- Email: [email protected]
Energy expenditure, energy balance, and composition of weight gain in low birth weight infants fed diets of different protein and energy content
FETAL AND NEONATAL MEDICINE
Energy expenditure, energy balance, and composition of weight gain in low birth weight infants fed diets of different protein and energy content The effect of energy and protein intakes on energy expenditure, energy balance, and amount and relative rate of both protein and fat deposition in new tissue was investigated in 19 low birth weight infants whose mean protein and energy intakes, respectively, were/2.24 g / k g / d and 113 k c a l / k g / d (formula A, n = 8), 3.6 g / k g / d and 115 k c a l / k g / d (formula B, n = 5), and 3.5 g / k g / d and 149 k c a l / k g / d (formula C, n = 6). The higher energy intake (formula C) but not the higher protein intake (formula B) resulted in greater energy expenditure. Both the higher protein (formula B vs formula A) and higher energy intakes (formula C vs formula B) resulted in greater weight gain secondary, in group B, to a greater absolute rate of protein deposition and, in group C, to a greater absolute rate of fat deposition. The relative composition of the new tissue deposited reflected the proportional intakes of protein and energy. The numerical value of the protein/fat ratio ( g / g ) of the new tissue deposited by infants fed formulas A a n d C, the protein contents of which were low relative to energy contents, were similar and significantly lower than the numerical value of the protein/fat ratio of the new tissue deposited by infants fed formula B, which had a higher protein content relative to energy content. These findings suggest that the composition of weight gain is related to both the absolute amounts and the proportions of dietary protein and energy; thus, both must be considered in formulation of nutritional regimens for LBW infants. (J PEDIATR 1987;110:753-9) Karl F. Schulze, M.D., Mark Stefanski, B.A., Julia Masterson, M.S., Regina Spinnazola, M.D., Rajasekhar Ramakrishnan, Eng.Sc.D., Ralph B. Dell, M.D., a n d William C. Heird, M.D. From the Department of Pediatrics, College of Physicians & Surgeons of Columbia University, and Babies Hospital (Presbyterian Hospital), New York
The intrauterine rate of weight gain, the standard by which the nutritional efficacy of dietary regimens for low birth weight infants is commonly judged, I can be achieved readily with diets providing several different protein and
Supported by Grants HD13030 and RR00645 from the National Institutes of Health. Submitted for publication June 2, 1986; accepted Nov. 25, 1986. Reprint requests: Karl F. Schulze, M.D., Department of Pediatrics, College of Physicians and Surgeons, Columbia University, 630 W. 168th St., New York, NY 10032.
See related letter, p. 815.
LBW ~'co2 ~'o2
I
Low birth weight Carbon dioxide consumption Oxygen consumption
]
I
energy intakes. 2-7 However, bone of the diets studied to date produces both the intrauterine rate of weight gain and the intrauterine composition of weight gain. In fact, the intakes shown to produce the intrauterine rate of weight 753
754
S c h u l z e et al.
The Journal o f Pediatrics May 1987
T a b l e I. Characteristics of study population
Birth weight (g) Gestational age (wk) Age desired intake tolerated (d) Age birth weight regained (d)
Group A (n = 8)
Group B (n = 5 )
Group C (n = 6 )
1448 + 237 31+2 15 _+ 3 18+4
1436 + 251 32_+3 19_+ 5 18+4
1502 -- 201 32+3 19_+ 8 14+6
Values representmean + SD.
gain appear to result in a disproportionate accretion of fat. Several studies suggest that the amount of protein deposited is closely related to protein intake and that the amount of fat deposited is closely related to energy intake. 27 However, many of these studies were conducted in infants fed different amounts of a single formula rather than in infants fed formulas in which protein and energy varied independently; others were conducted without concurrent measurements of both nitrogen and energy balance. Thus the possibility that the rate and composition of weight gain may depend on the proportion of protein to energy intake and on the absolute intakes of protein and energy has not been investigated. In addition to the effect of diet composition On the rate and composition of weight gain, there is thought to be an effect of diet on energy expenditure. This effect has been termed diet-induced thermogenesis or specific dynamic action. The independent effects of protein and energy intakes on energy expenditure also have not been well studied; most studies did not include independent variations of protein and energy intake. We have recently reported studies of growth, metabolic response, and retention of nitrogen and major electrolytes and minerals in LBW infants fed formulas providing var,ying intakes of protein and energy and varying proportions of each. s Energy expenditure and energy balance data collected concurrently in a subset of this population of infants are reported here. These data help to resolve some of the difficulties mentioned above. METHODS Data were obtained for 19 of the 27 LBW infants comprising the study population of the previously mentioned report? The clinical characteristics of these 19 patients are summarized in Table I. None had significant medical problems, and none was receiving medications other than vitamins. Infants enrolled in the larger study but not included in this study had incomplete energy balance data because of equipment problems (one infant), incomplete (<6 hours) measurement of energy expenditure
(two infants), or scheduling conflicts (five infants). No infant was excluded because of medical or dietary considerations. Shortly after birth the infants were assigned randomly and blindly to receive one of three color-coded formulas. These provided mean protein and energy intakes of 2.24 g/kg/d and 113 kcal/kg/d (formula A), 3.6 g/kg/d and 115 kcal/kg/d (formula B), and 3.5 g/kg/d and 149 kcal/kg/d (formula C). The protein content of the formulas was modified bovine milk protein with a 60:40 ratio of whey proteins and caseins; nonprotein energy consisted of roughly equicaloric amounts of lactose and a mixture of corn (~60%) and coconut oil (~40%). The electrolyte, mineral, and vitamin contents of all three formulas were similar.8 The assigned formula was started as soon as enteral feedings were tolerated by the infant, and the volume of formula fed was increased, as tolerated, until the desired intake of 180 ml/kg/d) was reached. This volume was then maintained throughout the study, that is, until weight reached 2200 g. During this period infants were nursed in servocontrolled, single-walled incubators at a thermoneutral temperature. Formulas were delivered either by orogastric tube inserted at the time of each feeding or, in older infants, by nipple. Vitamin E (25 IU) and a mixture of vitamins A (1500 IU), C (35 mg), and D (400 IU) were administered daily once feedings were established. Weight gain, nitrogen balance, and energy balance were monitored from the time full feeds were tolerated until discharge. The methods used for monitoring weight gain and nitrogen balance were described previously? Those for measuring energy balance are described below. The study was approved by the Institutional Review Board of the College of Physicians & Surgeons of Columbia University. Informed parental consent was obtained before an infant was enrolled. Nutrient balance studies. All balance data were determined during the first week of desired intake (180 ml/ kg-~/d -~) and every other week thereafter until weight reached 2200 g. Forty-eight-hour collections of stool and urine were obtained using the procedures described previ-
Volume 110 Number 5
Protein/energy intake re weight gain
755
T a b l e II. Components of energy balance
Gross energy intake (kcal/kg/d) Energy excreted (kcal/kg/d) Metabolizable energy (kcal/kg/d) Energy expenditure (kcal/kg/d) Energy stored (kcal/kg/d) Respiratory quotient
Group A (n = 8 )
Group B (n:5)
Group C (n=6)
113 -+ 3
115 • 3
149 _+ 6
7• 4 106 • 5
12 + 9 104 + 9
11 _+ 6 137 • 6*
60 • 4
58 _+ 3
69 -+ 8*
47 _ 4 0.95 • 0.05
46 • 6 0.91 + 0.04
69 • 10" 0.97 • 0.05
Values representmean-+ SD. *Difference fromother groups statisticallysignificant.
