Share this article on:

Effect of Formula Supplemented with Docosahexaenoic Acid andγ-Linolenic Acid on Fatty Acid Status and Visual Acuity in Term Infants

Jørgensen, Marianne Hørby*; Hølmer, Gunhild*; Lund, Pia; Hernell, Olle; Michaelsen, Kira Fleischer

Journal of Pediatric Gastroenterology & Nutrition: April 1998 - Volume 26 - Issue 4 - pp 412-421
Original Articles

Background: Docosahexaenoic acid is present in high concentration in retina and does not influence visual development in preterm infants. It is still under discussion whether docosahexaenoic acid is important for visual development in term infants.

Methods: Thirty-seven infants fed formula for a median of 14 days were randomized at median age of 25 days to three formulas: a) DHAGF: 0.3 wt% docosahexaenoic acid and 0.5 wt% γ-linolenic acid; b) DHAF 0.3 wt% docosahexaenoic acid; or c) STF: standard formula without long-chain polyunsaturated fatty acids and 17 breast-fed infants were observed, using blood samples and anthropometric measurements from 1 to 4 months of age. At 4 months, visual acuity was measured by swept steadystate visual evoked potential. A cross-sectional study on 25 breast-fed infants was carried out as a reference group for the analyses.

Results: Infants fed the two docosahexaenoic acid-supplemented formula had relative docosahexaenoic acid concentrations in red blood cell phospholipids almost as high as those in breast-fed infants, whereas infants in the standard formula group had significantly lower levels. The addition of γ-linolenic acid to the formula had a positive effect on red blood cell arachidonic acid levels, compared with levels obtained using fish oil only. Visual acuity was significantly different among all feeding groups (analysis of variance;p = 0.05, means ± standard deviation: breast-fed, 0.37± 0.06 logMAR; DHAF and DHAGF combined, 0.40 ± 0.07 logMAR; and standard formula 0.44 ± 0.07 logMAR. However, there was no statistical difference among the formula groups. In a multiple regression analysis including all formula-fed infants, weight at delivery (p = 0.002), but not type of formula, was significantly associated with visual acuity at 4 months of age.

Conclusions: The addition of docosahexaenoic acid resulted in concentrations in red blood cells at similar levels as those in breast-fed infants, whereas the increase in visual acuity did not reach significance. The addition ofγ-linolenic acid resulted in higher arachidonic acid concentrations in red blood cells.

Center for Advanced Food Studies and Research Department of Human Nutrition, Royal Veterinary and Agricultural University, Frederiksberg, Denmark; and *Center for Advanced Food Studies and Department of Biochemistry and Nutrition, Technical University of Denmark, Copenhagen, Denmark; and †Department of Pediatrics, Umeå, Sweden.

Received March 3, 1997; revised October 29, 1997; accepted November 5, 1997.

Address correspondence and reprint requests to Dr. M. Hørby Jørgensen, Research Department of Human Nutrition, The Royal Veterinary and Agricultural University, Rolighedsvej 30, DK-1958 Frederiksberg C, Denmark.

Docosahexaenoic acid (DHA) and arachidonic acid (AA) are incorporated in cell membrane phospholipids, particularly in the brain and retina, during the last trimester of pregnancy and in the first year of life. Human milk contains DHA and AA, whereas infants fed standard formula have to meet their needs by endogenous synthesis from the precursors α-linolenic acid and linoleic acid, respectively. Concentrations in blood (1,2) as well as in the brain (3,4), of long-chain polyunsaturated fatty acids (LCPUFA), particularly DHA, are significantly lower in formula-fed than in breast-fed infants. Preterm infants fed a formula supplemented with DHA have a more rapid development of visual acuity than do preterm infants fed a standard formula devoid of DHA(5,6). The functional consequences of these biochemical differences between formula-fed and breast-fed term infants is still under discussion. Results of two observational studies-one cross-sectional (7) and one prospective(2)-show that breast-fed term infants have more rapid development of visual acuity than do infants fed a formula devoid of DHA, whereas Innis et al. noted no differences in visual acuity between breast-fed and formula-fed term infants (8). Results of randomized studies by Makrides et al. (9) showed that infants, either breast-fed or fed a formula supplemented with DHA, had better visual acuity at 16 and 30 weeks of age than did infants fed a standard formula, whereas Carlson et al. (10) recorded better visual acuity at 2 months of age in breast-fed infants and in infants fed a DHA-supplemented formula than in infants fed a standard formula. However, at 4 months this difference had disappeared. The discrepancies in results among these studies could be related to the use of different methods to assess visual acuity (VEP vs. Teller Acuity Card) or different levels of DHA in the formula studied (0.3 wt% [9] vs. 0.1 wt%[10]).

