Human milk (HM) is universally accepted as the optimal nutrient for young infants: exclusive breast-feeding for around 6 months is considered to be a desirable goal (1). Some of the numerous advantages of breast-feeding over feeding formula are thought to be related to the presence of the principal long-chain polyunsaturated fatty acids (LCPUFAs), arachidonic acid (AA [C20:4n-6]), and docosahexaenoic acid (DHA [C22:6n-3]) in HM. However, contribution of AA and DHA to the fatty acid composition of HM exhibits considerable variability among both different populations (2,3) and the different periods of lactation (4,5).
The percentage contribution of AA and DHA to HM lipids was found to decrease with advancing duration of lactation in several studies studying small groups of lactating mothers (6–8). In our previous studies involving small groups of women investigated during the first month of lactation, we also found significant decreases in the percentage contribution of AA and DHA in mothers both of full-term (9) and preterm (10) infants. Although it was suggested that HM lipid contents increase during lactation and this may compensate for the declining percentages of AA and DHA (8), it is still questionable how stable the LCPUFA supply via breast-feeding is during the recommended 6 months of exclusive breast-feeding.
Clear delineation of time-related changes in the fatty acid composition of HM may contribute to better understanding of the role of AA and DHA in the development and well-being of the breast-fed infant. Therefore, we decided to measure fat content and fatty acid composition of HM in a sizeable cohort of lactating women investigated both at the sixth week and sixth month of lactation. Because the potentially untoward role of trans isomeric fatty acids has been discussed also in the perinatal period (11,12), we paid special attention to the analysis of HM trans fatty acid contents.
SUBJECTS AND METHODS
The data presented here originate from a large birth cohort study carried out in Germany. The detailed description of the study design has been published elsewhere (13). Briefly, women delivering their babies in the Department of Gynecology and Obstetrics at the University of Ulm participated in the study, with the exclusion of women with a baby of <32 gestational weeks, a baby of <2500 g birth weight, or a baby transferred to inpatient paediatric care immediately after delivery. Participation was voluntary and informed consent was obtained in each case. The study was approved by the ethics boards of the University of Ulm and of the physicians' boards of the states of Baden-Württemberg and Bavaria. Overall, 1066 women were included in this study (for ethnical distribution, see Table 1), of whom 462 (43.3%) still breast-fed both at the infants' ages of 6 weeks and 6 months. The vast majority of the still breast-feeding mothers (around 98%) provided samples at the sixth month of lactation.
Both at the sixth week and the sixth month of lactation, 10 mL HM was collected, immediately cooled, and frozen at −80°C within 24 hours. In the vast majority of cases (90%), milk samples were collected by a trained nurse visiting the women in their homes, whereas in rare cases milk samples were collected by the mothers themselves. Milk samples were collected from both breasts, mostly by manual expression before feeding, and in rare cases by means of a breast pump. Although every attempt was made to standardise the circumstances of milk collection, slight technical differences occurring during more than 900 visits to the homes of the mothers cannot be excluded.
Fat contents were measured by the creamatocrit method. The detailed description of analysis of fatty acid composition was published elsewhere (13). Briefly, fatty acid methyl esters were measured by high-resolution capillary gas-liquid chromatography with the use of a Finnigan 9001 chromatograph (Finnigan/Tremetrics Inc, Austin, TX) with split injection (1:15) and a flame ionisation detector. A 60-m cyanopropyl column (DB-23, J&W Scientific, Folsom, CA) was used. Data are reported for 28 fatty acids detected with chain lengths between 10 and 24 carbon atoms. Fatty acid values are presented in %weight/weight and as medians with the range of first-to-third quartile. We detected 3 trans isomers: trans hexadecenoic acid (C16:1n-7t), trans octadecenoic acid (C18:1n-7/9t), and trans octadecadienoic acid (C18:2n-6tt). The sum of trans fatty acids was calculated by summing up the values of these 3 trans isomers.
For statistical analysis we used SAS version 8, 1st ed (SAS Institute, Cary, NC). We carried out the Kruskal-Wallis test for the difference between the groups. The Wilcoxon paired test was used to detect the difference between the fatty acid composition of HM samples at the sixth week and the sixth month of lactation. Correlations between trans fatty acids and LCPUFAs were calculated as partial Spearman correlation coefficients, adjusted for nationality and place of birth. Results were considered statistically significant at P < 0.05.
