- Email: [email protected]
Growth and metabolic response of premature infants fed whey- or casein-dominant formulas after hospital discharge
Growth and metabolic response of premature infants fed whey- or casein-dominant formulas after hospital discharge Judy C. Bernbaum, MD, Sharon R. Sasanow, MA, RD, Helen R. Churella, PhD, a n d A n d r e a Daft, RN From the Neonatal Follow-up Program, The Children's Hospital of Philadelphia, Philadelphia, Pennsylvania, and the Medical Department, Ross Laboratories, Columbus, Ohio
We conducted a double-blind, randomized study to test the hypothesis that a whey-dominant formula permits a growth and metabolic a d v a n t a g e over a casein-dominant formula in preterm infants after hospital discharge. Nineteen low birth weight infants were studied for 6 months from the time of discharge. Ten received a casein-dominant formula, and nine received a whey-dominant formula. Growth (weight, length, head circumference, mid-arm circumference, and skin-fold thickness), biochemical measurements (alkaline phosphatase activity, acid-base status, and hemoglobin, serum total protein, albumin, and urea nitrogen levels), and quantity of formula intake did not differ significantly between the groups over a 6-month study period. Serum transthyretin and urea nitrogen concentrations differed significantly between the two feeding groups at the d a y of entry into the study only. The results indicate that, after hospital discharge, premature infants fed a whey-dominant formula do not differ in growth or biochemical measurements from those fed a casein-dominant formula. (J PEDIATR1989;115:652-6)
Premature infants who are not fed human milk typically receive formulas that contain whey proteins as the dominant protein source during the early neonatal period. The results of clinical research suggest that metabolic response to whey-dominant formulas is similar to that elicited by human milk feedings.~ Just before discharge from the hospital, formula-fed premature infants are usually given a standard 20 kcal/fl oz formula containing either casein or
whey proteins as the dominant protein source. Preterm infants have been reported to have a higher incidence of metabolic acidosis when fed casein-dominant than when fed whey-dominant formulas. I No such difference was found, however, in term infants receiving a casein- or whey-dominant formula. 2, 3 It is not known whether preterm infants
I MAC/HC Supported by a grant from Ross Laboratories. Presented in part before the Federation of American Societies for Experimental Biology, St. Louis, Mo., April 14, 1986. Submitted for publication Aug. 22, 1988; accepted April 19, 1989. Reprint requests: Judy C. Bernbaum, MD, Director, Neonatal Follow-Up Program, The Children's Hospital of Philadelphia, Philadelphia, PA 19104. 9/23/13364
652
Mid-arm circumference/head circumference (ratio)
I
at the time of hospital discharge or in the first 26 weeks of life have a difference in acid-base status when fed the two types of formulas. This study was undertaken to determine whether premature infants fed a whey-dominant formula differ in growth, acid-base status, and other biochemical measurements from
Volume 115 Number 4
Growth o f LB W infants after hospital discharge
infants fed a casein-dominant formula after hospital discharge.
Table I, Composition of study formulas (per deciliter)
METHODS
Subjects. Premature infants who weighed less than 1750 gm at birth, whose size was appropriate for gestational age, and who were free of genetic abnormalities or major cardiac, respiratory, or metabolic problems were considered for the study. To be selected for enrollment, however, the infants were required to receive the same oral feeding (Similac Special Care 24) for at least 1 week, be on full oral feedings of 110 to 120 kcal/kg/day, and weigh at least 1800 gm. Twenty infants who met these criteria were randomly assigned to the double-blind parallel study of two formula feedings. Informed consent was obtained from each infant's parents. The infants were only orally fed at the time of entry into the study. Of the infants enrolled, one was removed from the study before the first clinic visit (week 2) because of viral gastroenteritis that required dietary alteration. This infant's data were not included in the analysis. Two other infants were removed from the study, one before the second and one before the third visit, because of illness that was not formula related. A fourth infant was removed before the fourth visit because the parents changed the formula. Data for these infants were included up to the time of removal from the study. The study was approved by the institutional review board of the Children's Hospital of Philadelphia. Procedures. While still in the hospital, the infants eligible for the study were randomly assigned to be fed either a casein-dominant or a whey-dominant formula (20 kcal/fl oz). Infants received their assigned formula at least 3 days before hospital discharge and during the period of study. Infants were fed the assigned formula ad libitum by their parents or caretakers during the study period. Parents were instructed not to give their infants solid foods for at least 16 weeks. Infants who consumed more than 10% of energy from foods other than formula before 16 weeks of age were dropped from the study. Parents kept a record of the number of fluid ounces of formula and of the amount of solid food consumed by the infant for 3 consecutive days before each clinic visit. Spitting up and intolerance to formula were also recorded. Parents brought their infants to The Children's Hospital of Philadelphia Neonatal Follow-Up Program for testing 2, 6, 12, and 26 weeks after entry into the study. Just before receiving their assigned formulas on the first day of the study (time zero) and at the 2-, 6-, 12-, and 26week visits, infants underwent anthropometric measurements and, except at 12 weeks, the biochemical testing described below. Formulas. The test formulas were Similae with iron
Nutrient
Protein (gin) Fat (gin) Carbohydrate (gin) Energy (kcal) Calcium (rag) Phosphorus (rag) Sodium (rag) Potassium (mg) Chloride (mg) Iron (rag) Vitamin A (IU) Vitamin E (IU) Vitamin C (mg) Vitamin B6 (/~g)
653
Whey-dominant Casein-dominant formula formula
1.58 3.58 7.3 67.8 44.2 32.1 21.6 79.4 43.0 1.6 394.0 2.4 9.6 55.0
1.56 3.63 7.3 68.0 55.8 44.0 22.5 82.25 52.5 1.7 407.0 2.3 15.7 45.0
Both formulas were fortified to provide the following nutrients per deciliter: 0.5 mg zinc, 0.10 mg copper, 10 #g iodine, 3.4 #g manganese, 4.1 mg magnesium, 40 IU vitamin D, 65 #g B1, 100 #g vitamin B2, 0.70 mg niacin, 10 ~zg folic acid, 0.15 ~zgvitamin B i2, 5.5 #g vitamin Kb 0.3 mg pantothenic acid, and 1 ~g biotin.
