Nutritional: Needs of Low-Birth-Weight Infants
Nutritional: Needs of Low-Birth-Weight Infants
Nutritional: Needs of Low-Birth-Weight Infants
Committee on Nutrition
The goal of feeding regimens for low-birth- In practice, caloric intakes of 110 to 150
weight infants is to obtain a prompt postnatal kcal/kg/day enable most low-birth-weight in-
resumption of growth to a rate approximating fants to achieve satisfactory rates of growth. If
intrauterine growth because this is believed to infants fail to gain satisfactorily, a higher caloric
provide the best possible conditions for subse- intake may be offered.
quent normal development. This statement
reviews current opinion and practices as well as Caloric Density of the Formula-Water
earlier reviews1-5 of the feeding of the low-birth- Requirement
weight infant. Although human milk or formulas that provide
67 kcal/dl (20 kcal/oz) are recommended for
Caloric Requirement term infants, more concentrated formulas are
The basal metabolic rate of low-birth-weight often used for low-birth-weight infants to facili-
infants is lower than that of full-term infants tate increased caloric intakes in infants with
during the first week of life, but it reaches and limited gastric capacity. Several studies have
exceeds that of the full-term infant by the second shown that feeding low-birth-weight infants
week. Daily caloric requirements reach 50 to 100 formulas with higher caloric densities results in
kcal/kg by the end of the first week of life and faster rates of growth.8-13 Some nurseries now feed
usually increase to 110 to 150 kcal/kg in subse- formulas of 81 kcal/dl (24 kcal/oz) and in some
quent active growth. instances 91 kcal/dl (27 kcal/oz). The 81-kcal/dl
A partition of the daily minimum energy re- concentration supplies most of the water required
quirements is shown in Table 1.6 by the infant (150 ml/kg)14 and provides 120
There are considerable variations from these kcal/kg.
average values, depending on both biological and The increased protein and mineral levels in
environmental factors. Infants who are small for these more concentrated formulas increase the
gestational age tend to have a higher basal renal solute load. With the limited capability of
metabolic.rate than do premature infants of the the immature kidney for concentrating urine,
same weight.7 The degree of physical activity sufficient water may not be supplied if the
appears to be a characteristic of the individual formula is too concentrated. Infants consuming
infant. Environmental factors may have a greater less than a normal volume of formula are particu-
influence than biological variation in determining larly vulnerable because, under constant condi-
the total caloric requirements. The maximal tions of extrarenal water loss, the lower the
response to cold stress can increase the resting formula intake the greater the proportion of
rate of heat production up to 21/2 times.6 Calories water required for renal excretion.'5 Infants
expended for specific dynamic action and for whose water balance is threatened (e.g., infants
fecal losses are dependent on the composition of exposed to heat, phototherapy, or cold stress, and
the milk or formula fed, as well as on individual those with infection or diarrhea) should have
variations in absorption of nutrients, particularly formulas of low renal solute load and should not
fat. be fed formulas of caloric density greater than 81
Downloaded from www.aappublications.org/news at Indonesia:AAP Sponsored on October 5, 2019
PEDIATRICS Vol. 60 No. 4 October 1977 519
TABLE I IV fluid is discontinued or hyperglycemia and
ESTIMATED REQUIREMENTS FOR CALORIES IN A TYPICAL, hyperosmolality, which may be difficult to
GROWING PREMATURE INFANT' control.4 Serum glucose levels should be regularly
monitored and glucose infusion rates lowered to
Item kcal/kg/Day 0.4 g/kg/hr, or less if the serum glucose level
Resting caloric expenditure 50 exceeds 125 mg/dl.4 The first feeding should be
Intermittent activity 15 distilled water to avoid excessive damage to the
Occasional cold stress 10 lungs if vomiting occurs.
Specific dynamic action 8
Fecal loss of calories 12 Protein Requirement
Growth allowance 25
Total 120 The optimal protein intake for the low-birth-
weight infant has not been precisely defined;
'Data from Sinclair et al.6 however, it is between 2.25 and 5 g/kg/day for
cow's milk formulas. Human milk contains about
1.1 g of protein or less per deciliter, or 1.65 g/100
kcal.31 When fed at intakes of 120 kcal/kg/day,
keal/dl (24 kcal/oz).15 Preterm infants excrete human milk supplies almost 2 g of protein per
sodium well,16 17 and late metabolic acidosis seen kilogram per day. The feeding of human milk to
in low-birth-weight infants may be related to low premature infants was the preferred practice
mineral intake.16-19 Lactic acid-containing formu- until 25 years ago when Gordon et al.32 demon-
las should not be fed because they may produce strated that premature infants gained more
acidosis.20 weight and retained more nitrogen when fed
cow's milk formulas of higher protein content.
