Measurement of energy intake
Prospective assessment of energy intake was achieved by two steps to maximize accuracy. Recording the weight of the food consumed over a 3-day inpatient period at the beginning of study was followed by seven random day telephone calls over a 2-month period for recall of food consumed during the preceding 24 h. The energy content of the ingested food for each day was calculated from standard tables . The mean energy content of the 3-day inpatient study and the mean energy content of the 7-day food recall was averaged. This averaged number (cal/kg per day) and the expected energy intake for age and sex-matched norms  were used to arrive at a Z-score  for energy intake.
Measurement of energy expenditure
Resting energy expenditure was measured by indirect calorimetry with a pediatric ventilated hood and a metabolic monitor (Deltatrac; Dates Instrumentation, Helsinki, Finland). All resting measurements were performed in early morning as the subject was just waking up from sleep and overnight fast. The value obtained (kcal/day) and the expected value for age and weight-matched (in kg) norms  were used to arrive at a Z-score  for resting energy expenditure. Total energy expenditure was assessed by the doubly labeled water method . For this purpose, the children received an oral dose of 0.2 g/kg (body weight) of pyrogen-free deuterium and oxygen enriched water (Isotec, Inc. Miamisburg, Ohio, USA) with both heavy isotopes enriched to about 10 atomic percent. Urine samples were collected before the dose and on days 1, 2, 6, 7, and 8 after the dose. These urine samples were analyzed in duplicate by mass spectrometry (SIRA-12; VG Isogas, Coventry, UK) for isotope ratio determination of the dilution of tracers. Total energy expenditure was calculated from the difference between the elimination rate of deuterium, which is lost from the body in water, and that of oxygen-18, which is lost in both water and carbon dioxide. A measure of the mean rate of production of carbon dioxide and hence a measure of energy expenditure was then determined using standard calorimetric equations . The total energy expenditure value (kcal/day) obtained for each child and the expected age and sex-matched normal values  were used to arrive at a Z-score  for total energy expenditure.
Energy intake, resting energy expenditure, and total energy expenditure were adjusted for comparison by calculating Z-scores . Likewise, for comparison of IGF-1 levels, each child's IGF-1 value and the expected normal value for that child's age and sex  were used to arrive at an IGF-1 Z-score. Differences in the Z-scores between the group of HIV-infected children with growth impairment (≥ grade 1 weight/age or height/age deficit) and the group of HIV-infected children with normal growth parameters for age (no deficit in weight/age or height/age) were assessed by the Wilcoxon's rank sum analysis. Differences in other clinical characteristics and laboratory values with non-Gaussian distribution were also compared by rank sum analysis. Differences between the means of continuous data were analyzed by the Student's t test. Comparison of measured values in HIV-infected subjects to expected age and gender matched normal values utilized the paired Student's t test. Distribution of non-continuous data was compared by Fisher's exact test of probability. Analysis was performed with the aid of the Statistical software package (Jandel Scientific, San Rafael, California, USA).
Twenty-three congenitally HIV-infected children aged 1.3 to 13.2 years were enrolled in the study. Subjects included 15 girls and eight boys. All children studied were prepubertal. Eleven of 23 children exhibited a heterogeneous growth pattern abnormality based upon their weight and height measurements at the initiation of study. Four of the children were identified to be `failing-to-thrive' with symmetrical weight/age and height/age deficits grade 2 or 3. Three children were found to have `stunting' with height/age deficit but no weight/age deficit. One child was identified as `thin' since weight/age deficit was present without a height/age deficit. Three others were identified as having `wasting syndrome' with weight/age deficit greater than height/age deficit and ≥ 10% weight loss over the 3-month period prior to study. One of the three had fluconazole-resistant candidal esophagitis, another had Mycobacterium avium intracellulare infection, and the third child had protracted recurrent diarrhea of unknown etiology with malabsorption. The remaining twelve of 23 children were categorized as `normal growth' since no weight/age or height/age deficits were evident.
