Jørgensen, Marianne Hørby*; Hernell, Olle†; Hughes, Elizabeth L.‡; Michaelsen, Kim Fleischer*
*Research Department of Human Nutrition, The Royal Veterinary and Agricultural University, Frederiksberg, Denmark; the †Department of Clinical Sciences, Pediatrics, Umeå University, Sweden; and the ‡Department of Child Health, University of Aberdeen, Scotland
Accepted December 21, 2000.
Supported by grants from the Danish Research and Development Programme for Food Technology (FØTEK), and BASF Health and Nutrition A/S.
Address correspondence and reprint requests to Professor Kim Fleischer Michaelsen, Rolighedsvej 30, DK-1958 Frederiksberg C, Denmark (e-mail: KFM@kvl.dk).
Background: Docosahexaenoic acid (DHA), present in high concentrations in the brain and retina, has a role in visual development. DHA is present in human milk, but not in most infant formulas. It is, however, under discussion whether DHA should be added to formulas intended for term infants. The concentration of DHA in human milk, which is influenced by maternal diet, varies considerably, but it is unknown whether this variation affects visual development in term infants.
Methods: The authors investigated 39 4-month-old fully breast-fed term infants in a cross-sectional study. Visual acuity was measured by swept visual evoked potentials, milk DHA was determined by gas chromatography, and maternal fish intake was assessed by a frequency questionnaire.
Results: Frequency of fish intake correlated positively to the DHA level in breast-milk (P = 0.001). Mothers who ate fish the day before sampling had a milk DHA level higher than expected from habitual fish intake (P = 0.002). If this was taken into account, 57% of the variation in milk DHA could be explained by fish intake. Multiple linear regression analysis revealed a significant association between visual acuity and milk DHA (P = 0.02, R2 = 0.09).
Conclusion: This finding suggests a cause-and-effect relationship between infant milk DHA intake and visual acuity. If these data are confirmed, there is a need to consider the optimal intake of DHA for the lactating mother.
From the last trimester of pregnancy and through the first year of life, docosahexaenoic acid (22:6n-3; DHA) is incorporated rapidly into cell membranes of the brain and retina. DHA is also present in human milk, but not in most infant formulas. During the last decade it has been questioned whether formula-fed infants meet their needs of DHA for optimal development by endogenous elongation and desaturation of the precursor α-linolenic acid. Although there is general agreement that exogenous DHA is needed for optimal visual development in preterm infants until the capacity to synthesize is developed sufficiently (1,2), it is still discussed whether DHA is conditionally essential for term infants as well. Comparisons of visual acuity or neurodevelopment in intervention studies of term infants fed formulas with or without DHA have shown conflicting results (1,2).
The concentration of DHA in human milk varies considerably; for example, a mean of 0.2 wt% of total fatty acids in Western countries compared with a mean of 1.4 wt% in Inuit women (3). We have previously shown comparatively high, but variable concentrations of milk DHA among Danish mothers after 4 months of lactation (median, 0.30 wt%; range, 0.17–1.98 wt%; interquartile range, 0.24–0.39 wt%) (4).
The aim of the current study was to investigate whether the variation in milk DHA content between Danish mothers is large enough to cause differences in visual acuity in their healthy, term, 4-month-old infants, and to evaluate the influence of frequency of fish intake on the DHA level of the milk.
Seventy 4-month-old infants were recruited through the national birth registry as a random sample among infants living in an area close to the institute. Inclusion criteria were term delivery (37–42 gestational weeks); normal birth weight for gestational age; uncomplicated pregnancy, delivery, and neonatal period; Apgar score more than 8 points after 5 minutes; and fully breast-fed at time of examination (no energy drinks and less than 100 mL formula a day were allowed). One infant was excluded because he was being monitored by an ophthalmologist for strabismus, four were excluded because they were small for gestational age (below the 10th percentile for birthweight ), and one was operated for pyloric stenosis. Six infants did not complete the sweep visual evoked potential (VEP), five because of technical problems with the equipment and one because of lack of cooperation. Six infants were partially formula fed and 13 were exclusively formula fed. Thus, 39 fully breast-fed infants were included in the analysis. All 39 infants were exclusively breast-fed until 14 weeks of age. From 14 weeks of age, one infant had 150 mL formula twice a week.
