Nutrition 30 (2014) 685–689
Contents lists available at ScienceDirect
Nutrition journal homepage: www.nutritionjrnl.com
Applied nutritional investigation
Dietary polyunsaturated fatty acid intake during late pregnancy affects fatty acid composition of mature breast milk Renata Y. Nishimura M.S. a, Patricia Barbieiri a, Gabriela S.F. de Castro b, Ph.D. d, ~o Jr. Ph.D. c, Gleici da Silva Castro Perdona Alceu A. Jorda d, * Daniela S. Sartorelli Ph.D. ~o Preto Medical School at the Sa ~o Paulo University, Sa ~o Paulo, Brazil Program in Community Health in the Ribeira ~o Preto Medical School at the Sa ~o Paulo University, Sa ~o Paulo, Brazil Program in Internal Medicine in the Ribeira ~o Preto Medical School at the Sa ~o Paulo University, Sa ~o Paulo, Brazil Department of Internal Medicine in the Ribeira d ~o Preto Medical School at the Sa ~o Paulo University, Sa ~o Paulo, Brazil Department of Social Medicine in the Ribeira a
b c
a r t i c l e i n f o
a b s t r a c t
Article history: Received 24 September 2013 Accepted 13 November 2013
Objective: The aim of this study was to investigate how maternal polyunsaturated fatty acid intake at different periods during pregnancy affects the composition of polyunsaturated fatty acids in mature human milk. Methods: A prospective study was conducted involving 45 pregnant women, aged between 18 and 35 y, who had full-term pregnancies and practiced exclusive or predominant breast-feeding. Mature breast milk samples were collected after the 5th postpartum week by manual expression; fatty acid composition was determined by gas chromatography. Fatty acid intake during pregnancy and puerperium was estimated through multiple 24-h dietary recalls. Linear regression models, adjusted by postpartum body mass index and deattenuated, were used to determine associations between estimated fatty acids in maternal diet during each trimester of pregnancy and fatty acid content in mature human milk. Results: A positive association was identified between maternal intake of eicosapentaenoic acid (b, 1.873; 95% confidence interval [CI], 0.545, 3.203) and docosahexaenoic acid (b, 0.464; 95% CI, 0.212–0.714) during the third trimester of pregnancy, as well as the maternal dietary u-3 to u-6 ratio (b, 0.093; 95% CI, 0.016–0.170) during the second and third trimesters and postpartum period, with these fatty acids content in mature breast milk. Conclusions: The maternal dietary docosahexaenoic acid and eicosapentaenoic acid content during late pregnancy may affect the fatty acid composition of mature breast milk. Additionally, the maternal dietary intake of u-3 to u-6 fatty acid ratio, during late pregnancy and the postpartum period, can affect the polyunsaturated fatty acid composition of breast milk. Ó 2014 Elsevier Inc. All rights reserved.
Keywords: Human milk Dietary intake Pregnancy Polyunsaturated fatty acids n-3 fatty acids n-6 fatty acids Trans fatty acids
Introduction The long-chain polyunsaturated fatty acids (LC-PUFA), docosahexaenoic acid (DHA, 22:6 u-3), eicosapentaenoic acid (EPA, 20:5 u-3), and arachidonic acid (ARA, 20:4 u-6), are of great importance during child growth and development [1]. The RYN and DSS were responsible for the study design, data collection, and analysis. RYN developed the first version of the manuscript. GSFC and AJJ were responsible for data analysis and manuscript review. DSS coordinated the study and reviewed the final manuscript version. All authors have registered curriculum in the Lattes platform from CNPq. There is no conflict of interest to be disclosed. * Corresponding author. Tel.: þ55 360 227 1216; fax: þ55 363 313 8616. E-mail address: [email protected] (D. S. Sartorelli). 0899-9007/$ - see front matter Ó 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.nut.2013.11.002
content of these fatty acids in human milk, especially DHA, varies widely [2,3], possibly as a result of diverse food consumption among women and because fatty acid content in maternal diets varies among regions. Clinical trials suggest that consumption of fatty fish and fish oil supplements during puerperium are associated with an increased content of EPA and DHA in breast milk. However, in studies that employed labeled isotopes, only 30% of the variability in polyunsaturated fatty acid content (LC-PUFA) in breast milk is explained by the estimated dietary intake during the puerperium [4]. Randomized controlled trials (RCTs) suggest a direct relation between the use of u-3 supplements during the second and third trimesters of pregnancy and its content in mature breast milk [5, 6]. However, results reporting the effect of this supplement
686
R. Y. Nishimura et al. / Nutrition 30 (2014) 685–689
during pregnancy on EPA and a-linolenic acid (ALA) concentrations are controversial [5,6]. Several studies using u-3 (DHA, EPA, and ALA) and u-6 (ARA and linoleic acid [LA]) fatty acid supplementation during gestational or postpartum periods have found that it has distinct effects on fatty acid content in breast milk [5–11]. Studies investigating the influence of dietary fatty acid during pregnancy on breast milk fatty acid composition are scarce and those that have actually investigated the issue have been limited to food consumption during the third trimester only [12,13]. The association between maternal intake of fatty acids and their content in human milk may differ according to the period of intake, given that according to the gestational stage, the rate by which they are transported by the placenta may vary and maternal metabolism may change as well [14]. We hypothesized that maternal PUFA intake might have distinct effects on the PUFA composition of breast milk depending on the period of maternal intake. This hypothesis has not been tested in previous studies. This study aimed to investigate how fatty acid intake during different pregnancy periods affects the composition of PUFA in mature breast milk. Materials and methods Study population A prospective study was conducted to test the accuracy of a quantitative food frequency questionnaire involving 103 pregnant women, attendees at the Basic Health Units in Ribeir~ ao Preto, S~ ao Paulo, Brazil, aged between 18 and 35 y, with prepregnancy body mass index (BMI) between 18.5 and 24.9 kg/m2, had no diseases that may alter usual food consumption (diabetes mellitus, hypertension). For the validation study, 24-h dietary recalls (24 hRs) were obtained in each trimester of pregnancy. Seventy-five of these women completed the validation study and information on food intake (evaluated by the 24 hRs) throughout their pregnancies were available. Those women were invited to participate in the present study. The inclusion criteria were: Women who practiced exclusive breast-feeding (exclusive breast milk intake, with no other liquid or solid food intake) or predominant breast-feeding (breast milk intake and other water-based liquid food intake) after the fifth postpartum week, and mothers with full-term pregnancies (birth after the 37th gestational week). The study was approved by the Municipal Health Secretary of Ribeir~ ao Preto de-Escola and approved by the Research Ethics Committee from the Centro de Sau ~o Paulo, Brazil (Protocol number 378/CEP-CSE/FMRP-USP). at FMRP, USP, Sa
Socio-demographic characteristics and lifestyle A structured questionnaire applied at the first interview during the prospective study (first trimester of pregnancy) was used to collect data on age, education, assets, and skin color. The economic classification of each participant was assessed according to the Brazilian Economic Classification Criterion (CCEB), which defines classes from A (highest socioeconomic level) to E (lowest socioeconomic level) [15].
Assessment of maternal BMI, weight gain, and gestational age The nutritional status was assessed using the body weight of each participant in each trimester during prenatal and postpartum care appointments ~o Paulo-SP, using an electronic digital scale (Plenna, model MEA 07700, Sa Brazil) with a capacity of 150 kg and 100 g gradation. Heights were measured using an anthropometric scale ruler (Cauduro, model 101 PL, Cachoeira do SulRS, Brazil). Gestational weight gain was calculated as the difference between the weight at the last prospective data collection (third trimester) and prepregnancy weight (reported by the participant). Maternal pregestational and postpartum BMI were obtained using the height measured at the first prenatal visit and self-reported pregestational weight and measured postpartum weight, respectively. Gestational age was calculated preferably based on the date of the last menstrual cycle on record, which was correlated to data from the ultrasound exam performed up to the twentieth gestational week.
