nutritional needs of low-birth-weight...

AMERICAN ACADEMY OF PEDIATRICS

Committee on Nutrition

Nutritional Needs of Low-Birth-Weight Infants

The goal of feeding regimens for low-birth-weight infants is to obtain a prompt postnatalresumption of growth to a rate approximatingintrauterine growth because this is believed toprovide the best possible conditions for subse-quent normal development. This statementreviews current opinion and practices as well asearlier reviews1-5 of the feeding of the low-birth-weight infant.

Caloric RequirementThe basal metabolic rate of low-birth-weight

infants is lower than that of full-term infantsduring the first week of life, but it reaches andexceeds that of the full-term infant by the secondweek. Daily caloric requirements reach 50 to 100kcal/kg by the end of the first week of life andusually increase to 110 to 150 kcal/kg in subse-quent active growth.A partition of the daily minimum energy re-

quirements is shown in Table 1.6There are considerable variations from these

average values, depending on both biological andenvironmental factors. Infants who are small forgestational age tend to have a higher basalmetabolic.rate than do premature infants of thesame weight.7 The degree of physical activityappears to be a characteristic of the individualinfant. Environmental factors may have a greaterinfluence than biological variation in determiningthe total caloric requirements. The maximalresponse to cold stress can increase the restingrate of heat production up to 21/2 times.6 Caloriesexpended for specific dynamic action and forfecal losses are dependent on the composition ofthe milk or formula fed, as well as on individualvariations in absorption of nutrients, particularlyfat.

In practice, caloric intakes of 110 to 150kcal/kg/day enable most low-birth-weight in-fants to achieve satisfactory rates of growth. Ifinfants fail to gain satisfactorily, a higher caloricintake may be offered.

Caloric Density of the Formula-WaterRequirement

Although human milk or formulas that provide67 kcal/dl (20 kcal/oz) are recommended forterm infants, more concentrated formulas areoften used for low-birth-weight infants to facili-tate increased caloric intakes in infants withlimited gastric capacity. Several studies haveshown that feeding low-birth-weight infantsformulas with higher caloric densities results infaster rates of growth.8-13 Some nurseries now feedformulas of 81 kcal/dl (24 kcal/oz) and in someinstances 91 kcal/dl (27 kcal/oz). The 81-kcal/dlconcentration supplies most of the water requiredby the infant (150 ml/kg)14 and provides 120kcal/kg.The increased protein and mineral levels in

these more concentrated formulas increase therenal solute load. With the limited capability ofthe immature kidney for concentrating urine,sufficient water may not be supplied if theformula is too concentrated. Infants consumingless than a normal volume of formula are particu-larly vulnerable because, under constant condi-tions of extrarenal water loss, the lower theformula intake the greater the proportion ofwater required for renal excretion.'5 Infantswhose water balance is threatened (e.g., infantsexposed to heat, phototherapy, or cold stress, andthose with infection or diarrhea) should haveformulas of low renal solute load and should notbe fed formulas of caloric density greater than 81

PEDIATRICS Vol. 60 No. 4 October 1977 519 by guest on June 3, 2018www.aappublications.org/newsDownloaded from

TABLE I

ESTIMATED REQUIREMENTS FOR CALORIES IN A TYPICAL,GROWING PREMATURE INFANT'

Item kcal/kg/Day

Resting caloric expenditure 50Intermittent activity 15Occasional cold stress 10Specific dynamic action 8Fecal loss of calories 12Growth allowance 25Total 120

'Data from Sinclair et al.6

keal/dl (24 kcal/oz).15 Preterm infants excretesodium well,16 17 and late metabolic acidosis seenin low-birth-weight infants may be related to lowmineral intake.16-19 Lactic acid-containing formu-las should not be fed because they may produceacidosis.20

Alternate Feeding Procedures

When conventional feedings every few hoursdo not result in the attainment of an adequatenutrient intake, alternate methods of feeding suchas continuous nasogastric drip,21 nasojejunal feed-ing,22 23 intravenous (IV) administration ofnutrients supplemented by oral feeding,24 andtotal IV alimentation25 27 may be tried. However,the hazards and complexities of IV administrationpreclude its use in routine practice.28 Parenteraladministration of (1) 20% glucose and 2.5% aminoacid solution; (2) 12% glucose with 2.5% aminoacids and 10% soybean oil emulsion; and (3) 12%glucose, 2.5% amino acids, and 1% alcohol in avolume of 125 to 150 ml/kg/day all providedpositive nitrogen balance.29 Glucagon levels werelower and growth hormone levels higher ininfants given the fat-free mixtures. Parenteralfeedings appear to increase water retention.30

In a controlled study of the feeding of low-birth-weight infants by continuous nasogastricdrip,21 satisfactory growth and clinical progresswere reported with the feeding of human milkand a simulated human milk formula (67 kcal/dl).Feeding was started at the fourth hour of life atthe rate of 60 ml/kg/day and increased to 300ml/kg/day (200 kcal/kg/day) by the ninth day. Inpractice, the latter intake is difficult to achieve.

Although early administration of fluids isgenerally considered beneficial to prevent dehy-dration, excessive weight loss, hypoglycemia, andexcessive jaundice,4 5 use of 10% glucose parenter-ally may cause reactive hypoglycemia when the

IV fluid is discontinued or hyperglycemia andhyperosmolality, which may be difficult tocontrol.4 Serum glucose levels should be regularlymonitored and glucose infusion rates lowered to0.4 g/kg/hr, or less if the serum glucose levelexceeds 125 mg/dl.4 The first feeding should bedistilled water to avoid excessive damage to thelungs if vomiting occurs.

Protein RequirementThe optimal protein intake for the low-birth-

weight infant has not been precisely defined;however, it is between 2.25 and 5 g/kg/day forcow's milk formulas. Human milk contains about1.1 g of protein or less per deciliter, or 1.65 g/100kcal.31 When fed at intakes of 120 kcal/kg/day,human milk supplies almost 2 g of protein perkilogram per day. The feeding of human milk topremature infants was the preferred practiceuntil 25 years ago when Gordon et al.32 demon-strated that premature infants gained moreweight and retained more nitrogen when fedcow's milk formulas of higher protein content.Subsequent reports have confirmed this, but insome reports the increased weight gains with thehigher protein cow's milk formulas wereattributed in part to increased electrolyte intakeand subsequent water retention.33-37 However,Babson and Bramhall38 found no increase inweight gain when only minerals were added toformula providing 1.8 g of protein per 100 kcal.

In studies designed to determine the proteinrequirement of low-birth-weight infants, proteinhas been given at levels ranging from 1.7 to 9g/kg/day. The feedings have consisted of humanmilk and cow's milk formulas, with the proteincontent varied by dilution with carbohydrate orby the addition of casein or deionized milk orwhey. Because of the many variables in theformulas, including types and levels of fat andcarbohydrate and levels of vitamins and minerals,it is difficult to assess the nutritional adequacy ofthe various formulas used in the studies or toattribute the findings solely to dietary proteinlevel.