ously, including use of carmine red markers to delineate the stool collections? Energy balance, or energy stored, as calculated by the equation ES = ( G E - F E - UE) - EE = M E - EE. where ES is stored energy, GE is gross energy intake, FE is fecal energy loss, UE is urinary energy loss, ME is metabolizable energy (net energy available after subtraction of fecal and urinary energy losses from gross energy intake), and EE is energy expenditure. Gross energy intake was calculated as the product of the daily volumes of intake and the protein, carbohydrate, and fat concentrations of the formulas using energy equivalents of 5.65 kcal/g, 3.95 kcal/g, and 9.25 kcal/g, respectively, for protein, carbohydrate, and fat. 9 The energy content of the vitamin preparations received by the infants during the study was included as part of energy intake; for this purpose, the energy content of the preparations provided by the manufacturers was accepted. Fecal energy losses were calculated from the estimated nitrogen content and actual fat content of the stool, using the same energy equivalents for protein and fat listed above; stool carbohydrate losses, known to be small,2 were ignored. This method probably underestimates total fecal energy excretion. From the data of Brooke and Wood, ~~it appears that fecal fat accounts for only approximately 75% of fecal energy measured by ballistic bomb calorimetry. Application of this correction factor to the data reported below would reduce metabolizable energy and energy stored (Table II) of groups A, B, and C by 3, 5, and 4 kcal/kg/cl, respectively. Urinary energy losses were calculated as 5.4 kcal/g (heat of combustion of urea) of urinary nitrogen. H Energy expenditure was calculated in the usual manner from oxygen consumption, carbon dioxide production, and
urinary nitrogen excretion, using the constants of Lusk/2 ~r02 and Vc02 were determined for two Successive 3-hour periods using a traditional flow-through open-circuit system/ 3 Respiratory quotient was computed as the ratio of X/co2 to Vo2. During each study, the respiratory chamber was warmed indirectly by an overhead radiant heater servocontrolled to maintain abdominal skin temperature 36.5 ~ C. Changes in oxygen concentration across the respirometer were measured with a Servomex OA 184 paramagnetic oxygen analyzer (Taylor Sybron, Crowborough, Sussex, England). Differences in carbon dioxide concentration were measured with a Beckman LB-2 Infrared CO2 Analyzer (Beckman Instruments, Fullerton, Calif.). Air flow through the system was measured with a Matheson Linear Mass Flow Meter (Matheson, Secaucus, N.J.). Analog outputs from the analyzers and flow meter were processed electronically, on-line, to reduce the volume of data to 1-minute average values; these were then transferred to an Apple II+ computer for storage. The overall bench accuracy of this system is approximately ___2% for both Vo2 and Vc02. ~3 As described previously,8 nitrogen balance was computed as the difference between measured nitrogen intake and the sum of measured urinary nitrogen and estimated fecal nitrogen (i.e., 10% of intake). Reported fecal nitrogen losses from 16 balance studies in healthy LBW infants range from 8.6% to 19% of intake2.s,14-16and are independent of intake (r = -0.32); thus the estimated fecal nitrogen loss (10% of intake) seems reasonable. Moreover, an error of 100% in this estimate will not invalidate any of the conclusions drawn. Composition of weight gain. The protein content of weight gain (protein stored) was calculated as 6.25 times the nitrogen balance in g/kg-~/d-k The fat content of the weight gain (fat stored) was calculated as
756
Schulze et al.
The Journal of Pediatrics May 1987
T a b l e Ill. Composition of weight gained Group A (n = 8)
Weight gain (g/kg/d)* Protein stored (Prs)* g/kg/d g/g Fat stored (Fs) g/kg//d g/g PrJF~ (g/g)
14.4 _+ 2.59
Group B (n = 5)
16.8 • 2.5t
Group C (n = 6)
21.7 +_ 3.29
1.63 _+ 0.119 0.117 _+ 0.023
2.59 _+0.22 0.156 • 0.019"{"
2.63 --_ 0.11 0.122 • 0.018
4.11 + 0.50 0.290 +_0.042 0.40 _+ 0.07
3.34 _+ 0.57 0.205 • 0.0599 0.79 _+ 0.14
5.89 _+ 1.07t 0.278 • 0.064 0.46 _+ 0.10
Values representmean -+ SD. *Differencesin weightgain and nitrogenbalanceamonggroups,as detected by analysisof variance,reported previouslyr Differencesreported here basedon analysisof covarianee(see text and Letter to the Editor, p. 815). tDifference from other two groups statisticallysignificant(P <0.05).
