Original Research

A Comparison of Intakes of Breast-Fed and Bottle-Fed

Infants during the First Two Days of Life

Shaul Dollberg, MD, Sigalit Lahav, RN, Francis B. Mimouni, MD, FACN
The Department of Neonatology, Lis Maternity Hospital, Tel Aviv Sourasky Medical Center, Tel Aviv, ISRAEL
Key words: infants, term, appetite, breast-feeding

Objective: In the first days of life, breast-fed infants consume minimal amounts of milk; this may be
explained by substrate limitation (limited milk output) and/or by self-limitation (through low appetite and/or
suck-swallow competency). The spontaneous milk intake of unrestricted formula-fed infants has not been studied
to date. We compared the spontaneous formula intake of unrestricted formula-fed infants to that of breast-fed
infants over the first 48 hours of life. We hypothesized that 1) spontaneous formula intake of unrestricted infants
is much higher than that of breast-fed infants and 2) spontaneous formula intake correlates positively with
gestational age or birthweight.
Methods: We studied 43 healthy, term infants. By maternal choice, 15 infants were exclusively breast-fed
and 28 were formula-fed ad libitum every four hours. Breast-fed infants were weighed before and one hour after
initiation of feeding, and intake was calculated from the difference between the measurements and corrected
individually for the infants normal postnatal decrease in body weight. Bottles offered to formula-fed infants
contained 60 cc, and the remainder was carefully measured. Intakes were expressed as cc/kg/d, and weight
changes as % of birthweight. Statistical methods included Students t tests and stepwise regression analysis.
Results: Breast feeding on Day 1 was 9.610.3 (meanSD) vs. 18.59.6 cc/kg/d in formula-fed infants
(p0.011); on Day 2 it was 13.011.3 vs. 42.214.2 cc/kg/d (p0.001). Breast-fed infants lost significantly
more weight on Day 2 (p0.015). In multiple regression, when the dependent variable was the second-day
intake, the significant independent variables were group (higher intake in the formula-fed group), weight loss
(the higher the weight loss, the lower the intake), and first-day intake (the higher the first-day intake, the higher
the second-day intake).
Conclusion: Newborn infants offered formula ad libitum every four hours consumed much larger amounts
than breast-fed infants fed according to the same schedule. In addition, weight loss was more marked in
breast-fed infants on Day 2 of life.

INTRODUCTION infant is unknown. We aimed to evaluate the spontaneous

In the first days of life, breast-fed infants consume minimal
amounts of milk, as little as 13 mL/kg/day on Day 1 and 40
mL/kg/day on Day 2 of life [1]. In theory, such small amounts
may be due to low maternal colostrum output from a breast in
transition from pregnancy to lactation. It is known, however,
that the breast milk production adjusts itself to the infants
requirements [2]. Whether low colostrum or milk output of the
first two days of life is primarily due to a relative lack of
readiness to lactate or is secondary to low consumption by the
intake of infants fed formula (by maternal choice) ad libitum.
We hypothesized that such infants consume greater amounts of
milk in the first two days of life than breast-fed infants.
PATIENTS AND METHODS
We recruited 43 healthy, singleton term, (38 to 41 weeks
gestation), appropriate for gestational age (using the curves of
Lubchenco [3]) newborn infants concurrently born after un-
complicated pregnancy, labor and delivery; maternal analgesia

Address reprint requests to: Shaul Dollberg, MD, The Department of Neonatology, Lis Maternity Hospital, Tel Aviv Sourasky Medical Center, 6 Weizman Street, Tel Aviv
64239, ISRAEL. E-mail: mimouni@tasmc.health.gov.il
Presented in part at the Society for Pediatric Research Conference, San Francisco, CA, May 1999.

Journal of the American College of Nutrition, Vol. 20, No. 3, 209211 (2001)
Published by the American College of Nutrition