Infants fed 1.7 to 2.25 g of protein per kilogramper day either from human milk or a cow's milkformula did not increase in weight36-39 orlength38-39 as rapidly as those fed higher intakes,and some developed low levels of serumproteins.The feeding of relatively high levels of protein

(6 to 9 g/kg/day) was associated with hyperpy-rexia and lethargy,40 high BUN levels,36 diar-rhea,39 high urinary excretion of phenols,39 clin-ical edema,42 late metabolic acidosis, and

520 NUTRITION IN LOW-BIRTH-WEIGHT INFANTS by guest on June 3, 2018www.aappublications.org/newsDownloaded from

increased mortality.39 The weight gains obtainedwith the feeding of the high intakes of protein didnot exceed those obtained by the feeding ofmoderate levels.36'39.40 Elevated plasma aminoacid levels in low-birth-weight infants fed high-protein formulas suggest that the high proteinintake may present an amino acid load thatexceeds the metabolizing capability of the imma-ture enzyme systems. Elevated levels of plasmatyrosine and phenylalanine are not uncommon, afinding related to late maturing of p-hydroxy-phenylpyruvic oxidase.43 44 High plasma levels ofproline and methionine are also associated withhigh protein intake.45The amino acid composition of formulas for

premature infants deserves special attention.Low-birth-weight infants require some aminoacids that are not essential for the term infant. Inthe balance studies of Snyderman,46 the removalof either cystine or tyrosine from the diet resultedin an impairment of growth and nitrogen reten-tion, and a depression of the level of that partic-ular amino acid in the plasma. Infants requiringcystine also failed to show an increase in plasmacystine level after a methionine load, a finding inaccord with the lack of cystathionase in the liversof fetuses and premature infants reported bySturman et al.47'48 The high levels of cystathioninein the plasma49 and urine50 of premature infantsfed high-protein formulas also suggest thatconversion of methionine to cystine is not effi-cient until some time after birth.

Raiha et al.5152 fed low-birth-weight infantsfive formulas, including pooled breast milk. Thebreast milk supplied approximately 1.7 g ofprotein per kilogram per day, two formulassupplied 2.25 g of protein per kilogram per day,and two formulas supplied 4.50 g/kg/day. Oneformula at each protein level had a 60:40 ratio ofwhey/casein proteins, and the other two had an18:82 ratio of whey/casein proteins. All infantsgrew equally well when fed 117 kcal/kg/day;statistically, the breast-fed group gained at aslightly lower rate. Significant differences inplasma amino acid and ammonia levels werenoted. The lower ammonia, tyrosine, and phenyl-alanine levels were found in infants fed whey/casein of 60:40, and the highest levels were inthose fed the high-protein formula with caseinpredominant. Those fed the high-protein, casein-predominant formula developed late metabolicacidosis. Serum protein levels were lowest ininfants fed breast milk.A major difference between the formulas was

the higher content of cystine in the breast milkand high-whey protein formula. In addition, the

breast milk had higher taurine levels than theother formulas. This suggestive study requiresconfirmation.A review of the literature led Cox and Filer53 to

conclude that, with an adequate caloric intake,most low-birth-weight infants will grow satisfac-torily on cow's milk formulas supplying 2.25 to5.0 g/kg/day of cow's milk protein. Fomon andco-workers estimated from hypothetical consider-ations that the premature infant requires 3.0g/kg/day or 2.54 g of protein per 100 kcal,assuming an intake of 120 kcal/kg/day.54

If further studies confirm these findings,consideration of protein quality along the linesdiscussed here may be important in defining theoptimal protein quantity for low-birth-weightinfants. With adequate intakes, human milk maybe the superior feeding for low-birth-weightinfants.

FatThe ability of low-birth-weight infants to

absorb fat, particularly saturated fat such asbutterfat, is relatively poor.55-58 This limitation isassociated with liver immaturity and decreasedbile salt synthesis,59 and it is found to a lesserextent in full-term infants during the first fewweeks of life.606' When palmitic acid-a long-chain saturated fatty acid-is present in fat, itsabsorption depends on its position in the triglyc-eride molecule.6263 Early recommendations forthe feeding of low-birth-weight infants includedthe feeding of low-fat formulas.64'65 However, therecognition that the vegetable oils were muchbetter absorbed than butterfat and other satu-rated fats55 led to use in formulas of vegetable oils,or blends of vegetable oils and animal fats. Theseare absorbed well, as is human milk fat.66

Including medium chain triglycerides as part ofthe fat in the formula has been shown to improvefat absorption in low-birth-weight infants.67'0Medium-chain triglycerides have also been shownto increase weight gain70 and to enhance calciumabsorption69 and nitrogen retention.68

Fat in human milk supplies a major proportionof the caloric content. Formulas with 40% to 50%of calories from fat are recommended for thefeeding of low-birth-weight infants becauseformulas of a lower fat content may containhigher levels of protein which increase renalsolute load.To meet the normal infant's requirement for

essential fatty acids, it is recommended71 thatinfant feedings supply 3% of the total calories inthe form of linoleic acid or 300 mg of linoleic acidper 100 kcal. Proprietary infant formulas with

AMERICAN ACADEMY OF PEDIATRICS 521 by guest on June 3, 2018www.aappublications.org/newsDownloaded from

their high content of unsaturated fat supply agenerous allowance of linoleic acid.

CarbohydrateThe utilization of carbohydrate by the low-

birth-weight infant differs slightly from that ofthe full-term infant. Intestinal disaccharidasesdevelop early in fetal life72; maltase and sucrasereach mature values by the sixth to eighth month,and lactase reaches it at term. These data suggestthat the low-birth-weight infant can adequatelydigest disaccharides, although there is someevidence that lactose digestion may not be fullyefficient for the first few days of life.73 Low-birth-weight infants develop satisfactorily when fedformulas in which the lactose of the milk has beenaugmented with sucrose and when fed lactose-free formulas (i.e., formulas based on soy isolates,meat or protein hydrolysates and containingsucrose and/or dextrose, maltose, and dextrins asthe carbohydrate). Lactose, sucrose, and maltoseoral tolerance tests conducted on 2-week-old,low-birth-weight infants previously fed formulascontaining either lactose or sucrose as the solecarbohydrate revealed no significant differencesin the utilization of these three disaccharides,despite the presence or absence of the substratesugar in the diet for the two weeks preceding thetest.74 In another study of dietary sugars, infantsgrew equally well on soy-isolate formulascontaining sucrose or dextrose,75 and a slightlylesser rate with lactose.

Lactose, as the natural sugar of human milk,has been the usual choice for addition to a cow'smilk formula to increase the carbohydratecontent up to that of human milk. However, arecent study with cow's milk formulas found thatthe addition of sucrose rather than lactose to themilk base resulted in a lower incidence of diar-rhea and metabolic acidosis,76 which appears tosupport the findings of Boellner et al.73 Thus, theslight delay in the maturation of intestinal lactasemay be of physiological consequence in someinfants. Usually lactose enhances calcium absorp-tion in the small intestine77 and promotes afermentative, less putrefactive bacterial flora78'79and reduces the incidence of constipation.