Fat stored = [Energy stored - 5.65 (Protein stored)]/9.25. where the units for energy stored and protein stored are kcal/kg/d and g/kg/d, respectively, and the constants 5.65 and 9.25 are the energy equivalents for protein and fat (kcal/g) of dairy products (which do not differ substantially from the caloric equivalents for protein and fat of other foods and animal tissuesg). This calculation ignores energy stored as carbohydrate (i.e., from 0.45% of body weight in the 31-week fetus to 0.9% of body weight in the term neonate17,~s). To express the protein and fat components of weight gain as amounts per unit of weight gain, the respective values of protein stored and fat stored for each infant (g/kg/d) were divided by the mean rate of weight gain (g/kg/d) for that infant, and group means were calculated from individual subject means. Chemical methods. Total nitrogen content of the formulas and urine was measured by the Dumas method 19using a Coleman Model 129 automatic nitrogen analyzer. Lactose content of the formulas was determined from the galactose released following hydrolysis with lactase using a kit (Boehringer-Manheim, Houston). 2~ Fat content of the formulas and stools was determined by the method of Jeejeebhoy et al. 2j Data analysis. Inasmuch as there was no statistically significant time trend for any component of energy balance (gross energy intake, energy losses, metabolizable energy intake, energy expenditure, energy stored), the mean value of each variable for each patient was computed and the individual patient means used to compute group means for each variable. Analysis of covariance (with birth weight as a covariate) was used to test the significance of differences among the diet groups in weight gain, energy expenditure, energy stored, and protein stored; analysis of variance was used to test the significance of differences among groups in
variables related to the composition of weight gain. To control for multiple comparisons, a Bonferroni procedure2z was applied to two groups of variables. The first group included variables related to energy metabolism (metabolizable energy, energy expenditure, and respiratory quotient); for these, P <0.05/3 (F test of group differences) was required to denote statistical significance at the 5% level, and P <0.01/3 to denote statistical significance at the 1% level. The second group of variables included fat stored and the numerical value of the ratio of protein stored to fat stored; for these, P <0.05/2 and P <0.01/2, respectively, was required to denote statistical significance at the 5% and 1% levels. If the F test was significant, individual group differences (group A vs group B vs group C) were tested by the Duncan multiple range test. The statistical significance of the differences in weight gain and nitrogen retention (hence protein stored) among the three groups of infants as detected by one-way analysis of variance was reported previously,s The differences in both reported here are those detected in the larger population by analysis of covariance, which controls for subjectto-subject variation in birth weight. 23 The mean squared error plus the critical t values from a table of critical values for the Duncan test were used to construct 95% confidence limits of the group differences in energy expenditure. These confidence limits show the range of difference in energy expenditure that might be seen if the experiment were repeated. RESULTS The mean values of all components of energy balance (gross energy intake, energy losses, metabolizable energy, energy expenditure, and energy stored) in the three groups are shown in Table II. Gross energy intakes for groups A and B did not differ and, per study design, were less than
Volume 110 Number 5
that of group C. Because the differences in energy excretion among the groups were small, metabolizable energy intakes of groups A and B also did not differ and were significantly less than that of group C. There was no statistically significant relationship between either gross energy intake and energy excretion or energy excretion and postnatal age. Mean energy expenditure of group C was significantly higher than that of groups A and B, a reflection of the effect of energy intake on energy expenditure. There was no statistically significant difference in energy expenditure between groups A and B, a reflection of the absence of an effect of protein intake on energy expenditure. The 95% confidence limits for the difference in energy expenditure between group A and group B were - 7 . 4 to 11.0 kcal/ k g / / d ; for the group B vs group C difference, the limits were 0.2 to 20.8 kcal/kg/d. The greater energy intake of group C was not accompanied by greater energy excretion, and the difference in energy expenditure between this group and the other groups was small, so most of their greater intake was stored. Thus, energy stored by group C was significantly greater than that stored by groups A and B. As reported previously,8 nitrogen retention of group A (0.259 + 0.021 g / k g / d in the infants reported here) are less than that of groups B and C (0.409 _+ 0.034 and 0.417 ___0.024 g / k g / d , respectively, in the infants reported here). The rates of weight gain of the three groups of infants and both the absolute ( g / k g / d ) and relative (g/g) rates of protein and fat storage, estimated from nitrogen and energy balance data, are summarized in Table III. The rate of weight gain of all three groups differed from each other; that is, both protein and energy intake enhanced the rate of weight gain. This conclusion is based on statistical analysis of the data using birth weight as a covariate, and differs from that stated previously; that is, the rates of weight gain of groups B and C did not differ significantly and were greater than that of group A, 8 on the basis of statistical analysis of the data without using initial weight as a covariate) 3 The absolute rates of protein and fat deposition (g/kg/d) mirrored the absolute protein and energy contents of the three intakes studied; the relative rates of protein and fat deposition more closely mirrored the dietary proportions of protein and energy. In fact, the numerical values of the ratio of protein stored to fat stored of groups A and C, which received intakes with a relatively low proportion of protein to energy, were remarkably similar to each other and significantly less than that of group B, which received a higher proportion of dietary protein to energy.
Protein/energy intake re weight gain
757
DISCUSSION Energy expenditure of LBW infants fed formulas of constant composition is known to increase with increases i n formula intake). 5-7 This increase in energy expenditure is thought to reflect the energy cost of new tissue synthesis. Inasmuch as the energy cost of either synthesizing peptide bonds or catabolizing protein is greater than the energy cost of depositing fat, 24 it is commonly believed that the energy cost of tissue synthesis is related predominantly to the energy cost of protein synthesis) 5 Our data refute this common belief. There was no difference in energy expenditure between groups A and B despite a difference in both protein intake and nitrogen retention of approximately 60%. On the other hand, the 95% confidence limits of the difference in energy expenditure between groups A and B are wide ( - 7 , 4 to 11.0 kcal/kg/d). Thus a larger population or a greater difference in protein intake must be studied before it can be concluded that protein intake does not affect energy expenditure. In contrast, a statistically significant effect of energy intake on energy expenditure was observed. Energy expenditure of group C, the intake of which differed from that of group B only in energy content and the nitrogen retention of which did not differ from that of group B, was significantly greater than that of group B. Thus the greater energy expenditure of group C was not associated with greater protein synthesis but with greater storage of nonprotein energy as either glycogen or fat. Because little glycogen is stored during this time of life (only 0.45% to 0.9% of body weight ~7~8 and the energy cost of depositing fat is thought to be trivial,26 it is likely that the greater expenditure is explained by lipogenesis from carbohydrate. This conclusion is supported by the greater respiratory quotient of group C compared with group B (0.97 vs 0.91). Other studies also suggest that lipogenesis from carbohydrate is prominent in LBW infants fed high energy intakes. 27 Further, the energy expended in storing metabolizable energy as fat has been estimated from animal studies to be 0.36 kcal per kcal energy stored as fat. z83~ Group C stored roughly 20 kcal/kg/d more energy as fat than groups A and B did; thus, according to this estimate, their energy expenditure should exceed that of the other groups by 7.2 kcal/kg/d. The mean difference in energy expenditure observed was 11 kcal/kg/d. The clinical implications of the greater energy expenditure of infants fed the higher energy intake remain to be determined. Infants fed a hi~Ja energy intake are likely to have high rates of oxygen and carbon dioxide transport. Thus, infants whose metabolic gas exchange is limited
758
Schulze et al.
