Contents - Previous - Next

This is the old United Nations University website. Visit the new site at http://unu.edu

Vitamin A and breast-feeding

Vitamin A and breast-feeding: A comparison of data from developed and developing countries

Vicky Newman

Abstract

The vitamin A status of lactating women, its effect on the vitamin A content of human milk, and the adequacy of human milk as a source of vitamin A for the infant were assessed, comparing data from developing countries with those from developed countries. The vitamin A concentration in breast milk during the first two weeks of lactation is nearly double that at one month. It is even higher in preterm milk than in term milk during the first several months. Human milk alone provides sufficient vitamin A to prevent clinical deficiency throughout the first 12 months of life, even in presumably more poorly nourished populations in developing countries. However, it is not sufficient to allow liver storage after about six months of lactation.

Introduction

Vitamin A deficiency is a major public health problem in many developing countries, outnumbering all other causes of blindness in children [1]. It is estimated that it afflicts 7% of all children under the age of five years [2], and countless other children have vitamin A depletion, a condition associated with decreased resistance to infectious diseases and increased mortality [3, 4]. The World Health Organization (WHO) estimates that worldwide one child dies needlessly from vitamin A deficiency every minute [5]. For these reasons, elimination of the disorder by the year 2000 was targeted in the Declaration on Children endorsed by political leaders around the world at the World Summit for Children held in New York in September 1990.

Most newborn infants have a marginal vitamin A status, and those who are born early and/or whose mothers have inadequate vitamin A intakes appear to be at particular risk [6, 7]. In many growing children, reserves can maintain optimum levels for no more than a few weeks. This underscores the importance of adequate vitamin A in breast milk, and the fact that infants and young children, if not properly nourished, can rapidly develop the signs of deficiency.

Although the concentrations in human milk depend on the mother's vitamin A status, vitamin A deficiency is rare among breast-fed infants, even in parts of the world where the deficiency is endemic [8-11]. The protective effect appears to continue after breastfeeding is discontinued, presumably because some of the vitamin provided by human milk is stored in the child's liver [12].

To facilitate the integration of activities promoting breast-feeding with programmes to prevent vitamin A deficiency, I reviewed the relevant world literature from the last 40 years, compiling information on the vitamin A status of lactating women, its effect on the vitamin A content of human milk, and the adequacy of breast milk as a source of the vitamin and also assessing the impact of mothers' supplementation on the vitamin A content of breast milk and on the health of women and their infants [13].

Methods

There is wide variation in the terms used to describe vitamin A activity and in the units of measure used in published reports, making comparison among studies difficult. For example, vitamin A activity has been reported as vitamin A, as retinol, as carotene, as ß-carotene, and as carotenoids. Units of measure include micrograms, millimoles, and international units, and these are reported per millilitre, per decilitre, per litre, per cubic centimetre, per gram, and per day.

To facilitate comparison, all reports of vitamin A activity in the diet and in milk were converted to retinol equivalents (RE), which adjusts for the assumed biological effectiveness of the different forms of vitamin A (table 1). If not otherwise defined in the reports, vitamin A was assumed for conversion purposes to be retinol, and carotene was assumed to be ß-carotene. When the vitamin A content of human milk was presented as total carotenoids, only the portion assumed to be ß-carotene was converted to retinol equivalents, because ß-carotene is the most biologically active of the carotenoids [17]. Since ß-carotene reportedly makes up 25% of total carotenoids in human milk [17, 18], only 25% of total carotenoids were assumed to be ß-carotene for conversion purposes.

TABLE 1. Vitamin A conversion table

Adapted from refs. 14-16.

a. D. Finley, technical editor, American Journal of Clinical Nutrition, personal communication, 1992.

When the dietary intake of vitamin A was reported in international units (IU), it was assumed that 50% was from retinol and that ß-carotene made up all the provitamin A carotenoids in the other 50% [14, 15]. This results in a conversion factor of 1 RE for every 5 IU of dietary Vitamin A activity. This general assumption was considered valid for most regions of the world except Africa and Asia, where vitamin A from animal sources (retinol) is only about 20% and 15% respectively. Thus, the conversion factor used for Africa and Asia was 1 RE for every 7 IU of dietary vitamin A.

Assuming nutritional deprivation to be most likely in countries with high child mortality rates, data from countries with higher child mortality (developing) were compared with those from countries with lower rates (developed). "Developing" countries were defined as those with a mortality rate for children under five years old (U5MR) of 21 or over, and "developed" countries as those with a U5MR of 20 or under [19]. Although it might have been useful to divide developing countries further into those with moderate U5MR (21-70) and those with higher U5MR ( >=71), not enough studies on vitamin A during lactation were available from countries with moderate U5MRs to make meaningful comparisons (table 2).

Reported values for vitamin A in human milk were divided by the time after delivery during which the samples were obtained (1-6, 7-13, or 14-21 days; 1-2, 3-4, 5-6, 7-12,1324, or >24 months). Levels were divided further by whether the birth was term or preterm (i.e., <37 weeks' gestation).

Assessing vitamin A status

The vitamin A status of individuals or of populations can be assessed using a dietary, clinical, or biochemical approach; a combination of these is considered most accurate [20].