Anthropometric data of the mothers investigated both at the sixth week and sixth month of lactation are seen in Table 1. The age and the height of the Turkish mothers were significantly lower than in the 3 other groups; no other differences were seen.
Fat content of milk samples were significantly higher at the sixth month than at the sixth week of lactation (Table 2). Values expressed as %weight/weight of C10:0, C20:0, and C24:0 were significantly lower at the sixth month of lactation than at the sixth week. In contrast, percentage values of the C12:0, C14:0, C16:0, and C22:0 fatty acids and the sum of the saturated fatty acids were significantly higher at the sixth month of lactation. Percentage values of 16- and 18-carbon cis monounsaturated fatty acids and the sum of cis monounsaturated fatty acids were significantly lower at the sixth month than at the sixth week of lactation, whereas percentage contributions of 20-, 22-, and 24-carbon monounsaturated fatty acids were significantly higher. Percentage values of trans octadecenoic acid (C18:1n-7/9t) and the sum of trans fatty acids significantly decreased between the sixth week and the sixth month of lactation.
Values of the n-3 essential fatty acid, α-linolenic acid (ALA [C18:3n-3]), and the n-6 essential fatty acid, linoleic acid (LA [C18:2n-6]), increased significantly between the sixth week and sixth month of lactation. Similarly, percentage values of eicosatrienoic acid (C20:3n-3), eicosapentaenoic acid (C20:5n-3), docosapentaenoic acid (C22:5n-3), and the most important n-3 metabolite, DHA, were significantly higher at the sixth month than at the sixth week of lactation. With the advancement of lactation, we also observed significant increases in γ-linolenic acid (C18:3n-6) and eicosadienoic acid (C20:2n-6) values, as well as in the most important n-6 metabolite, AA and C22:4n-6.
Because of the significant difference in anthropometric parameters between subgroups, we statistically adjusted for nationality and place of birth at the correlation analysis. There were no significant inverse correlations between the 16-carbon trans isomer, trans hexadecenoic acid (C16:1n-7t), and the investigated n-3 and n-6 fatty acids (Table 3) (with the exception of C22:5n-3). In contrast, we found significant inverse correlations between C18:1n-7/9t and LA, C20:2n-6, dihomo-γ-linolenic acid (C20:3n-6), and AA. Similarly, values of ALA, C20:5n-3, and C22:5n-3 correlated significantly and inversely to C18:1n-7/9t. However, there were no significant negative correlations between the most important n-3 metabolite, DHA, and the investigated 18-carbon trans isomers. We found significant inverse correlations between C18:1n-7/9t and AA but not between C18:1n-7/9t and DHA. There were significant inverse correlations between the sum of trans isomeric fatty acids and LA, C20:3n-6, and AA (Table 3). Trans fatty acids correlated significantly and inversely to ALA and C22:5n-3, but not to DHA (Table 3). Trans isomeric fatty acids showed significant inverse correlations with n-6 polyunsaturated fatty acids and n-6 LCPUFAs (Table 3).
In the present study we compared fat contents and fatty acid composition of HM samples collected both at the sixth week and at the sixth month of lactation in the same lactating women participating in a large birth cohort study. We used sophisticated analytical methods that have been successfully applied by us in several previous studies (9,10,14); hence, the data obtained in the present study may offer reliable evaluation of the changes of lipid composition of HM during the period covering three fourths of the currently recommended duration of exclusive breast-feeding.
The considerable variability of our fatty acid compositional data may be seen as an apparent weakness of the present study; however, it can be assumed with good reason that the relatively broad interquartile ranges seen in the present study reflect real interindividual variability rather than biases in sample handling or laboratory analysis. First, the variability of data was most marked in the case of fatty acids that are more prone to short-term dietary influences, like eicosapentaenoic acid and DHA, whereas it was less marked in the case of saturated and pertinent cis monounsaturated fatty acids. Second, the variability of data was less pronounced at the sixth month than at the sixth week of lactation, which may be related to the increased stability of maternal diet with longer experiences with breast-feeding. Finally, in spite of the minuscule differences between medians and the large interquartile ranges, paired statistical analysis showed high level of significance in nearly all of the fatty acids analysed. This latter observation is in concert with the assumption of high degree of tracking of fatty acid profiles within the same mother (8).