(casein dominant) and Similac with whey plus iron (whey dominant). Both formulas contained approximately 68 kcal, about 1.5 gm protein, and similar amounts of fat per deciliter as fed. The fat in both formulas was 60% soy and 40% coconut oils (Table l). Calcium, phosphorus, and chloride concentrations were greater in the casein-dominant than in the whey-dominant formula. This difference was unavoidable because of the protein source. All other mineral and vitamin concentrations were similar in the two formulas. The formula was supplied by Ross Laboratories as concentrated liquid in 13 fl oz (383 ml) cans. Parents were given verbal and written instructions for preparing the formula to yield 20 kcal/fl oz (67.6 kcal/dl). Anthropometric measures. The infants' weight, length, head circumference, mid-arm circumference, and triceps skin-fold thickness were measured. Weights were measured with a scale accurate to _+5 gm. A recumbent infant length board (Infantometer, Seretex, Inc., Carlstadt, N.J.) was used for measuring length. A nonstretehable tape was used to measure head and mid-arm circumferences. Skin folds were measured with a Holtain Caliper (Holtain, Ltd., Crymych, United Kingdom). Two trained evaluators made all measurements using standardized techniques. Biochemical measurements. Boood was drawn by venipuncture irrespective of the fed or fasting state because none of the biochemical components measured are known to be acutely affected by feeding. A heparinized sample of blood was used to determine pH, Pco2, and hemoglobin. Base excess was estimated by means of the Siggaard-Andersen alignment chart (nomogram). 4 Serum from the nonhep-
654
Bernbaum et al.
The Journal of Pediatrics October 1989
T a b l e II. F o r m u l a intake of study infants Weeks of study Formula caloric intake' (kcal/kg/day)
Whey-dominant Casein-dominant
0
2
6
116 • 24 (n = 9) 122 • I8 (n = 10)
118 • 32 (n = 9) 135 • 23 (n = 10)
113 • (n = i23 • (n =
26 8) t4 9)
12
26
110 • 25 (n = 7) 10I • I1 (n = 8)
93 • 27 (n = 7) 85 • i0 (n = 7)
Values are expressed as mean _+ SD. *No significant difference in caloric intake of formula between feeding groups.
arinized blood specimen was analyzed for the following components: total protein, s albumin,6 urea nitrogen] and alkaline phosphatase. 8 The serum transthyretin (prealbumin) level was measured by radial immunodiffusion assay 9 (Calbiochem Behring, La Jolla, Call f). Statistical analysis, Birth weight, length, head circumference, and Apgar scores of the infants in the two feeding groups were compared with the Student t test. Subsequent anthropometric measurements and blood studies were evaluated by two-way repeated measures analysis of variance. The two formula groups were compared for homogeneity at study day 0, the day before they were to receive the assigned test formula. All anthropometric, biochemical, and formula intake measures that did not differ at baseline were analyzed by two-way repeated measures analysis of variance. Serum urea and transthyretin concentrations differed at baseline for the two feeding groups; therefore these two measures were evaluated by two-way analysis of covariance, with baseline values being covariates for correction for any differences. RESULTS The two groups of infants were similar in mean birth weigbt, length, head circumference, Apgar scores, gestational age, and weight 1 week before entry into study. At entry into the study, infants in the two randomized groups did not differ significantly in mean (__+SD) weight (2129 = 111 vs 2062 _+ 204 gm), weight gain (17.3 _+ 2.0 vs 18.6 _+ 1.4 gm/kg/day), or postconceptional age (38.2 + 1.6 vs 37.7 _+ 1.8 weeks). The groups did not differ significantly in mean weight, length, head circumference, or left mid-arm triceps skinfold thickness measurements at the time of study entry or at the subsequent 2-, 6-, 12-, and 26-week clinic visits. The ratio of mid-arm to head circumference and the skin-fold thickness gains were also similar. Daily gains in weight, length, and head circumference did not differ between the two feeding groups. (Data available on request from the authors.) Formula intake (expressed as kilocalories per kilogram per day) did not differ between the feeding groups at any
testing period (Table II). Although at 12 weeks the daily mean formula intake declined slightly in both groups, no infant was receiving solid foods. One week before the 26-week visit, however, records indicated that infants in both feeding groups were receiving about 15% of their caloric intake from solid foods. Incidences of spitting up and intolerance to formula did not differ between the two groups. At the time of entry into the study and at the four ages at which the infants were tested, acid-base status did not differ between the two feeding groups. No infant was judged to have metabolic acidosis, defined as a base excess less than -4.5 mEq/L. Hemoglobin values throughout the study were similar in the two groups. At the start of the study, concentrations of serum total protein, albumin, and alkaline phosphatase was similar in the two feeding groups (Table III). Serum transthyretin and serum urea concentrations, however, were significantly greater in the infants assigned to receive whey-dominant formula than those assigned to receive casein-dominant formula. There was no difference in caloric intake between the formula groups just before the start of the study and no other apparent explanation for this dissimilarity. At subsequent times, there were no significant differences between feeding groups in any of these indicators of protein status or in alkaline phosphatase activity. Mean concentrations of protein status indicators increased over time (Table III). Alkaline phosphatase activity was significantly lower at the 26-week than at the 6-week visit (p _< 0.001). DISCUSSION In this study, premature infants fed a formula containing casein as the predominant protein source grew similarly to, i and had serum biochemical measurements similar to, those fed a formula containing whey proteins as the predominant protein source. Most of the infants did not achieve weight or length at or above the 50th percentile for term infants of the same postconceptional age. This finding is consistent with those of Friel et al. 1~and Brandt, 11 who found that, on average, the weight gain of premature infants was retarded for as long as 21 months compared with full-term infants.