(those with active or resolving cardiac or respiratory insufficiency), who are often the targets of nutritional interventions designed to increase energy intake and hence weight gain, may be unable to meet the increased demand for gas transport imposed by a relatively high energy intake. Several studies in infants fed different amounts of a single formula, and hence a constant proportion of protein to energy, show that retention and storage of both nutrients are closely correlated to the respective intakes of each.S, ~s However, these studies leave open the possibility that retention of protein and fat might be affected also by the proportion of dietary protein to dietary energy. The macronutrient balance data reported here, although limited to some extent by the facts that fecal nitrogen excretion was estimated rather than measured, and fecal energy excretion was calculated from estimated fecal nitrogen and measured fecal fat rather than measured directly by calorimetry (see above), are the only such data obtained in infants fed formulas providing both different proportions and different absolute amounts of energy and protein. Moreover, the limitations of the data, as discussed above, are not sufficiently serious to affect the overall conclusions drawn from them. Like other studies, our data show that both the rate and composition of weight gain are closely related to the absolute intakes of both protein and energy---a greater intake of either led to a greater rate of weight gain. In infants fed a greater protein intake, the greater rate of weight gain was associated with a greater rate of protein deposition (groups B and C vs group A); in those fed a greater energy intake, the greater rate of weight gain was associated with a greater rate of fat deposition (group C vs group B). Our data also suggest that the proportion of dietary protein and energy affects the composition of weight gain. For example, although infants fed formulas A and B, which differed 0nly in protein content, stored energy at comparable rates, infants fed the higher protein intake (formula B) retained more nitrogen and gained weight more rapidly. Moreover, the weight gained by infants fed the higher protein intake was leaner. In contrast, infants fed formulas B and C, which differed only in energy content, deposited similar amounts of protein, but both the rate of weight gain and the amount of fat stored were greater in infants fed the higher energy intake (formula C). Thus the weight gained by infants fed formula C was fatter. It is interesting that the relative composition of weight gained by this group (protein/fat 2.6:5.9, or 0.46) and group A (protein/fat 1.6:4.1, or 0.40), the intakes of which provided a relatively lower proportion of protein to energy than that of group B, was remarkably similar. These findings suggest that it is possible to produce different rates of weight gain of either similar (groups A
The Journal of Pediatrics May 1987
and C) or dissimilar (groups B and C) composition simply by regulating the composition of dietary intake. This, in turn, has important implications for designing nutritional regimens for LBW infants. For example, the common practice of supplementing the intake of LBW infants with either carbohydrate or fat, in addition to increasing energy expenditure, will result in greater fat deposition, which may or may not be desirable. Should future goals for dietary management of the LBW infant include goals for the composition as well as the rate of weight gain, this finding must be considered. REFERENCES
1. American Academy of Pediatrics, Committee on Nutrition. Nutritional needs of low birth weight infants. Pediatrics 1985;75:976-86. 2. Reichman B, Chessex P, Putet G, et al. Diet, fat accretion and growth in premature infants. N Engl J Med 1981;305:14951500. 3. Whyte RK, Haslam R, Vlainic C, et al. Energy balance and nitrogen balance in growing low birth weight infants fed human milk or formula. Pediatr Res 1983;17:891-98. 4. Putet G, Senterre J, Rigo J, Salle B. Nutrient balance, energy utilization and composition of weight gain in very low birth weight infants fed pooled human milk or a preterm formula. J PEDIATR1984;105:79-85. 5. Brooke OG. Energy balance and metabolic rate in preterm infants fed with standard and high-energy formulas. Br J Nutr 1980; 40:13-23. 6. Reichman B, Chessex P, Verellen G, et al. Dietary composition and macronutrient storage. Pediatrics 1983;72:322-7. 7. Chessex P, Reichman B, Verellen G, et al. Influence of postnatal age, energy intake and weight gain on energy metabolism in the very low birth weight infant. J PEDIATR 1981;99:761-6. 8. Kashyap S, Forsyth M, Zucker C, et al. Effects of varying protein and energy intakes on growth and metabolic response in low birth weight infants. J PEDIATR1986;108:955-63. 9. Food and Agricultural Organization of the United Nations, Ad Hoc Committee. Energy and protein requirements. FAO Nutritional Series No. 7, Rome, 1973;1024. 10. Brooke OG, Wood C. Relation between fecal fat and energy in preterm infants. Arch Dis Child 1983;58:305-6. 11. Kleiber M. The fire of life: an introduction to animal energetics. Melbourne, Fla.: Krieger, 1974;104-30. 12. Lusk G. Basal metabolism'~standards.In: Scientific tables, 6th ed. Montreal: Documenta Geigy, 1961;628. 13. Schulze K, Kairam R, Stefanski M, et al. Continuous measurement of minute ventilation and gaseous metabolism of newborn infants. J Appl Physiol 1981;50:1098-1103. 14. Voyer M, Senterre J, Rigo J, et al. Human milk lactoengineering. Acta Paediatr Scand 1984;73:302-6. 15. Freymond D, Schutz Y, Decombaz J, et al. Energy balance, physical activity and thermogenic effect of feeding in premature infants. Pediatr Res 1986;20:638-45. 16. Catzeflis C, Schutz Y, Micheli JL, et al. Whole body protein synthesis and energy expenditure in very low birth weight infants. Pediatr Res 1985;19:679-87. 17. Shelley HJ. Carbohydrate reserves in the newborn infant. Br Med J 1964;1:273-5. 18. Widdowson EM. Changes in body proportion and composi-
Volume 110 Number 5
19. 20.
21.
22. 23.
24.
Protein/energy intake re weight gain
tion during growth. In: Davis JA, Dobbing J, eds. Scientific foundation of pediatrics. Philadelphia: Saunders, 1974;15364. Ebeling M. The Dumas method for nitrogen in feeds. J Assoc Off Anal Chem 1968;51:766-70. Kurz G, Wallenfels K. Lactose and other/3-galactosides. In: Bergmeyer HNU, ed. Methods of enzymatic analysis, vol 3. Orlando, Fla.: Academic Press, 1974;1180-4. Jeejeebhoy KN, Ahmad S, Kozak G. Determination of fecal fats containing both medium and long chain triglycerides and fatty acids. Clin Biochem 1970;3:157-63. Miller RG Jr. Simultaneous statistical inference. New York: McGraw-Hill, 1966. Kashyap S, Schulze KF, Ramakrishman R, Dell RB, Heird WC. Effects of protein and energy intakes on growth [Letter]. J PED1ATR 1987;110:815. Flatt JP. Energetics of intermediary metabolism. In: Kinney JM, ed. Energy metabolism in health and disease. Columbus, Ohio: Ross Laboratories, 1980;77-87.
75 9
25. Mestyan J, Jarai I, Fekete M. The total energy expenditure and its components in premature infants maintained under different nursing and environmental conditions. Pediatr Res 1968;2:161-71. 26. Hommes FA, Drost YM, Geraets WXM, Reijenga MAA. The energy requirement for growth: an application of Atkinson's metabolic price system. Pediatr Res 1975;9:51-5. 27. Van Aerde J, Sauer P, Heim T, et al. Growth, macronutrient oxidation and accretion in very low birth weight (VLBW) infants with variable energy intake and constant diet composition. Pediatr Res 1985;19:368A. 28. Van Es AJH. The energetics of fat deposition during growth. Nutr Metab 1977;21:88-104. 29. Pullar JD, Webster AJF. The energy cost of fat and protein deposition in the rat. Br J Nutr 1977;37:355-63. 30. Harvey GR, Tobin GL. Luxuskonsumption, diet induced thermogenesis and brown fat: a critical review. Clin Sci 1983;64:7-18.
FELLOWSHIPS Available fellowships in pediatric subspecialties and those for general academic pediatric training are listed once a year, in May, in The JournM of Pediatrics. Each October, forms for listing such fellowships are sent to the Chairman of the Department of Pediatrics at most major hospitals in the United States and Canada. Should you desire to list fellowships, a separate application must be made each year for each position. All applications must be returned to The C. V. Mosby Company by February 15 of the listing year to ensure publication. Additional forms will be supplied on request from the Journal Editing Department, The C. V. Mosby Company, 11830 Westline Industrial Drive, St. Louis, M O 63146/314-872-8370.