Minerals

Two thirds of the mineral content of the bodyof the full-term infant is deposited during the lasttwo months of gestation. The amounts of mineralsthe preterm infant must retain from the diet toachieve the mineral composition of the full-terminfant might be estimated from the differences in

mineral contents of the bodies of premature andterm infants.80'81 Body composition data canprovide a rough estimate, at best, of the increasein minerals that the low-birth-weight infantwould have accrued had he remained in utero.

Because infants with low stores may retain 50%to 70% of the nutrients they are fed, the levels ofnutrients supplied by formula must be 1.3 to 2times those required to meet their needs. Itappears that minimal levels in formula designedfor full-term infants could probably satisfy re-quirements of the low-birth-weight infant forsodium, potassium, chloride, and zinc; wouldprobably be borderline in copper; and wouldprobably be deficient in iron, calcium, and phos-phorus. Based on these calculations, some changesin mineral composition might be made informulas intended for use by premature infants toachieve a mineral retention equivalent to that inutero.

Calcium and PhosphorusBalance studies and roentgenographic findings

in low-birth-weight infants suggest that formulasmade from cow's milk with higher calcium andphosphorus levels provide greater retention ofcalcium and phosphorus and increased minerali-zation of the skeleton than does humanmilk.64'82-84 Although earlier studies85'86 suggestedthat the low phosphorus content of human milkwas the limiting factor in skeleton mineralization,work by Day et al.87 suggests that the underminer-alization of bone observed in premature infantsfed human milk may also be caused by aninadequate calcium intake. In the Day et al.87study of low-birth-weight infants (less than 1,300g), infants fed a proprietary formula supple-mented with calcium lactate (total calcium, 154mg/ 100 kcal) showed a better-defined bonetexture and wider cortices than infants fed unsup-plemented formula containing 63 mg of calciumper 100 kcal. Fomon et al.54 have calculated thatpremature infants require 132 mg of calcium per100 kcal.

Infant formulas fed to low-birth-weight infantsin the United States (Table II) contain higherlevels of calcium and phosphorus than thosesupplied by human milk, but they are generallylower than the level used by Day et al.87 or thatrecommended by Fomon et al.54 In clinicalstudies in which low-birth-weight infants werefed current proprietary formulas88'89 or earlierformulas of comparable calcium and phosphoruscontent,36'90 no apparent abnormalities of cal-cium/phosphorus metabolism were noted, offer-


TABLE II

NUTRIENT COMPOSITION OF HUMAN MILK AND PROPRIETARY INFANT FORMULAS AND RECOMMENDED LEVELS FOR FULL-TERMAND LOW-BIRTH-WEIGHT INFANTS'

Nutrient Minimum Human Enfamil PM60/40 Premature Similac SMA SimilacLevel Milk60 Formula (13, 24,Recoin- or 27 Calo-mendedt riesloz)

Protein, gFat, gCarbohydrate, gAsh, mgVitamin A, IUVitamin D, IUVitamin E, IUVitamin K, jgVitamin C, mgThiamin, jigRiboflavin, ,jgNiacin, ,ugVitamin B6, LgFolic acid, ugPantothenic acid, ,ugVitamin B12, lAgBiotin, ,tgInositol, mgCholine, mgCalcium, mgPhosphorus, mgMagnesium, mgIron, mgIodine, ,ugCopper, ugZinc, mgManganese, ugSodiummgmEq

1.8t3.3§

250400.3(0.7)1148

4060

250

35¶4

3000.151.547

5025#60.15005

600.55

1.3-1.6510.3

300250

30.327.8

2560

250154

3000.151.0

2013502560.14-9

600.51.5

2.35.5

10.3530250631.998.1

7894

1,2506316

4700.32.557

807070.2tt

10100

0.65160

2.35.2

11.1320370602.248.1

96147

1,074497.3

4400.221.77.5

18.8603060.46

620.595

2.85.1

11.5680250631.998.0

7894

1,2506316

4700.32.567

15678100.28

1000.65

160

2.35.310.6

530370602.2148.1

96147

1,030607.3

4400.221.55.01575576

Tracett15600.745

2.35.3

10.7370390631.498.6

105156780638

3100.163.05.513664981.9

10700.55

23

2.75.3

10.3580370602.248.1

96147

1,000607.3

4400.221.5

25102796

Tracett15600.745Manganese, ug 5 1.5 160 5 160 5 23 5

Sodiummg 20# 24 42 23 40 32 24 38mEq 0.9# 1.0 1.8 1.0 1.7 1.6 1.0 1.7

Potassiummg 80 81 102 85 110 103 83 126mEq 2.1 2.1 2.6 2.2 2.8 2.5 2.1 3.2

Chloridemg 55 55 80 66 85 79 55 94mEq 1.6 1.6 2.3 2.0 2.4 2.3 1.6 2.7

Renal solute load01 ... 11.3 16 14.3 18.1 16.2 13.6 18.4mOsm

°Per 100 kcal.tCommittee on Nutrition recommendations7" for formula for full-term infants per 100 kcal.VProtein of a nutritional quality equivalent to casein.§Including a minimum of 300 mg of essential fatty acids.IlCommittee on Nutrition recommendation different for low-birth-weight infants; also 1.0 IU/g linoleic acid.Minimum of 15 ,ug of vitamin B6 per gram of protein.*Some evidence for higher requirement for low-birth-weight [email protected] mg in iron-fortified formula.ttl.5 mg in iron-fortified formula.t$Calculated by the method of Ziegler and Fomon.?5


ing some assurance that current ranges ofconcentrations are not grossly inadequate for thefeeding of low-birth-weight infants.The hypocalcemia-hyperphosphatemia syn-

drome seen in normal term neonates fed undi-luted cow's milk has been attributed to immaturehomeostatic control of serum phosphate.88 Thissuggests that formulas for low-birth-weightinfants should have a calcium/phosphorus ratioapproaching that of human milk (2.0:1), or at leastbetween 1.1:1 and 2.0:1, as recommended forformulas fed to full-term infants.7

Magnesium

The recommended minimum requirement formagnesium in formula for the term infant wasbased on the amount present in human milk, 6mg/100 kcal.71 Clinical experience suggests thatthis amount also suffices for the low-birth-weightinfant. Average serum magnesium levels of theneonatal, low-birth-weight infant are similar toadult values.89 Low levels are not uncommon inthe first few days of life, particularly in the small-for-gestational-age group, but they rise to adultvalues within days after feeding of conventionalformulas.89Magnesium depletion has been observed in

infants suffering from the gross malnutrition ofkwashiorkor,9' and hypomagnesemia, as is truewith hypocalcemia, may result from high phos-phate feedings.92 But, there have been no reportsof magnesium deficiency in healthy, low-birth-weight infants fed formulas of magnesium contentequal to or in moderate excess over that in humanmilk. Low-birth-weight infants have been main-tained for prolonged periods on IV feedingcontaining 2 mg of magnesium per 100 kcal.2f Ifthe low-birth-weight infant can absorb 33% of themagnesium in the formula-a reasonable as-sumption-the 6 mg/100 kcal required in theformula would readily supply the 2 mg/ 100 kcalgiven intravenously.