Volume 115 Number 4
Growth o f LB W infants after hospital discharge
655
Table III. S e r u m t o t a l p r o t e i n , a l b u m i n , t r a n s t h y r e t i n , u r e a n i t r o g e n , a n d a l k a l i n e p h o s p h a t a s e c o n c e n t r a t i o n s in s t u d y i n f a n t s Weeks of Study 0 Total protein (gm/dl) Whey-dominant Casein-dominant Albumin (gm/dl) Whey-dominant Casein-dominant Transthyretin (mg/dl) Whey-dominant Casein-dominant Urea nitrogen (mg/dl) Whey-dominant Casein-dominant Alkaline phosphatase (U/L) Whey-dominant Casein-dominant
4.4 • (n = 4.2 • (n =
2
6
26
0.4 9) 0.2 10)
4.6 + 0.4 (n = 9) 4.6 • 0.5 (n = 10)
5.0 • (n = 4.8 • (n =
0.4 8) 0.4 9)
6.0 (n 6.0 (n
• = • =
0.4* 7) 0.4* 8)
3.5 • 0.3 (n = 9) 3.2 + 0.1 (n = 10)
3.7 • 0.2 (n = 9) 3.6 • 0.2 (n = 10)
4.0 + 0.3 (n = 8) 3.9 • 0.3 (n = 9)
4.5 (n 4.5 (n
• = + =
0.1" 7) 0.2* 8)
15.0 • 3.6 (n=9) 11.1 • 2.8~ (n = I0)
13.7 • 2.7 (n=9) 10.1 • 2.4 (n = 10)
12.0 • 3.7 (n=7) 14.7 • 3.8 (n = 9)
18.8 • 2.0t (n=5) 18.9 • 4.5* (n = 8)
4.7 • (n = 3.1 • (n =
1.7 9) 0.7~ 10)
3.8 • (n = 4.9 • (n =
1.3 8) 2.8 10)
5.6 • (n = 6.1 • (n =
1.7 8) 2.4 9)
9.3 (n 10.8 (n
• = • =
1.1" 7) 1.7" 8)
443 • (n = 335 • (n =
172 9) 93 10)
437 • (n = 372 • (n =
111 9) 149 10)
399 • (n = 323 • (n =
72 8) 67 9)
264 2 45w (n = 7) 245 • 63w (n = 8)
Values are expressedas mean _+SD. *Significantlygreater than values at 0 and 2 weeks (p < 0.001). ]'Significantlygreater than values at 0 and 2 weeks (p < 0.01). ~:Significantlyless than whey-dominant(p < 0.05). w less than values at 0 and 2 weeks (p < 0.001).
The head circumferences of premature infants from dis-
and M A C / H C ratio but not necessarily in skin-fold thick-
charge to 6 months after discharge were variable, with a
ness gain. These indicators of growth increased during the study period at a rate suggesting that the infants were re-
number being above the 50th percentile of the National Center for Health Statistics 12 ( N C H S ) curve. Brandt 11 found that head circumference growth in preterm infants is similar to that of term infants of the same postconceptional age. The M A C / H C ratio has been useful in assessing proteincalorie malnutritionl3; during marginal nutritional depri-
ceiving adequate nutrition. R/iih/i et al. 1 observed a higher incidence of metabolic acidosis in premature infants fed casein-dominant formulas within the first few weeks after birth than in those fed whey-dominant formulas. These authors found the mean base excess of infants fed the casein-dominant formula,
vation, head growth may continue at a normal rate at the expense of fat deposition. Skin-fold thickness is also used as
containing a protein concentration similar to that of our
an indicator of body fatness. In this study the two feeding groups had similar M A C / H C ratios and skin-fold thickness
weeks after birth. The mean base excess of infants fed the casein-dominant formula in our study was only - 1 . 0 m E q /
measurements. At discharge, when infants were similar in postconceptional age to term infants, the mean M A C / H C
tional age, when the infants were 9 and 15 weeks of post-
ratio was only slightly less than that found by Sasanow et al. 14 for full-term infants. Skin-fold thicknesses of our
natal age, respectively. The lack of effect of the caseindominant formula on acid-base status may be explained by
infants at initiation of the study were below the mean for newborn full-term infants reported by Oakley 15 but were
the fact that our infants had reached greater maturity when studied. J~irvenpg/i et al. 2 found no difference in acid-base
almost equivalent 6 weeks later. Regardless of the formula,
status of term infants fed whey- or casein-dominant for-
infants lagged behind full-term infants in weight, length,
mula.
study formula, to be less than - 4 m E q / L during the first 7
L at 2 weeks and - 2 . 0 m E q / L at 6 weeks of postconcep-
656
Bernbaum et al.