Alternate Feeding Procedures Subsequent reports have confirmed this, but in
When conventional feedings every few hours some reports the increased weight gains with the
do not result in the attainment of an adequate higher protein cow's milk formulas were
nutrient intake, alternate methods of feeding such attributed in part to increased electrolyte intake
as continuous nasogastric drip,21 nasojejunal feed- and subsequent water retention.33-37 However,
ing,22 23 intravenous (IV) administration of Babson and Bramhall38 found no increase in
nutrients supplemented by oral feeding,24 and weight gain when only minerals were added to
total IV alimentation25 27 may be tried. However, formula providing 1.8 g of protein per 100 kcal.
the hazards and complexities of IV administration In studies designed to determine the protein
preclude its use in routine practice.28 Parenteral requirement of low-birth-weight infants, protein
administration of (1) 20% glucose and 2.5% amino has been given at levels ranging from 1.7 to 9
acid solution; (2) 12% glucose with 2.5% amino g/kg/day. The feedings have consisted of human
acids and 10% soybean oil emulsion; and (3) 12% milk and cow's milk formulas, with the protein
glucose, 2.5% amino acids, and 1% alcohol in a content varied by dilution with carbohydrate or
volume of 125 to 150 ml/kg/day all provided by the addition of casein or deionized milk or
positive nitrogen balance.29 Glucagon levels were whey. Because of the many variables in the
lower and growth hormone levels higher in formulas, including types and levels of fat and
infants given the fat-free mixtures. Parenteral carbohydrate and levels of vitamins and minerals,
feedings appear to increase water retention.30 it is difficult to assess the nutritional adequacy of
In a controlled study of the feeding of low- the various formulas used in the studies or to
birth-weight infants by continuous nasogastric attribute the findings solely to dietary protein
drip,21 satisfactory growth and clinical progress level.
were reported with the feeding of human milk Infants fed 1.7 to 2.25 g of protein per kilogram
and a simulated human milk formula (67 kcal/dl). per day either from human milk or a cow's milk
Feeding was started at the fourth hour of life at formula did not increase in weight36-39 or
the rate of 60 ml/kg/day and increased to 300 length38-39 as rapidly as those fed higher intakes,
ml/kg/day (200 kcal/kg/day) by the ninth day. In and some developed low levels of serum
practice, the latter intake is difficult to achieve. proteins.
Although early administration of fluids is The feeding of relatively high levels of protein
generally considered beneficial to prevent dehy- (6 to 9 g/kg/day) was associated with hyperpy-
dration, excessive weight loss, hypoglycemia, and rexia and lethargy,40 high BUN levels,36 diar-
excessive jaundice,4 5 use of 10% glucose parenter- rhea,39 high urinary excretion of phenols,39 clin-
ally may cause reactive hypoglycemia when the ical edema,42 late metabolic acidosis, and
Downloaded from www.aappublications.org/news at Indonesia:AAP Sponsored on October 5, 2019
520 NUTRITION IN LOW-BIRTH-WEIGHT INFANTS
increased mortality.39 The weight gains obtained breast milk had higher taurine levels than the
with the feeding of the high intakes of protein did other formulas. This suggestive study requires
not exceed those obtained by the feeding of confirmation.
moderate levels.36'39.40 Elevated plasma amino A review of the literature led Cox and Filer53 to
acid levels in low-birth-weight infants fed high- conclude that, with an adequate caloric intake,
protein formulas suggest that the high protein most low-birth-weight infants will grow satisfac-
intake may present an amino acid load that torily on cow's milk formulas supplying 2.25 to
exceeds the metabolizing capability of the imma- 5.0 g/kg/day of cow's milk protein. Fomon and
ture enzyme systems. Elevated levels of plasma co-workers estimated from hypothetical consider-
tyrosine and phenylalanine are not uncommon, a ations that the premature infant requires 3.0
finding related to late maturing of p-hydroxy- g/kg/day or 2.54 g of protein per 100 kcal,
phenylpyruvic oxidase.43 44 High plasma levels of assuming an intake of 120 kcal/kg/day.54
proline and methionine are also associated with If further studies confirm these findings,
high protein intake.45 consideration of protein quality along the lines
The amino acid composition of formulas for discussed here may be important in defining the
premature infants deserves special attention. optimal protein quantity for low-birth-weight
Low-birth-weight infants require some amino infants. With adequate intakes, human milk may
acids that are not essential for the term infant. In be the superior feeding for low-birth-weight
the balance studies of Snyderman,46 the removal infants.