Iron

The low-birth-weight infant is especiallysusceptible to the development of iron deficiencyanemia because its stores of iron are much smallerthan those of a full-term infant, and they areinsufficient to last over a prolonged period whengrowth must be rapid. Erythrocyte and hemo-globin levels are high at birth; the hemoglobiniron released by destruction of old RBCs issalvaged and stored for future use. Active erythro-poiesis resumes between 1 and 2 months of age,and rapidly decreases the size of the iron reserve.

Without supplemental iron, the body stores ofiron will be depleted sometime after 2 months ofage rather than after 4 to 6 months of age, as inthe normal, full-term infant.93 Orally adminis-tered iron is well absorbed.94

Although the greater need for iron by the low-birth-weight infant has been interpreted to indi-cate that iron-fortified formulas be given as earlyas possible, recent findings show that iron-supple-mented formulas increase the susceptibility ofinfants to vitamin E deficiency and hemolyticanemia, especially when formulas are high inpolyunsaturated fatty acids.5 95 (See discussion ofvitamin E.) These studies leave unresolved thequestion of whether supplementary iron shouldbe started at 2 months of age or shortly afterbirth.93 They also suggest that formulas for low-birth-weight infants containing more than 1 mg ofiron per 100 kcal should contain moderateamounts of polyunsaturated fats (and amplevitamin E in an absorbable form), and that thosewith higher amounts of polyunsaturated fatsshould contain about 0.1 mg of iron per 100 kcal,'which is roughly equivalent to the iron present inbreast milk.

Another, more speculative reason for tempo-rarily delaying the use of iron-fortified formula inlow-birth-weight infants comes from recentevidence that two iron-binding proteins in humanmilk (lactoferrin and transferrin) lose their bacte-riostatic action when saturated with iron.93 Thebacteriostatic properties of these proteins may beespecially important to low-birth-weight infantsin the early weeks of life.

Thus, even though the Committee on Nutritioncontinues to recommend that low-birth-weightinfants receive 2 mg of iron per kilogram per daystarting at age 2 months or earlier, one cannotcategorically require that formulas provide thislevel of iron from birth. Thus, formulas for low-birth-weight infants may provide either 0.1 mg or1.5 mg of iron per 100 kcal. If they provide thehigher level of iron, they should contain amplevitamin E and a moderate level of polyunsatu-rated fatty acids.

CopperThe recommended level of copper in infant

formulas is 60 pgg/100 kcal.71 This level is based onearly data on human milk. Although copperdeficiency has not been noted in normal infantson customary feedings, several reports96-98 haveindicated that copper deficiency may develop insmall infants fed formulas not supplemented withcopper. Recent data suggest that an intake of 90


gtl100 kcal is desirable for low-birth-weightinfants.99

Iodine

The recommended minimum requirement ofnormal infants (5 ,ug of iodine per 100 keal)71 wasalso based on the iodine content of human milk.The uptake of radioactive iodine by the thyroidgland of premature infants has been found to be inthe normal range for children and adults100; there-fore, we can assume that 5 ,ug of iodine per 100kcal is also adequate for the low-birth-weightinfant.

Zinc and Manganese

The Committee on Nutrition has recentlyproposed that infant formulas for full-term infantssupply 0.5 mg of zinc and 5 ,ug of manganese per100 kcal.7' There is no basis for modifying theserecommended levels for low-birth-weight in-fants.

Other Trace Minerals

Although other minerals (such as cobalt,molybdenum, selenium, and chromium) areprobably essential in trace amounts for infants,there is no information on which to base recom-mendations at this time. Fluoride is usuallyprovided in supplements or as fluoridatedwater.

Sodium, Chloride, and PotassiumThe daily requirements of low-birth-weight

infants for sodium, chloride, and potassium canonly be roughly estimated from tissue composi-tion (Table II), because obligatory losses in theurine and feces and from the skin vary consider-ably.Minimum levels of sodium, chloride, and potas-

sium recommended by the Committee on Nutri-tion for new formula standards were based onlevels in human milk and should be sufficient forlow-birth-weight infants. There have been a fewreports'01-'03 of hyponatremia in low-birth-weightinfants fed a formula with a concentration ofsodium similar to that in human milk. Fomon etal.54 also suggest that the level of sodium inhuman milk is not sufficient for the prematureinfant; they recommend 30 mg of sodium per 100kcal.The Committee on Nutrition recommends that,

to prevent dehydration and disturbances of acid-base balances, minimum and maximum levels ofsodium, chloride, and potassium in formulas forlow-birth-weight infants be the same as those

recommended for full-term infants71 until furtherwork confirms a higher, suggested need.

VitaminsTable II shows the levels of vitamins recom-

mended by the Committee as minimum levels forinfant formulas for full-term infants.7' Theseapply to formulas based on milk and milk substi-tutes. The Committee recommends that the samelevels apply to formulas for low-birth-weightinfants, except for vitamin E.However, even with proprietary formulas

containing adequate levels of vitamins, prematureand low-birth-weight infants often consume muchless than 120 kcal/kg/day during the early weeksof life, and they may not receive sufficient vita-mins to prevent deficiency. For example, ricketshas been found in premature infants fed proprie-tary formulas containing 400 IU of vitamin D perliter,104 and. folate and vitamin B1, deficiencieshave also been noted'05-'08 when the infantsconsumed a small volume of formula. Therefore,the Committee recommends that low-birth-weight infants receive an intramuscular injectionof 1 to 2 mg of vitamin K at birth and a daily, oral,multivitamin supplement providing the recom-mended daily allowance of vitamins for infants asestablished by the Food and Drug Administra-tion.'09The vitamin E requirement of low-birth-weight

infants merits special consideration.1. Absorption of vitamin E by these infants is

poor; Gordon et al.1"0 showed that the levels ofvitamin E usually found in formulas-which areadequate to maintain normal serum tocopherollevels in term infants-were not sufficient to do soin premature infants. Recently, it was shown thatwater-soluble forms of vitamin E improve absorp-tion and result in higher' serum tocopherollevels."'

2. The requirement for vitamin E increases asthe level of polyunsaturated fats in the dietincreases.93 Thus, when formulas are high inpolyunsaturated fats, the infant needs morevitamin E.

3. When iron levels in the formula are high,the requirement for vitamin E by low-birth-weight infants also increases. For example, low-birth-weight infants receiving a formula supple-mented with iron (12 mg/liter) and high inpolyunsaturated fats have a greater incidence ofRBC hemolysis and lower serum tocopherol levelsthan infants receiving formula that has a low levelof iron and is low in polyunsaturated fats.94Therefore, the Committee on Nutrition recom-

AMERICAN ACADEMY OF PEDIATRICS 525

by guest on June 3, 2018www.aappublications.org/newsDownloaded from

mends that formula fed to premature infantsshould provide 0.7 IU of vitamin E per 100 kcaland at least 1.0 IU of vitamin E per gram oflinoleic acid. In addition, the multivitaminsupplement given to low-birth-weight infantsshould provide 5 IU of vitamin E, preferably inwater-soluble form.