Estimation of polyunsaturated fatty acid intake during pregnancy PUFA (u-3 fatty acids [DHA, EPA, and ALA] and u-6 fatty acids [ARA and LA]) intake during pregnancy were evaluated through three 24 hR applied throughout the pregnancy, one 24 hR in each trimester of pregnancy, and two 24 hRs during the postpartum period. A second 24 hR was applied in each trimester of pregnancy in a subsample of the participants to correct for intraindividual variability. The 24 hR were applied by trained nutritionists using the multiple pass technique [16]. The NutWin software (NutWin software, Nutrition Support Program, version ~o Paulo, Brazil, 2002) was used for the analysis 1.5, Escola Paulista de Medicina, Sa of PUFA content based on the 24 hR information, employing the Brazilian Food Composition Table-TACO [17], and the U.S. Department of Agriculture’s Food Composition Table [18]. Determination of fatty acids in breast milk Samples (5–10 mL) of mature breast milk (after postpartum week 5) were obtained by hand expression by the participant in the morning, immediately after the baby’s first feeding and before mother’s breakfast. Milk samples were stored at 80 C until analysis. Fatty acids were extracted from 800 mL of milk, to which methanol and chloroform (v/v, 1:1) were added according to a previously described method [19]. After lipid extraction, the chloroform phase was evaporated under nitrogen current and fatty acids were methylated with potassium hydroxide in methanol (0.5 M) and removed with hexane. One mL of fatty acid methyl esters was injected in a SHIMADZU GC-2014 gas chromatographer (Shimadzu Europe, Duisburg, Germany) equipped with AOC-20 i auto-injector (Shimadzu Europe, Duisburg, Germany) and separated with a polyethylene glycol-SUPLELCOWAX 30 m capillary column (30 m, 25 mm 0.25 mm; Supelco Inc., Bellefonte, PA, USA). Helium was used as the carrier gas at a flow rate of 1 mL/min. Synthetic air was used for flame ionization detection at 280 C. The injections were performed in split mode. The injector and detector temperatures were 250 C. The initial column temperature was maintained for 1 min at 100 C, followed by an increase of 13 C/min until reaching 195 C, when it was maintained for 5 min, and subsequently elevated to 240 C, at the rate of 15 C/min, and maintained at this temperature for 30 min. The identification standard used was composed of a mixture of fatty acid methyl esters from Supelco (Supelco 37 Component FAME Mix; Supelco Inc., Bellefonte, PA, USA) with the addition of 9 c, 11 t-CLA and 10 t, 12 c-CLA. The quantification was performed by area normalization and results presented in weight percentages. Data analysis Linear regression models were used employing the breast milk fatty acid content as the dependent variable and the maternal fatty acid intake, during each trimester of pregnancy and puerperium, as the independent variables. Univariate and multiple linear models were adjusted by postpartum BMI and deattenuation. Deattenuation consists of correction of the linear regression coefficient by the intra- and interpersonal variability error. It was calculated using the formula: Bt ¼ bo [1 þ (s intra/s inter)/n] [20], where bt is the true regression coefficient, bo is the observed regression coefficient, s intra and s inter are the intra- and interindividual variances, respectively, and n is the number of 24 hR replicates (in this case, n ¼ 2 per trimester). The confounding variables tested in the models, using the milk fatty acid content as the dependent variable, were age, family income, education, economic class, skin color, alcohol consumption, smoking, physical activity, prepregnancy BMI, postpartum BMI, gestational length, gestational weight gain, and baby’s weight at birth. However, only the postpartum BMI was directly related to the milk fatty acid content and therefore was employed as the adjusting variable in the final model. The time between the 24 hR applications and milk collection was tested as a covariate in the linear regression model; however, no association was observed. The statistical analyses were carried out using the SPSS software (SPSS Software, Version 17.0 SPSS Inc., Woking, Surrey, UK).
Results Seventy-five women completed the prospective study, of whom 20 were excluded. Eighteen were not practicing exclusive or predominant breast-feeding and two developed gestational diabetes. Among the 55 eligible women, 2 refused to participate in the study, 2 were not located, 5 relocated to another town, and 1 did not complete the 24 hR during puerperium; therefore, 45
R. Y. Nishimura et al. / Nutrition 30 (2014) 685–689 Table 1 Maternal characteristics and types of breast-feeding in women residing in ~o Preto, SP, Brazil, 2010 (N ¼ 45) Ribeira Characteristics
Statistics
Age (y) Pregestational BMI (kg/m2) Postpartum BMI (kg/m2) Skin color White Brown/mulata Others Education (years of study) 4 y 4-8 y >8 y Economic class A and B C, D, and E Type of breast-feeding In exclusive breast-feeding In predominant breast-feeding
Mean
SD
25 21.94 23.98 Frequency 23 16 6
4.48 1.79 2.42 % 51.11 35.55 13.33
2 11 32
4.44 24.44 71.11
5 32
11.11 88.88
23 22
51.11 48.88
BMI, body mass index
women were included in this study. The women were evaluated in first trimester (at 11 gestational wk average); second trimester (at 23 gestational wk average); and third trimester (at 30 gestational wk average). Milk samples were collected at 6.4 postpartum wk average (minimum of 5 wk and maximum of 12 wk). Table 1 presents demographic, socioeconomic, nutritional status, and breast-feeding activity characteristics of the study participants. The majority were of white skin color, married, with 8 y of education, and belong to the C, D, and E economic classes. The mean and SD of the dietary fatty acid estimates, in each gestational trimester and postpartum period, and fatty acid composition in mature breast milk are presented in Table 2. The b regression coefficients (95% CI) of the relation between dietary fatty acids in each trimester of pregnancy and postpartum period and milk fatty acid content are presented in Table 3. A positive association between the dietary EPA and DHA content during the third trimester of pregnancy and fatty acid content in mature milk was observed after postpartum BMI adjustment and regression coefficient deattenuation. Twentyfive and 33% of the EPA and DHA content in breast milk, respectively, was explained by the tested model. Positive association was observed between the dietary u-3 to u-6 ratio in the second and third trimesters of pregnancy and postpartum period and the fatty acid content in breast milk. However, the models only explained 22%, 15%, and 12% of the milk’s variability of the u-3 to u-6 ratio by the dietary intake during the second and third trimesters of pregnancy, and puerperium, respectively. No
687
association was observed between the other estimated dietary fatty acids (ALA, LA, total u-3, and total u-6) along the pregnancy and postpartum period and the fatty acid composition in breast milk.
Discussion To our knowledge, this study was the first to investigate the influence of the dietary fatty acid content, consumed during different gestational periods, on the fatty acid content in mature breast milk. The data demonstrate that the dietary EPA and DHA content during the third trimester of pregnancy is directly related to the content of these fatty acids in mature breast milk; however, no association was observed between dietary EPA and DHA content at the beginning of pregnancy or during the postpartum period, and milk composition. Additionally, a direct association was observed between the dietary u-3 to u-6 ratio in the second and third trimesters of pregnancy and the postpartum period and milk composition. The results of this study suggest that maternal body stocks of fatty acids have greater influence on breast milk fatty acid composition than the estimated diet during puerperium, corroborating previously reported results [4,21]. Moreover, only the diet during the third trimester of pregnancy was related to EPA and DHA content in breast milk. A possible explanation would be the increased weight gain and body fat storage observed in participants during late pregnancy [22]. The dietary EPA and DHA content in the third trimester of pregnancy explained 25% and 33% of the variability in the content of these fatty acids in breast milk, respectively, suggesting that there are other determinants contributing to the variability. Previous studies suggest that the degree of milk maturation, gestational length [23], and nutritional status of the mother [13, 24] are predictors of the fatty acid composition in breast milk. To minimize possible biases, this study’s sample was composed of women with prepregnancy BMI between 18.5 and 24.9 kg/m2 and presented full-term pregnancy, and only mature breast milk samples were used. The postpartum BMI was the only confounding variable among the ones investigated and associated with milk fatty acid composition, and therefore was employed as the adjusting variable in the models. A factor that might have influenced the results is the competition for the D6 desaturase enzyme between LA and ALA, which is responsible for the conversion of these fatty acids in ARA and LC-PUFA u-3 (EPA and DHA), respectively [25], considering that the Brazilian population’s diet is rich in LA [26] resulting from high consumption of vegetable oils. The endogenous synthesis of LC-PUFA u-3 may have suffered interference, resulting in low DHA and EPA content in maternal milk.
