in such infections is not clear. New medications or interventions to decrease transmucosal migration of HIV are needed. Gastrointestinal, Liver, and Pancreatic Dysfunction Gastrointestinal dysfunction in children with symptomatic HIV infection may involve a number of abnormalities of digestion and absorption which in conjunction with other factors contribute to weight loss, wasting, and, as noted above, disease transmission. Abnormalities in gastrointestinal function may not be clinically evident and frequently may exist in children with normal stool patterns. The intestinal dysfunction syndrome was recognized starting in the early 1990s. This led to the understanding of the central role of the intestine and other organs involved in the digestive-absorptive processes in the pathogenesis of HIV disease. Intestinal dysfunction includes several distinct features, which are listed in Table 1 together with the frequency found in children from industrialized countries, who are not taking antiretroviral combination therapy. Although intestinal dysfunction may be an early feature of HIV infection, its relationship to progression of HIV disease is not clear.TABLE 1: Features of intestinal dysfunctionPancreatic dysfunction may also be a feature of altered digestion. In approximately 30% of children with symptomatic HIV infection, reduced fecal elastase concentration correlates with fat malabsorption. Overall, approximately 40% to 60% of children with symptomatic HIV infection living in industrialized countries who are not on combination highly active antiretroviral therapy (HAART) have some degree of maldigestion or malabsorption. The frequency and severity of intestinal dysfunction are expected to be even higher in developing countries, but data are lacking. Hepatic dysfunction is not a common manifestation of HIV disease in children, although over 90% of HIV-infected children will have hepatosplenomegaly during the course of their illness. Hepatomegaly without other symptoms of systemic disease is most likely associated with nutritional deficiency; however, opportunistic infections may involve the liver and should be considered in the HIV-infected child. Cytomegalovirus may infect the hepatocytes but also has been associated with biliary tract disease. Cryptosporidium and Mycobacterium avium intracellulare also have been found to cause AIDS cholangiopathy. Most of the medications used to treat HIV disease are somewhat hepatotoxic. Nucleoside reverse transcriptase inhibitors (nRTls) have been associated with a potentially fatal syndrome of lactic acidosis with hepatic steatosis. The pathogenesis of intestinal dysfunction is unknown. Opportunistic agents have been suspected of inducing intestinal dysfunction, but evidence to support this hypothesis and morphologic correlation is lacking. Investigators have speculated that HIV itself may contribute to intestinal epithelial cell dysfunction, but this hypothesis remains unproved. Immune dysregulation probably plays a central role in the pathogenesis of intestinal dysfunction, but the exact cause–effect relationship is unknown, and the frequency and severity of intestinal dysfunction only weakly correlate with immune status. However, preliminary results suggest that restoration of the systemic immune system in children started on HAART undergo a rapid improvement in intestinal absorption (6). The duration of this response is presently unknown. In addition to enteric infection, immunodeficiency, and HIV itself, malnutrition may contribute to intestinal dysfunction. In non–HIV-infected children with severe malnutrition, permeability and brush border enzyme activity may be altered. The clinical observations that nutritional rehabilitation may be effective in partially restoring digestive-absorptive function support this hypothesis. Research is needed to more fully understand the relationship between alterations in mucosal immunity, epithelial cell function, and digestion of specific nutrients. Areas in Need of Investigation Clinical Issues. The impetus to obtain data on the prevalence and relevance of intestinal dysfunction in developing countries is to establish a practical and effective clinical approach to treat maldigestion, malabsorption, and eventual malnutrition. Evaluation of intestinal dysfunction should be available to all children with HIV infection at the time of diagnosis so that problems of malabsorption can be identified before the onset of undernutrition. Because resources in industrialized and nonindustrialized countries may not permit the delivery of such care, intestinal function tests should be performed as indicated in children with intestinal symptoms or weight loss. Ideally, these studies should be repeated every 6 months in children at risk for developing malabsorption. Investigations need to be simple to perform, noninvasive, inexpensive, and easily available. In regions with limited resources, frozen fecal specimens may be stored for noninvasive testing. Currently, the following tests may be performed: steatocrit, to assess fat malabsorption; fecal alpha-1-antitrypsin concentration to assess intestinal permeability and protein loss; and reducing substances to identify carbohydrate malabsorption. In the future, tests may be available in stool to determine nutrient concentration, microbiologic analysis, and mediators of inflammation. The clinical relevance of these