of either cystine or tyrosine from the diet resulted
in an impairment of growth and nitrogen reten- Fat
tion, and a depression of the level of that partic- The ability of low-birth-weight infants to
ular amino acid in the plasma. Infants requiring absorb fat, particularly saturated fat such as
cystine also failed to show an increase in plasma butterfat, is relatively poor.55-58 This limitation is
cystine level after a methionine load, a finding in associated with liver immaturity and decreased
accord with the lack of cystathionase in the livers bile salt synthesis,59 and it is found to a lesser
of fetuses and premature infants reported by extent in full-term infants during the first few
Sturman et al.47'48 The high levels of cystathionine weeks of life.606' When palmitic acid-a long-
in the plasma49 and urine50 of premature infants chain saturated fatty acid-is present in fat, its
fed high-protein formulas also suggest that absorption depends on its position in the triglyc-
conversion of methionine to cystine is not effi- eride molecule.6263 Early recommendations for
cient until some time after birth. the feeding of low-birth-weight infants included
Raiha et al.5152 fed low-birth-weight infants the feeding of low-fat formulas.64'65 However, the
five formulas, including pooled breast milk. The recognition that the vegetable oils were much
breast milk supplied approximately 1.7 g of better absorbed than butterfat and other satu-
protein per kilogram per day, two formulas rated fats55 led to use in formulas of vegetable oils,
supplied 2.25 g of protein per kilogram per day, or blends of vegetable oils and animal fats. These
and two formulas supplied 4.50 g/kg/day. One are absorbed well, as is human milk fat.66
formula at each protein level had a 60:40 ratio of Including medium chain triglycerides as part of
whey/casein proteins, and the other two had an the fat in the formula has been shown to improve
18:82 ratio of whey/casein proteins. All infants fat absorption in low-birth-weight infants.67'0
grew equally well when fed 117 kcal/kg/day; Medium-chain triglycerides have also been shown
statistically, the breast-fed group gained at a to increase weight gain70 and to enhance calcium
slightly lower rate. Significant differences in absorption69 and nitrogen retention.68
plasma amino acid and ammonia levels were Fat in human milk supplies a major proportion
noted. The lower ammonia, tyrosine, and phenyl- of the caloric content. Formulas with 40% to 50%
alanine levels were found in infants fed whey/ of calories from fat are recommended for the
casein of 60:40, and the highest levels were in feeding of low-birth-weight infants because
those fed the high-protein formula with casein formulas of a lower fat content may contain
predominant. Those fed the high-protein, casein- higher levels of protein which increase renal
predominant formula developed late metabolic solute load.
acidosis. Serum protein levels were lowest in To meet the normal infant's requirement for
infants fed breast milk. essential fatty acids, it is recommended71 that
A major difference between the formulas was infant feedings supply 3% of the total calories in
the higher content of cystine in the breast milk the form of linoleic acid or 300 mg of linoleic acid
and high-whey protein formula. In addition, the per 100 kcal. Proprietary infant formulas with
Downloaded from www.aappublications.org/news at Indonesia:AAP Sponsored on October 5, 2019
AMERICAN ACADEMY OF PEDIATRICS 521
their high content of unsaturated fat supply a mineral contents of the bodies of premature and
generous allowance of linoleic acid. term infants.80'81 Body composition data can
Carbohydrate
provide a rough estimate, at best, of the increase
in minerals that the low-birth-weight infant
The utilization of carbohydrate by the low- would have accrued had he remained in utero.