Formula Compositions

Table II shows the composition of majorproprietary formulas available for feeding full-term and low-birth-weight infants, the averagecomposition of human milk, and the recentrecommendations of the Committee on Nutri-tion71 for proposed standards for infant formula.This information is presented in units per 100kcal-which relate specific nutrient needs tocaloric requirements-and is particularly useful indiscussing formulas for low-birth-weight infantsbecause it allows easy comparison of formulas ofdiffering caloric densities.

In many instances, formulas for term infants areused for feeding low-birth-weight infants, and, inother instances, special formulas are available forpremature and low-birth-weight infants.

Discussions in this statement suggest that levelsof some nutrients in formulas for low-birth-weightinfants should be somewhat higher than theminimum levels proposed by the Committee forfull-term infants; however, all of these recom-mendations can be met within the proposedstandards for infant formulas.7

ConclusionsThe optimal diet for the low-birth-weight

infant may be defined as one that supports a rateof growth approximating that of the thirdtrimester of intrauterine life, without imposingstress on the developing metabolic or excretorysystems. Cell division and growth of all tissues inthe infant should proceed at a rapid rate; unduedelay in the resumption of growth may haveserious and lasting consequences. The attainmentof an adequate caloric intake is the primaryrequirement, and this may be facilitated by thefeeding of formulas of caloric density greater thanthat of human milk. However, the feeding of thistype of formula requires special attention to avoidtoo high an osmolar load and to provide sufficientwater. The use of continuous intragastric orintrajejunal drip with formulas providing 67 or 81kcal/dl also may be a safe and practical means ofincreasing caloric intake. Caloric intakes of about120 kcal/kg/day in formula volumes of 150 to 200

ml/kg/day will support the desired weight gainin most infants.An appropriate requirement for protein equiv-

alent to casein for the low-birth-weight infantwould appear to fall in the range of 2.5 to 5.0g/kg/day, or 2.25 to 4.5 g/100 kcal. A moreprecise statement of optimal protein quantityawaits the definition of optimal protein qualityfor low-birth-weight infants. Evidence has beenaccumulating that some amino acids considerednonessential for the normal infant are indispens-able to low-birth-weight infants. Thus, theapparent normal growth of infants fed breast milksupplying 1.7 g of protein per kilogram of bodyweight may be caused in part by its distribution ofamino acids. For the larger low-birth-weightinfant, recommendations similar to those for theterm infant, including the desirability of breastfeeding, apply.71

Fat mixtures in formulas currently in useinclude unsaturated vegetable oils and/or me-dium-chain triglycerides, which are well ab-sorbed. Although good absorption of fat is impor-tant-not only for energy requirements but also toenhance the absorption of fat-soluble vitaminsand certain minerals-other aspects must beconsidered in the selection of ideal formula fatcompositions for low-birth-weight infants. Thefatty acid composition of the diet influences thecomposition of body lipids, especially in low-birth-weight infants who have meager stores ofbody fat. Fat mixtures should not be too saturatedor too unsaturated.The occurrence of hemolytic anemia in low-

birth-weight infants has been related to thepolyunsaturated fat, vitamin E, and iron contentof the formula. The fortification of infantformulas with vitamin E, related to the polyunsat-urated fatty acid content, is particularly impor-tant for the low-birth-weight infant because poorabsorption of naturally occurring vitamin E bylow-birth-weight infants makes them moresusceptible to a deficiency. This is especiallyimportant if iron-supplemented formulas are usedin the early weeks of life.

Recent evidence indicates that some mineralrequirements (e.g., calcium, sodium, copper) ofthe low-birth-weight infant may be greater per100 kcal than for full-term infants. This suggeststhat slightly higher levels of these minerals bepresent in formulas for low-birth-weight infantsthan the minimum levels proposed by theCommittee for full-term infants.Low body stores of vitamins, possible defects in

absorption (particularly of fat-soluble vitamins),

526 NUTRITION IN LOW-BIRTH-WEIGHT INFANTS


and low intakes of formula in the first weeks oflife necessitate the use of vitamin supplements,even though a formula adequate for full-terminfants is used. A single injection of vitamin K1 atbirth and daily oral supplements of vitamins A, C,D, E, and all the B group are recommended.The long-term effects of early nutrition are

important and challenging aspects of infant nutri-tion. Early feeding of low-birth-weight infantsentails a special responsibility because this is acrucial period of development when inadequa-cies, excesses, or imbalances are most likely toinfluence permanent changes. Long-term studies,still in progress, are attempting to relate feedingpractices in the premature nursery to subsequentneurologic development, learning ability, behav-ioral characteristics, and mental development ingeneral. Other possible pathologic consequencesof improper early nutrition that are legitimateareas of concern for the pediatric nutritionistinclude atherosclerosis, obesity, hypertension,and renal disease.

COMMITTEE ON NUTRITION

Members: Lewis A. Barness, M.D., Chairman; Alvin M.Mauer, M.D., Vice-Chairman; Arnold S. Anderson, M.D.;Peter R. Dallman, M.D.; Gilbert B. Forbes, M.D.; James C.Haworth, M.D.; Mary Jane Jesse, M.D.; Buford L. Nichols,Jr., M.D.; Nathan J. Smith, M.D.; Myron Winick, M.D.

Consultants: William C. Heird, M.D.; 0. L. Kline, Ph.D.;Donough O'Brien, M.D.

REFERENCES1. Holt LE Jr, Snyderman SE: The feeding of premature

and newborn infants. Pediatr Clin North Am13:1103, 1966.

2. Davidson M: Formula feeding of normal term and lowbirth weight infants. Pediatr Clin North Am17:913, 1970.

3. Babson SG: Feeding the low-birth-weight infant. JPediatr 79:694, 1971.

4. Dweck HS: Feeding the prematurely born infant:Fluids, calories and methods of feeding during theperiod of extrauterine growth retardation. ClinPerinatol 2:183, 1975.

5. Barness LA: Nutrition for the low birth weight infant.Clin Perinatol 2:345, 1975.

6. Sinclair JC, Driscoll JM Jr, Heird WC, Winters RW:Supportive management of the sick neonate:Parenteral calories, water, and electrolytes. Pedi-atr Clin North Am 17:863, 1970.

7. Hill JR, Robinson DC: Oxygen consumption innormally grown, small-for-dates and large-for-dates new-born infants. J Physiol 199:685, 1968.

8. Hardy JB, Goldstein EO: The feeding of prematureinfants: The value of high caloric diets in reducingthe length of hospital stay. J Pediatr 38:154,1951.

9. Snyderman SE, Holt LE Jr: The effect of high caloricfeeding on the growth of premature infants. JPediatr 58:237, 1961.

10. Combes MA, Pratt EL: Premature infants andconcentrated feeding. Am J Dis Child 102:610,1961.

11. Flakner F, Steigman AJ, Cruise MO: The physicaldevelopment of the premature infant: Some stan-dards and certain relationships to caloric intake. JPediatr 60:895, 1962.

12. Keitel HG, Chu E: Premature infant feeding: I. Theclinical usefulness of caloric concentration offormulas, of early vs. late feeding and of lowstearic acid content formulas. Pediatr Clin NorthAm 12:309, 1965.

13. Fomon SJ, Filer LJ Jr, Thomas LN, et al: Relationshipbetween formula concentration and rate ofgrowth of normal infants. J Nutr 98:241, 1969.