Table 2 Mean (SD) of the estimates for dietary fatty acids in each trimester of pregnancy and postpartum period and fatty acid content in mature breast milk in lactating women ~o Preto, SP, Brazil, 2010 (N ¼ 45) residing in Ribeira Diet
ALA (% total fat) EPA (% total fat) DHA (% total fat) LA (% total fat) ARA (% total fat) u-3/u-6 ratio
Composition
First trimester
Second trimester
Third trimester
Postpartum
Breast milk
2.0274 0.0127 0.0635 17.3535 0.1356 0.1291
1.7181 0.0088 0.0499 15.1779 0.1260 0.1268
1.7081 0.0072 0.0480 14.4576 0.1587 0.1363
2.1065 0.0186 0.0545 17.1387 0.1208 0.1305
1.5348 0.0789 0.1023 20.7344 0.4746 0.0811
(1.20) (0.05) (0.11) (8.37) (0.20) (0.08)
(0.74) (0.04) (0.15) (6.83) (0.19) (0.06)
(0.86) (0.02) (0.13) (6.76) (0.17) (0.10)
(0.95) (0.03) (0.13) (6.67) (0.10) (0.05)
ALA, a-linolenic fatty acid; ARA, arachidonic fatty acid; DHA, docosahexaenoic fatty acid; EPA, eicosapentaenoic fatty acid; LA, linoleic fatty acid
(0.40) (0.04) (0.06) (4.27) (0.09) (0.01)
688
R. Y. Nishimura et al. / Nutrition 30 (2014) 685–689
Table 3 The b regression coefficients (95% CI) and r2 of the relation between dietary fatty acids in each trimester of pregnancy and postpartum period and fatty acid composition of mature breast milk, Ribeir~ ao Preto, SP, Brazil, 2010 (N ¼ 45)* First trimester
b ALA (%) Model 1y Model 2z EPA (%) Model 1y Model 2z DHA (%) Model 1y Model 2z LA (%) Model 1y Model 2z ARA (%) Model 1y Model 2z u–3/u–6 Model 1y Model 2z
Second trimester
Third trimester
Postpartum
95% CI
r2
b
95% CI
r2
b
95% CI
r2
0.091 0.133
0.041 to 0.223 0.060 to 0.327
0.044 –
0.138 0.213
0.314 to 0.037 0.481 to 0.056
0.058 –
0.061 0.078
0.104 to 0.225 0.134 to 0.255
0.014 –
0.078 0.078
0.297 to 0.141 0.296 to 0.140
0.105 –
0.026 0.025
0.279 to 0.227 0.278 to 0.227
0.094 –
0.690 1.873
0.201 to 1.180 0.545 to 3.203
0.066 0.133
0.079 to 0.212 0.160 to 0.430
0.117 –
0.022 0.060
0.155 to 0.112 0.429 to 0.310
0.101 –
0.234 0.464
0.071 0.122
0.051 to 0.193 0.088 to 0.334
0.036 –
0.058 0.085
0.184 to 0.067 0.272 to 0.099
0.025 –
0.049 0.080
0.083 to 0.180 0.136 to 0.297
0.025 –
0.004 0.004
0.147 to 0.139 0.159 to 0.151
0.036 0.058
0.016 to 0.088 0.026 to 0.143
0.067 –
0.094 0.117
0.035 to 0.152 0.043 to 0.190
b
95% CI
r2
0.115 0.140
0.040 to 0.271 0.044 to 0.299
0.051 –
0.245 –
0.044 0.055
0.338 to 0.251 0.422 to 0.313
0.095 –
0.107 to 0.360 0.212 to 0.714
0.330 –
0.015 0.021
0.121 to 0.151 0.170 to 0.212
0.100 –
0.041 0.052
0.079 to 0.161 0.101 to 0.207
0.016 –
0.139 0.176
0.016 to 0.294 0.020 to 0.372
0.077 –
0.012 –
0.080 0.126
0.061 to 0.221 0.094 to 0.352
0.043 –
0.135 0.196
0.092 to 0.362 0.133 to 0.525
0.045 –
0.218 –
0.051 0.093
0.009 to 0.093 0.016 to 0.170
0.145 –
0.074 0.086
0.006 to 0.143 0.006 to 0.166
0.124 –
ALA, a–linolenic acid; ARA, arachidonic acid; BMI, body mass index; CI, confidence interval; DHA, docosahexaenoic acid; EPA, eicosapentaenoic acid; u–3, total u–3 fatty acids; LA, linoleic acid; u–6: total u–6 fatty acids; u–3/u–6: u–3/u–6 ratio * Significant coefficients are in bold. y Model 1: Linear model considering breast milk fatty acid content as dependent variable and maternal fatty acids intake as independent variable, adjusted by postpartum BMI. z Model 2: Model 1 after deattenuation of the b and the 95% CI.