Energy expenditure, energy balance, and composition of weight gain in low birth weight infants fed diets of different protein and energy content The effect of energy and protein intakes on energy expenditure, energy balance, and amount and relative rate of both protein and fat deposition in new tissue was investigated in 19 low birth weight infants whose mean protein and energy intakes, respectively, were/2.24 g / k g / d and 113 k c a l / k g / d (formula A, n = 8), 3.6 g / k g / d and 115 k c a l / k g / d (formula B, n = 5), and 3.5 g / k g / d and 149 k c a l / k g / d (formula C, n = 6). The higher energy intake (formula C) but not the higher protein intake (formula B) resulted in greater energy expenditure. Both the higher protein (formula B vs formula A) and higher energy intakes (formula C vs formula B) resulted in greater weight gain secondary, in group B, to a greater absolute rate of protein deposition and, in group C, to a greater absolute rate of fat deposition. The relative composition of the new tissue deposited reflected the proportional intakes of protein and energy. The numerical value of the protein/fat ratio ( g / g ) of the new tissue deposited by infants fed formulas A a n d C, the protein contents of which were low relative to energy contents, were similar and significantly lower than the numerical value of the protein/fat ratio of the new tissue deposited by infants fed formula B, which had a higher protein content relative to energy content. These findings suggest that the composition of weight gain is related to both the absolute amounts and the proportions of dietary protein and energy; thus, both must be considered in formulation of nutritional regimens for LBW infants. (J PEDIATR 1987;110:753-9) Karl F. Schulze, M.D., Mark Stefanski, B.A., Julia Masterson, M.S., Regina Spinnazola, M.D., Rajasekhar Ramakrishnan, Eng.Sc.D., Ralph B. Dell, M.D., a n d William C. Heird, M.D. From the Department of Pediatrics, College of Physicians & Surgeons of Columbia University, and Babies Hospital (Presbyterian Hospital), New York
The intrauterine rate of weight gain, the standard by which the nutritional efficacy of dietary regimens for low birth weight infants is commonly judged, I can be achieved readily with diets providing several different protein and
Supported by Grants HD13030 and RR00645 from the National Institutes of Health. Submitted for publication June 2, 1986; accepted Nov. 25, 1986. Reprint requests: Karl F. Schulze, M.D., Department of Pediatrics, College of Physicians and Surgeons, Columbia University, 630 W. 168th St., New York, NY 10032.
See related letter, p. 815.
LBW ~'co2 ~'o2
I
Low birth weight Carbon dioxide consumption Oxygen consumption
]
I
energy intakes. 2-7 However, bone of the diets studied to date produces both the intrauterine rate of weight gain and the intrauterine composition of weight gain. In fact, the intakes shown to produce the intrauterine rate of weight 753
754
S c h u l z e et al.
The Journal o f Pediatrics May 1987
T a b l e I. Characteristics of study population
Birth weight (g) Gestational age (wk) Age desired intake tolerated (d) Age birth weight regained (d)
Group A (n = 8)
Group B (n = 5 )
Group C (n = 6 )
1448 + 237 31+2 15 _+ 3 18+4
1436 + 251 32_+3 19_+ 5 18+4
1502 -- 201 32+3 19_+ 8 14+6
Values representmean + SD.
gain appear to result in a disproportionate accretion of fat. Several studies suggest that the amount of protein deposited is closely related to protein intake and that the amount of fat deposited is closely related to energy intake. 27 However, many of these studies were conducted in infants fed different amounts of a single formula rather than in infants fed formulas in which protein and energy varied independently; others were conducted without concurrent measurements of both nitrogen and energy balance. Thus the possibility that the rate and composition of weight gain may depend on the proportion of protein to energy intake and on the absolute intakes of protein and energy has not been investigated. In addition to the effect of diet composition On the rate and composition of weight gain, there is thought to be an effect of diet on energy expenditure. This effect has been termed diet-induced thermogenesis or specific dynamic action. The independent effects of protein and energy intakes on energy expenditure also have not been well studied; most studies did not include independent variations of protein and energy intake. We have recently reported studies of growth, metabolic response, and retention of nitrogen and major electrolytes and minerals in LBW infants fed formulas providing var,ying intakes of protein and energy and varying proportions of each. s Energy expenditure and energy balance data collected concurrently in a subset of this population of infants are reported here. These data help to resolve some of the difficulties mentioned above. METHODS Data were obtained for 19 of the 27 LBW infants comprising the study population of the previously mentioned report? The clinical characteristics of these 19 patients are summarized in Table I. None had significant medical problems, and none was receiving medications other than vitamins. Infants enrolled in the larger study but not included in this study had incomplete energy balance data because of equipment problems (one infant), incomplete (<6 hours) measurement of energy expenditure
(two infants), or scheduling conflicts (five infants). No infant was excluded because of medical or dietary considerations. Shortly after birth the infants were assigned randomly and blindly to receive one of three color-coded formulas. These provided mean protein and energy intakes of 2.24 g/kg/d and 113 kcal/kg/d (formula A), 3.6 g/kg/d and 115 kcal/kg/d (formula B), and 3.5 g/kg/d and 149 kcal/kg/d (formula C). The protein content of the formulas was modified bovine milk protein with a 60:40 ratio of whey proteins and caseins; nonprotein energy consisted of roughly equicaloric amounts of lactose and a mixture of corn (~60%) and coconut oil (~40%). The electrolyte, mineral, and vitamin contents of all three formulas were similar.8 The assigned formula was started as soon as enteral feedings were tolerated by the infant, and the volume of formula fed was increased, as tolerated, until the desired intake of 180 ml/kg/d) was reached. This volume was then maintained throughout the study, that is, until weight reached 2200 g. During this period infants were nursed in servocontrolled, single-walled incubators at a thermoneutral temperature. Formulas were delivered either by orogastric tube inserted at the time of each feeding or, in older infants, by nipple. Vitamin E (25 IU) and a mixture of vitamins A (1500 IU), C (35 mg), and D (400 IU) were administered daily once feedings were established. Weight gain, nitrogen balance, and energy balance were monitored from the time full feeds were tolerated until discharge. The methods used for monitoring weight gain and nitrogen balance were described previously? Those for measuring energy balance are described below. The study was approved by the Institutional Review Board of the College of Physicians & Surgeons of Columbia University. Informed parental consent was obtained before an infant was enrolled. Nutrient balance studies. All balance data were determined during the first week of desired intake (180 ml/ kg-~/d -~) and every other week thereafter until weight reached 2200 g. Forty-eight-hour collections of stool and urine were obtained using the procedures described previ-
Volume 110 Number 5
Protein/energy intake re weight gain
755
T a b l e II. Components of energy balance
Gross energy intake (kcal/kg/d) Energy excreted (kcal/kg/d) Metabolizable energy (kcal/kg/d) Energy expenditure (kcal/kg/d) Energy stored (kcal/kg/d) Respiratory quotient
Group A (n = 8 )
Group B (n:5)
Group C (n=6)
113 -+ 3
115 • 3
149 _+ 6
7• 4 106 • 5
12 + 9 104 + 9
11 _+ 6 137 • 6*
60 • 4
58 _+ 3
69 -+ 8*
47 _ 4 0.95 • 0.05
46 • 6 0.91 + 0.04
69 • 10" 0.97 • 0.05
Values representmean-+ SD. *Difference fromother groups statisticallysignificant.