14. Gordon HH, Levine SZ: The metabolic basis for theindividualized feeding of infants, premature andfull-term. J Pediatr 25:464, 1944.

15. Ziegler EE, Fomon SJ: Fluid intake, renal solute load,and water balance in infancy. J Pediatr 78:561,1971.

16. Aperia A, Broberger 0, Thodenius K, Zetterstrom R:Developmental study of the renal response to anoral salt load in preterm infants. Acta PaediatrScand 63:517, 1974.

17. Aperia A, Broberger 0, Thodenius K, Zetterstrom R:Renal response to an oral sodium load in newbornfull term infants. Acta Paediatr Scand 61:670,1972.

18. Radde IC, Chance GW, Bailey K, et al: Growth andmineral metabolism in very low birth weightinfants: I. Comparison of the effects of two modesof NaHCO3 treatment of late metabolic acidosis.Pediatr Res 9:564, 1975.

19. Suljok E: Sodium homoeostasis in preterm infants.Lancet 1:930, 1975.

20.. Ballabriga A, Conde C, Gallart-Catala A: Metabolicresponse of prematures to milk formulas withdifferent lactic acid isomers or citric acid. HelvPaediatr Acta 25:25, 1970.

21. Valman HB, Heath CD, Brown RJK: Continuousintragastric milk feeds in infants of low birthweight. Br Med J 3:547, 1972.

22. Rhea JW, Kilby JO: A nasojejunal tube for infantfeeding. Pediatrics 46:36, 1970.

23. Cheek JA Jr, Staub GF: Nasojejunal alimentation forpremature and full-term newborn infants. JPediatr 82:955, 1973.

24. Benda GIM, Babson SG: Peripheral intravenousalimentation of the small premature infant. JPediatr 79:494, 1971.

25. Wilmore DW, Dudrick SJ: Growth and developmentof an infant receiving all nutrients exclusively byvein. JAMA 203:860, 1968.

26. Peden VH, Karpel JT: Total parenteral nutrition inpremature infants. J Pediatr 81:137, 1972.

27. Driscoll JM Jr, Heird WC, Schullinger JN, et al: Totalintravenous alimentation in low-birth-weightinfants: A preliminary report. J Pediatr 81:145,1972.

28. Committee on Nutrition: Parenteral feeding-A noteof caution. Pediatrics 49:776, 1972.

29. Asch MJ, Sperling M, Fiser R, et al: Metabolic and


hormonal studies comparing three parenteralnutrition regimens in infants. Ann Surg 182:62,1975.

30. Brans YW, Sumners JE, Dweck HS, Cassady G:Feeding the low birth weight infant: Orally orparenterally? Preliminary results of a comparativestudy. Pediatrics 54:15, 1974.

31. Lonnerdal B, Forsum E, Hambraeus L: The proteincontent of human milk: I. Transversal study ofSwedish normal material. Nutr Rep Int 13:125,1976.

32. Gordon HH, Levine SZ, McNamara H: Feeding ofpremature infants: A comparison of human andcow's milk. Am J Dis Child 73:442, 1947.

33. Kagan BM, Stanincova V, Felix NS, et al: Bodycomposition of premature infants: Relation tonutrition. Am J Clin Nutr 25:1153, 1972.

34. Goldman HI, Karelitz S, Acs H, Seifter E: Therelationship of the sodium, potassium, and chlo-ride concentration of the feeding to the weightgain of premature infants. Pediatrics 30:909,1962.

35. Kagan BM, Felix N, Molander CW, et al: Body waterchanges in relation to nutrition of prematureinfants. Ann NY Acad Sci 110:830, 1963.

36. Davidson M, Levine SZ, Bauer CH, Dann M: Feedingstudies in low birth weight infants: Relationshipsof dietary protein, fat, and electrolyte to rates ofweight gain, clinical courses and serum chemicalconcentrations. J Pediatr 70:695, 1967.

37. Levin B, Mackay HMM, Neill CA, et al: WeightGains, Serum Protein Levels and Health of BreastFed and Artificially Fed Infants, special reportseries No. 296. London, Medical Research Coun-cil, 1959.

38. Babson SG, Bramhall JL: Diet and growth in thepremature infant: The effect of different dietaryintakes of ash-electrolyte and protein on weightgain and linear growth. J Pediatr 74:890, 1969.

39. Omans WB, Barness LA, Rose CS, Gyorgy P:Prolonged feeding studies in premature infants. JPediatr 59:951, 1961.

40. Goldman HI, Freudenthal R, Holland B, Karelitz S:Clinical effects of two different levels of proteinintake on low birth weight infants. J Pediatr74:881, 1969.

41. Nichols MM, Danford BH: Feeding premature infants:A comparison of effects on weight gain, blood andurine of two formulas with varying protein andash composition. South Med J 59:1420, 1966.

42. Snyderman SE, Boyer A, Kogut MD, Holt LE Jr: Theprotein requirement of the premature infant: I.The effect of protein intake on the retention ofnitrogen. J Pediatr 74:872, 1969.

43. Levine SZ, Gordon HH, Marples E: Defect in themetabolism of tyrosine and phenylalanine inpremature infants: II. Spontaneous occurrenceand eradication by vitamin C. J Clin Invest20:209, 1941.

44. Menkes JH, Welcher DW, Levi HS, et al: Relationshipof elevated blood tyrosine to the ultimate intellec-tual performance of premature infants. Pediatrics49:218, 1972.

45. Snyderman SE, Holt LE Jr, Norton PM, PhansalkarSV: Protein requirement of the premature infant:II. Influence of protein intake on free amino acid

content of plasma and red blood cells. Am J ClinNutr 23:890, 1970.

46. Snyderman SE: The protein and amino acid require-ments of the premature infant, in Jonxis JHP,Vesser HRA, Troelstra JA (eds): MetabolicProcesses in the Foetus and Newborn Infant.Baltimore, Williams and Wilkins, 1971.

47. Sturman JA, Gaull G, Raiha NCR: Absence of cysta-thionase in human fetal liver: Is cystine essential?Science 169:74, 1970.

48. Gaull G, Sturman JA, Raiha NCR: Development ofmammalian sulfur metabolism: Absence of cysta-thionase in human fetal tissues. Pediatr Res 6:538,1972.

49. Valman HB, Brown RJK, Palmer T, et al: Proteinintake and plasma amino acids of infants of lowbirth weight. Br Med J 4:789, 1971.

50. Przyrembel H, Bremer HJ: Cystathioninuria in prema-ture infants. Clin Chim Acta 41:95, 1972.

51. Raiha NCR, Heinonen K, Rassin DK, Gaull GE: Milkprotein quantity and quality in low-birthweightinfants: I. Metabolic responses and effects ongrowth. Pediatrics 57:659, 1976.

52. Gaull GE, Rassin DK, Raiha NCR, Heinonen K: Milkprotein quantity and quality in low birth weightinfants: III. Effects on sulfur amino acids inplasma and urine. J Pediatr 90:348, 1977.

