BREASTFEEDING 2001, P&RT I:

THE EVIDENCE FOR BREASTFEEDING 0031-3955/01 $15.00 + .OO

NUTRITION, GROWTH, AND

COMPLEMENTARY FEEDING OF
THE BREASTFED INFANT
Kathryn G. Dewey, PhD

Human milk is the ideal food for infants because of its unique
nutritional characteristics.30,35 For example, in terms of protein content,
human milk has a high ratio of whey to casein, a relatively high propor-
tion of nonprotein nitrogen, and high concentrations of certain specific
proteins.14 These components serve nutritional and nonnutritional func-
tions.30Also, human milk is rich in some fatty acids essential for brain
development and contains several nonlactose carbohydrates that have a
role in resistance to infection. Furthermore, human milk changes in
composition as infants mature.
After a certain age, however, human milk alone no longer can
supply all of an infant’s nutritional requirements, and complementary
foods are needed to ensure adequate nutrition and growth. It is com-
monly assumed that an increased need for energy and protein is the
primary factor dictating complementary feeding, but some of the micro-
nutrients are likely to become limiting sooner than the macronutrients.
If a mother nurses on demand and is well nourished, her milk supply
probably can keep pace with her infant’s energy needs for considerably
longer than 6 months. By contrast, the amount of iron provided by
human milk may become insufficient even before 6 months if an infant
has suboptimal iron reserves at birth.
This article reviews the contribution of human milk to nutritional
needs during the first 2 years of life, growth patterns of breastfed infants,
and recommendations regarding the age of introduction and optimal
nutrient density of complementary foods. The focus is on healthy, term

From the Department of Nutrition, University of California, Davis, Davis, California

PEDIATRIC CLINICS OF NORTH AMERICA

VOLUME 48 * NUMBER 1 * FEBRUARY 2001 87

 88 DEWEY

infants; other articles in this issuezover feeding of preterm infant^.^' For

detailed discussion of specific nutritional needs of breastfed infants, the
reader is referred to articles on iron,= polyunsaturated fatty acidsz9and
vitamins.24

CONTRIBUTION OF HUMAN MILK TO NUTRIENT

NEEDS DURING THE FIRST 2 YEARS OF LIFE

A report on complementary feeding published in 19984 presented

an in-depth examination of the nutrient needs of breastfed infants. If a
mother is well nourished, exclusive breastfeeding can meet all of her
infant’s nutrient requirements for at least 6 months, with the possible
exception of vitamin D in some population^^^ and iron in infants with
relatively low birth weight.”
The amounts of nutrients provided by human milk can be estimated
by multiplying average human milk intake by the concentration of
each nutrient in human milk. By subtracting these values from total
recommended nutrient intakes, one can derive estimates of the amounts
of nutrients needed from complementary foods after 6 months of age.
The 1998 report provided these estimates for infants in developing
countries. Using the same approach, Table 1 provides these data for
infants in industrialized countries, based on slightly different assump-
tions about the quantities of human milk consumed at various ages and
the average nutrient content of the milk. In this table, the recommended
nutrient intakes are based on the Dietary Reference Values from the
United Kingdom,13 except for energy, protein, folate, iron, and zinc.
Energy and protein requirements are taken from a more recent report:
14, 38 folate and iron requirements are based on Food and Agriculture
Organization of the United Nations/World Health Organization (FAO/
WHO) estimates,= and zinc requirements are based on the calculations
in Annex I11 of the 1998 report4 The estimated amount of each nutrient
provided by human milk is based on average milk volumes of women
in industrialized countries during each of the age intervals ( M m o ,
9-11 mo, 12-23 mo post partum) using pooled data from 15 studies
compiled in the 1998 report. Nutrient concentrations in human milk
were taken from the Institute of Medicine report,30except for zinc, for
which more recent data from Krebs et aP2were used.
The first row of Table 1 shows the total energy requirements and
the estimated amounts of energy from human milk and complementary
foods at each age. The values for human milk consumption are based
on intakes of infants who were already receiving complementary foods.
Because such foods tend to displace human milk intake, at least partially,
it is somewhat circular to describe the gap between human milk intake
I and total energy requirements as the energy required from complemen-
tary foods. Infants exclusively breastfed beyond 6 months are likely to
have considerably higher human milk intakes, but because few data are
available for such cases, to calculate the true energy gap (if any) under
 Table 1. RECOMMENDED NUTRIENT INTAKES, AVERAGE AMOUNT PROVIDED BY BREAST MILK, AND AMOUNT NEEDED FROM
COMPLEMENTARY FOODS F W INFANTS IN DEVELOPED COUNTRIES AT 6-8,9-11, AND 12-23 MO
6-8 mo Sll mo 12-23 mo
Amount Amount Amount Amount Amount Amount
From Needed % From Needed % From Needed %
Breast From from Breast From from Breast From from
Nutrient RNI* Milkt CF CF RNI' Milkt CF CF RNP Milkt CF* CF
Energy (kcal/d) 682 486 196 29 830 375 455 55 1092 313 779 71
Protein (g/d) 9.1 7.2 1.9 21 9.6 5.6 4.0 42 10.9 4.7 6.2 57
Vitamin A (pg RE/d) 350 461 0 0 350 354 0 0 400 300 100 25
Folate (pg/d) 32 58 0 0 32 45 0 0 50 38 12 24
Niacin (mg/d) 4.0 1.0 3.0 75 5.0 0.8 4.2 84 8.0 0.7 7.3 91
Riboflavin (mg/d) 0.40 0.24 0.16 40 0.40 0.19 0.21 53 0.60 0.16 0.44 73
Thiamin (mg/d) 0.20 0.14 0.06 30 0.30 0.11 , 0.19 63 0.50 0.10 0.40 80
Vitamin 8, (mg/d) 0.30 0.06 0.24 80 0.40 0.05 0.35 88 0.70 0.04 0.66 94
Vitamin B,, (pg/d) 0.40 0.67 0 0 0.40 0.51 0 0 0.50 0.47 0.03 6
Vitamin C (mg/d) 25 28 0 0 25 21 4 16 30 18 12 40
Vitamin D (kg/d) 7.0 0.4 6.6 94 7.0 0.3 6.7 96 7.0 0.2 6.8 97
Calcium (mg/d) 525 193 332 63 525 148 377 72 350 125 225 64
Chloride (mg/d) 500 289 211 42 500 222 278 56 800 188 612 77
Copper (mg/d) 0.30 0.17 0.13 43 0.30 0.13 0.17 57 0.40 0.11 0.29 73
Iodine (pg/d) 60 76 0 0 60 58 2 3 70 49 21 30
Irons (mg/d) 7.0 0.2 6.8 97 7.0 0.2 6.8 97 4.0 0.1 3.9 98
Magnesium (mg/d) 75 24 51 68 80 19 61 76 85 16 69 81
Phosphorus (mg/d) 400 96 304 76 400 74 324 81 270 63 207 77
Potassium (mg/d) 700 361 339 48 700 278 422 60 800 235 565 71
Selenium (pg/d) 10 14 0 0 10 11 0 0 15 10 5 33
Sodium (mg/d) 320 124 196 61 350 95 255 73 500 81 419 84
Zinc (me;/d) 2.8 0.6 2.2 79 2.8 0.4 2.4 86 2.8 0.3 2.5 89