ously, including use of carmine red markers to delineate the stool collections? Energy balance, or energy stored, as calculated by the equation ES = ( G E - F E - UE) - EE = M E - EE. where ES is stored energy, GE is gross energy intake, FE is fecal energy loss, UE is urinary energy loss, ME is metabolizable energy (net energy available after subtraction of fecal and urinary energy losses from gross energy intake), and EE is energy expenditure. Gross energy intake was calculated as the product of the daily volumes of intake and the protein, carbohydrate, and fat concentrations of the formulas using energy equivalents of 5.65 kcal/g, 3.95 kcal/g, and 9.25 kcal/g, respectively, for protein, carbohydrate, and fat. 9 The energy content of the vitamin preparations received by the infants during the study was included as part of energy intake; for this purpose, the energy content of the preparations provided by the manufacturers was accepted. Fecal energy losses were calculated from the estimated nitrogen content and actual fat content of the stool, using the same energy equivalents for protein and fat listed above; stool carbohydrate losses, known to be small,2 were ignored. This method probably underestimates total fecal energy excretion. From the data of Brooke and Wood, ~~it appears that fecal fat accounts for only approximately 75% of fecal energy measured by ballistic bomb calorimetry. Application of this correction factor to the data reported below would reduce metabolizable energy and energy stored (Table II) of groups A, B, and C by 3, 5, and 4 kcal/kg/cl, respectively. Urinary energy losses were calculated as 5.4 kcal/g (heat of combustion of urea) of urinary nitrogen. H Energy expenditure was calculated in the usual manner from oxygen consumption, carbon dioxide production, and
urinary nitrogen excretion, using the constants of Lusk/2 ~r02 and Vc02 were determined for two Successive 3-hour periods using a traditional flow-through open-circuit system/ 3 Respiratory quotient was computed as the ratio of X/co2 to Vo2. During each study, the respiratory chamber was warmed indirectly by an overhead radiant heater servocontrolled to maintain abdominal skin temperature 36.5 ~ C. Changes in oxygen concentration across the respirometer were measured with a Servomex OA 184 paramagnetic oxygen analyzer (Taylor Sybron, Crowborough, Sussex, England). Differences in carbon dioxide concentration were measured with a Beckman LB-2 Infrared CO2 Analyzer (Beckman Instruments, Fullerton, Calif.). Air flow through the system was measured with a Matheson Linear Mass Flow Meter (Matheson, Secaucus, N.J.). Analog outputs from the analyzers and flow meter were processed electronically, on-line, to reduce the volume of data to 1-minute average values; these were then transferred to an Apple II+ computer for storage. The overall bench accuracy of this system is approximately ___2% for both Vo2 and Vc02. ~3 As described previously,8 nitrogen balance was computed as the difference between measured nitrogen intake and the sum of measured urinary nitrogen and estimated fecal nitrogen (i.e., 10% of intake). Reported fecal nitrogen losses from 16 balance studies in healthy LBW infants range from 8.6% to 19% of intake2.s,14-16and are independent of intake (r = -0.32); thus the estimated fecal nitrogen loss (10% of intake) seems reasonable. Moreover, an error of 100% in this estimate will not invalidate any of the conclusions drawn. Composition of weight gain. The protein content of weight gain (protein stored) was calculated as 6.25 times the nitrogen balance in g/kg-~/d-k The fat content of the weight gain (fat stored) was calculated as
756
Schulze et al.
The Journal of Pediatrics May 1987
T a b l e Ill. Composition of weight gained Group A (n = 8)
Weight gain (g/kg/d)* Protein stored (Prs)* g/kg/d g/g Fat stored (Fs) g/kg//d g/g PrJF~ (g/g)
14.4 _+ 2.59
Group B (n = 5)
16.8 • 2.5t
Group C (n = 6)
21.7 +_ 3.29
1.63 _+ 0.119 0.117 _+ 0.023
2.59 _+0.22 0.156 • 0.019"{"
2.63 --_ 0.11 0.122 • 0.018
4.11 + 0.50 0.290 +_0.042 0.40 _+ 0.07
3.34 _+ 0.57 0.205 • 0.0599 0.79 _+ 0.14
5.89 _+ 1.07t 0.278 • 0.064 0.46 _+ 0.10
Values representmean -+ SD. *Differencesin weightgain and nitrogenbalanceamonggroups,as detected by analysisof variance,reported previouslyr Differencesreported here basedon analysisof covarianee(see text and Letter to the Editor, p. 815). tDifference from other two groups statisticallysignificant(P <0.05).