The major biological finding of the present study is the significant increase of all n-3 and n-6 fatty acid values (with the exception of C20:3n-6) from the sixth week to the sixth month of lactation. It may be disputed whether some of the small differences seen (eg, in AA values or in the sum of n-6 LCPUFAs) are biologically relevant; nevertheless, median DHA values and the sum of n-3 LCPUFAs increased by about 35% and 55%, respectively.
The present data are in some opposition to the general view that the percentage contribution of LCPUFAs to HM lipids decreases with increasing duration of breast-feeding. Indeed, several studies investigated the fatty acid composition of mature HM at different points of lactation and reported increasing levels of the essential fatty acids but decreasing values of the most important LCPUFAs, AA, and DHA (7,8,15–17). Agostoni et al (8) carried out a longitudinal study analysing HM samples obtained in the same Italian women recruited after delivery of full-term infants (n = 10) at the first, third, sixth, ninth, and 12th months of lactation. They found decreasing percentage contributions of LCPUFAs together with increasing milk fat contents and suggested that the secretion of AA and DHA remains stable during lactation. Our present data may allow speculating a step further: at least stable or increasing LCPUFA percentages together with the significant increase of total fat contents indicate that the LCPUFA intake of the breast-fed infant may remain stable throughout the period of exclusive breast-feeding, in spite of the obvious decrease of milk intake referred to units of body weight.
The major nutritional finding of the present study is the demonstration of significant inverse correlations between 18-carbon trans isomeric fatty acids and LA, ALA, C18:1n-7/9t, and AA at the sixth month of lactation. These findings are in accordance with previous reports describing significant inverse correlations between 18-carbon trans fatty acids and various PUFAs (including AA and DHA) in venous cord blood lipids in healthy infants with an atopic trait (18), in cord vessel wall lipids in healthy full-term infants (19), and in plasma lipids of healthy children ranging from 1 to 15 years of age (20). It is to be noted that this relation was seen at relatively low levels of the contribution of trans fatty acids to HM lipids. Although the HM samples analysed in the present study were collected a few years ago, we assume that similar inverse relations between trans fatty acids and LCPUFA may be characteristic to HM also today.
We found similar inverse associations between trans fatty acids and AA and DHA in the total pool of HM samples (n = 769) donated at the sixth week of lactation by the women participating in the present study (13). The weaker, albeit still significant inverse associations between trans fatty acids and LCPUFAs at the sixth month compared with the sixth week of gestation may be explained by the significant decrease of the contribution of trans octadecenoic acid to the fatty acid composition of HM at the sixth month compared with that at the sixth week of lactation. This finding is in interesting concert with our recent observation that the contribution of trans isomeric fatty acids to erythrocyte membrane phosphatidylcholine and phosphatidylethanolamine lipids was significantly higher in venous cord blood than in blood samples obtained in infancy (21). Decrease of the contribution of trans fatty acids to both HM and infantile lipids during the first months of life may be related to the decreased maternal dietary intakes of trans fatty acids during this period.
It is reasonable to assume that there should be some threshold level in the exposure to trans isomeric fatty acids that is needed to interfere with the availability of LCPUFAs. In the present study many of the inverse associations between trans isomeric fatty acid and LCPUFAs were still present at the sixth month of lactation, in spite of the significant decrease in trans fatty acid values between the sixth week and sixth month of lactation. This inverse association indicates that LCPUFA content of HM may benefit from reduced maternal exposure to trans fatty acids. Mojska et al (17) investigated the relation between the diet of lactating mothers and the fatty acid composition of their milk and found that trans isomeric fatty acid contents in breast milk samples was determined primarily by the maternal diet; hence, modification of maternal lipid intakes may be effective in reducing maternal trans fatty acid exposure. Decreasing perinatal trans isomeric fatty acid exposure may offer direct benefits to the infants because recent data indicate that trans isomeric fatty acids in umbilical vein and artery wall lipids were inversely related to neurologic optimality score determined at the age of 18 months (22).
In summary, we found increasing fat contents with stable or increasing percentage contribution of LCPUFAs in HM samples compared between the sixth week and at the sixth month of lactation. We speculate that these compositional changes may allow at least stable LCPUFA intake to the exclusively breast-fed infant throughout lactation; however, trans isomeric fatty acids may interfere with the availability of LCPUFAs even at the sixth month of lactation.