209
Formula Intake during the First Two Days

during labor was solely by epidural bupivacaine. All infants Table 1. Characteristics and Milk Intake in Breast-Fed and
had an Apgar score of more than 7 at 1 and at 5 minutes and Formula-Fed Infants
were delivered by spontaneous vaginal vertex delivery.
Formula-
By maternal choice, 28 were formula-fed and 15 breast-fed; Breast-Fed
Fed Significance
in both groups, infants were fed every four hours, by hospital n15
n28
protocol, starting as early as two and as late as ten hours, as
Gender (M:F) 7:8 13:15 ns
soon as the mother was ready to feed the infant. No baby was Birthweight (g) 3258408 3348348 ns
roomed-in. Formula-fed infants received ready-to-feed Similac Gestational age (wks) 39.60.9 39.71.2 ns
(20 cal/oz, Ross laboratories, Columbus OH). Any remainder Parity 2.31.3 2.40.9 ns
of the 60 mL bottle was carefully measured, and results were Apgar score 1 min 9.30.5 9.40.7 ns
Apgar score 5 min 10.00.0 9.90.4 ns
rounded to the closest mL. In the breast-fed group, every
Weight loss day 1 (g) 14996 13056 ns
infants was weighed, fully clothed, before and one hour after Weight loss day 2 (g) 6758 2146 p0.015
initiation of feeding at the breast using an electronic scale (Seca Age at first feed (h) 7.82.6 4.92.7 p0.019
Model 727, Hamburg, Germany). Intake was calculated from Intake day 1 (mL/kg) 9.610.3 18.59.6 p0.011
the difference between the two measurements. Correction for Intake day 2 (mL/kg) 13.011.3 42.214.2 p0.001
the normal postnatal decrease in body weight during the time
spent at the breast, suggested by Salarya et al. [4], was based
infants (p0.001). In backward stepwise regression analysis
upon the daily weight loss of each individual infant. Specifi-
where intake on Day 1 was the dependent variable and birth-
cally, we used the 24-hour weight loss of each child and
weight (or gestational age), Apgar Score at five minutes, age at
divided it by the time spent at the breast in order to provide an
first feed, 24 hours weight loss and group (formula vs. breast
adequate correction. In both groups, patients were weighed
feeding) were the independent variables, only the group was
every 24 hours using an electronic scale. All regurgitations
significant (higher intakes in the formula-fed group). When the
were noted on a collection form.
dependent variable was the second-day intake and the indepen-
dent variables were all the above-mentioned variables plus the
Statistical Analyses first 24 hours intake and the first 24-hour intake, the variables
Intakes were calculated as mL/kg/day and weight changes that remained significant in the equation were group (higher
as % of birthweight. Results are expressed as meanSD. intake in the formula-fed group), weight loss (the higher the
Paired t tests (for normally distributed variables), were used to weight loss, the lower the intake), and first-day intake (the
test the difference between Day 1 and Day 2 intakes. The higher the first-day intake, the higher the second-day intake).
Kruskal-Wallis test was used for non-normally distributed vari-
ables. The chi-square test was used to determine the difference
between the groups for discrete variables (such as gender).
DISCUSSION
Backward stepwise regression analysis was used to study the
effect of selected variables on Day 1 or Day 2 intakes. We showed that formula-fed infants offered formula ad
libitum on Day 1 and Day 2 of life have an intake of 18.59
mL/kg/day on Day 1, which increases to 4214 mL/kg/day on
RESULTS Day 2. Significant regurgitations were not reported in either
group, and we do not believe that unreported regurgitations
Twenty boys and 23 girls were recruited in the study. All may have biased significantly our results. Nevertheless, these
mothers were middle-class, healthy women aged 24 to 41 years values are significantly higher than those found in our control,
with an average parity of 2.4 (range 1 to 6). All infants were breast-fed group. This observation may be interpreted possibly
Israeli Jews, most of them of mixed Sepharadic and Ashkenazi as being due to substrate limitation in breast-fed infants or,
origin. There were no significant differences between the alternatively, to overfeeding in formula-fed infants. While it
breast-fed and the formula-fed group in terms of gestational may very well be that infants fed formula ad libitum are
age, birth weight or Apgar scores (Table 1). overfed when compared to breast-fed infants, it is possible
Formula feeding increased from 18.59 mL/kg/day on Day that reduced maternal milk production in our hospital settings
1 to 42.214.2 mL/kg/day on Day 2 (p0.001). In breast-fed may have played a role. Indeed, none of the infants in our study
infants, milk intake was calculated to be 9.610.3 mL/kg/day was roomed-in, and, although they were fed ad libitum, it was
on Day 1, increasing to 13.011.3 mL/kg/day on Day 2 according to a strict every-four-hours schedule. In such circum-
(p0.005). The age at first feed was significantly (by approx- stances, establishment of colostrum production may be slower
imately three hours) lower in bottle-fed infants than in breast- and less efficient than when the infant is roomed-in [5]. In fact,
fed infants. Both on Day 1 and on Day 2 the intake of formula- the values found in formula-fed infants were close to those
fed infants was significantly higher than that of the breast-fed found by Casey et al. (1316 mL/kg/day on Day 1 and 4023