Fat stored = [Energy stored - 5.65 (Protein stored)]/9.25. where the units for energy stored and protein stored are kcal/kg/d and g/kg/d, respectively, and the constants 5.65 and 9.25 are the energy equivalents for protein and fat (kcal/g) of dairy products (which do not differ substantially from the caloric equivalents for protein and fat of other foods and animal tissuesg). This calculation ignores energy stored as carbohydrate (i.e., from 0.45% of body weight in the 31-week fetus to 0.9% of body weight in the term neonate17,~s). To express the protein and fat components of weight gain as amounts per unit of weight gain, the respective values of protein stored and fat stored for each infant (g/kg/d) were divided by the mean rate of weight gain (g/kg/d) for that infant, and group means were calculated from individual subject means. Chemical methods. Total nitrogen content of the formulas and urine was measured by the Dumas method 19using a Coleman Model 129 automatic nitrogen analyzer. Lactose content of the formulas was determined from the galactose released following hydrolysis with lactase using a kit (Boehringer-Manheim, Houston). 2~ Fat content of the formulas and stools was determined by the method of Jeejeebhoy et al. 2j Data analysis. Inasmuch as there was no statistically significant time trend for any component of energy balance (gross energy intake, energy losses, metabolizable energy intake, energy expenditure, energy stored), the mean value of each variable for each patient was computed and the individual patient means used to compute group means for each variable. Analysis of covariance (with birth weight as a covariate) was used to test the significance of differences among the diet groups in weight gain, energy expenditure, energy stored, and protein stored; analysis of variance was used to test the significance of differences among groups in
variables related to the composition of weight gain. To control for multiple comparisons, a Bonferroni procedure2z was applied to two groups of variables. The first group included variables related to energy metabolism (metabolizable energy, energy expenditure, and respiratory quotient); for these, P <0.05/3 (F test of group differences) was required to denote statistical significance at the 5% level, and P <0.01/3 to denote statistical significance at the 1% level. The second group of variables included fat stored and the numerical value of the ratio of protein stored to fat stored; for these, P <0.05/2 and P <0.01/2, respectively, was required to denote statistical significance at the 5% and 1% levels. If the F test was significant, individual group differences (group A vs group B vs group C) were tested by the Duncan multiple range test. The statistical significance of the differences in weight gain and nitrogen retention (hence protein stored) among the three groups of infants as detected by one-way analysis of variance was reported previously,s The differences in both reported here are those detected in the larger population by analysis of covariance, which controls for subjectto-subject variation in birth weight. 23 The mean squared error plus the critical t values from a table of critical values for the Duncan test were used to construct 95% confidence limits of the group differences in energy expenditure. These confidence limits show the range of difference in energy expenditure that might be seen if the experiment were repeated. RESULTS The mean values of all components of energy balance (gross energy intake, energy losses, metabolizable energy, energy expenditure, and energy stored) in the three groups are shown in Table II. Gross energy intakes for groups A and B did not differ and, per study design, were less than
Volume 110 Number 5
that of group C. Because the differences in energy excretion among the groups were small, metabolizable energy intakes of groups A and B also did not differ and were significantly less than that of group C. There was no statistically significant relationship between either gross energy intake and energy excretion or energy excretion and postnatal age. Mean energy expenditure of group C was significantly higher than that of groups A and B, a reflection of the effect of energy intake on energy expenditure. There was no statistically significant difference in energy expenditure between groups A and B, a reflection of the absence of an effect of protein intake on energy expenditure. The 95% confidence limits for the difference in energy expenditure between group A and group B were - 7 . 4 to 11.0 kcal/ k g / / d ; for the group B vs group C difference, the limits were 0.2 to 20.8 kcal/kg/d. The greater energy intake of group C was not accompanied by greater energy excretion, and the difference in energy expenditure between this group and the other groups was small, so most of their greater intake was stored. Thus, energy stored by group C was significantly greater than that stored by groups A and B. As reported previously,8 nitrogen retention of group A (0.259 + 0.021 g / k g / d in the infants reported here) are less than that of groups B and C (0.409 _+ 0.034 and 0.417 ___0.024 g / k g / d , respectively, in the infants reported here). The rates of weight gain of the three groups of infants and both the absolute ( g / k g / d ) and relative (g/g) rates of protein and fat storage, estimated from nitrogen and energy balance data, are summarized in Table III. The rate of weight gain of all three groups differed from each other; that is, both protein and energy intake enhanced the rate of weight gain. This conclusion is based on statistical analysis of the data using birth weight as a covariate, and differs from that stated previously; that is, the rates of weight gain of groups B and C did not differ significantly and were greater than that of group A, 8 on the basis of statistical analysis of the data without using initial weight as a covariate) 3 The absolute rates of protein and fat deposition (g/kg/d) mirrored the absolute protein and energy contents of the three intakes studied; the relative rates of protein and fat deposition more closely mirrored the dietary proportions of protein and energy. In fact, the numerical values of the ratio of protein stored to fat stored of groups A and C, which received intakes with a relatively low proportion of protein to energy, were remarkably similar to each other and significantly less than that of group B, which received a higher proportion of dietary protein to energy.
Protein/energy intake re weight gain
757
DISCUSSION Energy expenditure of LBW infants fed formulas of constant composition is known to increase with increases i n formula intake). 5-7 This increase in energy expenditure is thought to reflect the energy cost of new tissue synthesis. Inasmuch as the energy cost of either synthesizing peptide bonds or catabolizing protein is greater than the energy cost of depositing fat, 24 it is commonly believed that the energy cost of tissue synthesis is related predominantly to the energy cost of protein synthesis) 5 Our data refute this common belief. There was no difference in energy expenditure between groups A and B despite a difference in both protein intake and nitrogen retention of approximately 60%. On the other hand, the 95% confidence limits of the difference in energy expenditure between groups A and B are wide ( - 7 , 4 to 11.0 kcal/kg/d). Thus a larger population or a greater difference in protein intake must be studied before it can be concluded that protein intake does not affect energy expenditure. In contrast, a statistically significant effect of energy intake on energy expenditure was observed. Energy expenditure of group C, the intake of which differed from that of group B only in energy content and the nitrogen retention of which did not differ from that of group B, was significantly greater than that of group B. Thus the greater energy expenditure of group C was not associated with greater protein synthesis but with greater storage of nonprotein energy as either glycogen or fat. Because little glycogen is stored during this time of life (only 0.45% to 0.9% of body weight ~7~8 and the energy cost of depositing fat is thought to be trivial,26 it is likely that the greater expenditure is explained by lipogenesis from carbohydrate. This conclusion is supported by the greater respiratory quotient of group C compared with group B (0.97 vs 0.91). Other studies also suggest that lipogenesis from carbohydrate is prominent in LBW infants fed high energy intakes. 27 Further, the energy expended in storing metabolizable energy as fat has been estimated from animal studies to be 0.36 kcal per kcal energy stored as fat. z83~ Group C stored roughly 20 kcal/kg/d more energy as fat than groups A and B did; thus, according to this estimate, their energy expenditure should exceed that of the other groups by 7.2 kcal/kg/d. The mean difference in energy expenditure observed was 11 kcal/kg/d. The clinical implications of the greater energy expenditure of infants fed the higher energy intake remain to be determined. Infants fed a hi~Ja energy intake are likely to have high rates of oxygen and carbon dioxide transport. Thus, infants whose metabolic gas exchange is limited
758
Schulze et al.