At entrance, body length, weight, and head circumference were measured; visual acuity of the infant was assessed; and a questionnaire on frequency of fish intake of the mother was administered. The mothers were also asked about any complications they experienced during pregnancy, weight gain during pregnancy, weight and length of the child at birth, Apgar score, complications during the first months of life, and feeding mode. The educational level of the mothers was recorded and classified in four groups from low to high.
Visual acuity was assessed as a steady-state swept VEP using the NuDiva system (6,7). All infants were tested under identical conditions. Vertical sine wave gratings, contrast reversed at a rate of 6.6 Hz, were presented at 80% contrast at a space average luminance of 160 cd/m2. The trial duration was 10 seconds, during which grating was increased in 10 linear steps from 1 to 16 cycles/deg. Five recording sites were used. Visual acuity was estimated by extrapolating the VEP amplitude versus spatial frequency to zero amplitude (8). Five trials were recorded and the records were scored automatically by computer and were checked manually for errors. The infant's visual acuity was taken as the highest score in the second harmonic. A detailed description of the method has been published previously (9). Values are given as LogMAR (logarithm10 to the minimal angle of resolution), and low values are equivalent to a better visual acuity.
Comparing two sweep VEP tests from the same infant within 3 days (n = 25; age, 4–12 months) showed that reliability was high. Only 8.5% of the day-to-day variation was caused by variation within the subject (unpublished data), which is in accordance with others (7).
Maternal fish intake was determined by a simple food frequency questionnaire adapted from Olsen et al (10). The questionnaires showed that in Danish pregnant women, 44% of the variation in red blood cell n-3 fatty acids could be explained by the answers from three simple questions (Table 1), whereas using a self-administered, more detailed questionnaire explained 48%, and combined with a structured interview to quantify daily fish intake explained 54% of the variation. Because intake of n-3 fatty acids is reflected in milk fatty acid composition 7 to 8 hours after a meal (11–13), a question about fish intake during the preceding day was added to the questionnaire (Table 1). Mothers were asked to express manually 4 mL breast milk after the first feed on the morning of the day of examination. The samples were frozen immediately and kept frozen until transported to the lab. On arrival at the institute, the milk was stored at −80°C until analysis. Total lipids were extracted according to Bligh and Dyer (14), and fatty acid methyl esters were extracted and analyzed by gas chromatography as described (15). In short, a 50 m × 0.25 mm FAME column from Chrompack (CP Sil 88) was used. Standards from NuCheck (nos. 87, 90, and 96) were used to identify the response. The response factor was set at 1.00. A pooled sample of human milk was used as a daily control of the extraction.
Logarithm10 milk DHA values were used in all analyses because the distribution was skewed. A general linear model was performed to investigate possible relationships between milk DHA (dependent variable) and frequency of fish intake (analyzed as a class variable). Multivariate linear regression analysis was performed to investigate possible relationships between visual acuity, milk DHA, and fish intake the day before (dummy variable, 0–1 variable). SPSS 7.5 for Windows was used for statistical analysis.
Characteristics of the infants included are shown in Table 2. The median DHA content of the milk was 0.31 wt% of total fatty acids (range, 0.12–1.20 wt%). Table 3 shows the fatty acid composition of the milk. According to the questionnaire none of the mothers took fish oil supplements regularly. Only one of the nine mothers who ate fish the day before the milk sample was taken ate lean fish, the remaining mothers ate fatty fish. In a general linear model including frequency of consumption of lean and fatty fish, and fatty fish intake the day before sampling, all three variables were associated positively with milk DHA (frequency of lean fish intake, P = 0.02; frequency of fatty fish intake, P = 0.02; fatty fish intake the day before, P = 0.002; total model, P = 0.001). This model explained 57% of the variation in milk DHA (adjusted R2). Milk DHA levels of mothers who ate fatty fish the day before sampling was higher than expected from their habitual fish intake. No interaction terms were significant, indicating a strong influence of fatty fish intake the day before on milk DHA, independent of the frequency of fish intake. There was no significant difference between the effect of frequency of lean fish intake on milk DHA and frequency of fatty fish intake (P = 0.26).