studies in developed and developing patient populations should be validated. More importantly, resources should be allocated for the study of gastrointestinal function in regions with a high prevalence of pediatric HIV disease. Other more sensitive tests, although more expensive and requiring technology that is not universally available, include the lactose breath test and the determination of specific activity of disaccharidases in intestinal tissue. Exocrine pancreatic function may be explored by the determination of fecal elastase. The hypothesis that combination therapy is effective in the short term in restoring intestinal dysfunction, or in preventing dysfunction, needs to be confirmed in larger sample size and in longer follow-up. The effects of nutritional rehabilitation on impaired intestinal function should be investigated along with the efficacy of enteral and parenteral nutrition or selected nutrient supplements to restore and maintain intestinal absorptive processes. Ongoing protocols evaluating the impact of the early introduction of calorically increased infant formula on the growth of HIV-exposed babies may provide an insight into the role of nutrition in modifying sequelae of HIV disease. Alternative strategies, including the use of biotherapeutic agents, also called functional foods, fibers, and prebiotics need to be explored. Research Issues. The pathogenesis of intestinal dysfunction is a major question in HIV-infected children. The lack of appropriate animal models has resulted in studies in human subjects. Animal and cellular models should be developed to investigate the potential interrelationships among nutrition, infection, gastric acid production, motility, and the function of the immunologic, neurologic, and endocrine systems with respect to gastrointestinal tract function. Most of the data on intestinal tract function is derived from adults with HIV disease. Resources should be allocated to determine whether HIV-infected children have similar pathophysiology. Diarrhea or intestinal dysfunction may affect the absorption and metabolism of anti-retroviral drugs. Because intestinal function in HIV-infected children may differ from that of HIV-infected adults, pharmacokinetics and pharmacodynamics of antiretroviral agents should be tested in children. Preliminary data have shown efficacy of growth hormone (GH) in the HIV-infected child who is not growing or gaining weight. In addition to its effect on bone growth and lean body mass, GH has trophic as well as antisecretory properties on the enterocyte. Thus, GH may be an ideal drug to treat chronic diarrhea in HIV-infected children. This hypothesis needs to be investigated in both in vitro as well as in vivo studies. Opportunistic Infections Diarrhea is a frequent problem in HIV disease. Although the incidence of diarrhea is thought to be increased in HIV-infected children living in industrialized countries, this dogma is not based on solid data. Opportunistic microorganisms cause few or no symptoms in healthy subjects, but may result in serious illness in the immunocompromised/malnourished children. Globally, Cryptosporidium parvum is the major enteric opportunistic agent responsible for severe and protracted diarrhea in HIV-infected adults. In children, enteric cryptosporidiosis causes chronic diarrhea, dehydration, and severe weight loss and is often associated with advanced HIV disease or with rapid immunologic deterioration. Other opportunistic agents that may cause diarrhea include Enterocytozoon bieneusi or microsporidium (which however seems to be restricted to adult subjects, at least in Western countries), cytomegalovirus, and Mycobacterium avium intracellulare (7). In Africa, several protozoa, such as Cyclospora cayetanensis, Schistosoma duodenalis, and Isospora belli, frequently are found in HIV-infected subjects. Nevertheless, common enteric pathogens, such as Rotavirus, also cause life-threatening diarrhea in HIV-infected children. Although there is little evidence, restoration of immune function using combination therapy might protect children from opportunistic infections. In regions without access to effective HIV therapy, treatment of specific intestinal infections needs to be aggressive. Antibiotics or immunotherapy (8) may be options, but frequently effective therapy is lacking. Areas in Need of Investigation Clinical. More effective therapies are needed for control of opportunistic and chronic enteric infections. The role of helmenthiasis and malaria as confounding factors should be explored. The risk–benefit ratio of vaccination is optimally assessed in the populations at greatest risk. Clinical trials should include the relationship between nutritional status, especially vitamin A and folic acid deficiency, and enteric infection so that the impact of early nutritional intervention can be determined. Research. The pathophysiology of intestinal cryptosporidiosis is especially important in HIV-infected children because of the frequency and severity of diarrhea induced by this protozoon. Several hypotheses have been raised, including the production of enterotoxic substances, a cytotoxic effect or the induction of a prostanoid response causing intestinal secretion. Research involving the pathogenesis of cryptosporidial-induced diarrhea will help develop more effective therapeutic agents. Passive immune therapy with hyperimmunoglobulin