S e r u m urea nitrogen concentrations of infants in both feeding groups were lower than those observed in other studies in p r e t e r m or term infants fed whey- or casein-dominant formula or in breast-fed infants. As observed by others,2, 16 the quantity of milk protein intake, not the type of milk protein, primarily influences serum urea nitrogen concentrations. ~2 S e r u m transthyretin, an indicator of protein status in p r e m a t u r e infants, iv is rapidly responsive to differences in adequacy of protein and energy i n t a k e J 8' I9 T h e r e were no significant differences in serum t r a n s t h y r e t i n concentrations between the two feeding groups during the course of the study. S e r u m transthyretin concentrations were similar to those observed by Kashyap et al. 2~ and Moskowitz et al. 17 in p r e t e r m infants before hospital discharge when protein intake was similar. T h e values in our infants were also similar to transthyretin values of t e r m infants at birth. 21 T r a n s t h y r e t i n concentrations increased significantly over the study period in both feeding groups. Thus the caseind o m i n a n t formula was as effective as the whey-dominant formula in improving or m a i n t a i n i n g transthyretin concentrations and other indexes of protein status of infants. W e do not ascribe any clinical significance to the differences observed in transthyretin and serum urea nitrogen between the two formula groups at the start of the study. Energy intakes of the study infants during the first 12 weeks were greater than those recorded by Dewey a n d L6nnerda122 for breast-fed infants of similar postconceptional age. Energy intakes were similar, however, to intakes of t e r m infants fed formula. 23' 24 Friel e t a l ) ~ found t h a t p r e m a t u r e infants after discharge consumed significantly more formula t h a n term infants. In spite of the slightly higher energy intake, the m e a n weights and lengths of prem a t u r e infants in this study, regardless of feeding, r e m a i n e d below the 50th percentiles of the N C H S growth charts at each testing period. T h e findings of this study suggest ' i-at both casein- and w h e y - d o m i n a n t formulas promote gr~)wth and m a i n t a i n satisfactory protein and metabolic status in p r e m a t u r e infants from discharge to 6 months of adjusted age. We thank Joel Siegman for his thorough statistical evaluation of the data. REFERENCES
1. Rfiih~i NCR, Heinonen K, Rassin DK, Gaull GE. Milk protein quantity and quality in low-birthweight infants. I. Metabolic responses and effects on growth. Pediatrics 1976;57:659-74. 2. J~irvenp~ig.AL, Rfiih~i NCR, Rassin DK, Gaull GE. Milk protein quantity and quality in the term infant. I. Metabolfc responses and effects on growth. Pediatrics i982;70:214-20. 3. Jonas LM, Picciano MF, Hatch TF. Indices ef protein metabolism in term infants fed human milk, whey-predominant formula or cow's milk formula. Pediatrics 1985;75:775-84. 4. Siggaard-Andersen O. An acid-base chart for arterial blood
The Journal of Pediatrics October 1989
with normal and pathophysiological reference areas. Scand J Clin Lab Invest 1971;27:239-45. 5. Koch TR, Johnson GF, Chilcote ME. Kinetic determination of total serum protein with a centrifugal analyzer. Clin Chem t974;20:392-4. 6. Doumas BT, Watson WA, Biggs HG. Albumin standards and the measurement of serum albumin with bromocresol green. Clin Chim Acta 1971 ;31:87-96. 7. Spayd R, Bruschi B, Burdick BA, et al. Multilayer film elements for clinical analysis: applications to representative chemical determinations. Clin Chem 1978;24:1343-50. 8. Babson AL, Greeley S J, Coleman CM, Phillips GE. Phenolphthalein monophosphate as a substrate for serum alkaline phosphatase. Clin Chem 1966;12:482-90. 9. Mancini G, Carbonara AO, Heremans JF. Immunochemical quantitation of antigens by single radial immunodiffusion. Int J Immunochem 1965;2:235-54. 10. Friel JK, Gibson RS, Kawash GK, Watts J. Dietary zinc intake and growth during infancy. J Pediatr Gastroenterol Nutr 1985;4:746-51. 11. Brandt I. Growth dynamics of low-birth-weight infants. Acta Paediatr Scand 1985;319(suppl):38-47. 12. Hamil PVV, Drizd TA, Johnson CL, et al. Physical growth: National Center for Health Statistics percentiles. Am J Clin Nutr 1979;32:607-29. 13. Kanawati AA, McLaren DS. Assessment of marginal malnutrition. Nature 1970;228:573-4. 14. Sasanow SR, Georgieff MK, Pereira GR. Mid-arm circumference and mid-arm/head circumference ratios: standard curves for anthropometric assessment of neonatal nutritional status. J PEDIATR 1986;109:311-5. 15. Oakley JR- Differences in subcutaneous fat in breast- and formula-fed infants. Arch Dis Child 1977;52:79-80. 16. Davies DP, Saunders R. Blood urea normal values in early infancy related to feeding practices. Arch Dis Child 1973;48:5635. 17. Moskowitz SR, Pereira G, Spitzer A, Heaf L, Amsel J, Watkins JB. Prealbumin as a biochemical marker of nutritional adequacy in premature infants. J PEI)IA'rR 1983;102:749-53. 18. Smith FR, Goodman PS, Zaklama MS, etal. Serum vitamin A, retinol-binding protein, and prealbumin concentrations in protein-calorie malnutrition. I. A functional defect in hepatic retinol release. Am J Clin Nutr 1973;26:973-81. 19. ingenbleek Y, DeVisscher M, DeNayer P. Measurement of prealbumin as index of protein-calorie malnutrition. Lancet 1972;2:106-8. 20. Kashyap S, Forsyth M, Zucher C, Ramakrishnan R, Dell RB, Heird WC. Effects of varying protein and energy intakes on growth and metabolic response in low-birth-weight infants. J PEDIATR 1986;108:955-63. 2I. Sasanow SR~ Spitzer AR, Pereir'a GR, Heaf L, Watkins JB. Effect of gestational age upon prealbumin and retinol binding protein in preterm and term infants. J Pediatr Gastroenterol Nutr 1986;5:111-5. 22. Dewey KG, L6nnerdal B. Milk and nutrient intake of breastfed infants from 1 to 6 months: relation to growth and fatness. J Pediatr Gastroenterol Nutr 1983;2:497-506. 23. Montandon CM, Wills C, Garza C, Smith EO, Nichols BL. Formula intake of 1- and 4-month old infants. J Pediatr Gastroen~eroi Nutr 1986;5:434-43. 24. Fomon S J, Thomas LN, Filer LJ Jr, Ziegler EE, Leonard MT. Food consumption and growth of normal infants fed milkbased formulas. Acta Paediatr Scand Suppl 1971;223:36.