The mean visual acuity was 0.37 LogMAR (range, 0.28–0.56 LogMAR). In a multiple linear regression analysis, there was a significant association between visual acuity of the infant and the mother's milk DHA (P = 0.02), controlling for intake of fatty fish the day before sampling (Table 4, Fig. 1). The visual acuity of infants of mothers who ate fish the day before sampling did not differ from the rest of the group (P = 0.96, t-test). Neither did arachidonic acid, eicosapentaenoic acid, linoleic, or α-linolenic acid correlate with visual acuity, nor did any of the anthropometric data, such as gestational age or age at examination (data not shown). No association was found between educational level of the mother and visual acuity or educational level and milk DHA (P = 0.60 and 0.57, respectively; analysis of variance).
Maternal Diet and Milk DHA
We have shown that frequency of fish intake was strongly associated with DHA levels in breast-milk. Although this has not been shown before, it was not unexpected because regular intake of fish oil capsules (12,16), DHA as single-cell oil (14), or egg yolk (18) all increase the DHA content of human milk.
The answers from four simple questions explained 57% of the variation in milk DHA. The fact that we showed an influence of recent fatty fish intake on milk DHA has implications for studies in which milk DHA is used as a parameter, especially in populations with a highly variable fish intake. A spot milk sample is influenced markedly by recent fish intake. However, despite this, a single milk sample will still give a reasonable reflection of the average DHA intake of the infant. Furthermore, we have shown previously that there is substantial tracking of milk DHA. Mothers having a high DHA at 1 month after delivery also tended to have high DHA levels 4 months after delivery (4).
Surprisingly, the effect of intake of lean and fatty fish on milk DHA did not differ. We speculate that it could be because lean fish in Denmark is more often eaten for supper, and thus in larger amounts, than fatty fish, which is eaten mainly as a bread spread for lunch. Furthermore, intake at supper will also have a stronger influence on levels in morning samples because of the shorter time interval.
Milk DHA and Visual Acuity
This study is the first to report a positive association between the DHA level in human milk and visual acuity in breast-fed infants. Our findings suggest a dose–response relationship between milk DHA and visual acuity. Gibson et al. (19) supplemented lactating mothers with different levels of DHA as single-cell oils, which resulted in graded increases in milk DHA. Despite this, they did not observe any effect on visual acuity when the mothers were randomized to five different levels of DHA intake. They commented that the group sizes were small (5–10 infants/group) because of a high frequency of unsuccessful VEP measurements, raising the possibility of a type II error. Furthermore, the study by Gibson et al. (19) differs from the current study in type of visual test performed (transient VEP versus sweep VEP in the current study). With transient VEP the infant has to be alert to the video monitor five to seven times longer, and the data analysis of the electroencephalogram differs in several respects. Transient VEP is therefore less sensitive than sweep VEP. Interestingly, in the same study, Gibson et al. (19) found a positive relation between DHA levels in both infant plasma, red blood cells, and mothers' milk at 3 months, and Bailey developmental score at 1 year of age, supporting that DHA intake of the breast-fed infant affects early neurologic development.
Because the current study is cross-sectional, we do not have information on maternal fish intake, and thereby DHA status, during pregnancy. It is reasonable to believe that the mothers' fish intake during pregnancy does not differ markedly from their fish intake during lactation. It is therefore also likely that milk DHA levels, to some degree, also reflect the DHA status of the mother during pregnancy. Fish intake during pregnancy influences gestational age and birthweight (20). We have shown previously a strong relation between weight at delivery and visual acuity at 4 months, but only in formula-fed infants (8). Neither in the current study nor in previous studies did we find significant correlations between anthropometric data and visual acuity in breast-fed infants. We speculate that the DHA status of mothers during pregnancy (which is likely to reflect the DHA status of the infant at birth) is more important for formula-fed infants lacking an exogenous DHA supply immediately after birth than for breast-fed infants. This is supported by our study of infants whose mothers were supplemented with fish oil during pregnancy only. Eighty-eight percent of these infants were exclusively breast-fed for more than 1 month, and we could not find an effect of DHA supplementation during pregnancy on visual acuity at 6 to 10 months (21). We thereby find it reasonable to believe that our finding of an effect on visual acuity is mostly the result of DHA intake of the infant during lactation and less the result of DHA endowment during pregnancy.
If it can be confirmed that the DHA intake of the breast-fed infant, and thereby the DHA status of the mother, influences visual development, there is a need to consider the optimal intake of DHA for women during lactation.