Arachidonic acid and eicosapentaenoic acid (EPA) are both precursors of eicosanoids. In adults (11) and preterm infants(12), high intakes of EPA from fish oil reduce the level of AA in red blood cells (RBCs). Further, it has been suggested that the AA level is a predictor of growth: In one study, results showed that preterm infants who received fish oil with a comparatively high EPA:DHA ratio had significantly lower z-scores for weight, length, and head circumference than did infants receiving standard formula without fish oil supplementation (13). This result was paralleled by a positive correlation between growth rate and AA status(14) during the first year of life. However, this effect on growth was not confirmed in the results of another study on preterm infants supplemented with fish oil (15). The only study on the effect of fish oil supplementation on growth in term infants showed no association in its results (9).

γ-Linolenic acid (GLA, 18:3n-6) is a precursor of AA. Feeding GLA rather than linoleic acid bypasses the delta-6 desaturase reaction and might therefore be more effective in promoting AA synthesis. Infants fed a standard formula without LCPUFA had 30% higher RBC AA than did infants fed a formula enriched with DHA, as fish oil, and GLA at 4 months of age(9), probably because of the presence of EPA in the fish oil. However, the study did not include a group of infants supplemented only with DHA from fish oil. Therefore, it is not known whether the addition of GLA to a fish oil-supplemented formula has a positive effect on the AA level.

The purpose of the current study was, in a randomized design, to investigate the effect of DHA on fatty acid status, growth, and visual development in term formula-fed infants weaned within the first month of life, and to compare the results with data on breast-fed infants. Visual acuity and RBC fatty acid composition were considered as the primary outcome, and the main hypothesis was that even in initially breast-fed infants in a population with a comparably high fish intake, feeding a formula supplemented with DHA would improve visual acuity. Secondly, we wanted to investigate whether further supplementation of the formula with GLA would increase RBC AA levels.

Back to Top | Article Outline


Thirty-nine formula-fed infants were included in a randomized, prospective study that began in October 1993. Only in fants completely weaned at the time of the first visit, 1 month after delivery, were included. Two infants failed to attend examinations at 2 and 4 months of age and were excluded. The remaining 37 infants completed the study. A group of 17 breast-fed infants recruited during a period beginning 6 months before the start of the formula study and lasting for 18 months served as control subjects. Because the time difference between data collection in the two studies is small and overlaps, we hypothesized that this difference in timing was not likely to disturb the results. Some of the data on breast-fed infants have been published previously(2). Because the swept steady-state visual-evoked potential (SWEEP VEP) method to measure visual acuity was not established in our laboratory when the breast-fed infants were observed, we included a group of 25 breast-fed infants, examined only at the age of 4 months, for comparison of SWEEP VEP data with that of the formula-fed groups. Infants in this group were examined during the same period as the formula-fed infants.

The infants randomized to the experimental formula were either recruited through the maternity ward or by health visitors within the first 30 days after birth. The breast-fed infants, observed longitudinally, were recruited through maternity wards, and the breast-fed infants enrolled in the cross-sectional group were recruited through the National Birth Registry at the age of 3.5 months. Inclusion criteria for both formula- and breast-fed infants were: uncomplicated pregnancy, term delivery (gestational age, 37-42 weeks), birth weight between 2700 and 4500 g, Apgar score higher than 7 after 5 minutes, and no neonatal diseases. Inclusion criteria for formula-fed infants was termination of breast-feeding before 30 days of age, without using a DHA-supplemented formula. Mothers were first asked to participate when they had stopped breast-feeding completely. This design was chosen because of the high breast-feeding frequency in Denmark, in which 98% of infants are breast-fed at discharge from the maternity ward. Exclusion criteria were admission of the infant to the hospital, serious illness during the study period, or suspicion of intolerance to the formula (vomiting, diarrhea). No infants fulfilled the exclusion criteria. Basic characteristics of the feeding groups were comparable (Table 1). The only significant difference was higher maternal age in the breast-fed group compared with the combined formula fed group (Student's t-test; p = 0.01). Within the formula groups, there was a trend toward a lower birth weight in the group supplemented with only DHA (DHAF group; p = 0.06).