*Recommended nutrient intakes from reference 13, except for energy (reference 7), protein (reference 14), folate (reference 23), iron (reference 23), and zinc (reference 4).
tBased on average milk volume of 688, 529, and 448 mL/d for 6-8, 9-11, and 12-23 mo, respebively,' and milk nutrient concentrations from reference 30 (except for
zinc, taken from reference 32).
OD $Assuming high bioavailability of iron.
\o CF = complementary foods.
these circumstances is not currently possible. Thus, the values for energy
shown in the column for the ”amount needed from complementary
foods” should be considered as desc7ipfive of usual patterns rather than
as prescriptive of what should be consumed. With this caveat in mind,
the average expected energy intake from complementary foods is ap-
proximately 840 kJ (1 kcal = 4.2 kJ) at 6 to 8 months, 1890 kJ at 9 to 11
months and 3276 kJ at 12 to 23 months. These represent 29%, 55%, and
71% of total energy needs, respectively, coinciding with the decreased
intake of human milk at older ages. These values differ if an infant is
consuming more or less human milk than the average. The 1998 report
provides estimates for three levels of human milk intake: (1)low (mean
- 2 [standard deviation]), (2) average (mean), and (3) high (mean + 2
SD). Table 1 shows only the values corresponding to average human
milk intake.
The second row of Table 1 shows the same estimates for protein.
Assuming average human milk intake, the amount of protein needed
from complementary foods increases from approximately 2 g/d at 6 to
8 months to approximately 6 g/d at 12 to 23 months, with the percentage
from complementary foods increasing from 21% to 57%. The remaining
rows show the estimates for micronutrients for which an established
recommended nutrient intake value exists (the 1998 report also includes
estimates for pantothenic acid, vitamin K, fluoride, and manganese). For
several nutrients (i.e., vitamin A, folate, vitamin B,, vitamin C, iodine,
and selenium), the amount needed from complementary foods before 12
months is zero (or close to zero) because human &lk contains generous
amounts of these nutrients if a mother is well nourished. For others, the
percentage of the total requirement needed from complementary foods
ranges from 30% to 97%. In the case of vitamin D, this percentage is
very high (>94%) because relatively little vitamin D is present in human
milk; however, adequate exposure to sunlight can meet infants’ needs
for vitamin D even if complementary foods are not rich in this nutrient.24
The values for niacin needed from complementary foods also appear
high (7591% of requirements), but because niacin needs also can be met
by the contribution of tryptophan in the diet, niacin is not likely to be a
limiting nutrient among infants in industrialized countries who receive
adequate protein.
Complementary foods must provide relatively large proportions of
iron, zinc, phosphorus, magnesium, calcium, and vitamin B6. In the case
of iron, complementary foods are expected to provide nearly all of the
daily requirement because the amount in human milk is low (although
what is present is well absorbed). The values for iron shown in Table 1
are based on the assumption that the complementary foods provided
have high iron bioavailability, which means generous amounts of meat,
fish, poultry, and foods rich in vitamin C. When this is not the case
(for example, among vegetarians), the amount of iron needed from
complementary foods is even higher (as shown in the 1998 report).
For many nutrients, the values in Table 1 should be taken as an
approximation, given the level of uncertainty about nutrient require-
ments during infancy. Nonetheless, they provide some insight into the
 NUTRITION, G R O F , AND COMPLEMENTARY FEEDING OF BREASTFED INFANTS 91

contribution of human milk to nutrient needs and the nutrients that

may be deficient in the diets of infants and children who are breastfed.
The last section of this article returns to the latter question.