In the growth-impaired group, three children were in clinical stage A, three in B, and five in C whereas in the normal growth group, two children were in clinical stage N, five in A, four in B and one in C (P < 0.05 for N/A/B versus C between the two groups). In the growth-impaired group, three children were in immunologic stage 2 and eight in stage 3 (mean CD4 % ± SD: 11.7 ± 9.7) and for the normal growth group, four children were in stage 1, five in stage 2, and three in stage 3 (mean CD4 %: 27.3 ± 10.3). The CD4 percentages between the two groups were significantly different (P < 0.05).
Evidence for increased energy expenditure was not found (Table 1) for HIV-infected children when total energy expenditure and resting energy expenditure measurements for all HIV-infected subjects (n = 23) were compared to normal values for age and sex. However, decreased energy intake was observed in HIV-infected children when compared to expected normal energy intake values (P < 0.04).
Results obtained and compared between the 11 HIV-infected children with abnormal growth and 12 HIV-infected children with normal growth for total energy expenditure and resting energy expenditure showed no evidence for increased energy expenditure in children with abnormal growth (Table 1). Likewise comparison of energy intake Z-scores in these two groups was not significantly different. However, the subgroup of children (n = 7) who were `failing to thrive' or `stunted' (i.e. children with height/age deficit equal or greater than weight/age deficit and indicative of more chronic growth failure) had significantly greater energy intake (P < 0.05 by rank sum analysis) when compared with the subgroup of children (n = 3) who were `wasting' (i.e. children with more acute growth impairment).
The mean ± SD values of viral load, serum leptin level, iron metabolism markers (iron level, total iron binding capacity, percent saturation), protein metabolism markers (total protein, globulin, albumin) and lipid metabolism markers (cholesterol, triglyceride) of children with growth impairment were compared to those values of HIV-infected children with normal growth (Table 2). Significant differences were found for log viral load measurement (P < 0.05) and total protein level (P < 0.01) between the groups. IGF-1 levels in HIV-infected children were significantly less than those expected normal levels for age and gender (P < 0.002). IGF-1 Z-scores in HIV-infected children with normal growth were higher (P < 0.05) than Z-scores of HIV-infected children with poor growth.
Serum IL-6 levels for the HIV-infected children with poor growth tended to be higher than levels in children with normal growth. However, differences between mean levels were not significant. IL-6 levels after whole blood lipopolysaccharide stimulation of cells for the children with poor growth were significantly greater than levels produced by cells from children with normal growth (mean ± SD: 1248.3 ± 1069.8 versus 546.5 ± 345.2 pg/ml, P < 0.05).
Results presented here demonstrate that total energy expenditure by HIV-infected children is similar to the expected energy expenditure of non-infected children. This finding is consistent with results of adult patients with HIV infection . However, unlike observations made in adult patients [16–17], increased resting energy expenditure was not found in HIV-infected children when compared with normal children. Moreover, differences in energy expenditure were not evident when HIV-infected children with impaired growth were compared with HIV-infected children with normal growth. This finding is similar to a recent report where prepubescent HIV-infected children with and without growth failure were compared . Together with published observations in adults and children, the observations reported here further support that a hypermetabolic state is not the basis for growth impairment in HIV-infection.