210 VOL. 20, NO. 3


mL/kg/day on Day 2) in strictly breast-fed infants, although the
rooming-in situation of these infants was not described [1]. In
Caseys study, all infants were breast-fed for the first time
within the first two hours of life, a circumstance which may
also have led to greater human milk production than in our
patients, whose age at first feed, a variable which proved
influential on intake, ranged between two and ten hours of life.
Although the difference in age at first feed between the two
groups was significantly different, in regression analysis, this
difference of approximately three hours did not significantly
affect the intake on Day 1 or Day 2.
Regardless of the feeding group, the values found in this
study were much lower than those reported between the ages of
eight and 30 days, when the median intake is 196 mL/kg/day
[6]. There can be no suspicion that the formula-fed group was
substrate-limited. Thus, lower intake in the first days of life
must be interpreted as due to self-limitation by the infant. This
self limitation could possibly be due to lower appetite and/or
thirst, decreased sucking and swallowing competency or de-
creased alertness level and may be aggravated by iatrogenic
factors such as opiate analgesics. In our study, the latter is not
likely, as all infants were delivered after bupivacaine epidural
analgesia, which has little effect on neonatal behavior [7].
The fluid intakes reported in this study as well as in that of
Casey are well below what is advised in major neonatology
textbooks, which suggest a true water requirement of 60 to 65
mL/kg/day at birth [8,9]. Thus, published recommendations
may exceed the neonates physiological needs. However, when
fluid or caloric requirements are increased, such as in hypogly-
cemia or hyperviscosity, the infants spontaneous intake may
be insufficient. Nevertheless, while weight loss was similar on
Day 1 of life in breast and formula-fed infants, it was more
marked in breast-fed infants on the second day of life. Second-
day weight loss correlated inversely with first-day intake,
which itself correlated with age at first feed. Thus, it may very
well be that, regardless of the group, one way to reduce exces-
sive postnatal weight loss is to introduce an early feeding
JOURNAL OF THE AMERICAN COLLEGE OF NUTRITION
Formula Intake during the First Two Days
policy. In our opinion, the fact that first-day intake had an
influence on the second-day intake is in support of the specu-
lation that the infant experience (competency in sucking and
swallowing) might be a factor even more important than ap-
petite.
REFERENCES
1. Casey CE, Neifert MR, Seacat JM, Neville MC: Nutrient intake by
breast-fed infants during the first five days after birth. Am J Dis
Child 140:933936, 1986.
2. Daly SE, Hartmann PE: Infant demand and milk supply. Part 1:
Infant demand and milk production in lactating women. J Hum Lact
11:2126, 1995.
3. Lubchenco LO, Hansman C, Dressler M, Boyd E: Intrauterine
growth as estimated from live-born birth weight data at 24 to 42
weeks gestation. Pediatrics 32:793799, 1963.
4. Salarya EM, Easton PM, Cater JL: Duration of breast-feeding after
early initiation and frequent feeding. Lancet 2:11411142, 1978.
5. Perez-Escamilla R, Pollitt E, Lonnerdal B, Dewey KG: Infant feed-
ing policies in maternity wards and their effect on breast-feeding
success: an analytical overview. Am J Public Health 84:8997,
1994.
6. Fomon SJ, Owen GM, Thomas LN: Milk or formula volume in-
gested by infants fed ad libitum. Am J Dis Child 108:601604,
1964.
7. Rosenblatt DB, Belsey EM, Lieberman BA, Redshaw M, Caldwell
J, Notarianni L, Smith RL, Beard RW: The influence of maternal
analgesia on neonatal behavior: II. Epidural bupivacaine. Br J Ob-
stet Gynaecol 88:407413, 1981.
8. Bell EF, Oh W: Fluid and electrolyte management. In Avery GB,
Fletcher MA, MacDonald MG (eds): Neonatology, 4th ed. Phil-
adelphia: J.B. Lippincott, pp 312339, 1994.
9. Liechty EA: Water requirements. In Polin RA, Fox WW (eds):
Fetal and Neonatal Physiology, 2nd ed. Philadelphia: W.B. Saun-
ders, pp 305307, 1998.
Received January 25, 2000; revision accepted August 3, 2000.
211