GROWTH PAlTERNS OF BREASTFED INFANTS

Assessing the growth of the breastfed infant is a common method

for advising when to introduce complementary foods and for evaluating
the adequacy of the diet; however, the conclusions drawn from plotting
the growth of a breastfed infant on a standard growth chart may be
erroneous if the growth chart does not adequately reflect normal growth.
For this reason, it is useful to review this issue herein.
The most widely used growth reference data have been the growth
charts that were based on data compiled by the US. National Center
for Health Statistics (NCHS) in 1977 and subsequently adopted for
international use by the WHO. For children less than 24 months of age,
the 1977 NCHS charts were based on data from a single community
(Yellow Springs, OH), collected between 1929 and 1975 as part of the
Fels Longitudinal Study. These reference data have several limitations:
(1) the sample was relatively homogeneous with regard to ethnic and
racial background; (2) measurements were made only every 3 months,
which is too infrequent to adequately characterize the pattern of growth
in early infancy; (3) technical problems occurred related to variable
sample sizes and outdated curve-fitting procedures; and (4) few infants
in the Fels study were breastfed for more than 3 months. When growth
of a breastfed infant is compared with the 1977 NCHS growth chart, the
typical pattern observed is a relatively rapid gain in the first 2 or 3
months, followed by a downward trend in percentile ranking thereaf-
ter.20,21
The downward trend after the first few months leads some health
care providers to conclude that growth faltering has occured and to
suggest that human milk intake is inadequate. This suggestion may lead
to the premature introduction of supplementary formula or complemen-
tary foods.
Several studies have demonstrated that the growth pattern men-
tioned above is found consistently among breastfed infants in various
countries.21,33, 4o In general, greater deviation exists in weight for age
than in length for age, and, as a result, breastfed infants are often lower
than average in weight for length by 12 months of age. Part of the
discrepancy between the growth patterns of breastfed infants and the
1977 NCHS growth chart is caused by the technical problems mentioned
earlier, particularly the curve-fitting procedures used for early infancy.
Even for formula-fed infants, the 1977 growth chart does not accurately
represent the observed pattern of growth in the first few months. Some
of these technical problems have been rectified by NCHS, and new
growth charts were released in May 2000, which are based primarily on
national child growth surveys; however, the national survey data include
relatively few infants who were breastfed for more than a few months.
92 DEWEY

13
9P
95th
12 90th

11 75th

50th
10
25”

-s
cn
9

a
10th
5m
3rd

E
.-cn
s 7

2
0 1 2 3 4 5 6 7 8 9 1 0 1 1 12

Figure 1. Mean weight for age of male (A) and female (6)infants who were breastfed for
at least 12 months, from a pooled data set from seven studies in North America and
northern Europe,2i, 39 plotted against the revised NCHS growth reference published in
May, 2000.
Illustration continued on opposite page

Notable differences still exist in later infancy (9-12mo) between the

average weight (although not length) of infants still breastfed and the
50th percentile of the new NCHS growth charts (Figs. 1 and 2).
Regardless of the growth chart used, considerable evidence shows
differences in growth rates of breastfed and formula-fed infants. A
review of 19 studies that directly compared growth of breastfed and
formula-fed infants in affluent populations15concluded that nearly all of
the studies showed greater weight gain among the latter than among
the former. The difference in average attained weight at 12 months was
approximately 600 g to 650 g. The results for length gain were much
less consistent than those for weight gain, with more than half of the
studies showing no significant differences by feeding mode. Because the
feeding-mode differences in weight were much greater than were the
differences in length, breastfed infants in most studies were lower in
weight for length and other indices of fatness than were formula-fed
 NUTRITION, GROWTH, AND COMPLEMENTARY FEEDING OF BREASTFED INFANTS 93

13

12
97th
95*
11 90m

75m
10
50m
9 25*
10"
8 5'h
3m

2
0 1 2 3 4 5 6 7 8 9 1 0 1 1 12

B
Figure 1 (Continued).

infants. The difference in fatness is most evident at 9 to 12 months.19

There is no evidence of any long-term difference in adult stature caused
by feeding mode in infancy, but whether breastfeeding is protective
against obesity later in life is contr~versial.~
The finding that breastfed infants do not show maximal growth has
raised questions regarding the adequacy of human milk intake or nutri-
ent content; however, in affluent populations, human milk intake is not
associated with maternal characteristics, such as nutritional status, but
rather is determined primarily by infant demand.17,30 Also, the most
striking deviation in growth relative to formula-fed infants occurs during
the latter part of infancy, when breastfed infants are receiving comple-
mentary foods. If these infants were receiving insufficient breast milk,
they presumably could compensate for this lack by consuming more
energy from other foods, but the data indicate that breastfed infants
regularly leave a portion of the complementary food offered to them
unconsumed. These and other data suggest that breastfed infants self-
regulate their energy intake at a lower level than consumed by formula-
fed i n f a r ~ t s .The
~ , ~reasons
~ for this difference are unclear, but it is known
94 DEWEY

85

9P
95th
80 90th
751b
50ih
75
25Ih
10th
5m
70 3d
h

v6
5 65
m
I
=
a,
-1
60

55

50
..
. ._.