We conducted a double-blind, randomized study to test the hypothesis that a whey-dominant formula permits a growth and metabolic a d v a n t a g e over a casein-dominant formula in preterm infants after hospital discharge. Nineteen low birth weight infants were studied for 6 months from the time of discharge. Ten received a casein-dominant formula, and nine received a whey-dominant formula. Growth (weight, length, head circumference, mid-arm circumference, and skin-fold thickness), biochemical measurements (alkaline phosphatase activity, acid-base status, and hemoglobin, serum total protein, albumin, and urea nitrogen levels), and quantity of formula intake did not differ significantly between the groups over a 6-month study period. Serum transthyretin and urea nitrogen concentrations differed significantly between the two feeding groups at the d a y of entry into the study only. The results indicate that, after hospital discharge, premature infants fed a whey-dominant formula do not differ in growth or biochemical measurements from those fed a casein-dominant formula. (J PEDIATR1989;115:652-6)
Premature infants who are not fed human milk typically receive formulas that contain whey proteins as the dominant protein source during the early neonatal period. The results of clinical research suggest that metabolic response to whey-dominant formulas is similar to that elicited by human milk feedings.~ Just before discharge from the hospital, formula-fed premature infants are usually given a standard 20 kcal/fl oz formula containing either casein or
whey proteins as the dominant protein source. Preterm infants have been reported to have a higher incidence of metabolic acidosis when fed casein-dominant than when fed whey-dominant formulas. I No such difference was found, however, in term infants receiving a casein- or whey-dominant formula. 2, 3 It is not known whether preterm infants
I MAC/HC Supported by a grant from Ross Laboratories. Presented in part before the Federation of American Societies for Experimental Biology, St. Louis, Mo., April 14, 1986. Submitted for publication Aug. 22, 1988; accepted April 19, 1989. Reprint requests: Judy C. Bernbaum, MD, Director, Neonatal Follow-Up Program, The Children's Hospital of Philadelphia, Philadelphia, PA 19104. 9/23/13364
652
Mid-arm circumference/head circumference (ratio)
I
at the time of hospital discharge or in the first 26 weeks of life have a difference in acid-base status when fed the two types of formulas. This study was undertaken to determine whether premature infants fed a whey-dominant formula differ in growth, acid-base status, and other biochemical measurements from
Volume 115 Number 4
Growth o f LB W infants after hospital discharge
infants fed a casein-dominant formula after hospital discharge.
Table I, Composition of study formulas (per deciliter)
METHODS
Subjects. Premature infants who weighed less than 1750 gm at birth, whose size was appropriate for gestational age, and who were free of genetic abnormalities or major cardiac, respiratory, or metabolic problems were considered for the study. To be selected for enrollment, however, the infants were required to receive the same oral feeding (Similac Special Care 24) for at least 1 week, be on full oral feedings of 110 to 120 kcal/kg/day, and weigh at least 1800 gm. Twenty infants who met these criteria were randomly assigned to the double-blind parallel study of two formula feedings. Informed consent was obtained from each infant's parents. The infants were only orally fed at the time of entry into the study. Of the infants enrolled, one was removed from the study before the first clinic visit (week 2) because of viral gastroenteritis that required dietary alteration. This infant's data were not included in the analysis. Two other infants were removed from the study, one before the second and one before the third visit, because of illness that was not formula related. A fourth infant was removed before the fourth visit because the parents changed the formula. Data for these infants were included up to the time of removal from the study. The study was approved by the institutional review board of the Children's Hospital of Philadelphia. Procedures. While still in the hospital, the infants eligible for the study were randomly assigned to be fed either a casein-dominant or a whey-dominant formula (20 kcal/fl oz). Infants received their assigned formula at least 3 days before hospital discharge and during the period of study. Infants were fed the assigned formula ad libitum by their parents or caretakers during the study period. Parents were instructed not to give their infants solid foods for at least 16 weeks. Infants who consumed more than 10% of energy from foods other than formula before 16 weeks of age were dropped from the study. Parents kept a record of the number of fluid ounces of formula and of the amount of solid food consumed by the infant for 3 consecutive days before each clinic visit. Spitting up and intolerance to formula were also recorded. Parents brought their infants to The Children's Hospital of Philadelphia Neonatal Follow-Up Program for testing 2, 6, 12, and 26 weeks after entry into the study. Just before receiving their assigned formulas on the first day of the study (time zero) and at the 2-, 6-, 12-, and 26week visits, infants underwent anthropometric measurements and, except at 12 weeks, the biochemical testing described below. Formulas. The test formulas were Similae with iron
Nutrient
Protein (gin) Fat (gin) Carbohydrate (gin) Energy (kcal) Calcium (rag) Phosphorus (rag) Sodium (rag) Potassium (mg) Chloride (mg) Iron (rag) Vitamin A (IU) Vitamin E (IU) Vitamin C (mg) Vitamin B6 (/~g)
653
Whey-dominant Casein-dominant formula formula
1.58 3.58 7.3 67.8 44.2 32.1 21.6 79.4 43.0 1.6 394.0 2.4 9.6 55.0
1.56 3.63 7.3 68.0 55.8 44.0 22.5 82.25 52.5 1.7 407.0 2.3 15.7 45.0
Both formulas were fortified to provide the following nutrients per deciliter: 0.5 mg zinc, 0.10 mg copper, 10 #g iodine, 3.4 #g manganese, 4.1 mg magnesium, 40 IU vitamin D, 65 #g B1, 100 #g vitamin B2, 0.70 mg niacin, 10 ~zg folic acid, 0.15 ~zgvitamin B i2, 5.5 #g vitamin Kb 0.3 mg pantothenic acid, and 1 ~g biotin.