Dietary intake

Habitual dietary inadequacy is a useful early indicator of vitamin A depletion. Recommended dietary intakes (RDI), also called allowances, are used to measure dietary adequacy. The RDIs for vitamin A published by the FAO/WHO and the US National Research Council differ somewhat (table 3). The FAO/WHO recommendations use a two-tier system in which the basal amount corresponds to intake that prevents signs of deficiency and the safe amount corresponds to that with an appropriate liver reserve (20 µg/g). The suggested RDI in the United States corresponds to the FAO/WHO's safe amount. The different recommended amounts for adult women refer to the different weights of the reference persons, 55 kg for women in the FAO/WHO report and 62 kg in the US report [17].

TABLE 2. Countries listed by under-five child mortality rate (U5MR)

Adapted from ref. 19.

Clinical assessment

Clinical assessment includes documenting the presence or absence of functional signs of insufficiency. Early signs in humans are growth failure, loss of appetite, and impaired immune response with lowered resistance to infection [22]. Night blindness develops when liver reserves of the vitamin are nearly exhausted. Later, ocular lesions include conjunctival xerosis (abnormal dryness of the conjunctive), corneal xerosis (abnormal dryness of the cornea), Bitot's spots (triangular, shiny, grey spots on the conjunctive), and keratomalacia (irreversible corneal lesions associated with partial or total blindness) [22]. Collectively, these ocular lesions are referred to as xerophthalmia.

TABLE 3. Recommended dietary intakes of vitamin A in retinol equivalents (RE)

A retinol equivalent is defined as 1 µg of retinol or 6 µg of ß-carotene.

a. 6-10 years.
b. Extrapolated from refs. 16 and 21.

Biochemical measurements

Unlike dietary and clinical assessments, which are subjective, biochemical assessment has the advantage of being objective. Biochemical measurements used to assess vitamin A status include the vitamin's activity in blood, the liver, and breast milk. Because more than 90% of the vitamin A in the body is stored in the form of retinyl ester in the liver [15], a measure of liver stores is the best index [22]. Liver biopsies are impractical in population studies, however, and thus other measures are used.

Biochemical indicators of vitamin A status are shown in table 4. In this table, measurements in the deficient range correlate with the presence of clinical indicators of deficiency (e.g., night blindness, xerophthalmia), and those in the adequate range correlate with the absence of these indicators. Although blood plasma is not technically the same as serum, the terms are used interchangeably in this report, as they are generally in the articles reviewed. It is assumed that the indicators for preschool children are valid for infants, and that those for adults are valid for lactating women, as no separate indicators for infants and lactating women have been published to date.

TABLE 4. Biochemical indicators of vitamin A status

a. Values for children from refs.15 and 17; values from ref.17, published as µmol, have been multiplied by 286.44 (the molecular weight of retinol) to convert them to µg. Values for adults from ref.16.
b. Values for children extrapolated from ref. 23. High- and moderate-risk values for adults not available; low-risk value from ref. 22.
c. Values for children from refs. 15 and 17 (cf. note a above). Values for adults from ref.15.
d. Also ref.24.
e. These values have been assumed to apply to lactating women as no separate values for lactating women are currently available.
f. Also ref.25.

Guidelines for assessing vitamin A status, as well as information on additional measurements, such as the relative dose response test and conjunctival impression cytology, are discussed elsewhere [17, 22, 25].

Nutrient Interactions and other factors affecting vitamin A status

Vitamin A status can be affected by other nutrients and environmental and other factors. For example, it is dependent on an adequate dietary intake of nutrients such as fat, protein, vitamin E, and zinc. Dietary fibre appears to reduce the bioavailability of ß-carotene. Season affects the availability of many fruits and vegetables, and thus status in populations that depend on these foods for most of their vitamin A. Tobacco smoke, air pollution, and sunlight may affect vitamin A status by increasing the need for ß-carotene. Other important factors are stress, acute and chronic diseases, estrogens, age, sex, and race [22].

Vitamin A status of infants

Because of limited placental transfer, newborns usually have lower serum retinol and carotene levels than their mothers [26, 27]. Several studies reported that preterm infants of less than 36 weeks' gestation have lower plasma retinol and retinol-binding protein (RBP) values than term infants [24, 27, 28]. Blood levels of retinal fluctuate during the first week of life regardless of dietary intake, probably due to fluctuating levels of RBP [29]. Until adolescence, the levels are about half those of adults (J. A. Olson, personal communication, 1992).

The ß-carotene level in term and preterm infants at birth was only about one-eighth the level in maternal serum [26]. Human breast milk, particularly colostrum, contains very high concentrations. Thus the breast-fed term infant attains serum levels of ß-carotenes comparable to those in the adult within four to six days of breast-feeding [26]. Breast-fed infants in Canada had significantly higher serum carotene concentrations than those who were fed breast-milk substitutes during the first three months of life, probably because the substitutes did not contain carotene [30].

Liver stores of retinol are low during the early months of life, even in infants born at term. The preterm infant has virtually no such reserves and is at particular risk for vitamin A deficiency [7]. In wellnourished populations, the liver stores begin to increase between three and six months of age, and continue to rise through four years of age, whereas in poorly nourished populations the stores decrease throughout the first four years [31].

Continue

Contents - Previous - Next