45
0 1 2 3 4 5 6 7 8 9 1 0 1 1 1 2

A Age (mo)

Figure 2. Mean length for age of male (A) and female (B) infants who were breastfed for
at least 12 months, from a pooled dataset from seven studies in North America and
northern Europe,*’. 39 plotted against the revised NCHS growth reference published in
May, 2000.
Illustration continued on opposite page

that body temperature and minimal observable metabolic rate are lower
in breastfed than in formula-fed infants.6 It could be argued that the
greater weight gain of formula-fed infants represents excessive growth,
given that no deleterious functional outcomes are associated with the
slower weight gain of breastfed infants.ls Breastfed infants typically
exhibit less morbidity and superior cognitive development compared
with formula-fed infants.26It is possible, however, that the combination
of human milk and complementary foods typically fed (even in affluent
populations) may be inadequate in certain nutrients, which could lead
to suboptimal growth in some infants. At present there is no evidence
to indicate that growth patterns of breastfed infants in industrialized
countries are associated with complementary feeding practices, but this
issue merits further research.
The health advantages associated with breastfeeding have led many
to consider the growth pattern of the breastfed infant as the biological
 NUTRITION, GROWTH: AND COMPLEMENTARYFEEDING OF BREASTFEDINFANTS 95

85

80 97th
95th
90m
75"
75
50th
25m
70 10th
5*
3d

5
0
65
C

60

55

50

45 v I I I 1 I I I I 1 1 . 2
0 1 2 3 4 5 6 7 8 9 1 0 1 1 1 2

B Age (mo)

Figure 2 (Continued).

norm. This consideration is the rationale behind the current effort by

WHO to develop a new international growth reference based on growth
of healthy infants breastfed throughout the first year of life.12This new
growth chart is expected to be available within 4 years. In the meantime,
practitioners who work with breastfed infants may wish to consult a
previous report published by WHO that includes growth charts devel-
oped from a pooled data set of breastfed infants from seven studies in
North America and northern Europe.39These charts are not intended for
widespread use because they are based on a limited sample of infants
with relatively high birth weights and from predominantly white back-
grounds; however, they may be useful as an adjunct to other information
in cases of suspected growth faltering.

AGE OF INTRODUCTION OF COMPLEMENTARY

FOODS

There has been considerable debate regarding the optimal duration

of exclusive breastfeeding. Nearly all authorities agree that a minimum
96 DEWEY

of 4 months of exclusive breastfeeding is desirable and that continuing

much beyond 6 months is neither practical nor desirable in most popula-
tions, but disagreement exists about the exact wording of what to recom-
mend between 4 and 6 months. Since 1979, WHO has stated that “all
infants should be fed exclusively on breast milk from birth to 4 to 6
months of age.”l An age range of 4 to 6 months was thought to be
advisable to allow breastfed infants to adjust to consuming complemen-
tary foods; however, some investigators have suggested that the wording
of this recommendation is confusing and prompts parents to begin
providing complementary foods at (or before) 4 months. In 1992 and
1994, the wording of the feeding recommendations from the World
Health Assembly included statements that appropriate complementary
feeding should be fostered ”from the age of about six months,” wording
that UNICEF adopted for use in its publications. For many years the
American Academy of Pediatrics used the wording ”4 to 6 months,” but
in 1997 the statement on breastfeeding stated that “exclusive breastfeed-
ing is ideal nutrition and sufficient to support optimal growth and
development for approximately the first 6 months after birth.”2 This
section briefly reviews the scientific evidence regarding the recommen-
dation to exclusively breastfeed for “about six months.”
The nutrient needs of full-term, normal birth weight infants typi-
cally can be met by human milk alone for the first 6 months if the
mother is well nourished; however, in some circumstances, some of the
micronutrients may become limiting before 6 months. In the case of iron,
for example, the reserves present at birth are a critical factor determining
the risk for anemia during infancy because the iron concentration of
human milk is low. In normal birth weight infants, who typically have
adequate liver iron reserves, the risk for anemia is small with exclusive
breastfeeding before 9 months of age, although biochemical indices of
low iron status may occur in some infants between 6 and 9 months.25
The risk for iron deficiency is much greater in low birth weight infants
because their iron stores at birth are much smaller. In Honduras, infants
with birth weights of less than 3000 g were at risk for anemia at 6 months
even if they received iron-fortified complementary foods between ages
4 and 6 monthsz It is advised-by the WHO, the Institute of Medicine,
and other agencies-that low birth weight infants receive medicinal iron
drops beginning at 2 or 3 months of age. Iron drops also may be a more
effective alternative than complementary foods for breastfed infants with
birth weights of 2500 g to 3000 g.
The zinc concentration of human milk is also relatively low, al-
though its bioavailability is high. Low hepatic stores of zinc at birth may
be a factor predisposing some infants to zinc deficiency;l although this
deficiency is much less recognized than the situation with iron. Term, low
birth weight infants who were fed both human milk and infant formula
demonstrated more rapid growth when given zinc supplements, but
infants who were exclusively breastfed grew well regardless of whether
they were given zinc? In Honduras, breastfed infants who were given
complementary foods between 4 and 6 months of age received twice
 NUTRITION, GROWTH, AND COMPLEMENTARY FEEDING OF BREASTFED INFANTS 97