The authors thank laboratory technician Kirsten Ebbesen for her valuable help in connection with milk sampling and analyses.
1. Gibson RA, Makrides M. The role of long chain polyunsaturated fatty acids (LCPUFA) in neonatal nutrition. Acta Paediatr 1998; 87:1017–22.
2. Jørgensen MH, Lauritzen L, Michaelsen KF. The role of long-chain polyunsaturated fatty acids in neonatal nutrition [Letter]. Acta Paediatr 1999; 88:916–7.
3. Jørgensen MH, Lauritzen L, Michaelsen KF. Does human milk DHA level affect functional outcome in infants? J Hum Lact 1999; 15:3–6.
4. Jørgensen MH, Hernell O, Lund P, Hølmer G, Michaelsen KF. Visual acuity and erythrocyte docosahexaenoic acid status in breast-fed and formula-fed term infants during the first four months of life. Lipids 1996; 31:99–105.
5. Greisen G, Michaelsen KF. Perinatal growth. A practical perinatal growth curve [in Danish]. Vgeskrift for Laeger 1989;151:1831–15.
6. Norcia AM, Tyler CW. Spatial frequency sweep VEP: visual acuity during the first year of life. Vision Res 1985; 25:1399–408.
7. Norcia AM, Tyler CW. Infant VEP acuity measurements: analysis of individual differences and measurement error. Electroencephalogr Clin Neurol 1985; 61:359–69.
8. Norcia AM, Clarke M, Tyler CW. Digital filtering and robust regression techniques for estimating sensory thresholds from the evoked potential. IEEE
9. Jørgensen MH, Hølmer G, Lund P, Hernell O, Michaelsen KF. Effect of formula supplemented with docosahexaenoic acid and gamma-linolenic acid on fatty acid status and visual acuity in term infants. J Pediatr Gastroenterol Nutr 1998; 26:412–21.
10. Olsen SF, Hansen HS, Sandström B, Jensen B. Erythrocyte levels compared with reported dietary intake of marine n-3 fatty acids in pregnant women. Br J Nutr 1995; 28:1185–92.
11. Hachey DL, Thomas MR, Emken EA. Human lactation: maternal transfer of dietary triglycerides labeled with stable isotopes. J Lipid Res 1987; 28:1185–92.
12. Henderson RA, Jensen RG, Lammi–Keefe CJ, Ferris AM, Dardick KR. Effect of fish oil on the fatty acid composition of human milk and maternal and infant erythrocytes. Lipids 1992; 27:863–9.
13. Lauritzen L, Jørgensen MH, Michaelsen KF. Fluctuations in the docosahexaenoic acid (DHA)-content of human milk in relation to the ingestion of fish. FASEB J 1998; 12:A201.
14. Bligh EG, Dyer WP. A rapid method of total lipid extraction and purification. Can J Biochem Physiol 1959; 37:911–8.
15. Jørgensen MH, Lassen A, Michaelsen KF. Fatty acid composition in Danish infant formula compared to human milk. Scand J Nutr 1995; 39:50–4.
16. Harris WS, Connor WE, Lindsey S. Will dietary omega-3 fatty acids change the composition of human milk? Am J Clin Nutr 1984; 40:780–5.
17. 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.
18. Cherian G, Sim JS. Changes in the breast milk fatty acids and plasma lipids of nursing mothers following consumption of n-3 polyunsaturated fatty acid enriched eggs. Nutrition 1996; 12:8–12.
19. Gibson RA, Neumann MA, Makrides M. Effect of increasing breast milk docosahexaenoic acid on plasma and erythrocyte phospholipid fatty acids and neural indices of exclusively breast fed infants. Eur J Clin Nutr 1997; 51:1–7.
20. Olsen SF, Sørensen JD, Secher NJ, Hedegaard M, Henriksen TB, Hansen HS, Grant A. Randomised controlled trial of effect of fish-oil supplementation on pregnancy duration. Lancet 1992; 339:1003–7.
21. Olsen SF, Jørgensen MH, Michaelsen KF, Secher NJ. Fish oil in pregnancy and infant visual acuity abstract: a randomized controlled trial. Prostagland Leuk Essent Fatty 1997; 57:196.
© 2001 Lippincott Williams & Wilkins, Inc.