Inadequate energy intake has been reported to be the chief determinant of negative energy balance in HIV-infected adults . Our experience with HIV-infected children is similar. The accuracy of data used to calculate energy intake depends on accurate reporting by caretakers. We used multiple assessments of 24-h telephone food recall to complement the 3-day inpatient caloric evaluations. We felt that this inpatient/outpatient check method produced reliable results. The inability to detect differences in energy intake between our cohort of HIV-infected children with impaired growth when compared with HIV-infected children with normal growth was most likely due to the heterogeneous nature of our growth-impaired group. This was demonstrated when the analysis of energy intake by the subgroups of children within the growth-impaired group did show significant differences. As anticipated, the children in the abnormal growth group who were acutely weight/age deficient (`wasting') had very low energy intake. However, children in the abnormal growth group with symmetrical weight/age and height/age deficiency (`failure-to-thrive') and predominant height/age deficiency (`stunted') had much higher calculated energy intake than anticipated. We found that this was due primarily to nutritional supplements in their diets which are an integral prescribed regimen in our care of these children. We want to emphasize the fact that the measurements of energy balance for this study are a one-time assessment value when the subjects were not acutely ill. It maybe the case that for the subgroup of children with more chronic patterns of growth impairment (`failure to thrive' and `stunted'), they are able to ingest ample calories (catch up energy intake) at times of wellness (as was measured in our study time point). However, poor energy intake maybe occurring at times of illness. Multiple episodes of illness (significantly more children in our growth-impaired group were found to have more advanced clinical stage) with inadequate energy intake at times of illness, combined with energy intake at time of wellness which is still insufficient to catch up growth, may have led to growth impairment of these children over time.
Specific nutritional deficiencies related to iron metabolism  lipid metabolism  and protein metabolism  have been reported to contribute to impaired growth of HIV-infected children. In our analysis, total serum protein levels in HIV-infected growth-impaired children were lower than levels in HIV-infected children with normal growth. Differences in other markers were not significant. These findings support the hypothesis that poor growth is associated with failure to maintain protein balance. Taken together with the information that this imbalance was not likely to be caused by decreased intake, these results suggest an abnormal utilization of protein. Ongoing studies aimed at defining substrate utilization in HIV-infected children may help define whether a causal relationship exists between dysregulation of protein metabolism and poor growth.
A strong correlation between adipose tissue mRNA levels of leptin and anorexia has been demonstrated [31,32]. Increased leptin production resulting from chronic or recurrent infections has been proposed as a mechanism for anorexia and wasting associated with these diseases. In the group of children we studied, we could not demonstrate an association between increased serum levels of leptin and poor growth. Previously work had shown that the leptin levels in HIV-infected children were not significantly different from leptin levels in age-matched non-infected children . Nevertheless, our study did not test the hypothesis that intermittent elevations of leptin levels could contribute to anorexia at times of acute opportunistic infections in HIV-infected children and that this would lead to poor growth. Children with HIV infection did have significantly lower calorie intake compared with normal intake values. Although we were unable to demonstrate that this decreased intake was associated with higher serum leptin levels, this experience with 23 children may have been too small to establish this association. Alternatively, measurement of tissue leptin mRNA levels and longer periods of observation may be required to define this relationship.
Cytokine derangement characterized by elevated proinflammatory cytokine (tumor necrosis factor-α, IL-1, and IL-6) activity has been postulated to play a role in HIV wasting in adults . Clinical trials using anti-cytokine therapies such as thalidomide  in HIV-infected adults are aimed at testing this hypothesis. Previous experience analyzing serum levels of IL-6 in HIV-infected children  led us to focus on IL-6 as a measure of proinflammatory cytokine activity. Peripheral blood cells of our subjects stimulated with endotoxin demonstrated increased plasma levels of IL-6 in the impaired growth group when compared to the normal growth group. This finding was of particular interest since we also found significant differences in IGF-1 levels of the same children. It has been proposed that chronic elevations of IL-6 may cause decreased IGF-1 responses and lead to poor growth. Previous studies have found low levels of IGF-1 in HIV-infected children . Recently, a transgenic mice model was used to demonstrate that IL-6 causes growth impairment (stunting in particular) through a decrease in IGF-1. A negative correlation between IL-6 and IGF-1 levels in juvenile rhematoid arthritis patients (with stunted growth) has been shown . In our patients, we have observed significantly lower IGF-1 levels in HIV-infected children compared with age- and gender-matched controls. Moreover, HIV-infected children with growth impairment had lower levels of IGF-1 when compared with HIV-infected children with normal growth. Thus, a relationship between increased IL-6 activity and decreased IGF-1 may exist and contribute to growth impairment in HIV-infected children. Studies which examine proinflammatory cytokine activity and IGF-1 levels at times of acute opportunistic infections in HIV-infected children should further delineate their contribution to growth impairment of these children.