(casein dominant) and Similac with whey plus iron (whey dominant). Both formulas contained approximately 68 kcal, about 1.5 gm protein, and similar amounts of fat per deciliter as fed. The fat in both formulas was 60% soy and 40% coconut oils (Table l). Calcium, phosphorus, and chloride concentrations were greater in the casein-dominant than in the whey-dominant formula. This difference was unavoidable because of the protein source. All other mineral and vitamin concentrations were similar in the two formulas. The formula was supplied by Ross Laboratories as concentrated liquid in 13 fl oz (383 ml) cans. Parents were given verbal and written instructions for preparing the formula to yield 20 kcal/fl oz (67.6 kcal/dl). Anthropometric measures. The infants' weight, length, head circumference, mid-arm circumference, and triceps skin-fold thickness were measured. Weights were measured with a scale accurate to _+5 gm. A recumbent infant length board (Infantometer, Seretex, Inc., Carlstadt, N.J.) was used for measuring length. A nonstretehable tape was used to measure head and mid-arm circumferences. Skin folds were measured with a Holtain Caliper (Holtain, Ltd., Crymych, United Kingdom). Two trained evaluators made all measurements using standardized techniques. Biochemical measurements. Boood was drawn by venipuncture irrespective of the fed or fasting state because none of the biochemical components measured are known to be acutely affected by feeding. A heparinized sample of blood was used to determine pH, Pco2, and hemoglobin. Base excess was estimated by means of the Siggaard-Andersen alignment chart (nomogram). 4 Serum from the nonhep-
654
Bernbaum et al.
The Journal of Pediatrics October 1989
T a b l e II. F o r m u l a intake of study infants Weeks of study Formula caloric intake' (kcal/kg/day)
Whey-dominant Casein-dominant
0
2
6
116 • 24 (n = 9) 122 • I8 (n = 10)
118 • 32 (n = 9) 135 • 23 (n = 10)
113 • (n = i23 • (n =
26 8) t4 9)
12
26
110 • 25 (n = 7) 10I • I1 (n = 8)
93 • 27 (n = 7) 85 • i0 (n = 7)
Values are expressed as mean _+ SD. *No significant difference in caloric intake of formula between feeding groups.
arinized blood specimen was analyzed for the following components: total protein, s albumin,6 urea nitrogen] and alkaline phosphatase. 8 The serum transthyretin (prealbumin) level was measured by radial immunodiffusion assay 9 (Calbiochem Behring, La Jolla, Call f). Statistical analysis, Birth weight, length, head circumference, and Apgar scores of the infants in the two feeding groups were compared with the Student t test. Subsequent anthropometric measurements and blood studies were evaluated by two-way repeated measures analysis of variance. The two formula groups were compared for homogeneity at study day 0, the day before they were to receive the assigned test formula. All anthropometric, biochemical, and formula intake measures that did not differ at baseline were analyzed by two-way repeated measures analysis of variance. Serum urea and transthyretin concentrations differed at baseline for the two feeding groups; therefore these two measures were evaluated by two-way analysis of covariance, with baseline values being covariates for correction for any differences. RESULTS The two groups of infants were similar in mean birth weigbt, length, head circumference, Apgar scores, gestational age, and weight 1 week before entry into study. At entry into the study, infants in the two randomized groups did not differ significantly in mean (__+SD) weight (2129 = 111 vs 2062 _+ 204 gm), weight gain (17.3 _+ 2.0 vs 18.6 _+ 1.4 gm/kg/day), or postconceptional age (38.2 + 1.6 vs 37.7 _+ 1.8 weeks). The groups did not differ significantly in mean weight, length, head circumference, or left mid-arm triceps skinfold thickness measurements at the time of study entry or at the subsequent 2-, 6-, 12-, and 26-week clinic visits. The ratio of mid-arm to head circumference and the skin-fold thickness gains were also similar. Daily gains in weight, length, and head circumference did not differ between the two feeding groups. (Data available on request from the authors.) Formula intake (expressed as kilocalories per kilogram per day) did not differ between the feeding groups at any
testing period (Table II). Although at 12 weeks the daily mean formula intake declined slightly in both groups, no infant was receiving solid foods. One week before the 26-week visit, however, records indicated that infants in both feeding groups were receiving about 15% of their caloric intake from solid foods. Incidences of spitting up and intolerance to formula did not differ between the two groups. At the time of entry into the study and at the four ages at which the infants were tested, acid-base status did not differ between the two feeding groups. No infant was judged to have metabolic acidosis, defined as a base excess less than -4.5 mEq/L. Hemoglobin values throughout the study were similar in the two groups. At the start of the study, concentrations of serum total protein, albumin, and alkaline phosphatase was similar in the two feeding groups (Table III). Serum transthyretin and serum urea concentrations, however, were significantly greater in the infants assigned to receive whey-dominant formula than those assigned to receive casein-dominant formula. There was no difference in caloric intake between the formula groups just before the start of the study and no other apparent explanation for this dissimilarity. At subsequent times, there were no significant differences between feeding groups in any of these indicators of protein status or in alkaline phosphatase activity. Mean concentrations of protein status indicators increased over time (Table III). Alkaline phosphatase activity was significantly lower at the 26-week than at the 6-week visit (p _< 0.001). DISCUSSION In this study, premature infants fed a formula containing casein as the predominant protein source grew similarly to, i and had serum biochemical measurements similar to, those fed a formula containing whey proteins as the predominant protein source. Most of the infants did not achieve weight or length at or above the 50th percentile for term infants of the same postconceptional age. This finding is consistent with those of Friel et al. 1~and Brandt, 11 who found that, on average, the weight gain of premature infants was retarded for as long as 21 months compared with full-term infants.