Back to Top | Article Outline

Study Outline

The infants were examined at 1, 2, and 4 months of age. At the first examination, the mothers were asked about any complications during pregnancy and delivery. At each visit, weight, length, and head circumference were measured. In the formula-fed infants, a blood sample was taken by a heel prick at each visit, and visual acuity was measured at the age of 4 months. Breast-fed infants followed prospectively had only blood samples taken, and breast-fed infants followed cross-sectionally had only visual acuity assessed. A milk sample (4 ml) was collected from all breast-feeding mothers at each visit. The mothers expressed the milk either the day before or on the day of examination and kept the sample in the freezer at home until transportation to the department. For fatty acid analysis, RBCs were separated from plasma and washed three times with 150 mM NaCl and 1 M EDTA. Two milliliters methanol containing 0.01% butylated hydroxytoluene was added as an antioxidant. Both blood and milk samples were stored at -80°C until analysis.

Back to Top | Article Outline

Feeding Groups

The formula-fed infants were randomized within the first month of life to one of the three formulas (Table 2). The mothers and observers were blind to the code of randomization. All formulas had a linoleic acid to α-linolenic acid ratio between 10 and 11, within the range of 5 to 15 recommended by ESPGHAN (16). The only difference in composition among the formulas was the content of LCPUFA. One formula(DHAF) contained DHA (0.3 wt%) and EPA (0.4 wt%) in fish oil; another (DHAGF) contained DHA and EPA at the same concentrations but also GLA (0.5 wt%) in borage oil; and a standard formula (STF) was not supplemented with LCPUFA. At the time the study was designed, no low-EPA and high-DHA oils were available. The fish oil and borage oil, microencapsulated with gelatin and corn starch to prevent oxidation and to allow homogenization with the formula powder, were added by the formula manufacturers. All formulas were produced in one batch. Tocopherol was added to the oils in a concentration of 1 mg/g oil as antioxidant, which is equivalent to approximately 0.75 and 1.75 mg tocopherol per liter in the two experimental formulas. In all three formulas, the whey:casein ratio was 60:40, protein content 15 g/l, fat energy 47%, lactose 72 g/l, and iron 8 mg/l.

Only seven of the formula-fed infants had never received human milk. The remaining infants had been breast fed for a median of 14 days (interquartile range, 7-21 days; range, 1-28 days). The infants were given the experimental formula beginning on the 25th day postpartum (median, interquartile range, 13-32 days; range, 6-41 days). Thus, infants who received DHA in the formula received a standard formula without DHA for a median period of 10 days before enrollment (interquartile range, 6-17 days; range, 3-29 days).

Small amounts of supplementary food (vegetable mashes and cereal-based gruel, given for no more than one meal a day) were introduced to one breast-fed infant, who was observed prospectively, and five formula-fed infants (three receiving DHAGF and two receiving STF) from the age of 3 months.

Back to Top | Article Outline

Visual Acuity

Binocular visual acuity was measured by a swept spatial frequency technique based on digital filtering of a SWEEP VEP(17,18). All infants were tested under the same conditions. The infant was placed on the parent's lap 100 cm from a M2400 high-resolution monochrome monitor (Dotronix, Eau Claire, WI, U.S.A). Vertical sine-wave gratings were presented at an 80% contrast at a space average luminance of approximately 160 cd/m2. The gratings were contrast reversed at a rate of 6.62 Hz in 10-second trials. During each 10-second trial, the spatial frequency of the gratings was increased in 10 linear steps, a range spanning the expected acuity. The infant's attention was attracted to the screen by small toys or bells. The trial was interrupted if the infant's gaze moved off the stimulus. Recording was resumed when the infant refixated.

An electroencephalogram was recorded at 397 Hz over a 1 to 100 Hz band pass filtering using Grass Instruments P511 amplifiers (Grass, Quincy, MA, U.S.A). Five recording sites were used, as shown in Figure 1. The amplitude and phase of the evoked response were analyzed at the first four harmonics of the stimulus frequency. Only the second and fourth harmonics showed consistent stimulus-related activity. The second harmonic was the largest component recorded, and all analyses were derived from this component. Visual acuity was estimated by extrapolating the VEP amplitude versus spatial frequency function to zero amplitude using the methods described by Norcia et al. (19) The records were scored automatically by the computer and were checked manually for errors. The infant's visual acuity was taken as the highest score in the second harmonic in all five channels. Visual acuity was given as the logarithm10 to the minimal angle of resolution(logMAR). A better visual acuity is reflected in lower values.