as much zinc as did those who were exclusively breastfed (and zinc
bioavailability of the foods should have been satisfactory based on their
low phytate to zinc molar ratio), but no significant difference was found
between groups in weight or length gain9 Thus, although zinc intake of
breastfed infants from 6 to 12 months of age may be deficient in some
populations, little evidence shows that zinc deficiency is problematic
among exclusively breastfed infants before 6 months of age.
In some circumstances, exclusively breastfed infants may have low
intakes of certain vitamins: (1) vitamin D deficiency may occur among
infants who do not receive much exposure to sunlight, (2) signs of
vitamin B,, deficiency may be exhibited in infants of mothers who are
complete vegetarians, (3) vitamin A status may be low among infants
whose mothers’ vitamin A reserves are depleted, and (4)riboflavin and
vitamin B, status may be low in infants of mothers with low intakes of
these vitamins. Otherwise, vitamin deficiencies are unlikely in exclu-
sively breastfed infants during the first 6 months of life. In all of these
situations, improving the mother’s diet or giving her supplements (for
vitamin A, B,, B12, and riboflavin) or giving vitamin drops directly to the
infant (for vitamin D) is the recommended treatment, rather than provid-
ing complementary foods to the infant.
Does any evidence show that introducing complementary foods
before age 6 months is beneficial? One way to evaluate this question is
to compare growth rates at 4 to 6 months between exclusively and
partially breastfed infants. In a pooled sample of breastfed infants from
six industrialized countries,2l those given only breast milk (n = 200)
gained a similar amount of weight and length from 4 to 6 months as
did infants given solid foods in addition to breast milk (n = 122).4In
developing countries, weight and length gain of infants exclusively or
predominantly breastfed between 4 and 6 months of age were similar to
or greater than those of partially breastfed infants given complementary
foods4; however, the lack of random assignment to groups in such
studies makes interpretation of the results
Only two randomized trials relevant to this question have been
conducted, both in Honduras. In the first low-income primipa-
rous mothers who had exclusively breastfed for 4 months were randomly
assigned to one of three groups: (1) continued exclusive breastfeeding
to 6 months (EBF); (2) introduction of complementary foods at 4 months
with ad libitum nursing 4 to 6 months (SF); and (3) introduction of
complementary foods at 4 months, with maintenance of baseline nursing
frequency (SF-M). Commercially prepared, nutritionally adequate baby
foods were provided to the SF and SF-M groups from 4 to 6 months.
Introduction of complementary foods caused breast milk intake to de-
crease, even in the SF-M group; as a result, no significant differences in
energy intake or growth in weight or length were found among groups,
nor was any long-term effect of the intervention on growth or infant
food acceptance found.
In the second study,16mothers of infants weighing 1500 to 2500 g at
birth and with a gestational age of 37 weeks or more were recruited in
98 DEWEY

hospital and assisted with exclusive breastfeeding during the first 4

months. At 4 months, they were randomly assigned to one of two
groups: (1) continued exclusive breastfeeding to 6 months (EBF) or (2)
complementary feeding plus breastfeeding from 4 to 6 months, with
mothers encouraged to maintain baseline nursing frequency (SF). Baby
foods in jars (i.e., rice cereal, chicken, fruits, vegetables) were fed twice
daily to all subjects (59 EBF, 60 SF). At 4 and 6 months, breast milk and
total energy intake were measured for a nonrandom subsample (those
who could stay overnight in a central unit: 32 EBF, 31 SF). Despite
significant displacement of breast milk by solid foods, change in total
energy intake from 4 to 6 months was significantly greater in the SF
than the EBF group in the subsample; however, no significant differences
in weight or length gain during the intervention were found between
groups, nor were any longer term (4-12 mo) growth differences observed.
The results of these two studies suggest that complementary foods
generally displace breast milk between 4 and 6 months and that intro-
ducing them at this age has no growth advantage, even in small-for-
gestational age infants. Strong evidence shows that complementary feed-
ing before 6 months is linked with increased morbidity in disadvantaged
p~pulations.~, 36 The risk for diarrhea in such populations is 2-fold to
13-fold higher when breastfed infants are given complementary foods
between 4 and 6 months than when they are exclusively breastfed. For
these reasons, the authors of the 1998 report concluded that the evidence
supports the recommendation that infants be exclusively breastfed for
approximately 6 months.

MEETING NUTRIENT NEEDS DURING THE PERIOD OF

COMPLEMENTARY FEEDING

Table 1 illustrates that breastfed infants need considerable amounts

of certain vitamins and minerals (particularly iron, zinc, phosphorus,
magnesium, calcium, and vitamin B6) from complementary foods after 6
months of age. How likely is it that the foods they are given will satisfy
these requirements? One way to estimate dietary quality is to calculate
the nutrient density of the complementary foods typically offered (i.e.,
amount of nutrient per 420 kJ) and to compare it with the desired nutrient
density. Desired nutrient density can be calculated by dividing the
amount of each nutrient needed from complementary foods by the
amount of energy expected to come from complementary foods (as
shown in Table 1).Table 2 shows these comparisons based on dietary
intake of several selected nutrients by 46 infants in California who were
breastfed for at least 12 Every 3 months, 4-day weighed
records of all foods and liquids consumed by the infant were completed.
This population was of generally high socioeconomic status and high
maternal educational levels. Table 2 illustrates that protein needs were
easily met in this population: mean protein density of the complemen-
tary food diet was twofold to threefold greater than required. Average
Table 2. NUTRIENT DENSITIES (PER 420 KJ [lo0KCALI) OF DIETS CONSUMED BY BREASTFED INFANTS IN CALIFORNIA
6-8 mo 9-11 mo