Similarly, it has been observed that women living in the coastal regions of China, where the diet is rich in LC-PUFA u-3, presented higher milk DHA content than Chinese countryside residents. However, the EPA content was similar in breast milk of women from both regions [12]. RCTs found an increase in DHA content in breast milk after DHA or fish oil supplementation during pregnancy and postpartum [5–8]. However, the effect of this supplementation on EPA and ALA content is controversial [5,6]. Such inconsistent findings could be attributed to the type of administered supplement (DHA capsules derived from algae or fish oil) and/or the supplemented quantity (2 g or 400 mg). In the present study, no association was observed between the u-6 fatty acid (LA and ARA) content in maternal diet and milk fatty acid composition. However, a direct relationship was verified between the dietary u-3 to u-6 ratio during the second and third trimesters of pregnancy and postpartum period and the milk fatty acid composition. Vegetable oils are the main source of u-6 in a diet and an accurate estimate of consumption of these oils is difficult to obtain because of difficulty in accounting for amounts used during food preparation; errors in these estimated measurements could have mitigated the associations in this study. Previous studies using labeled isotopes revealed that the LA and ARA content in breast milk 72 h after labeling was 30% and 10%, respectively [27], which resulted from direct intestinal absorption from the maternal diet during the postpartum period. These results suggest that these fatty acids do not reach the mammary glands directly after intestinal absorption, but are temporarily stored in the body and subsequently released into circulation, indicating that maternal storage (from previous intake, which could be during the pregnancy) is the main source of u-6, in particular of ARA [27]. RCTs demonstrated that long-term supplementation with ARA (8 wk) during postpartum or between the second gestational trimester and postpartum week 12, associated or not with DHA and EPA, led to an ARA increase in breast milk [9,10]. However, short-term ARA supplementation (1 wk) during
postpartum had no effect on ARA content in breast milk [11], suggesting that dietary ARA only influences breast milk after prolonged supplementation because it becomes part of the body’s fatty acid storage. Among the limitations of this study, the reduced number of 24 hR obtained in each period is the most significant, and could have attenuated the associations, particularly because the main dietary source of EPA and DHA (fish) is not eaten on daily basis. The models used explain a small proportion of the variability in breast milk fatty acid composition, suggesting the existence of other predictors that were not considered in the linear regression models. Considering the importance of LC-PUFA u-3 for adequate infant development, the results from this study are relevant for planning interventions to increase the fatty acid content in breast milk. In particular, the consumption of food rich in u-3 should be encouraged during the third trimester of pregnancy. However, this was a small observational study, and RCTs are needed to confirm this hypothesis. The present study did not consider the children’s intake of PUFA, which depends not only on fatty acid composition of human milk, but also on the milk production and fat content of the milk delivered [28]. Also important to note is that some of the mothers’ genetic variants (FADS1 and FADS2) could influence the fatty acid composition of human milk [29], which was not addressed in the present study. Conclusion The dietary DHA and EPA composition and u-3 to u-6 ratio during late pregnancy and the postpartum period directly influence the DHA and EPA content in mature breast milk. Acknowledgments The authors received funds from the Teaching, Research, and Assistance Support Foundation from the FAEPA for the study. RYN is in receipt of a scholarship from FAPESP (2010/12320-1) to
R. Y. Nishimura et al. / Nutrition 30 (2014) 685–689
References [1] Innis SM. Fatty acids and early human development. Early Hum Dev 2007;83:761–6. [2] Torres AG, Trugo NM. Evidence of inadequate docosahexaenoic acid status in Brazilian pregnant and lactating women. Rev Saude Publica 2009;43:359–68. [3] Schmeits BL, Cook JA, VanderJagt DJ, Magnussen MA, Bhatt SK, Bobik EG Jr, et al. Fatty acid composition of the milk lipids of women in Nepal. Nutrition Research 1999;19:1339–48. [4] Koletzko B, Rodriguez-Palmero M, Demmelmair H, Fidler N, Jensen R, Sauerwald T. Physiological aspects of human milk lipids. Early Hum Dev 2001;65:3–18. [5] Dunstan JA, Mitoulas LR, Dixon G, Doherty DA, Hartmann PE, Simmer K, et al. The effects of fish oil supplementation in pregnancy on breast milk fatty acid composition over the course of lactation: a randomized controlled trial. Pediatr Res 2007;6:689–94. [6] Imhoff-Kunsch B, Stein AD, Villalpando S, Martorell R, Ramakrishnan U. Docosahexaenoic acid supplementation from mid-pregnancy to parturition influenced breast milk fatty acid concentrations at 1 month postpartum in Mexican women. J Nutr 2011;141:321–6. [7] Jensen CL, Maude M, Anderson RE, Heird WC. Effect of docosahexaenoic acid supplementation of lactating women on the fatty acid composition of breast milk lipids and maternal and infant plasma phospholipids. Am J Clin Nutr 2000;71:292–9. [8] Makrides M, Neumann MA, Gibson RA. Effect of maternal docosahexaenoic acid (DHA) supplementation on breast milk composition. Eur J Clin Nutr 1996;50:352–7. [9] van Goor SA, Dijck-Brouwer DA, Hadders-Algra M, Doornbos B, Erwich JJ, Schaafsma A, et al. Human milk arachidonic acid and docosahexaenoic acid contents increase following supplementation during pregnancy and lactation. Prostaglandins Leukot Essent Fatty Acids 2009;80:65–9. [10] Weseler AR, Dirix CE, Bruins MJ, Hornstra G. Dietary arachidonic acid dosedependently increases the arachidonic acid concentration in human milk. J Nutr 2008;138:2190–7. [11] Smit EN, Koopmann M, Boersma ER, Muskiet FA. Effect of supplementation of arachidonic acid (AA) or a combination of AA plus docosahexaenoic acid on breastmilk fatty acid composition. Prostaglandins Leukot Essent Fatty Acids 2000;62:335–40. [12] Peng Y, Zhou T, Wang Q, Liu P, Zhang T, Zetterström R, et al. Fatty acid composition of diet, cord blood and breast milk in Chinese mothers with different dietary habits. Prostaglandins Leukot Essent Fatty Acids 2009;81:325–30. [13] Mäkelä J, Linderborg K, Niinikoski H, Yang B, Lagström H. Breast milk fatty acid composition differs between overweight and normal weight women: the STEPS Study. Eur J Nutr 2013;52:727–35.
[14] Haggarty P. Effect of placental function on fatty acid requirements during pregnancy. Eur J Clin Nutr 2004;58:1559–70. ~o Brasileira de Pesquisas (ABEP). Dados com base no levanta[15] Associac¸a mento socioeconomico 2006 e 2007. Disponıvel em . Acesso em 28 jan 2008. [16] Johnson RK, Soultanakis RP, Matthews DE. Literacy and body fatness are associated with underreporting of energy intake in US low-income women using the multiple-pass 24-hour recall: a doubly labeled water study. J Am Diet Assoc 1998;98:1136–40. cleo de Estudos e Pesquisa em Alimentac¸a ~o. Tabela Brasileira de [17] Nu Composic¸~ ao de Alimentos – TACO, Vers~ ao 2. 2nd ed. Campinas: Universidade Estadual de Campinas; 2006. [18] United States Department of Agriculture. Nutrient Database for Standard Reference. Beltsville, MD: Agricultural Research Service, USDA; 2001. [19] Bligh EG, Dyer WJ. A rapid method of total lipid extraction and purification. Can J Biochem Physiol 1959;37:911–7. [20] Beaton GH, Milner J, Corey P, McGuire V, Cousins M, Stewart E, et al. Sources of variance in 24-hour dietary recall data: implications for nutrition study design and interpretation. Am J Clin Nutr 1979;32: 2546–59. [21] Torres AG, Ney JG, Meneses F, Trugo NM. Polyunsaturated fatty acids and conjugated linoleic acid isomers in breast milk are associated with plasma non-esterified and erythrocyte membrane fatty acid composition in lactating women. Br J Nutr 2006;95:517–24. [22] Haggarty P. Fatty acid supply to the human fetus. Annu Rev Nutr 2010;30:237–55. ny O, Wahle J, Koletzko B. Fatty acid composition of hu[23] Genzel-Borovicze man milk during the 1st month after term and preterm delivery. Eur J Pediatr 1997;156:142–7. [24] Marın MC, Sanjurjo A, Rodrigo MA, de Alaniz MJ. Long-chain polyunsaturated fatty acids in breast milk in La Plata, Argentina: relationship with maternal nutritional status. Prostaglandins Leukot Essent Fatty Acids 2005;73:355–60. [25] Arterburn LM, Hall EB, Oken H. Distribution, interconversion, and dose response of n-3 fatty acids in humans. Am J Clin Nutr 2006;83:1467–76. [26] Instituto Brasileiro de Geografia e Estatıstica (Brasil). Pesquisa de lise do Consumo Alimentar Orc¸amentos Familiares 2008 to 2009: Ana Pessoal no Brasil. Rio de Janeiro: IBGE; 2011. [27] Del Prado M, Villalpando S, Elizondo A, Rodrıguez M, Demmelmair H, Koletzko B. Contribution of dietary and newly formed arachidonic acid to human milk lipids in women eating a low-fat diet. Am J Clin Nutr 2001;74:242–7. [28] Mitoulas LR, Gurrin LC, Doherty DA, Sherrifi JL, Hartmann PE. Infant intake of fatty acids from human milk over the first year of lactation. Br J Nutr 2003;90:979–86. [29] Xie L, Innis SM. Genetic variants of the FADS1 FADS2 gene cluster are associated with altered (n-6) and (n-3) essential fatty acids in plasma and erythrocyte phospholipids in women during pregnancy and in breast milk during lactation. J Nutr 2008;138:2222–8. ˇ
attend graduate school to obtain a master’s degree and GSFC is in receipt of a scholarship from FAPESP (2010/00408-1) as a doctoral fellow.