obtained from immunized animals may be effective but requires clinical trials in larger numbers of children. Anti-secretory medications may be effective in reducing stool output, but data on safety and efficacy in the pediatric population are lacking. Passive immunotherapy has the potential to save the lives of children with severe Rotaviral diarrhea and efforts are needed to develop a standard, effective, and inexpensive preparation of immunoglobulin of either human or animal origin to be administered orally. However, the newly discovered mechanism of Rotavirus diarrhea involving the enterotoxic effect of the NSP4, a nonstructural viral protein and the involvement of neuronal mediated intestinal secretion may lead to alternative approaches to treat viral diarrhea. Efforts to develop safe and effective vaccines for children against common and severe pathogens will ultimately improve the nutritional status and growth of HIV-infected children. The pathogenic role of HIV infection on the development of intestinal opportunistic infections and intestinal function such as permeability is not known. Additional basic research is needed to clarify the mechanism of this critical aspect of intestinal dysfunction. Growth Failure and Body Composition: Energy and Protein Metabolism in HIV Disease The importance of understanding the metabolic effects of HIV infection stems from observations that mortality in these patients is related to loss of lean body mass or weight (9) (Fig. 1). Progressive weight loss is also a known risk factor for opportunistic infection and consequent mortality (10). Additionally, one of the most common presentations of AIDS is undernutrition (failure to thrive in young children and wasting syndrome in older patients). While the cause of the undernutrition is not clear, aberrant energy and protein metabolism appears to play a role.FIG. 1.: Pathways of Growth Failure in Children With HIV InfectionInitial investigations suggested that HIV-infected adults had preferential loss of lean body mass. However, more recent studies observe fat depletion to be much more significant compared with lean body mass. In toto, the data obtained from studies of HIV-infected patients support the premise that, as body fat stores diminish, there is an appropriate response to starvation in that more lean body mass is lost. Studies in adults assessing protein turnover rates have shown varying results. Symptomatic, weight-stable HIV-infected patients have decreased protein flux, while protein turnover is increased in patients with advanced stages of HIV infection. A recent study of protein turnover (by 15N-glycine) in HIV-infected children showed slightly but not significantly increased protein turnover among those children with growth retardation (11). The energy needs of infants and children with HIV are largely unknown. The amounts of additional energy needed in HIV-infected children for both catch-up growth (if depletion has occurred) and for maintenance of an adequate nutritional state are largely unknown. Current recommendations for average daily energy requirements in healthy infants and young children are largely based on estimates of energy intakes. Energy needs are higher in HIV-infected adults with superimposed intestinal disorders and infections. Moreover, studies in adults have yielded data showing elevated resting energy requirements (by 8%–9%) in asymptomatic HIV-positive patients compared with controls. One study correlated IL6 levels with elevated resting energy expenditure (REE) even before CD4 T-cell numbers declined. Calculated values suggested that elevated fat oxidation rates accounted for this difference. In contrast, lower REE values found in patients with AIDS were attributed to adaptation to malabsorption. In malnourished HIV-infected adult patients, mean REE has been measured to be 8% to 12% above control values, an inadequate response to undernutrition, and a possible mechanism for progressive wasting in these patients. Grunfeld et al reported elevated REE values in HIV-positive adult subjects (11%), AIDS patients (25%), and AIDS patients with secondary infection (29%). Subsequent studies suggest that HAART may not decrease REE (12). Patients with superinfections showed a significant decline in oral intake and rapid weight loss, in contrast with stable intake and weight in the first two groups. These results are consistent with the possibility that rapid weight loss and anorexia may be early signs of secondary infection in AIDS. Only one report has been published to date with measurements of energy expenditure studies in HIV-infected children (11). HIV-infected patients were smaller and had less lean body mass compared with normal subjects. Although resting energy expenditure (REE) was lower among HIV-infected growth-retarded children, REE/kg body weight or lean body mass was not different. Energy intake did not differ among the three groups. From Grunfeld et al. and Macallan et al., it appears that the principle determinant for weight loss in HIV-infected adults is not hypermetabolism but diminished and insufficient nutrient intake. The data from Henderson et al. do not support this supposition in children, although the impact of opportunistic infection was not evaluated in this study. Although failure to thrive is a marker of poor outcome in HIV-infected children, weight loss does not appear to be a predictor for