Average Mean Median Minimum Maximum Average Mean Median Minimum Maximum
Desired Observed Observed Observed Observed Desired Observed Observed Observed Observed
Protein (g) 1.0 2.6 2.3 0.9 6.4 * 0.9 3.3 3.1 0.4 6.8
Calcium (mg) 169 73 59 8 233 83 56 53 14 193
Iron* (mg) 3.5 3.5 2.7 0.3 11.0 1.5 1.9 1.4 0.4 6.1
Zinc (mg) 1.1 0.4 0.4 0.1 1.0 0.5 0.4 0.4 0.1 0.9
Riboflavin (mg) 0.08 0.20 0.20 0.03 0.45 0.05 0.13 0.12 0.05 0.30
Thiamin (mg) 0.03 0.14 0.11 0.03 0.38 0.04 0.11 0.10 0.04 0.29
Niacin (mg) 1.5 1.5 1.4 0.2 3.8 0.9 1.3 1.1 0.3 3.0
Vitamin B6 0.12 0.10 0.09 0.02 0.16 0.08 0.10 0.10 0.04 0.24

*Assuming high bioavailability of iron.

Data from Heinig MJ, Nommsen LA, Peerson JM, et al: Energy and protein intakes of breast-fed and formula-fed infants during the first year of life and their association
with growth velocity: The DARLING study. Am J Clin Nutr 58152-161,1993; with permission.
vitamin content of the diet was generally satisfactory, as shown in the
table for riboflavin, thiamin, niacin, and vitamin B, (although the average
intake of vitamin B, was slightly below the desired density at 6-8 mo).
The situation for iron, zinc, and calcium is more problematic. At 6 to 8
months, the mean iron density of the complementary food diet just
barely met the average desired density (the median did not). At 9 to 11
months, the desired iron density is considerably lower than at 6 to
8 months because infants are expected to obtain more energy from
complementary foods; as a result, the mean iron density was slightly
more than the desired density. Most of the iron consumed came from
iron-fortified infant cereals; without such products, the mean nutrient
density of the diet would have been considerably lower than the desired
density. This is the best case scenario because it is assumed that the
foods had high iron bioavailability. With regard to zinc, the mean nutri-
ent density was approximately 0.4 mg/420 kJ at both ages, which was
much less than the desired density at 6 to 8 months and slightly less
than it at 9 to 11 months. For calcium, the mean density was much less
than the desired density at both ages.
This comparison illustrates that the nutrients most likely to be
deficient in the diets of breastfed infants are minerals; however, the
requirements for some of these nutrients during infancy are uncertain.
For example, the most recent revision of the US Dietary Reference
Intakes lists a value of 270 mg/d as the Adequate Intake of calcium for
infants 6 to 12 months of which is considerably less than the value
of 525 mg/d taken as the requirement for Table 1. In this case, however,
the adequate intake value is based not on any functional outcomes but
only on the observed intakes of breastfed infants in the United States,
with the assumption that these infants typically are well nourished.
Insufficient information is available on the level of calcium intake that
would be optimal during infancy with regard to subsequent bone den-
sity or other outcomes. The revised Dietary Reference Intakes for iron
and zinc are not yet published. Iron requirements during infancy are
relatively well understood and are based on factorial estimates of the
needs for growth and maintenance, so the new estimates are unlikely to
change much. Zinc requirements are not as well understood, but the
values used for Table 1 are based on the most recent metabolic data
available and are already well below the previous Recommended Di-
etary Allowance (RDA) of 5mg/d at 6 to 12 months of age. Thus, one
may conclude that, at least for iron and zinc, it is difficult to meet
requirements based on the typical complementary foods given to breast-
fed infants at 6 to 12 months of age.
How is it possible that human infants survived throughout our
evolutionary past if it is so difficult to meet micronutrient needs through
complementary foods? Based on dietary patterns of hunter-gatherers
and of nonhuman primates, paleoanthropologists believe that the diet
of humans during most of our evolutionary history was extraordinarily
rich in micronutrients from game, fish, nuts, and wild plant foods.
Infants presumably were fed premasticated foods typical of the overall
 NUTRITION, GROWTH, AND COMPLEMENTARY FEEDING OF BREASTFED INFANTS 101