689
Contents lists available at ScienceDirect
Nutrition journal homepage: www.nutritionjrnl.com
Applied nutritional investigation
Dietary polyunsaturated fatty acid intake during late pregnancy affects fatty acid composition of mature breast milk Renata Y. Nishimura M.S. a, Patricia Barbieiri a, Gabriela S.F. de Castro b, Ph.D. d, ~o Jr. Ph.D. c, Gleici da Silva Castro Perdona Alceu A. Jorda d, * Daniela S. Sartorelli Ph.D. ~o Preto Medical School at the Sa ~o Paulo University, Sa ~o Paulo, Brazil Program in Community Health in the Ribeira ~o Preto Medical School at the Sa ~o Paulo University, Sa ~o Paulo, Brazil Program in Internal Medicine in the Ribeira ~o Preto Medical School at the Sa ~o Paulo University, Sa ~o Paulo, Brazil Department of Internal Medicine in the Ribeira d ~o Preto Medical School at the Sa ~o Paulo University, Sa ~o Paulo, Brazil Department of Social Medicine in the Ribeira a
b c
a r t i c l e i n f o
a b s t r a c t
Article history: Received 24 September 2013 Accepted 13 November 2013
Objective: The aim of this study was to investigate how maternal polyunsaturated fatty acid intake at different periods during pregnancy affects the composition of polyunsaturated fatty acids in mature human milk. Methods: A prospective study was conducted involving 45 pregnant women, aged between 18 and 35 y, who had full-term pregnancies and practiced exclusive or predominant breast-feeding. Mature breast milk samples were collected after the 5th postpartum week by manual expression; fatty acid composition was determined by gas chromatography. Fatty acid intake during pregnancy and puerperium was estimated through multiple 24-h dietary recalls. Linear regression models, adjusted by postpartum body mass index and deattenuated, were used to determine associations between estimated fatty acids in maternal diet during each trimester of pregnancy and fatty acid content in mature human milk. Results: A positive association was identified between maternal intake of eicosapentaenoic acid (b, 1.873; 95% confidence interval [CI], 0.545, 3.203) and docosahexaenoic acid (b, 0.464; 95% CI, 0.212–0.714) during the third trimester of pregnancy, as well as the maternal dietary u-3 to u-6 ratio (b, 0.093; 95% CI, 0.016–0.170) during the second and third trimesters and postpartum period, with these fatty acids content in mature breast milk. Conclusions: The maternal dietary docosahexaenoic acid and eicosapentaenoic acid content during late pregnancy may affect the fatty acid composition of mature breast milk. Additionally, the maternal dietary intake of u-3 to u-6 fatty acid ratio, during late pregnancy and the postpartum period, can affect the polyunsaturated fatty acid composition of breast milk. Ó 2014 Elsevier Inc. All rights reserved.
Keywords: Human milk Dietary intake Pregnancy Polyunsaturated fatty acids n-3 fatty acids n-6 fatty acids Trans fatty acids
Introduction The long-chain polyunsaturated fatty acids (LC-PUFA), docosahexaenoic acid (DHA, 22:6 u-3), eicosapentaenoic acid (EPA, 20:5 u-3), and arachidonic acid (ARA, 20:4 u-6), are of great importance during child growth and development [1]. The RYN and DSS were responsible for the study design, data collection, and analysis. RYN developed the first version of the manuscript. GSFC and AJJ were responsible for data analysis and manuscript review. DSS coordinated the study and reviewed the final manuscript version. All authors have registered curriculum in the Lattes platform from CNPq. There is no conflict of interest to be disclosed. * Corresponding author. Tel.: þ55 360 227 1216; fax: þ55 363 313 8616. E-mail address: [email protected] (D. S. Sartorelli). 0899-9007/$ - see front matter Ó 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.nut.2013.11.002
content of these fatty acids in human milk, especially DHA, varies widely [2,3], possibly as a result of diverse food consumption among women and because fatty acid content in maternal diets varies among regions. Clinical trials suggest that consumption of fatty fish and fish oil supplements during puerperium are associated with an increased content of EPA and DHA in breast milk. However, in studies that employed labeled isotopes, only 30% of the variability in polyunsaturated fatty acid content (LC-PUFA) in breast milk is explained by the estimated dietary intake during the puerperium [4]. Randomized controlled trials (RCTs) suggest a direct relation between the use of u-3 supplements during the second and third trimesters of pregnancy and its content in mature breast milk [5, 6]. However, results reporting the effect of this supplement
686
R. Y. Nishimura et al. / Nutrition 30 (2014) 685–689
during pregnancy on EPA and a-linolenic acid (ALA) concentrations are controversial [5,6]. Several studies using u-3 (DHA, EPA, and ALA) and u-6 (ARA and linoleic acid [LA]) fatty acid supplementation during gestational or postpartum periods have found that it has distinct effects on fatty acid content in breast milk [5–11]. Studies investigating the influence of dietary fatty acid during pregnancy on breast milk fatty acid composition are scarce and those that have actually investigated the issue have been limited to food consumption during the third trimester only [12,13]. The association between maternal intake of fatty acids and their content in human milk may differ according to the period of intake, given that according to the gestational stage, the rate by which they are transported by the placenta may vary and maternal metabolism may change as well [14]. We hypothesized that maternal PUFA intake might have distinct effects on the PUFA composition of breast milk depending on the period of maternal intake. This hypothesis has not been tested in previous studies. This study aimed to investigate how fatty acid intake during different pregnancy periods affects the composition of PUFA in mature breast milk. Materials and methods Study population A prospective study was conducted to test the accuracy of a quantitative food frequency questionnaire involving 103 pregnant women, attendees at the Basic Health Units in Ribeir~ ao Preto, S~ ao Paulo, Brazil, aged between 18 and 35 y, with prepregnancy body mass index (BMI) between 18.5 and 24.9 kg/m2, had no diseases that may alter usual food consumption (diabetes mellitus, hypertension). For the validation study, 24-h dietary recalls (24 hRs) were obtained in each trimester of pregnancy. Seventy-five of these women completed the validation study and information on food intake (evaluated by the 24 hRs) throughout their pregnancies were available. Those women were invited to participate in the present study. The inclusion criteria were: Women who practiced exclusive breast-feeding (exclusive breast milk intake, with no other liquid or solid food intake) or predominant breast-feeding (breast milk intake and other water-based liquid food intake) after the fifth postpartum week, and mothers with full-term pregnancies (birth after the 37th gestational week). The study was approved by the Municipal Health Secretary of Ribeir~ ao Preto de-Escola and approved by the Research Ethics Committee from the Centro de Sau ~o Paulo, Brazil (Protocol number 378/CEP-CSE/FMRP-USP). at FMRP, USP, Sa
Socio-demographic characteristics and lifestyle A structured questionnaire applied at the first interview during the prospective study (first trimester of pregnancy) was used to collect data on age, education, assets, and skin color. The economic classification of each participant was assessed according to the Brazilian Economic Classification Criterion (CCEB), which defines classes from A (highest socioeconomic level) to E (lowest socioeconomic level) [15].
Assessment of maternal BMI, weight gain, and gestational age The nutritional status was assessed using the body weight of each participant in each trimester during prenatal and postpartum care appointments ~o Paulo-SP, using an electronic digital scale (Plenna, model MEA 07700, Sa Brazil) with a capacity of 150 kg and 100 g gradation. Heights were measured using an anthropometric scale ruler (Cauduro, model 101 PL, Cachoeira do SulRS, Brazil). Gestational weight gain was calculated as the difference between the weight at the last prospective data collection (third trimester) and prepregnancy weight (reported by the participant). Maternal pregestational and postpartum BMI were obtained using the height measured at the first prenatal visit and self-reported pregestational weight and measured postpartum weight, respectively. Gestational age was calculated preferably based on the date of the last menstrual cycle on record, which was correlated to data from the ultrasound exam performed up to the twentieth gestational week.