53. Cox WM Jr, Filer LJ Jr: Protein intake for low-birthweight infants. J Pediatr 74:1016, 1969.

54. Fomon S, Ziegler E, Vazquez H: Human milk and thesmall premature infant. Am J Dis Child 131:463,1977.

55. Tidwell HC, Holt LE Jr, Farrow HL, Neale S: Studiesin fat metabolism: III. Fat absorption in prema-ture infants and twins. J Pediatr 6:481, 1935.

56. Gordon HH, Levine SZ, Wheatley MA, Marples E:Respiratory metabolism in infancy and in child-hood: XX. The nitrogen metabolism in prematureinfants-comparative studies of human and cow'smilk. Am J Dis Child 54:1030, 1937.

57. Soderhjelm L: Fat absorption studies in children: I.Influence of heat treatment of milk on fat reten-tion by premature infants. Acta Paediatr 41:207,1952.

58. Davidson M, Bauer CH: Patterns of fat excretion infeces of premature infants fed various prepara-tions of milk. Pediatrics 25:375, 1960.

59. Watkins JB, Szczepanik P, Gould JB, et al: Bile saltmetabolism in the human premature infant.Gastroenterology 69:706, 1975.

60. Fomon SJ: Infant Nutrition, ed 2. Philadelphia, WBSaunders Co, 1974, p 101.

61. Widdowson EM: Absorption and excretion of fat,nitrogen and minerals from "filled" milks bybabies one week old. Lancet 2:1099, 1965.

62. Freeman CP, Jack EL, Smith LM: Intramolecularfatty acid distribution in the milk fat triglyceridesof several species. J Dairy Sci 48:853, 1965.

63. Mattson FH, Volpenhein RA: Rearrangement of glyc-eride fatty acids during digestion and absorption. JBiol Chem 237:53, 1962.

64. Gordon HH, McNamara H: Fat excretion of prema-ture infants: I. Effect on fecal fat of decreasing fatintake. Am J Dis Child 62:328, 1941.

65. Powers GF: Some observations on the feeding ofpremature infants, based on twenty years' experi-



ence at the New Haven hospital. Pediatrics 1:145,1948.

66. Shaw JCL: Evidence for defective skeletal mineraliza-tion in low-birthweight infants: The absorption ofcalcium and fat. Pediatrics 57:16, 1976.

67. Tantibhedhyangkul P, Hashim SA: Clinical and physi-ologic aspects of medium-chain triglycerides:Alleviation of steatorrhea in premature infants.Bull NY Acad Med 47:17, 1971.

68. Tantibhedhyangkul P, Hashim SA: Medium-chain tri-glyceride feeding in premature infants: Effects onfat and nitrogen absorption. Pediatrics 55:359,1975.

69. Andrews BF, Lorch V: Improved fat and Ca absorp-tion in L.B.W. infants fed a medium chain trigly-ceride containing formula, abstracted. Pediatr Res8:378, 1974.

70. Roy CC, Ste-Marie M, Chartrand L, et al: Correctionof the malabsorption of the preterm infant with amedium-chain triglyceride formula. J Pediatr86:446, 1975.

71. Committee on Nutrition: Commentary on breast-feeding and infant formulas, including proposedstandards for formulas. Pediatrics 57:278, 1976.

72. Auricchio S, Rubino A, Murset G: Intestinal glycosi-dase activities in the human embryo, fetus andnewborn. Pediatrics 35:944, 1965.

73. Boellner SE, Beard AG, Panos TC: Impairment ofintestinal hydrolysis of lactose in newborn infants.Pediatrics 36:542, 1965.

74. Jarrett EC, Holman GH: Lactose absorption in thepremature infant. Arch Dis Child 41:525, 1966.

75. Andrews BF, Cook LN: Low birth-weight infants fed anew carbohydrate-free formula with differentsugars: Growth and clinical course. Am J ClinNutr 22:845, 1969.

76. Fosbrooke AS, Wharton BA: "Added lactose" and"added sucrose" cow's milk formulae in nutritionof low birth-weight babies. Arch Dis Child 50:409,1975.

77. Duncan DL: The physiological effects of lactose. NutrAbst Rev 25:309, 1955.

78. Barbero GJ, Runge G, Fischer D, et al: Investigationson the bacterial flora, pH, and sugar content inthe intestinal tract of infants. J Pediatr 40:152,1952.

79. Cornely DA, Barness LA, Gyorgy P: Effect of lactoseon nitrogen metabolism and phenol excretion ininfants. J Pediatr 51:40, 1957.

80. Shohl AT: Mineral Metabolism. New York, ReinholdPublishing Co, 1939, p 19.

81. Widdowson E: Chemical analysis of the body, inBrozek J (ed): Human Body Composition:Approaches and Applications. Oxford, England,Pergamon Press, 1965.

82. Benjamin HR, Gordon HH, Marples E: Calcium andphosphorus requirements of premature infants.Am J Dis Child 65:412, 1943.

83. Hoevels 0, Thilenius OG, Krafczyk S: Untersuchenzum Calcium und Phosphatstoffwechsel Fruhge-borener. Monatsschr Kinderheilkd 108:112,1960.

84. Eek S, Gabrielsen LH, Halvorsen S: Prematurity andrickets. Pediatrics 20:63, 1957.

85. Widdowson EM, McCance RA, Harrison GE, SuttonA: Effect of giving phosphate supplements to

breast-fed babies on absorption and excretion ofcalcium, strontium, magnesium, and phosphorus.Lancet 2:1250, 1963.

86. Von Sydow G: A study of the development of rickets inpremature infants. Acta Paediatr Scand (Suppl)2:33, 1946.

87. Day GM, Chance GW, Radde IC, et al: Growth andmineral metabolism in very low birth weightinfants: II. Effects of calcium supplementation ongrowth and divalent cations. Pediatr Res 9:568,1975.

88. Oppe TE, Redstone D: Calcium and phosphorus levelsin healthy newborn infants given various types ofmilk. Lancet 1:1045, 1968.

89. Tsang RC, Oh W: Serum magnesium levels in lowbirth weight infants. Am J Dis Child 120:44,1970.

90. Barness LA, Omans WB, Rose CS, and Gyorgy P:Progress of premature infants fed a formulacontaining demineralized whey. Pediatrics 32:52,1963.

91. Caddell JL, Goddard DR: Studies in protein-caloriemalnutrition: I. Chemical evidence of magnesiumdeficiency. N Engl J Med 276:533, 1967.

92. Tsang RC: Neonatal magnesium disturbances. Am JDis Child 124:282, 1972.

93. Dallman P: Iron, vitamin E, and folate in the preterminfant. J Pediatr 85:742, 1974.

94. Gorten MK, Cross ER: Iron metabolism in prematureinfants: II. Prevention of iron deficiency. J Pediatr64:509, 1964.

95. Williams ML, Shott RJ, O'Neal PL, Oski FA: Role ofdietary irons and fat on vitamin E deficiencyanemia of infancy. N Engl J Med 292:887, 1975.

96. Griscom NT, Craig JN, Neuhauser EBD: Systemicbone disease developing in small prematureinfants. Pediatrics 48:883, 1971.

97. Seely JR, Humphrey GB, Matter BJ: Copper defi-ciency in a premature infant fed an iron fortifiedformula, abstracted. Clin Res 20:107, 1972.