Lastly, we found that increased viral load, advanced clinical disease and advanced immunological disease were associated with poor growth in HIV-infected children. The association between increased viral load and growth impairment has been reported recently for young children with perinatal HIV infection  indicating that average viral loads are higher in HIV-infected children with growth failure than those without. HIV-infected children with increased viral burden and/or categorized into more severe clinical and immunological illness are more likely to have had increased number and severity of opportunistic infections than HIV-infected children with lower viral burden and mild clinical and immune suppression. We propose that these recurrent opportunistic infections would have exposed these children to multiple stimulatory events of proinflammatory cytokine activation leading to catabolic pathways such as inappropriate substrate utilization, anorexia, and/or decreased IGF-1/growth hormone effect; the net result over time being growth impairment.
In summary, we have identified several clinical and laboratory determinants which were associated with growth impairment in HIV-infected children. Advanced HIV clinical disease, severe immune suppression, increased viral burden, increased IL-6 activity, decreased total serum protein and decreased IGF-1 levels were more likely to be found in HIV-infected children with growth impairment as compared to HIV-infected children with normal growth. Energy consumed by our cohort of HIV-infected children was found to be substandard when compared to expected normal energy intake. We postulate that these multiple associations identified are inexorably linked in the chain of events which lead to growth impairment in these children. Examining the relationships between energy intake, viral burden, proinflammatory cytokine activity, substrate utilization, and IGF-1/GH axis with growth parameters at times of acute opportunistic infections and quantitating longitudinally the level of fluctuations over times of stable medical condition in individual patients will be important in outlining the cascade of events which results in abnormal growth. Finally, in contrast to previous hypotheses which implicate a chronic `hypermetabolic state' as a major determinant of poor growth in HIV-infected children, we found no evidence of increased resting or total energy expenditures in our cohort of HIV-infected children at times of stable medical condition. This study is limited by the small number of children and the fact that these children were studied at one time point of wellness. Further studies involving a larger cohort of children and studies aimed at times of acute illnesses are needed to truly determine the contribution of energy expenditure to growth impairment in HIV-infected children.
We wish to thank Aloka Mitra, Natalie Seamon, and Dr.Marise McNeely for their excellent technical assistance. We are also grateful for the support and assistance provided to the study subjects by the following staff members of the New York Hospital Program for Children with AIDS: Dr. Florence Marshall, Dr. Joseph Stavola, Dr. Murli Purswani, Ann-Margaret Dunn, Claudia Grassey, Tony Hinds, and Marie Francis.
1. McKinney RE, Robertson WR. Effect of human immunodeficiency virus infection on the growth of young children. J Pediatr 1993, 123: 579–82.
2. Peters VB, Rosh JR, Mugrditchian L. et al
. Growth failure as the first expression of malnutrition in children with human immunodeficiency virus infection
. The Mount Sinai Journal of Medicine 1998, 65: 1–4.
3. McKinney RE, Wilfert C. Growth as a prognostic indicator in children with human immunodeficiency virus infection treated with zidovudine. AIDS Clinical Trials Group Protocol 043 Study Group.
J Pediatr 1994, 125: 728–33.
4. Scott GB, Buck BE, Leterman JG. et al
. Acquired immuno-deficiency syndrome in infants. N Engl J Med 1984, 310: 76–80.
5. Moye J, Rich KC, Kalish LA. et al
. Natural history of somatic growth in infants born to women infected by human immunodeficiency virus. Women and Infants Transmission Study Group.
J Pediatr 1996, 128: 58–69.