Estimation of polyunsaturated fatty acid intake during pregnancy PUFA (u-3 fatty acids [DHA, EPA, and ALA] and u-6 fatty acids [ARA and LA]) intake during pregnancy were evaluated through three 24 hR applied throughout the pregnancy, one 24 hR in each trimester of pregnancy, and two 24 hRs during the postpartum period. A second 24 hR was applied in each trimester of pregnancy in a subsample of the participants to correct for intraindividual variability. The 24 hR were applied by trained nutritionists using the multiple pass technique [16]. The NutWin software (NutWin software, Nutrition Support Program, version ~o Paulo, Brazil, 2002) was used for the analysis 1.5, Escola Paulista de Medicina, Sa of PUFA content based on the 24 hR information, employing the Brazilian Food Composition Table-TACO [17], and the U.S. Department of Agriculture’s Food Composition Table [18]. Determination of fatty acids in breast milk Samples (5–10 mL) of mature breast milk (after postpartum week 5) were obtained by hand expression by the participant in the morning, immediately after the baby’s first feeding and before mother’s breakfast. Milk samples were stored at 80 C until analysis. Fatty acids were extracted from 800 mL of milk, to which methanol and chloroform (v/v, 1:1) were added according to a previously described method [19]. After lipid extraction, the chloroform phase was evaporated under nitrogen current and fatty acids were methylated with potassium hydroxide in methanol (0.5 M) and removed with hexane. One mL of fatty acid methyl esters was injected in a SHIMADZU GC-2014 gas chromatographer (Shimadzu Europe, Duisburg, Germany) equipped with AOC-20 i auto-injector (Shimadzu Europe, Duisburg, Germany) and separated with a polyethylene glycol-SUPLELCOWAX 30 m capillary column (30 m, 25 mm 0.25 mm; Supelco Inc., Bellefonte, PA, USA). Helium was used as the carrier gas at a flow rate of 1 mL/min. Synthetic air was used for flame ionization detection at 280 C. The injections were performed in split mode. The injector and detector temperatures were 250 C. The initial column temperature was maintained for 1 min at 100 C, followed by an increase of 13 C/min until reaching 195 C, when it was maintained for 5 min, and subsequently elevated to 240 C, at the rate of 15 C/min, and maintained at this temperature for 30 min. The identification standard used was composed of a mixture of fatty acid methyl esters from Supelco (Supelco 37 Component FAME Mix; Supelco Inc., Bellefonte, PA, USA) with the addition of 9 c, 11 t-CLA and 10 t, 12 c-CLA. The quantification was performed by area normalization and results presented in weight percentages. Data analysis Linear regression models were used employing the breast milk fatty acid content as the dependent variable and the maternal fatty acid intake, during each trimester of pregnancy and puerperium, as the independent variables. Univariate and multiple linear models were adjusted by postpartum BMI and deattenuation. Deattenuation consists of correction of the linear regression coefficient by the intra- and interpersonal variability error. It was calculated using the formula: Bt ¼ bo [1 þ (s intra/s inter)/n] [20], where bt is the true regression coefficient, bo is the observed regression coefficient, s intra and s inter are the intra- and interindividual variances, respectively, and n is the number of 24 hR replicates (in this case, n ¼ 2 per trimester). The confounding variables tested in the models, using the milk fatty acid content as the dependent variable, were age, family income, education, economic class, skin color, alcohol consumption, smoking, physical activity, prepregnancy BMI, postpartum BMI, gestational length, gestational weight gain, and baby’s weight at birth. However, only the postpartum BMI was directly related to the milk fatty acid content and therefore was employed as the adjusting variable in the final model. The time between the 24 hR applications and milk collection was tested as a covariate in the linear regression model; however, no association was observed. The statistical analyses were carried out using the SPSS software (SPSS Software, Version 17.0 SPSS Inc., Woking, Surrey, UK).
Results Seventy-five women completed the prospective study, of whom 20 were excluded. Eighteen were not practicing exclusive or predominant breast-feeding and two developed gestational diabetes. Among the 55 eligible women, 2 refused to participate in the study, 2 were not located, 5 relocated to another town, and 1 did not complete the 24 hR during puerperium; therefore, 45
R. Y. Nishimura et al. / Nutrition 30 (2014) 685–689 Table 1 Maternal characteristics and types of breast-feeding in women residing in ~o Preto, SP, Brazil, 2010 (N ¼ 45) Ribeira Characteristics
Statistics
Age (y) Pregestational BMI (kg/m2) Postpartum BMI (kg/m2) Skin color White Brown/mulata Others Education (years of study) 4 y 4-8 y >8 y Economic class A and B C, D, and E Type of breast-feeding In exclusive breast-feeding In predominant breast-feeding
Mean
SD
25 21.94 23.98 Frequency 23 16 6
4.48 1.79 2.42 % 51.11 35.55 13.33
2 11 32
4.44 24.44 71.11
5 32
11.11 88.88
23 22
51.11 48.88
BMI, body mass index
women were included in this study. The women were evaluated in first trimester (at 11 gestational wk average); second trimester (at 23 gestational wk average); and third trimester (at 30 gestational wk average). Milk samples were collected at 6.4 postpartum wk average (minimum of 5 wk and maximum of 12 wk). Table 1 presents demographic, socioeconomic, nutritional status, and breast-feeding activity characteristics of the study participants. The majority were of white skin color, married, with 8 y of education, and belong to the C, D, and E economic classes. The mean and SD of the dietary fatty acid estimates, in each gestational trimester and postpartum period, and fatty acid composition in mature breast milk are presented in Table 2. The b regression coefficients (95% CI) of the relation between dietary fatty acids in each trimester of pregnancy and postpartum period and milk fatty acid content are presented in Table 3. A positive association between the dietary EPA and DHA content during the third trimester of pregnancy and fatty acid content in mature milk was observed after postpartum BMI adjustment and regression coefficient deattenuation. Twentyfive and 33% of the EPA and DHA content in breast milk, respectively, was explained by the tested model. Positive association was observed between the dietary u-3 to u-6 ratio in the second and third trimesters of pregnancy and postpartum period and the fatty acid content in breast milk. However, the models only explained 22%, 15%, and 12% of the milk’s variability of the u-3 to u-6 ratio by the dietary intake during the second and third trimesters of pregnancy, and puerperium, respectively. No
687
association was observed between the other estimated dietary fatty acids (ALA, LA, total u-3, and total u-6) along the pregnancy and postpartum period and the fatty acid composition in breast milk.
Discussion To our knowledge, this study was the first to investigate the influence of the dietary fatty acid content, consumed during different gestational periods, on the fatty acid content in mature breast milk. The data demonstrate that the dietary EPA and DHA content during the third trimester of pregnancy is directly related to the content of these fatty acids in mature breast milk; however, no association was observed between dietary EPA and DHA content at the beginning of pregnancy or during the postpartum period, and milk composition. Additionally, a direct association was observed between the dietary u-3 to u-6 ratio in the second and third trimesters of pregnancy and the postpartum period and milk composition. The results of this study suggest that maternal body stocks of fatty acids have greater influence on breast milk fatty acid composition than the estimated diet during puerperium, corroborating previously reported results [4,21]. Moreover, only the diet during the third trimester of pregnancy was related to EPA and DHA content in breast milk. A possible explanation would be the increased weight gain and body fat storage observed in participants during late pregnancy [22]. The dietary EPA and DHA content in the third trimester of pregnancy explained 25% and 33% of the variability in the content of these fatty acids in breast milk, respectively, suggesting that there are other determinants contributing to the variability. Previous studies suggest that the degree of milk maturation, gestational length [23], and nutritional status of the mother [13, 24] are predictors of the fatty acid composition in breast milk. To minimize possible biases, this study’s sample was composed of women with prepregnancy BMI between 18.5 and 24.9 kg/m2 and presented full-term pregnancy, and only mature breast milk samples were used. The postpartum BMI was the only confounding variable among the ones investigated and associated with milk fatty acid composition, and therefore was employed as the adjusting variable in the models. A factor that might have influenced the results is the competition for the D6 desaturase enzyme between LA and ALA, which is responsible for the conversion of these fatty acids in ARA and LC-PUFA u-3 (EPA and DHA), respectively [25], considering that the Brazilian population’s diet is rich in LA [26] resulting from high consumption of vegetable oils. The endogenous synthesis of LC-PUFA u-3 may have suffered interference, resulting in low DHA and EPA content in maternal milk.