(those with active or resolving cardiac or respiratory insufficiency), who are often the targets of nutritional interventions designed to increase energy intake and hence weight gain, may be unable to meet the increased demand for gas transport imposed by a relatively high energy intake. Several studies in infants fed different amounts of a single formula, and hence a constant proportion of protein to energy, show that retention and storage of both nutrients are closely correlated to the respective intakes of each.S, ~s However, these studies leave open the possibility that retention of protein and fat might be affected also by the proportion of dietary protein to dietary energy. The macronutrient balance data reported here, although limited to some extent by the facts that fecal nitrogen excretion was estimated rather than measured, and fecal energy excretion was calculated from estimated fecal nitrogen and measured fecal fat rather than measured directly by calorimetry (see above), are the only such data obtained in infants fed formulas providing both different proportions and different absolute amounts of energy and protein. Moreover, the limitations of the data, as discussed above, are not sufficiently serious to affect the overall conclusions drawn from them. Like other studies, our data show that both the rate and composition of weight gain are closely related to the absolute intakes of both protein and energy---a greater intake of either led to a greater rate of weight gain. In infants fed a greater protein intake, the greater rate of weight gain was associated with a greater rate of protein deposition (groups B and C vs group A); in those fed a greater energy intake, the greater rate of weight gain was associated with a greater rate of fat deposition (group C vs group B). Our data also suggest that the proportion of dietary protein and energy affects the composition of weight gain. For example, although infants fed formulas A and B, which differed 0nly in protein content, stored energy at comparable rates, infants fed the higher protein intake (formula B) retained more nitrogen and gained weight more rapidly. Moreover, the weight gained by infants fed the higher protein intake was leaner. In contrast, infants fed formulas B and C, which differed only in energy content, deposited similar amounts of protein, but both the rate of weight gain and the amount of fat stored were greater in infants fed the higher energy intake (formula C). Thus the weight gained by infants fed formula C was fatter. It is interesting that the relative composition of weight gained by this group (protein/fat 2.6:5.9, or 0.46) and group A (protein/fat 1.6:4.1, or 0.40), the intakes of which provided a relatively lower proportion of protein to energy than that of group B, was remarkably similar. These findings suggest that it is possible to produce different rates of weight gain of either similar (groups A
The Journal of Pediatrics May 1987
and C) or dissimilar (groups B and C) composition simply by regulating the composition of dietary intake. This, in turn, has important implications for designing nutritional regimens for LBW infants. For example, the common practice of supplementing the intake of LBW infants with either carbohydrate or fat, in addition to increasing energy expenditure, will result in greater fat deposition, which may or may not be desirable. Should future goals for dietary management of the LBW infant include goals for the composition as well as the rate of weight gain, this finding must be considered. REFERENCES
1. American Academy of Pediatrics, Committee on Nutrition. Nutritional needs of low birth weight infants. Pediatrics 1985;75:976-86. 2. Reichman B, Chessex P, Putet G, et al. Diet, fat accretion and growth in premature infants. N Engl J Med 1981;305:14951500. 3. Whyte RK, Haslam R, Vlainic C, et al. Energy balance and nitrogen balance in growing low birth weight infants fed human milk or formula. Pediatr Res 1983;17:891-98. 4. Putet G, Senterre J, Rigo J, Salle B. Nutrient balance, energy utilization and composition of weight gain in very low birth weight infants fed pooled human milk or a preterm formula. J PEDIATR1984;105:79-85. 5. Brooke OG. Energy balance and metabolic rate in preterm infants fed with standard and high-energy formulas. Br J Nutr 1980; 40:13-23. 6. Reichman B, Chessex P, Verellen G, et al. Dietary composition and macronutrient storage. Pediatrics 1983;72:322-7. 7. Chessex P, Reichman B, Verellen G, et al. Influence of postnatal age, energy intake and weight gain on energy metabolism in the very low birth weight infant. J PEDIATR 1981;99:761-6. 8. Kashyap S, Forsyth M, Zucker C, et al. Effects of varying protein and energy intakes on growth and metabolic response in low birth weight infants. J PEDIATR1986;108:955-63. 9. Food and Agricultural Organization of the United Nations, Ad Hoc Committee. Energy and protein requirements. FAO Nutritional Series No. 7, Rome, 1973;1024. 10. Brooke OG, Wood C. Relation between fecal fat and energy in preterm infants. Arch Dis Child 1983;58:305-6. 11. Kleiber M. The fire of life: an introduction to animal energetics. Melbourne, Fla.: Krieger, 1974;104-30. 12. Lusk G. Basal metabolism'~standards.In: Scientific tables, 6th ed. Montreal: Documenta Geigy, 1961;628. 13. Schulze K, Kairam R, Stefanski M, et al. Continuous measurement of minute ventilation and gaseous metabolism of newborn infants. J Appl Physiol 1981;50:1098-1103. 14. Voyer M, Senterre J, Rigo J, et al. Human milk lactoengineering. Acta Paediatr Scand 1984;73:302-6. 15. Freymond D, Schutz Y, Decombaz J, et al. Energy balance, physical activity and thermogenic effect of feeding in premature infants. Pediatr Res 1986;20:638-45. 16. Catzeflis C, Schutz Y, Micheli JL, et al. Whole body protein synthesis and energy expenditure in very low birth weight infants. Pediatr Res 1985;19:679-87. 17. Shelley HJ. Carbohydrate reserves in the newborn infant. Br Med J 1964;1:273-5. 18. Widdowson EM. Changes in body proportion and composi-
Volume 110 Number 5
19. 20.
21.
22. 23.
24.
Protein/energy intake re weight gain
tion during growth. In: Davis JA, Dobbing J, eds. Scientific foundation of pediatrics. Philadelphia: Saunders, 1974;15364. Ebeling M. The Dumas method for nitrogen in feeds. J Assoc Off Anal Chem 1968;51:766-70. Kurz G, Wallenfels K. Lactose and other/3-galactosides. In: Bergmeyer HNU, ed. Methods of enzymatic analysis, vol 3. Orlando, Fla.: Academic Press, 1974;1180-4. Jeejeebhoy KN, Ahmad S, Kozak G. Determination of fecal fats containing both medium and long chain triglycerides and fatty acids. Clin Biochem 1970;3:157-63. Miller RG Jr. Simultaneous statistical inference. New York: McGraw-Hill, 1966. Kashyap S, Schulze KF, Ramakrishman R, Dell RB, Heird WC. Effects of protein and energy intakes on growth [Letter]. J PED1ATR 1987;110:815. Flatt JP. Energetics of intermediary metabolism. In: Kinney JM, ed. Energy metabolism in health and disease. Columbus, Ohio: Ross Laboratories, 1980;77-87.
75 9
25. Mestyan J, Jarai I, Fekete M. The total energy expenditure and its components in premature infants maintained under different nursing and environmental conditions. Pediatr Res 1968;2:161-71. 26. Hommes FA, Drost YM, Geraets WXM, Reijenga MAA. The energy requirement for growth: an application of Atkinson's metabolic price system. Pediatr Res 1975;9:51-5. 27. Van Aerde J, Sauer P, Heim T, et al. Growth, macronutrient oxidation and accretion in very low birth weight (VLBW) infants with variable energy intake and constant diet composition. Pediatr Res 1985;19:368A. 28. Van Es AJH. The energetics of fat deposition during growth. Nutr Metab 1977;21:88-104. 29. Pullar JD, Webster AJF. The energy cost of fat and protein deposition in the rat. Br J Nutr 1977;37:355-63. 30. Harvey GR, Tobin GL. Luxuskonsumption, diet induced thermogenesis and brown fat: a critical review. Clin Sci 1983;64:7-18.
FELLOWSHIPS Available fellowships in pediatric subspecialties and those for general academic pediatric training are listed once a year, in May, in The JournM of Pediatrics. Each October, forms for listing such fellowships are sent to the Chairman of the Department of Pediatrics at most major hospitals in the United States and Canada. Should you desire to list fellowships, a separate application must be made each year for each position. All applications must be returned to The C. V. Mosby Company by February 15 of the listing year to ensure publication. Additional forms will be supplied on request from the Journal Editing Department, The C. V. Mosby Company, 11830 Westline Industrial Drive, St. Louis, M O 63146/314-872-8370.