diet. Only during the past 10,000-15,000 years, since the Agricultural
Revolution, have humans in most parts of the world subsisted on a
largely grain-based diet, which provides meager quantities of bioavaila-
ble micronutrients.
What guidelines can be offered to parents with regard to meeting
nutritional needs of breastfed infants during complementary feeding
(e.g., 6-24mo)? The first is to continue to breastfeed as often as the
infant desires, if possible, which is important to avoid excessive displace-
ment of breast milk by other foods and to ensure that the nutritional
and immunologic benefits of breastfeeding are maximized. As shown in
Table 1, human milk is an excellent source of protein, fat, and most
vitamins and can make a valuable nutritional contribution well beyond
the first year of life. The second is to aim for for a variety of complemen-
tary foods, with fruits, vegetables, and animal products (e.g., meat, fish,
poultry, or egg) offered daily. Iron-fortified infant cereals are a good
source of iron, but meats also can provide adequate iron if consumed in
sufficiently large quantities, and they have the added advantage of being
rich in zinc. Adequate calcium can be obtained from cheese, yogurt, and
other dairy products (although fresh bovine milk is not recommended
before 12 mo). Although some vegetables (e.g., spinach) are relatively
high in calcium, the bioavailability from such sources is questionable,
and the quantities required to meet calcium needs are unrealistic. Parents
should avoid giving too much juice to infants (no more than 120 mL/d
before 12mo, and no more than 240mL/d thereafter) because they
typically provide little more than calories and can displace more nutrient-
dense foods. The third guideline is to be alert to any signs that an
infant’s appetite, growth, or development is impaired. These are often
the first indicators of subtle nutritional deficits (e.g., zinc deficiency).
When an infant refuses to eat a varied diet or eats little or no animal
products (e.g., if the parents are vegetarians), a balanced vitamin-
mineral supplement is advisable. The standard chewable formulations
are usually well accepted and can be crushed into a powder and mixed
with foods if necessary. Lastly, parents should make mealtimes enjoyable
and avoid force-feeding or coercion. Modeling enjoyment of a varied,
nutritious diet is the best way to ensure good dietary practices through-
out childhood and beyond.

SUMMARY

Although additional research is needed on many of the issues dis-

cussed herein, the following conclusions are well substantiated by the
evidence available to date:
Breast milk alone can meet nutrient needs during the first 6
months, with the possible exception of vitamin D in certain popu-
lations and iron in infants of relatively low birth weight. Comple-
mentary foods offered before 6 months of age tend to displace
102 DEWEY

breast milk and do not confer any growth advantage over exclu-
sive breastfeeding.
Breast milk continues to provide substantial amounts of key nutri-
ents well beyond the first year of life, especially protein, fat, and
most vitamins.
Breastfed infants tend to gain less weight and usually are leaner
than are formula-fed infants in the second half of infancy. This
difference does not seem to be the result of nutritional deficits but
rather infant self-regulation of energy intake. New growth charts
based on infants breastfed throughout the first year of life are
being developed by WHO.
The nutrients most likely to be limiting in the diets of breastfed
infants are minerals, such as iron, zinc, and calcium. Using the
following guidelines can help to ensure that the nutrient needs of
the breastfed child are met:
Continue to breastfeed as often as the infant desires.
Aim for a variety of complementary foods, with fruits, vegeta-
bles, and animal products (e.g., meat, fish, poultry, or egg)
offered daily. Iron-fortified cereals and meats can provide
adequate iron. Calcium can be obtained from cheese, yogurt,
and other dairy products (although fresh cow’s milk is not
recommended before 12 mo). Avoid giving too much juice.
Be alert to any signs that the child’s appetite, growth, or
development is impaired. When in doubt, a balanced vitamin-
mineral supplement is advisable.
Make mealtimes enjoyable.

References
1. Akre J: Infant feeding: The physiological basis. Bull WHO 67(suppl):1-108,1989
2. American Academy of Pediatrics Work Group on Breastfeeding: Breastfeeding and the
Use of Human Milk. Pediatrics 1001035-1039, 1997
3. Brown KH, Black RE, Lopez de Romaiia G , et a1 Infant feeding practices and their
relationship with diarrheal and other diseases. Pediatrics 83:31-40,1989
4. Brown KH, Dewey KG, Allen LH: Complementary feeding of young children in
developing countries: a review of current scientific knowledge. World Health Organiza-
tion, Geneva WHO/NUT/98.1,1998
5. Butte NF The role of breastfeeding in obesity. Pediatr Clin North Am 48:189-198,2001
6. Butte NF, OBrian Smith E, Garza C Energy utilization of breast-fed and formula fed
infants. Am J Clin Nutr 51:350-358, 1990
7. Butte NF: Energy requirements of infants. Eur J Clin Nutr 5O(suppl):24-36, 1996
8. Castillo-Duran C, Rodriguez A, Venegas G, et a1 Zinc supplementation and growth of
infants born small for gestational age. J Pediatr 127206-211,1995
9. Cohen RJ, Brown KH, Canahuati J, et a1 Effects of age of introduction of complemen-
tary foods on infant breast milk intake, total energy intake, and growth: A randomised
intervention study in Honduras. Lancet 344288-293,1994
10. Cohen RJ, Brown KH, Canahuati J, et al: Determinants of growth from birth to 12
months among breast-fed Honduran infants in relation to age of introduction of
complementary foods. Pediatrics 96504-510, 1995
11. Cohen RJ, Landa Rivera L, Canahuati J, et al: Delaying the introduction of complemen-
 NUTRITION, GROWTH AND COMPLEMENTARY FEEDING OF BREASTFED INFANTS 103