Table 2 Mean (SD) of the estimates for dietary fatty acids in each trimester of pregnancy and postpartum period and fatty acid content in mature breast milk in lactating women ~o Preto, SP, Brazil, 2010 (N ¼ 45) residing in Ribeira Diet
ALA (% total fat) EPA (% total fat) DHA (% total fat) LA (% total fat) ARA (% total fat) u-3/u-6 ratio
Composition
First trimester
Second trimester
Third trimester
Postpartum
Breast milk
2.0274 0.0127 0.0635 17.3535 0.1356 0.1291
1.7181 0.0088 0.0499 15.1779 0.1260 0.1268
1.7081 0.0072 0.0480 14.4576 0.1587 0.1363
2.1065 0.0186 0.0545 17.1387 0.1208 0.1305
1.5348 0.0789 0.1023 20.7344 0.4746 0.0811
(1.20) (0.05) (0.11) (8.37) (0.20) (0.08)
(0.74) (0.04) (0.15) (6.83) (0.19) (0.06)
(0.86) (0.02) (0.13) (6.76) (0.17) (0.10)
(0.95) (0.03) (0.13) (6.67) (0.10) (0.05)
ALA, a-linolenic fatty acid; ARA, arachidonic fatty acid; DHA, docosahexaenoic fatty acid; EPA, eicosapentaenoic fatty acid; LA, linoleic fatty acid
(0.40) (0.04) (0.06) (4.27) (0.09) (0.01)
688
R. Y. Nishimura et al. / Nutrition 30 (2014) 685–689
Table 3 The b regression coefficients (95% CI) and r2 of the relation between dietary fatty acids in each trimester of pregnancy and postpartum period and fatty acid composition of mature breast milk, Ribeir~ ao Preto, SP, Brazil, 2010 (N ¼ 45)* First trimester
b ALA (%) Model 1y Model 2z EPA (%) Model 1y Model 2z DHA (%) Model 1y Model 2z LA (%) Model 1y Model 2z ARA (%) Model 1y Model 2z u–3/u–6 Model 1y Model 2z
Second trimester
Third trimester
Postpartum
95% CI
r2
b
95% CI
r2
b
95% CI
r2
0.091 0.133
0.041 to 0.223 0.060 to 0.327
0.044 –
0.138 0.213
0.314 to 0.037 0.481 to 0.056
0.058 –
0.061 0.078
0.104 to 0.225 0.134 to 0.255
0.014 –
0.078 0.078
0.297 to 0.141 0.296 to 0.140
0.105 –
0.026 0.025
0.279 to 0.227 0.278 to 0.227
0.094 –
0.690 1.873
0.201 to 1.180 0.545 to 3.203
0.066 0.133
0.079 to 0.212 0.160 to 0.430
0.117 –
0.022 0.060
0.155 to 0.112 0.429 to 0.310
0.101 –
0.234 0.464
0.071 0.122
0.051 to 0.193 0.088 to 0.334
0.036 –
0.058 0.085
0.184 to 0.067 0.272 to 0.099
0.025 –
0.049 0.080
0.083 to 0.180 0.136 to 0.297
0.025 –
0.004 0.004
0.147 to 0.139 0.159 to 0.151
0.036 0.058
0.016 to 0.088 0.026 to 0.143
0.067 –
0.094 0.117
0.035 to 0.152 0.043 to 0.190
b
95% CI
r2
0.115 0.140
0.040 to 0.271 0.044 to 0.299
0.051 –
0.245 –
0.044 0.055
0.338 to 0.251 0.422 to 0.313
0.095 –
0.107 to 0.360 0.212 to 0.714
0.330 –
0.015 0.021
0.121 to 0.151 0.170 to 0.212
0.100 –
0.041 0.052
0.079 to 0.161 0.101 to 0.207
0.016 –
0.139 0.176
0.016 to 0.294 0.020 to 0.372
0.077 –
0.012 –
0.080 0.126
0.061 to 0.221 0.094 to 0.352
0.043 –
0.135 0.196
0.092 to 0.362 0.133 to 0.525
0.045 –
0.218 –
0.051 0.093
0.009 to 0.093 0.016 to 0.170
0.145 –
0.074 0.086
0.006 to 0.143 0.006 to 0.166
0.124 –
ALA, a–linolenic acid; ARA, arachidonic acid; BMI, body mass index; CI, confidence interval; DHA, docosahexaenoic acid; EPA, eicosapentaenoic acid; u–3, total u–3 fatty acids; LA, linoleic acid; u–6: total u–6 fatty acids; u–3/u–6: u–3/u–6 ratio * Significant coefficients are in bold. y Model 1: Linear model considering breast milk fatty acid content as dependent variable and maternal fatty acids intake as independent variable, adjusted by postpartum BMI. z Model 2: Model 1 after deattenuation of the b and the 95% CI.