6. Gillin JS, Shike M, Alcock N. et al
. Malabsorption and mucosal abnormalities of the small intestine in the acquired immunodeficiency syndrome. Ann Intern Med 1985, 102: 619–622.
7. Miller TL, Orav EJ, Martin SR, Cooper ER, McIntosh K, Winter HS. Malnutrition and carbohydrate malabsorption in children with vertically transmitted human immuno-deficiency virus 1 infection. Gastroenterology 1991, 100: 1296–1302.
8. Yolken RH, Hart W, Oung I. et al
. Gastrointestinal dysfunction and disaccharide intolerance in children infected with human immunodeficiency virus. J Pediatr 1991, 118: 359–63.
9. Laue L, Pizzo PA, Butler K, Cutler GB. Growth and neuro endocrine dysfunction in children with acquire immunodeficiency syndrome. J Pediatr 1990, 117: 541–545.
10. Kaufman F, Gertner J, Sleeper L, et al
. Growth hormone secretion in HIV positive versus HIV-negative Hemophiliac males with abnormal growth and pubertal development: the Hemophilia growth and development study [abstract]. Pediatr Res
11. Gurram M, Chirmule N, Wang XP, Ponugoti N, Pahwa S. Increase spontaneous secretion of interleukin 6 and tumor necrosis factor alpha by peripheral blood lymphocytes of human immunodeficiency virus-infected children. Pediatr Infect Dis J 1994, 13: 496–501.
12. Cunningham-Rundles S, Kin SH, Dnistrian A. et al
. Micronutrient and cytokine interaction in congenital pediatric HIV infection. J Nutr 1996, 126: 2674S–2679S.
13. Dunn AM, Cervia J, Noel GJ. Failure to thrive in the HIV-infected child. XIth International Conference of AIDS.
Vancouver, July 1996 [abstract 2312].
14. Hommes MJT, Romjin JA, Godried MH. et al
. Increased resting energy expenditure in HIV-infected men. Metabolism 1990, 39: 1186–1190.
15. Melchoir JC, Salmon D, Rigaud D. et al
. Resting energy expenditure is increased in stable, malnourished HIV-infected patients. Am J Clin Nutr 1991, 53: 437–441.
16. Laue L, Cutler GB. Abnormalities in growth and development.
In:Pediatric AIDS, The Challenge of HIV Infection in Infants, Children, and Adolescents
, 2nd edn. Edited by Pizzo PA, Wilfert CM. Baltimore, Maryland: Williams & Wilkins; 1994. pp. 575–590.
17. Miller TL. Nutritional aspects of pediatric HIV infection.
In:Nutrition in Pediatrics, Basic Science and Clinical Applications
, 2nd edn. Edited by Walker WA, Watkins JB. Hamilton, London: B.C. Decker Inc.; 1997. pp. 534–550.
18. McCallan DC, Noble C, Baldwin C. et al
. Energy expenditure and wasting in human immunodeficiency virus infection. N Engl J Med 1995, 333: 83–88.
19. Centers for Disease Control and Prevention. 1994 revised classification system for human immunodeficiency virus infection in children less than 13 years of age. MMWR
1994, 43 (RR-12)
20. U.S. Dept. of Agriculture. Manual of food codes and conversions of measures to gram-weight for use with individual food intake data. Nationwide Food Consumption Survey, Working Copy-1979.
Science and Education Administration, Human Nutrition Center, Consumer and Food Economics Institute. Riverdale, MD: US Department of Agriculture, Agricultural Research Service.
21. Report of a Joint FAO/WHO/UNU Expert Consultation. Infants, children, & adolescents – energy requirements. Technical Report Series No. 724.
Geneva: World Health Organization; 1985. pp. 90–98.
22. Dibley MJ, Staehling NS, Nieburg P, Trowbridge FL. Interpretation of Z-score anthropometric indicators derived from the international growth reference. Am J Clin Nutr 1987, 46: 749–62.