Volume 115 Number 4
Growth o f LB W infants after hospital discharge
655
Table III. S e r u m t o t a l p r o t e i n , a l b u m i n , t r a n s t h y r e t i n , u r e a n i t r o g e n , a n d a l k a l i n e p h o s p h a t a s e c o n c e n t r a t i o n s in s t u d y i n f a n t s Weeks of Study 0 Total protein (gm/dl) Whey-dominant Casein-dominant Albumin (gm/dl) Whey-dominant Casein-dominant Transthyretin (mg/dl) Whey-dominant Casein-dominant Urea nitrogen (mg/dl) Whey-dominant Casein-dominant Alkaline phosphatase (U/L) Whey-dominant Casein-dominant
4.4 • (n = 4.2 • (n =
2
6
26
0.4 9) 0.2 10)
4.6 + 0.4 (n = 9) 4.6 • 0.5 (n = 10)
5.0 • (n = 4.8 • (n =
0.4 8) 0.4 9)
6.0 (n 6.0 (n
• = • =
0.4* 7) 0.4* 8)
3.5 • 0.3 (n = 9) 3.2 + 0.1 (n = 10)
3.7 • 0.2 (n = 9) 3.6 • 0.2 (n = 10)
4.0 + 0.3 (n = 8) 3.9 • 0.3 (n = 9)
4.5 (n 4.5 (n
• = + =
0.1" 7) 0.2* 8)
15.0 • 3.6 (n=9) 11.1 • 2.8~ (n = I0)
13.7 • 2.7 (n=9) 10.1 • 2.4 (n = 10)
12.0 • 3.7 (n=7) 14.7 • 3.8 (n = 9)
18.8 • 2.0t (n=5) 18.9 • 4.5* (n = 8)
4.7 • (n = 3.1 • (n =
1.7 9) 0.7~ 10)
3.8 • (n = 4.9 • (n =
1.3 8) 2.8 10)
5.6 • (n = 6.1 • (n =
1.7 8) 2.4 9)
9.3 (n 10.8 (n
• = • =
1.1" 7) 1.7" 8)
443 • (n = 335 • (n =
172 9) 93 10)
437 • (n = 372 • (n =
111 9) 149 10)
399 • (n = 323 • (n =
72 8) 67 9)
264 2 45w (n = 7) 245 • 63w (n = 8)
Values are expressedas mean _+SD. *Significantlygreater than values at 0 and 2 weeks (p < 0.001). ]'Significantlygreater than values at 0 and 2 weeks (p < 0.01). ~:Significantlyless than whey-dominant(p < 0.05). w less than values at 0 and 2 weeks (p < 0.001).
The head circumferences of premature infants from dis-
and M A C / H C ratio but not necessarily in skin-fold thick-
charge to 6 months after discharge were variable, with a
ness gain. These indicators of growth increased during the study period at a rate suggesting that the infants were re-
number being above the 50th percentile of the National Center for Health Statistics 12 ( N C H S ) curve. Brandt 11 found that head circumference growth in preterm infants is similar to that of term infants of the same postconceptional age. The M A C / H C ratio has been useful in assessing proteincalorie malnutritionl3; during marginal nutritional depri-
ceiving adequate nutrition. R/iih/i et al. 1 observed a higher incidence of metabolic acidosis in premature infants fed casein-dominant formulas within the first few weeks after birth than in those fed whey-dominant formulas. These authors found the mean base excess of infants fed the casein-dominant formula,
vation, head growth may continue at a normal rate at the expense of fat deposition. Skin-fold thickness is also used as
containing a protein concentration similar to that of our
an indicator of body fatness. In this study the two feeding groups had similar M A C / H C ratios and skin-fold thickness
weeks after birth. The mean base excess of infants fed the casein-dominant formula in our study was only - 1 . 0 m E q /
measurements. At discharge, when infants were similar in postconceptional age to term infants, the mean M A C / H C
tional age, when the infants were 9 and 15 weeks of post-
ratio was only slightly less than that found by Sasanow et al. 14 for full-term infants. Skin-fold thicknesses of our
natal age, respectively. The lack of effect of the caseindominant formula on acid-base status may be explained by
infants at initiation of the study were below the mean for newborn full-term infants reported by Oakley 15 but were
the fact that our infants had reached greater maturity when studied. J~irvenpg/i et al. 2 found no difference in acid-base
almost equivalent 6 weeks later. Regardless of the formula,
status of term infants fed whey- or casein-dominant for-
infants lagged behind full-term infants in weight, length,
mula.
study formula, to be less than - 4 m E q / L during the first 7
L at 2 weeks and - 2 . 0 m E q / L at 6 weeks of postconcep-
656
Bernbaum et al.