Similarly, it has been observed that women living in the coastal regions of China, where the diet is rich in LC-PUFA u-3, presented higher milk DHA content than Chinese countryside residents. However, the EPA content was similar in breast milk of women from both regions [12]. RCTs found an increase in DHA content in breast milk after DHA or fish oil supplementation during pregnancy and postpartum [5–8]. However, the effect of this supplementation on EPA and ALA content is controversial [5,6]. Such inconsistent findings could be attributed to the type of administered supplement (DHA capsules derived from algae or fish oil) and/or the supplemented quantity (2 g or 400 mg). In the present study, no association was observed between the u-6 fatty acid (LA and ARA) content in maternal diet and milk fatty acid composition. However, a direct relationship was verified between the dietary u-3 to u-6 ratio during the second and third trimesters of pregnancy and postpartum period and the milk fatty acid composition. Vegetable oils are the main source of u-6 in a diet and an accurate estimate of consumption of these oils is difficult to obtain because of difficulty in accounting for amounts used during food preparation; errors in these estimated measurements could have mitigated the associations in this study. Previous studies using labeled isotopes revealed that the LA and ARA content in breast milk 72 h after labeling was 30% and 10%, respectively [27], which resulted from direct intestinal absorption from the maternal diet during the postpartum period. These results suggest that these fatty acids do not reach the mammary glands directly after intestinal absorption, but are temporarily stored in the body and subsequently released into circulation, indicating that maternal storage (from previous intake, which could be during the pregnancy) is the main source of u-6, in particular of ARA [27]. RCTs demonstrated that long-term supplementation with ARA (8 wk) during postpartum or between the second gestational trimester and postpartum week 12, associated or not with DHA and EPA, led to an ARA increase in breast milk [9,10]. However, short-term ARA supplementation (1 wk) during
postpartum had no effect on ARA content in breast milk [11], suggesting that dietary ARA only influences breast milk after prolonged supplementation because it becomes part of the body’s fatty acid storage. Among the limitations of this study, the reduced number of 24 hR obtained in each period is the most significant, and could have attenuated the associations, particularly because the main dietary source of EPA and DHA (fish) is not eaten on daily basis. The models used explain a small proportion of the variability in breast milk fatty acid composition, suggesting the existence of other predictors that were not considered in the linear regression models. Considering the importance of LC-PUFA u-3 for adequate infant development, the results from this study are relevant for planning interventions to increase the fatty acid content in breast milk. In particular, the consumption of food rich in u-3 should be encouraged during the third trimester of pregnancy. However, this was a small observational study, and RCTs are needed to confirm this hypothesis. The present study did not consider the children’s intake of PUFA, which depends not only on fatty acid composition of human milk, but also on the milk production and fat content of the milk delivered [28]. Also important to note is that some of the mothers’ genetic variants (FADS1 and FADS2) could influence the fatty acid composition of human milk [29], which was not addressed in the present study. Conclusion The dietary DHA and EPA composition and u-3 to u-6 ratio during late pregnancy and the postpartum period directly influence the DHA and EPA content in mature breast milk. Acknowledgments The authors received funds from the Teaching, Research, and Assistance Support Foundation from the FAEPA for the study. RYN is in receipt of a scholarship from FAPESP (2010/12320-1) to
R. Y. Nishimura et al. / Nutrition 30 (2014) 685–689
References [1] Innis SM. Fatty acids and early human development. Early Hum Dev 2007;83:761–6. [2] Torres AG, Trugo NM. Evidence of inadequate docosahexaenoic acid status in Brazilian pregnant and lactating women. Rev Saude Publica 2009;43:359–68. [3] Schmeits BL, Cook JA, VanderJagt DJ, Magnussen MA, Bhatt SK, Bobik EG Jr, et al. Fatty acid composition of the milk lipids of women in Nepal. Nutrition Research 1999;19:1339–48. [4] Koletzko B, Rodriguez-Palmero M, Demmelmair H, Fidler N, Jensen R, Sauerwald T. Physiological aspects of human milk lipids. Early Hum Dev 2001;65:3–18. [5] Dunstan JA, Mitoulas LR, Dixon G, Doherty DA, Hartmann PE, Simmer K, et al. The effects of fish oil supplementation in pregnancy on breast milk fatty acid composition over the course of lactation: a randomized controlled trial. Pediatr Res 2007;6:689–94. [6] Imhoff-Kunsch B, Stein AD, Villalpando S, Martorell R, Ramakrishnan U. Docosahexaenoic acid supplementation from mid-pregnancy to parturition influenced breast milk fatty acid concentrations at 1 month postpartum in Mexican women. J Nutr 2011;141:321–6. [7] Jensen CL, Maude M, Anderson RE, Heird WC. Effect of docosahexaenoic acid supplementation of lactating women on the fatty acid composition of breast milk lipids and maternal and infant plasma phospholipids. Am J Clin Nutr 2000;71:292–9. [8] Makrides M, Neumann MA, Gibson RA. Effect of maternal docosahexaenoic acid (DHA) supplementation on breast milk composition. Eur J Clin Nutr 1996;50:352–7. [9] van Goor SA, Dijck-Brouwer DA, Hadders-Algra M, Doornbos B, Erwich JJ, Schaafsma A, et al. Human milk arachidonic acid and docosahexaenoic acid contents increase following supplementation during pregnancy and lactation. Prostaglandins Leukot Essent Fatty Acids 2009;80:65–9. [10] Weseler AR, Dirix CE, Bruins MJ, Hornstra G. Dietary arachidonic acid dosedependently increases the arachidonic acid concentration in human milk. J Nutr 2008;138:2190–7. [11] Smit EN, Koopmann M, Boersma ER, Muskiet FA. Effect of supplementation of arachidonic acid (AA) or a combination of AA plus docosahexaenoic acid on breastmilk fatty acid composition. Prostaglandins Leukot Essent Fatty Acids 2000;62:335–40. [12] Peng Y, Zhou T, Wang Q, Liu P, Zhang T, Zetterström R, et al. Fatty acid composition of diet, cord blood and breast milk in Chinese mothers with different dietary habits. Prostaglandins Leukot Essent Fatty Acids 2009;81:325–30. [13] Mäkelä J, Linderborg K, Niinikoski H, Yang B, Lagström H. Breast milk fatty acid composition differs between overweight and normal weight women: the STEPS Study. Eur J Nutr 2013;52:727–35.
[14] Haggarty P. Effect of placental function on fatty acid requirements during pregnancy. Eur J Clin Nutr 2004;58:1559–70. ~o Brasileira de Pesquisas (ABEP). Dados com base no levanta[15] Associac¸a mento socioeconomico 2006 e 2007. Disponıvel em . Acesso em 28 jan 2008. [16] Johnson RK, Soultanakis RP, Matthews DE. Literacy and body fatness are associated with underreporting of energy intake in US low-income women using the multiple-pass 24-hour recall: a doubly labeled water study. J Am Diet Assoc 1998;98:1136–40. cleo de Estudos e Pesquisa em Alimentac¸a ~o. Tabela Brasileira de [17] Nu Composic¸~ ao de Alimentos – TACO, Vers~ ao 2. 2nd ed. Campinas: Universidade Estadual de Campinas; 2006. [18] United States Department of Agriculture. Nutrient Database for Standard Reference. Beltsville, MD: Agricultural Research Service, USDA; 2001. [19] Bligh EG, Dyer WJ. A rapid method of total lipid extraction and purification. Can J Biochem Physiol 1959;37:911–7. [20] Beaton GH, Milner J, Corey P, McGuire V, Cousins M, Stewart E, et al. Sources of variance in 24-hour dietary recall data: implications for nutrition study design and interpretation. Am J Clin Nutr 1979;32: 2546–59. [21] Torres AG, Ney JG, Meneses F, Trugo NM. Polyunsaturated fatty acids and conjugated linoleic acid isomers in breast milk are associated with plasma non-esterified and erythrocyte membrane fatty acid composition in lactating women. Br J Nutr 2006;95:517–24. [22] Haggarty P. Fatty acid supply to the human fetus. Annu Rev Nutr 2010;30:237–55. ny O, Wahle J, Koletzko B. Fatty acid composition of hu[23] Genzel-Borovicze man milk during the 1st month after term and preterm delivery. Eur J Pediatr 1997;156:142–7. [24] Marın MC, Sanjurjo A, Rodrigo MA, de Alaniz MJ. Long-chain polyunsaturated fatty acids in breast milk in La Plata, Argentina: relationship with maternal nutritional status. Prostaglandins Leukot Essent Fatty Acids 2005;73:355–60. [25] Arterburn LM, Hall EB, Oken H. Distribution, interconversion, and dose response of n-3 fatty acids in humans. Am J Clin Nutr 2006;83:1467–76. [26] Instituto Brasileiro de Geografia e Estatıstica (Brasil). Pesquisa de lise do Consumo Alimentar Orc¸amentos Familiares 2008 to 2009: Ana Pessoal no Brasil. Rio de Janeiro: IBGE; 2011. [27] Del Prado M, Villalpando S, Elizondo A, Rodrıguez M, Demmelmair H, Koletzko B. Contribution of dietary and newly formed arachidonic acid to human milk lipids in women eating a low-fat diet. Am J Clin Nutr 2001;74:242–7. [28] Mitoulas LR, Gurrin LC, Doherty DA, Sherrifi JL, Hartmann PE. Infant intake of fatty acids from human milk over the first year of lactation. Br J Nutr 2003;90:979–86. [29] Xie L, Innis SM. Genetic variants of the FADS1 FADS2 gene cluster are associated with altered (n-6) and (n-3) essential fatty acids in plasma and erythrocyte phospholipids in women during pregnancy and in breast milk during lactation. J Nutr 2008;138:2222–8. ˇ
attend graduate school to obtain a master’s degree and GSFC is in receipt of a scholarship from FAPESP (2010/00408-1) as a doctoral fellow.
689