23. Appendix 2, Table A–11. Basal metabolic rates: infants and children.
In:Nutrition in Pediatrics, Basic Science and Clinical Applications
, 2nd edn. Edited by: Walker WA, Watkins JB. Hamilton, London: B.C. Decker Inc.; 1997. p. 829.
24. Schoeller DA. Measurement of energy expenditure in free-living humans by using doubly labeled water. J Nutr 1998, 118: 1278–1289.
25. Schoeller DA, Ravussin E, Schutz Y, Acheson KJ, Baertschi P, Jequier E. Energy expenditure by doubly labeled water: validation in humans and proposed calculations. Am J Physiol 1986, 250: 823–830.
26. Juul A, Banng P, Hertel NT. et al
. Serum insulin-like growth factor-1 in 1030 healthy children, adolescents, and adults: relation to age, sex, stage of puberty, testicular size, and body mass index. J Clin Endocrinol Metab 1994, 78: 744–752.
27. Arpadi SA, Cuff PA, Kotler DP, et al
. Energy balance, viral replication and growth in HIV-infected children [abstract]. Pediatr Res
28. Blumberg BS, Han HW, Mildran D, et al
. Iron and iron binding proteins in persistent generalized lymphadenopathy and AIDS [letter]. Lancet
29. Hellerstein MK, Grunfeld C, Wu K. et al
. Increased de novo hepatic lipogenesis in HIV-infected humans. J Clin Endocrinol Metab 1993, 76: 559–565.
30. Stein TP, Nutinsky C, Condoluci D. et al
. Protein and energy substrate metabolism in AIDS patients. Metabolism 1990, 39: 876–881.
31. Grunfeld C, Zhao C, Fuller J. et al
. Endotoxin and cytokines induce expression of leptin, the ob
gene product, in hamsters. J Clin Invest 1996, 97: 2152–2157.
32. Sarraf P, Frederick R, Turner E. et al
. Multiple cytokines and acute inflammation raise mouse leptin levels: potential role of inflammatory anorexia. J Exp Med 1997, 185: 171–175.
33. Johann-Liang R, Smith A, Licholai T, et al
. Leptin levels in HIV-infected hildren: correlation with body mass index. 37th Interscience Conference on Antimicrobial Agents and Chemotherapy.
Toronto, September–October 1997 [abstract I–25].
34. Navikas V, Link J, Persson T. et al
. Increased mRNA expression of IL-6, IL-10, TNF-α, and perforin in blood mononuclear cells in human HIV infection. J Acquir Immune Defic Syndr Hum Retrovirol 1995, 9: 484–489.
35. Landman D, Sarai A, Sathe S. Use of pentoxifylline therapy for patients with AIDS-related wasting: pilot study. Clin Infect Dis 1994, 18: 97–99.
36. Johann-Liang R, Cervia J, Noel GJ. Serum Levels of Interleukin-6 and failing-to thrive in children infected with human immunodeficiency virus [abstract]. Pediatr Res
37. Matarazzo P, Palomba E, Lala R. et al
. Growth impairment, IGF-1 hyposecretion and thyroid dysfunction in children with perinatal HIV-1 infection. Acta Paediatr 1994, 83: 1029–1034.
38. De Benedetti F, Alonzi T, Noretta A. et al
. Interleukin 6 causes growth impairment in transgenic mice through a decrease in insulin-like growth factor-1. J Clin Invest 1997, 99: 643–650.
39. Pollack H, Glasberg H, Lee H. et al
. Impaired early growth of infants perinatally infected with human immunodeficiency virus: Correlation with viral load. J Pediatr 1997, 130: 915–922.
Keywords:© 2000 Lippincott Williams & Wilkins, Inc.
resting metabolic rate; total metabolic rate; energy intake; failure-to-thrive; wasting; stunting; doubly-labeled water; cytokine