tary foods until 6 months does not affect appetite or mother’s report of food acceptance
of breastfed infants from 6 to 12 months in a low-income, Honduran population. J
Nutr 1252787-2792, 1995
12. de Onis M, Garza C, Habicht J P Time for a new growth reference. Pediatrics 100e8,
1998
13. Department of Health, Report on Health and Social Subjects: Dietary reference values
for food energy and nutrients for the United Kingdom. London, Her Majesty’s Statio-
nery Office, 1991
14. Dewey KG, Ekaton G, Fjeld C, et a1 Protein requirements of infants and children.
Proceedings of the International Dietary Energy Consultative Group. Eur J Clin Nutr
l(~~ppl):119-150, 1996
15. Dewey KG: Growth characteristics of breast-fed compared to formula-fed infants. Biol
Neonate 74396105, 1998
16. Dewey K G Cohen RJ, Brown KH, et a1 Age of introduction of complementary food
and growth of term, low birth weight breastfed infants: A randomized intervention
study in Honduras. Am J Clin Nutr 69:679486,1999
17. Dewey KG, Heinig J, Nommsen LA, et a1 Maternal versus infant factors related
to breast milk intake and residual milk volume: The DARLING study. Pediatrics
1987829-837, 1991
18. Dewey KG, Heinig MJ, Nommsen LA, et al: Adequacy of energy intake among breast-
fed infants in the DARLING study: Relationships to growth velocity, morbidity, and
activity levels. J Pediatr 119538-547, 1991
19. Dewey KG, Heinig MJ, Nommsen LA, et al: Breast-fed infants are leaner than formula-
fed infants at 1 y of age: The DARLING study. Am J Clin Nutr 57140-145,1993
20. Dewey KG, Heinig MJ, Nommsen LA, et al: Growth of breast-fed and formula-fed
infants from 0 to 18 months: The DARLING study. Pediatrics 89:1035-1041,1992
21. Dewey KG, Peerson JM, Brown KH, et a 1 WHO Working Group on Infant Growth
Growth of breast fed infants deviates from current reference data: A pooled analysis
of US, Canadian, and European data sets. Pediatrics 96:495-503, 1995
22. Dewey KG, Cohen RJ, Landa Rivera L, et a1 Effects of age of introduction of comple-
mentary foods on iron status of breastfed infants in Honduras. Am J Clin Nutr
678784384,1998
23. Food and Agriculture Organization (FAO), World Health Organization (WHO): Re-
quirements of vitamin A, iron, folate and vitamin BI2. Rome, Food and Agriculture
Organization, 1988
24. Greer FR: Do breastfed infants need supplemental vitamins? Pediatr Clin North Am
48:401413, 2001
25. Griffin IJ: Iron and breastfeeding 2001 (in press)
26. Heinig MJ, Dewey K G Health advantages of breastfeeding for infants: A critical
review. Nutr Res Rev 989-110,1996
27. Heinig MJ, Nommsen LA, Peerson JM, et a1 Energy and protein intakes of breast-fed
and formula-fed infants during the first year of life and their association with growth
velocity: The DARLING study. Am J Clin Nutr 58152-161, 1993
28. Heinig MJ, Nommsen LA, Peerson JM, et a1 Intake and growth of breast-fed and
formula-fed infants in relation to the timing of introduction of complementary foods:
The DARLING study. Acta Paediatr 82999-1006,1993
29. Heird WC: The role of polyunsaturated fatty acids in term and preterm infants and
breastfeeding mothers. Pediatr Clin North Am 48:173-188,2001
30. Institute of Medicine: Nutrition During Lactation. Washington, DC,National Academy
Press, 1991
31. Institute of Medicine: Dietary Reference Intakes for Calcium, Phosphorus, Magnesium,
Vitamin D and Fluoride. National Academy Press, Washington, DC, 1999
32. Krebs NF, Reidinger CJ, Hartley S, et al: Zinc supplementation during lactation: Effects
on maternal status and milk zinc concentrations. Am J Clin Nutr 61:1030-1036,1995
33. Mo-Suwan L, Junjana C: Breast-feeding and infant growth in the first six months. J
Med Assoc Thailand 74386-390, 1991
34. Pathak A, Shah N, Tataria A Growth of exclusively breastfed infants. Indian Pediatrics
30:1291-1300, 1993
35. Picciano MF. Nutrient composition of human milk. Pediatr Clin North Am 4853-67,
2001
36. Popkin BM, Adair L, Akin JS, et al: Breast-feeding and diarrheal morbidity. Pediatrics
86874-882,1990 .
37. Schanler RJ: The use of human milk for premature infants. Pediatr Clin North Am
48:207-219,2001
38. Torun B, Davies PSW, Livingstone MBE, et ak Energy requirements and dietary energy
recommendations for children and adolescents 1 to 18 years old. Eur J Clin Nutr
50(~~ppl):S37-S80, 1996
39. WHO Working Group on Infant Growth, Nutrition Unit: An evaluation of infant
g r o w A summary of analyses performed in preparation for the WHO Expert Com-
mittee on "Physical Status: the use and interpretation of antluopometry." Doc WHO/
NUT/94.8. Geneva, World Health Organization, 1994
40. Yoneyama K, Nagata H, Asano H: Growth of Japanese breast-fed and bottle-fed infants
from birth to 20 months. Ann Hum Biol21:59748,1994
41. Zlotkin SH, Cherian MG. Hepatic metallothionein as a source of zinc and cysteine
during the first year of life. Pediatr Res 24:326-329, 1988

Address reprint requests to

Kathryn G. Dewey, PhD
Department of Nutrition
University of California, Davis
One Shields Avenue
Davis, CA 95616-8669

e-mail: kgdewey@ucdavis.edu