birth-weight infant differs slightly from that of Because infants with low stores may retain 50%
the full-term infant. Intestinal disaccharidases to 70% of the nutrients they are fed, the levels of
develop early in fetal life72; maltase and sucrase nutrients supplied by formula must be 1.3 to 2
reach mature values by the sixth to eighth month, times those required to meet their needs. It
and lactase reaches it at term. These data suggest appears that minimal levels in formula designed
that the low-birth-weight infant can adequately for full-term infants could probably satisfy re-
digest disaccharides, although there is some quirements of the low-birth-weight infant for
evidence that lactose digestion may not be fully sodium, potassium, chloride, and zinc; would
efficient for the first few days of life.73 Low-birth- probably be borderline in copper; and would
weight infants develop satisfactorily when fed probably be deficient in iron, calcium, and phos-
formulas in which the lactose of the milk has been phorus. Based on these calculations, some changes
augmented with sucrose and when fed lactose- in mineral composition might be made in
free formulas (i.e., formulas based on soy isolates, formulas intended for use by premature infants to
meat or protein hydrolysates and containing achieve a mineral retention equivalent to that in
sucrose and/or dextrose, maltose, and dextrins as utero.
the carbohydrate). Lactose, sucrose, and maltose
oral tolerance tests conducted on 2-week-old,
low-birth-weight infants previously fed formulas Calcium and Phosphorus
containing either lactose or sucrose as the sole Balance studies and roentgenographic findings
carbohydrate revealed no significant differences in low-birth-weight infants suggest that formulas
in the utilization of these three disaccharides, made from cow's milk with higher calcium and
despite the presence or absence of the substrate phosphorus levels provide greater retention of
sugar in the diet for the two weeks preceding the calcium and phosphorus and increased minerali-
test.74 In another study of dietary sugars, infants zation of the skeleton than does human
grew equally well on soy-isolate formulas milk.64'82-84 Although earlier studies85'86 suggested
containing sucrose or dextrose,75 and a slightly that the low phosphorus content of human milk
lesser rate with lactose. was the limiting factor in skeleton mineralization,
Lactose, as the natural sugar of human milk, work by Day et al.87 suggests that the underminer-
has been the usual choice for addition to a cow's alization of bone observed in premature infants
milk formula to increase the carbohydrate fed human milk may also be caused by an
content up to that of human milk. However, a inadequate calcium intake. In the Day et al.87
recent study with cow's milk formulas found that study of low-birth-weight infants (less than 1,300
the addition of sucrose rather than lactose to the g), infants fed a proprietary formula supple-
milk base resulted in a lower incidence of diar- mented with calcium lactate (total calcium, 154
rhea and metabolic acidosis,76 which appears to mg/ 100 kcal) showed a better-defined bone
support the findings of Boellner et al.73 Thus, the texture and wider cortices than infants fed unsup-
slight delay in the maturation of intestinal lactase plemented formula containing 63 mg of calcium
may be of physiological consequence in some per 100 kcal. Fomon et al.54 have calculated that
infants. Usually lactose enhances calcium absorp- premature infants require 132 mg of calcium per
tion in the small intestine77 and promotes a 100 kcal.
fermentative, less putrefactive bacterial flora78'79 Infant formulas fed to low-birth-weight infants
and reduces the incidence of constipation. in the United States (Table II) contain higher
levels of calcium and phosphorus than those
Minerals
supplied by human milk, but they are generally
lower than the level used by Day et al.87 or that
Two thirds of the mineral content of the body recommended by Fomon et al.54 In clinical
of the full-term infant is deposited during the last studies in which low-birth-weight infants were
two months of gestation. The amounts of minerals fed current proprietary formulas88'89 or earlier
the preterm infant must retain from the diet to formulas of comparable calcium and phosphorus
achieve the mineral composition of the full-term content,36'90 no apparent abnormalities of cal-
infant might be estimated from the differences in cium/phosphorus metabolism were noted, offer-
Downloaded from www.aappublications.org/news at Indonesia:AAP Sponsored on October 5, 2019
Updated Information & including high resolution figures, can be found at:
Services http://pediatrics.aappublications.org/content/60/4/519
Permissions & Licensing Information about reproducing this article in parts (figures, tables) or in its
entirety can be found online at:
http://www.aappublications.org/site/misc/Permissions.xhtml
Reprints Information about ordering reprints can be found online:
http://www.aappublications.org/site/misc/reprints.xhtml
The online version of this article, along with updated information and services, is located on
the World Wide Web at:
http://pediatrics.aappublications.org/content/60/4/519
Pediatrics is the official journal of the American Academy of Pediatrics. A monthly publication, it has
been published continuously since 1948. Pediatrics is owned, published, and trademarked by the
American Academy of Pediatrics, 141 Northwest Point Boulevard, Elk Grove Village, Illinois, 60007.
Copyright © 1977 by the American Academy of Pediatrics. All rights reserved. Print ISSN: 1073-0397.