S e r u m urea nitrogen concentrations of infants in both feeding groups were lower than those observed in other studies in p r e t e r m or term infants fed whey- or casein-dominant formula or in breast-fed infants. As observed by others,2, 16 the quantity of milk protein intake, not the type of milk protein, primarily influences serum urea nitrogen concentrations. ~2 S e r u m transthyretin, an indicator of protein status in p r e m a t u r e infants, iv is rapidly responsive to differences in adequacy of protein and energy i n t a k e J 8' I9 T h e r e were no significant differences in serum t r a n s t h y r e t i n concentrations between the two feeding groups during the course of the study. S e r u m transthyretin concentrations were similar to those observed by Kashyap et al. 2~ and Moskowitz et al. 17 in p r e t e r m infants before hospital discharge when protein intake was similar. T h e values in our infants were also similar to transthyretin values of t e r m infants at birth. 21 T r a n s t h y r e t i n concentrations increased significantly over the study period in both feeding groups. Thus the caseind o m i n a n t formula was as effective as the whey-dominant formula in improving or m a i n t a i n i n g transthyretin concentrations and other indexes of protein status of infants. W e do not ascribe any clinical significance to the differences observed in transthyretin and serum urea nitrogen between the two formula groups at the start of the study. Energy intakes of the study infants during the first 12 weeks were greater than those recorded by Dewey a n d L6nnerda122 for breast-fed infants of similar postconceptional age. Energy intakes were similar, however, to intakes of t e r m infants fed formula. 23' 24 Friel e t a l ) ~ found t h a t p r e m a t u r e infants after discharge consumed significantly more formula t h a n term infants. In spite of the slightly higher energy intake, the m e a n weights and lengths of prem a t u r e infants in this study, regardless of feeding, r e m a i n e d below the 50th percentiles of the N C H S growth charts at each testing period. T h e findings of this study suggest ' i-at both casein- and w h e y - d o m i n a n t formulas promote gr~)wth and m a i n t a i n satisfactory protein and metabolic status in p r e m a t u r e infants from discharge to 6 months of adjusted age. We thank Joel Siegman for his thorough statistical evaluation of the data. REFERENCES
1. Rfiih~i NCR, Heinonen K, Rassin DK, Gaull GE. Milk protein quantity and quality in low-birthweight infants. I. Metabolic responses and effects on growth. Pediatrics 1976;57:659-74. 2. J~irvenp~ig.AL, Rfiih~i NCR, Rassin DK, Gaull GE. Milk protein quantity and quality in the term infant. I. Metabolfc responses and effects on growth. Pediatrics i982;70:214-20. 3. Jonas LM, Picciano MF, Hatch TF. Indices ef protein metabolism in term infants fed human milk, whey-predominant formula or cow's milk formula. Pediatrics 1985;75:775-84. 4. Siggaard-Andersen O. An acid-base chart for arterial blood
The Journal of Pediatrics October 1989
with normal and pathophysiological reference areas. Scand J Clin Lab Invest 1971;27:239-45. 5. Koch TR, Johnson GF, Chilcote ME. Kinetic determination of total serum protein with a centrifugal analyzer. Clin Chem t974;20:392-4. 6. Doumas BT, Watson WA, Biggs HG. Albumin standards and the measurement of serum albumin with bromocresol green. Clin Chim Acta 1971 ;31:87-96. 7. Spayd R, Bruschi B, Burdick BA, et al. Multilayer film elements for clinical analysis: applications to representative chemical determinations. Clin Chem 1978;24:1343-50. 8. Babson AL, Greeley S J, Coleman CM, Phillips GE. Phenolphthalein monophosphate as a substrate for serum alkaline phosphatase. Clin Chem 1966;12:482-90. 9. Mancini G, Carbonara AO, Heremans JF. Immunochemical quantitation of antigens by single radial immunodiffusion. Int J Immunochem 1965;2:235-54. 10. Friel JK, Gibson RS, Kawash GK, Watts J. Dietary zinc intake and growth during infancy. J Pediatr Gastroenterol Nutr 1985;4:746-51. 11. Brandt I. Growth dynamics of low-birth-weight infants. Acta Paediatr Scand 1985;319(suppl):38-47. 12. Hamil PVV, Drizd TA, Johnson CL, et al. Physical growth: National Center for Health Statistics percentiles. Am J Clin Nutr 1979;32:607-29. 13. Kanawati AA, McLaren DS. Assessment of marginal malnutrition. Nature 1970;228:573-4. 14. Sasanow SR, Georgieff MK, Pereira GR. Mid-arm circumference and mid-arm/head circumference ratios: standard curves for anthropometric assessment of neonatal nutritional status. J PEDIATR 1986;109:311-5. 15. Oakley JR- Differences in subcutaneous fat in breast- and formula-fed infants. Arch Dis Child 1977;52:79-80. 16. Davies DP, Saunders R. Blood urea normal values in early infancy related to feeding practices. Arch Dis Child 1973;48:5635. 17. Moskowitz SR, Pereira G, Spitzer A, Heaf L, Amsel J, Watkins JB. Prealbumin as a biochemical marker of nutritional adequacy in premature infants. J PEI)IA'rR 1983;102:749-53. 18. Smith FR, Goodman PS, Zaklama MS, etal. Serum vitamin A, retinol-binding protein, and prealbumin concentrations in protein-calorie malnutrition. I. A functional defect in hepatic retinol release. Am J Clin Nutr 1973;26:973-81. 19. ingenbleek Y, DeVisscher M, DeNayer P. Measurement of prealbumin as index of protein-calorie malnutrition. Lancet 1972;2:106-8. 20. Kashyap S, Forsyth M, Zucher C, Ramakrishnan R, Dell RB, Heird WC. Effects of varying protein and energy intakes on growth and metabolic response in low-birth-weight infants. J PEDIATR 1986;108:955-63. 2I. Sasanow SR~ Spitzer AR, Pereir'a GR, Heaf L, Watkins JB. Effect of gestational age upon prealbumin and retinol binding protein in preterm and term infants. J Pediatr Gastroenterol Nutr 1986;5:111-5. 22. Dewey KG, L6nnerdal B. Milk and nutrient intake of breastfed infants from 1 to 6 months: relation to growth and fatness. J Pediatr Gastroenterol Nutr 1983;2:497-506. 23. Montandon CM, Wills C, Garza C, Smith EO, Nichols BL. Formula intake of 1- and 4-month old infants. J Pediatr Gastroen~eroi Nutr 1986;5:434-43. 24. Fomon S J, Thomas LN, Filer LJ Jr, Ziegler EE, Leonard MT. Food consumption and growth of normal infants fed milkbased formulas. Acta Paediatr Scand Suppl 1971;223:36.