the onset or evolution of HIV infection to AIDS (13). Total energy expenditure (TEE) in HIV-infected adult patients has been studied with doubly labeled water. Macallan et al. found that TEE correlated with weight change, the most significant being a decline in TEE among the HIV-infected patients who were losing weight. More recently, Heijligenberg et al. did not find any difference in TEE between controls and weight-stable HIV-infected patients. Difficulty replenishing lost lean body mass has been recognized since Kotler's initial report demonstrating increase fat mass rather than lean body mass with parenteral nutrition. The inference from these data is that HIV-infected patients have alterations in metabolism and energy expenditure that are not the prime determinants of subsequent weight loss. Rather, it appears that morbidity and mortality actually stem from weight loss that is caused by cytokine-induced metabolic alterations, opportunistic infections, or tumors. Areas in Need of Investigation Clinical. Studies are needed to define the effects of HIV infection on alterations of body composition and growth patterns in children and adolescents. The effects of HIV infection on energy and protein metabolism and the relation of these observations to consequent growth in HIV-infected pediatric patients remain to be investigated. In undernourished children and adolescents with HIV, the effectiveness on nutrition rehabilitation needs to be determined, particularly with respect to the effects of nutrition support on measurements of growth and body composition and on energy and protein metabolism. Anthropometric measurements should be collected from developing countries and contrasted with those from developed countries to better define the scope of the nutritional problems of HIV-infected pediatric patients throughout the world. Some attempt should be made to determine whether metabolic observations in children living in developed countries are applicable to children living in different socioeconomic conditions. Micronutrient status with a focus on deficiency states requires further investigation. Research. The relationships of energy and protein metabolism and body composition with the generation of cytokines and other intermediaries of inflammatory responses require study. The investigation of alterations in cellular metabolism should be more clearly defined and related to the clinical observations in children and adolescents with HIV infection. Joint projects between researchers in major medical centers and field investigators should be promoted. Measurement techniques should be validated for field study. The growth patterns of HIV-infected patients should be defined and correlated with the degree of HIV infection and with the ultimate outcome of the infected pediatric patients. The effects of treatments, specifically HAART, on energy and protein metabolism and growth requires investigation. The hormonal milieu of HIV-infected pediatric patients requires definition. Nutrition The triad of human immunodeficiency virus (HIV) infection, nutritional status, and immune function are intimately interrelated (14). Without potent anti-retroviral drugs or adequate nutrition, an HIV-infected child experiences recurrent infection, malabsorption, and ultimately growth retardation, developmental delay, and death. The pattern of clinical malnutrition is changing in Africa, Asia, and Latin America because of the influence of HIV disease. From a public health viewpoint, nutritional support is one of the most important initial therapeutic interventions for the HIV-infected child. Prevention of transmission due to breast milk using replacement of formula requires new knowledge about the components in formula that will promote optimal growth in the HIV-infected child. Although standard formula contains 20 kcal per ounce, higher caloric density may provide additional energy needed for growth. Furthermore, hydrolyzed formulas or specific vitamin or nutrient additives may reduce infection and promote growth. Guidelines for nutritional support for the HIV-infected child include: Maintenance of good nutritional status during pregnancy by nutritional counseling, nutritional supplementation and treatment of infections. Try to provide 1.5 to 2 times the RDA if growth velocity is not appropriate by selecting nutrient dense foods. Control fever, because there is a 7% increase in basal energy expenditure for every degree (in Fahrenheit) increase in temperature. Enhance immune function through a balanced approach to nutrition Stress the importance of good hand washing and clean utensils. Teach food preparation that maximizes nutrient content and encourages foods high in antioxidants, vitamins, and minerals. Manage diarrhea by determining the cause if possible. Avoid disaccharides if intolerant. Monitor Vitamin A in both pregnant women and in growing children. Assess the adequacy of methionine, cysteine, copper, selenium, zinc, and vitamins C and E in children with malnutrition. Low plasma selenium is an independent predictor of mortality, and appears to be associated with faster disease progression (15). Areas in Need of Investigation Clinical. Clinical trials are needed to assess the role that micronutrients and antioxidants play in the progression of HIV disease, to evaluate the potential use of nutritional therapy as adjunctive treatment in HIV-infected