98. Ashkenazi A, Levin S, Djaldetti M, et al: Thesyndrome of neonatal copper deficiency. Pediat-rics 52:525, 1973.

99. Cordano A: The role played by copper in the physio-pathology and nutrition of the infant and thechild. Ann Nestle 33:2, 1974.

100. Martmer EE, Corrigan KE, Charbeneau HP, Sosin A:A study of the uptake of iodine (1-131) by thethyroid of premature infants. Pediatrics 17:503,1956.

101. Honour JW, Shackleton CHL, Valman HB: Sodiumhomoeostasis in preterm infants. Lancet 2:1147,1974.

102. Willis DM, Roy NR, Chance GW, et al: Growth ofvery low birth weight (VLBW) infants: Effects ofacidosis, caloric intake and hyponatremia, ab-stracted. Pediatr Res 8:452, 1974.

103. Ackerman I, Chance G, Day G, et al: The associationof reduced growth with hyponatremia in very lowbirth weight infants (VLBW) fed "improved" milkformula, abstracted. Clin Res 21:1020, 1973.

104. Levin PK, Reid M, Reilly BJ, et al: latrogenic rickets inlow-birth-weight infants. J Pediatr 78:207, 1971.

105. Strelling MK, Blackledge GD, Goodall HB, WalkerCHM: Megaloblastic anaemia and whole-bloodfolate levels in premature infants. Lancet 1:898,


1966. 50:584, 1972.106. Vanier TM, Tyas JF: Folic acid status in premature 109. Federal Register S125.1: 20717 (August 2) 1973.

infants. Arch Dis Child 42:57, 1967. 110. Gordon HH, Nitowsky HM, Tildon JT, Levin S:107. Roberts PM, Arrowsmith DE, Rau SM, Monk-Jones Studies of tocopherol deficiency in infants and

ME: Folate state of premature infants. Arch Dis children: V. An interim summary. PediatricsChild 44:637, 1969. 21:673, 1958.

108. Pathak A, Godwin HA, Prudent LM: Vitamin B,2 and 111. Gross S, Melhom DK: Vitamin E-dependent anemia infolic acid values in premature infants. Pediatrics the premature infants. J Pediatr 85:753, 1974.

TWO CASES OF SUDDEN AND UNEXPLAINED DEATH OF CHILDREN DURINGSLEEP AS REPORTED IN 1834

Published case reports of unexpected deaths of infants during sleep prior tothe early decades of the 20th century routinely attributed the infant's deatheither to overlaying of the mother or wet-nurse, or after the 1840s, tosuffocation by an enlarged thymus. The two cases in letter form reportedbelow are of historic interest because they are among the first to infer thatunexpected infant death during sleep might be due to natural causes.To the Editor of THE LANCET.

Sir,-I have lately been called upon to examine two children, who, without having beenpreviously indisposed, were found'dead in bed.

In the first case the child was about six months old, and was lying in bed with its mother, whodiscovered in the middle of the night that it was dead. An inquest was held upon the body, and Iwas directed, in the absence of anything like testimony as to the cause of its dissolution, to makea post-mortem investigation. I should mention that the mother stated positively that the childhad not lain near her, .and that it was impossible it could have been suffocated, either from itsmouth having been applied to any part of her person or to the bed linen.

I found nothing unusual in the cavity of the skull,-no engorgement of the vessels,-nosanguineous or serous effusion. The viscera of the belly were in every respect of healthyappearance, and there was nothing in the stomach to indicate that it had come by its deathunfairly. In the chest, however, I found, upon the surface of the thymus gland, numerous spots ofextravasated blood, similar spots upon the surface of the lower and back parts of each lung, andmany patches of ecchymosis upon the margin of the right ventricle of the heart, and along thecourse of the trunk of the coronary vein. There was no engorgement, however, of the pulmonaryvessels, of the coronaries, or of the vessels of the thymus.

In the second case the child was five months old. It had been pretty well, had been suckled byits mother, and laid in bed upon its side, and in about an hour and a half afterwards wasdiscovered to be dead. There was some frothy matter in and about the mouth, and its hands werefirmly clenched. From the position in which it was found it was impossible it could have beensmothered.The appearances exhibited in the autopsy were strikingly the same as in the first case. The

contents of the skull and belly were in a perfectly natural condition. The extravasated spots uponthe thymus gland were more numerous than in the first case, and those upon the heart and thesurface of the lungs were fewer in number. There was about half an ounce of serous fluid in thepericardium.

In these cases one naturally asks,-what was the cause of death? The similarity of the post-mortem appearances would lead one to suppose that the cause must in each case have been thesame.

In the first case I was strongly disposed to think, in spite of the evidence of the mother, thatthe child must have been destroyed by overlaying it; but after the occurrence of the last case,where, from all the testimony that could be obtained, it seemed impossible that the child couldhave been suffocated, as it was lying in bed by itself, and was not obstructed in its breathing bythe bed-clothes, I confess that the opinion I had formed was a good deal shaken, and that Ibecame almost entirely at a loss how to account for death in either. In both cases there seems tohave been, from some cause or other, a sudden and violent action of the heart,-and numeroussmall vessels, from the increased force of its contraction, appear as a consequence to have givenway. But so trifling a lesion could hardly, in either instance, be supposed to be of itself sufficientto produce death, and it is with the hope that some of your correspondents who may have seensimilar cases, and who may be better able to offer an explanation of the phenomena they presentthan I am, will take the' trouble of enlightening me upon the subject, that I am induced toforward you this communication.Derby, Oct. 19, 1834 Saml W. Fearn

Noted by T. E. C., Jr., M.D.From Feam SW: Sudden and unexplained death in children. Lancet 1:246, 1834.



1977;60;519Pediatrics Nutritional Needs of Low-Birth-Weight Infants

ServicesUpdated Information &

http://pediatrics.aappublications.org/content/60/4/519including high resolution figures, can be found at:

Permissions & Licensing

http://www.aappublications.org/site/misc/Permissions.xhtmlentirety can be found online at: Information about reproducing this article in parts (figures, tables) or in its

Reprintshttp://www.aappublications.org/site/misc/reprints.xhtmlInformation about ordering reprints can be found online:


http://http://pediatrics.aappublications.org/content/60/4/519

http://www.aappublications.org/site/misc/Permissions.xhtml

http://www.aappublications.org/site/misc/reprints.xhtml

1977;60;519Pediatrics Nutritional Needs of Low-Birth-Weight Infants

http://pediatrics.aappublications.org/content/60/4/519the World Wide Web at:

The online version of this article, along with updated information and services, is located on

Copyright © 1977 by the American Academy of Pediatrics. All rights reserved. Print ISSN: 1073-0397. American Academy of Pediatrics, 141 Northwest Point Boulevard, Elk Grove Village, Illinois, 60007.been published continuously since 1948. Pediatrics is owned, published, and trademarked by the Pediatrics is the official journal of the American Academy of Pediatrics. A monthly publication, it has


http://pediatrics.aappublications.org/content/60/4/519

nutritional needs of low-birth-weight...

Documents