1. Agostoni C, Decsi T, Fewtrell M, et al
. Complementary feeding: a commentary by the ESPGHAN Committee on Nutrition. J Pediatr Gastroenterol Nutr 2008; 46:99–110.
2. Koletzko B, Thiel I, Abiodun PO. The fatty acid composition of human milk in Europe and Africa. J Pediatr 1992; 120:S62–S70.
3. Brenna JT, Varamini B, Jensen RG, et al
. Docosahexaenoic and arachidonic acid concentrations in human breast milk worldwide. Am J Clin Nutr 2007; 85:1457–1464.
4. Jensen RG. Lipids in human milk. Lipids 1999; 34:1243–1271.
5. Koletzko B, Rodriguez-Palmero M, Demmelmair H, et al
. Physiological aspects of human milk lipids. Early Hum Dev 2001; 65:S3–S18.
6. Guesnet P, Antoine J, Rochette de Lempdes J, et al
. Polyunsaturated fatty acid composition of human milk in France: changes during the course of lactation and regional differences. Eur J Clin Nutr 1993; 47:700–710.
7. Luukkainen P, Salo MK, Nikkari T. Changes in the fatty acid composition of preterm and term human milk from 1 week to 6 months of lactation. J Pediatr Gastroenterol Nutr 1994; 18:355–360.
8. Agostoni C, Marangoni F, Lammardo AM, et al
. Long-chain polyunsaturated fatty acid concentrations in human hindmilk are constant throughout twelve months of lactation. Adv Exp Med Biol 2001; 501:157–161.
9. Minda H, Kovács A, Funke S, et al
. Changes of fatty acid composition of human milk during the first month of lactation: a day-to-day approach on the first week. Ann Nutr Metab 2004; 48:202–209.
10. Kovács A, Funke S, Marosvölgyi T, et al
. Fatty acids in early human milk after preterm and full-term delivery. J Pediatr Gastroenterol Nutr 2005; 41:454–459.
11. Carlson SE, Clandinin MT, Cook HW, et al
Fatty acids: infant and fetal development. Am J Clin Nutr 1997; 66:717S–736S.
12. Morrison JA, Glueck CJ, Wang P. Dietary trans
fatty acid intake is associated with increased fetal loss. Fertil Steril 2008; 90:385–390.
13. Szabó E, Boehm G, Beermann C, et al
Octadecenoic acid and trans
octadecadienoic acid are inversely related to long-chain polyunsaturates in human milk: results of a large birth cohort study. Am J Clin Nutr 2007; 85:1320–1326.
14. Decsi T, Oláh Sz, Molnár Sz, et al
. Low contribution of docosahexaenoic acid to the fatty acid composition of mature human milk in Hungary. Adv Exp Med Biol 2000; 478:413–414.
15. Gibson RA, Kneebone GM. Fatty acid composition of human colostrum and mature breast milk. Am J Clin Nutr 1981; 34:252–257.
16. Genzel-Boroviczény O, Wahle J, Koletzko B. Fatty acid composition of human milk during the 1st month after term and preterm delivery. Eur J Pediatr 1997; 156:142–147.
17. Mojska H, Socha P, Socha J, et al
fatty acids in human milk in Poland and their association with breastfeeding mothers' diets. Acta Paediatr 2003; 92:1381–1387.
18. Decsi T, Burus I, Molnár Sz, et al
. Inverse association between trans
isomeric and long-chain polyunsaturated fatty acids in cord blood lipids in full-term infants. Am J Clin Nutr 2001; 74:364–368.
19. Decsi T, Boehm G, Tjoonk R, et al
. 18-carbon trans
isomeric fatty acids are inversely related to arachidonic and docosahexaenoic acids and positively related to Mead acid in cord blood vessel wall lipids in full-term infants. Lipids 2002; 37:959–965.
20. Decsi T, Koletzko B. Do trans
fatty acids impair linoleic acid metabolism in healthy children? Ann Nutr Metab 1995; 39:36–41.
21. Jakobik V, Burus I, Decsi T. Fatty acid composition of erythrocyte membrane lipids from birth to young adulthood. Eur J Pediatr 2009; 168:41–147.
22. Bouwstra H, Dijck-Brouwer J, Decsi T, et al
. Neurological condition of healthy term infants at 18 months: positive association with venous umbilical DHA status and negative association with trans
-fatty acids. Pediatr Res 2006; 60:334–339.