children living in developed or developing countries, and to determine appropriate enteral formulae or nutritional supplements for children of various ages and different stages of HIV infection. The efficacy of immunoactive foods (e.g., arginine, nucleotides, omega-3 FAs, DHA) on progression of HIV disease and the occurrence of opportunistic infection could provide new therapeutic strategies for control of disease activity. Programs should be developed for hospice care and support for children living in health care settings with limited access to nutritional resources. Research. Basic research is needed in nutrition to provide data on which to base rational decisions for intervention with specific nutrients. III. PROPOSED PLAN TO ACHIEVE GOALS Research Issues in Prevention of Vertical Transmission Planned studies on the efficacy of an intervention on maternal–infant transmission should include an attempt to obtain information on the role of nutritional status on disease progression. Food companies could be approached to collaborate in these trials. The critical question in developing countries is the use of artificial formula and the timing of the introduction of transitional foods. Although the scientific questions are of prime importance, local customs, availability of resources, and traditions need to be considered. Collaboration between anthropologists, leaders who can represent the interests of the local population, research scientists, and clinical investigators is needed to apply new knowledge to specific trials. Agencies such as UNICEF may be of benefit in bringing these groups together. Gastrointestinal, Liver, and Pancreatic Dysfunction The problem of intestinal dysfunction and opportunistic infections exists primarily in developing countries where simple clinical protocols can be developed to systematically evaluate absorption and gastrointestinal symptoms. Serial microbiologic investigation will provide information on the cause–effect relationship between enteric infections and intestinal dysfunction. New knowledge is needed to understand the role of HIV in altering mucosal immune function and the ultimate impact this may have on intestinal dysfunction. Basic laboratory research should be funded in research centers with expertise in intestinal function, enteric infections, microbiology, mucosal immunology and nutrition. Professional societies and private foundations such as the Pediatric AIDS Foundation could be enlisted to focus this effort and encourage peer-reviewed proposals. Cooperation with providers and scientists in the developing countries is important to bring this information to the patients. Through support of focused and interactive international meetings, a closer relationship between basic scientists, clinician-investigators, and health care providers could be established. Economic resources and scientific cooperation are essential for the application of laboratory-based observations to the clinical setting. UNAIDS, an agency that is working at the international level, could bring together professionals working in microbiology, mucosal immunity, nutrition, and gastroenterology to implement and coordinate research in the field. As new therapies and vaccines are developed, clinical trials that include the evaluation of intestinal function, nutritional status, and body composition should be made available to all HIV-infected children. Opportunistic Infections To decrease opportunistic infections, effective vaccines and new therapeutic strategies need to be developed. Because the greatest need is in developing countries, financial incentives should be provided to industry to enable them to develop and produce medications needed to treat children. Testing these medications in the countries in which they will be most beneficial is critical to determining the value of new interventions. Situations in which the financial success of a product is dependent on sales in a developed country should be avoided, as risk–benefit ratios may differ among regions. Growth Failure and Body Composition The mechanisms for the observed changes in body composition resulting in growth failure remain unknown. Laboratory-based research on the impact of HIV infection on cellular metabolism as well as new knowledge about the relationship between immunoregulatory dysfunction and growth may provide much needed information. These data will enable clinicians to devise therapeutic strategies to modify the growth of children. Until this information is available, clinical trials involving agents such as human growth hormone, should be pursued. Support in the United States using Orphan Drug status with the collaboration of industry may provide support tor these trials. Nutrition Agencies such as the World Bank, World Health Organization, and professional organizations of nutritionists and gastroenterologists should share their expertise to develop region specific plans using local expertise and available food sources to provide appropriate nutrition for HIV-infected children. Economic infrastructures are needed to enable communities to provide sufficient nutritional resources to disrupt the cycle of malnutrition, immune deficiency, and chronic infection. New knowledge is needed to advise international organizations which foods or supplements are optimal to support growth and improve immune function for undernourished HIV-infected children.