Back to Top | Article Outline

Fatty Acid Composition

After total lipid extraction (20) and separation of lipid classes by thin-layer chromatography (21), methyl esters were prepared (22) from the phosphatidylcholine(PC) and phosphatidylethanolamine (PE) fractions of the RBCs. Fatty acid composition of the PC and PE fractions were then determined by gas liquid chromatography (GC) on a Hewlett-Packard 5890 instrument with on-column injection and a 50-m × 0.25-mm fused silica capillary column coated with a 0.2-μm film of CP-Sil 88 (Chrompack, The Netherlands). The oven temperature was programmed from 90°C to 160°C at a rate of 40°C/min, then to 200°C at 4°C/min, and finally after 10 minutes to 220°C at a rate of 4°C/min. Lipids from the milk were extracted(20) and methyl esters of total lipids prepared by direct methylation (23). The fatty acid composition was determined on a corresponding gas chromatography instrument with split injection and a 30-m × 0.32-mm fused silica column with a 0.2-μm film of SP2380 (Supelco, Bellefonte, PA, U.S.A). The temperature program started at 80°C, was increased to 110°C at a rate of 30°C/min, and then was increased to 200°C at 4°C/min.

Identification and qualification of the fatty acids were based on standards of fatty acid methyl esters (GLC 68 A and GLC 411) purchased from NuChek Prep(Elysian, MN, U.S.A.).

Back to Top | Article Outline

Statistical Analyses

Group means were compared by a Student's t-test, one-way analysis of variance (ANOVA) or chi-square test for single measurements for each infant. With repeated measurements from the same infant group means, time, and interaction were measured by a two-way ANOVA for repeated measurements. Because many infants had only a trace level of GLA in the RBCs, nonparametric statistics (Kruskal-Wallis one-way ANOVA) was used when comparing these data. A paired Student's t-test, comparing data from 1 and 4 months was used to analyze the time difference throughout the study period. Backward multiple regression was used to examine the influence of cofactors.

Before the study, power calculation was performed concerning SWEEP VEP data. A relevant clinical difference between the two groups was set to one standard deviation. With a power of 80% and a significance level at 0.05, the group sizes should be 15. Unfortunately, we realized after breaking the code, that the group receiving standard formula did not reach this number.

Back to Top | Article Outline


Ethical approval was obtained from the local ethics committee of the Copenhagen and Frederiksberg municipalities. Written, informed consent was obtained from the parents before enrollment.

Back to Top | Article Outline


Fatty Acid Composition

The fatty acid composition of RBCs is listed inTables 3 (PC) and 4 (PE). Results from the two-way ANOVA for repeated measurements are shown in Table 5.

The relative concentration of RBC DHA in PE and PC is shown inFigure 2. There were significant differences among feeding groups in both phospholipid classes, with the lowest levels recorded in infants receiving the standard formula. Furthermore, there was a significant depletion with time in these infants (PC: p = 0.01; PE:p < 0.001). Comparing only the two DHA-supplemented groups and the BF group, the latter had the highest levels, but this was only significant for PE (p = 0.04).

A difference between feeding groups in RBC EPA was noted in both the PC and PE fraction, with the highest levels in the two groups receiving DHA in fish oil (Fig. 2). When including these two groups only in the two-way ANOVA, there was no significant difference between the groups. At 4 months of age, the DHAF group had a significantly higher EPA concentration than did the DHAGF group within the PE fraction (Student's t-test;p = 0.006). There was a significant increase in 22:5n3 PE in all feeding groups with time, from a mean of 1.2 wt% to 3.5 wt% (p < 0.001; Table 4). We did not find, however, any difference between groups in 22:5n3 PE. In the PC fraction, the concentrations are low(Table 3), and no difference with time was seen.

The RBC AA level differed among feeding groups in both PC and PE fractions, with higher relative concentrations among STF infants and BF infants(Fig. 2). Arachidonic acid in RBCs decreased in both DHA-supplemented groups in PC with time (paired Student's t-tests: DHAGF, p = 0.01; DHAF, p = 0.05). Including only these two groups in the analysis, there was a significant difference between the groups(p = 0.01), with higher levels in the DHAGF group(Fig. 2).

No differences among feeding groups was recorded in RBC GLA concentration(Kruskal-Wallis one-way ANOVA, at 1, 2, or 4 months). However, GLA was barely detectable in most of the blood samples (Table 3 and 4). For RBC levels of dihomo-γ-linolenic acid (20:3n-6), there was a difference among groups in both PC and PE fractions with the highest levels in the DHAGF infants (Table 3). In PC, there was a decrease with time in all feeding groups, except in the DHAGF groups (paired Student'st-test; DHAF, p = 0.009; STF, p = 0.009; and BF,p < 0.001); and in PE, an increase in the DHAGF group (paired Student's t-test; p = 0.02). Excluding the DHAGF group in the two-way ANOVA, no differences were recorded among the other three groups.

Comparing the sum of all n-6 fatty acids with 20 carbon atoms or longer(LCPUFA n-6) between the DHAGF and DHAF groups, we could show a difference between the groups (p < 0.001 in both PC (Table 3) and PE (Table 4), with a higher level in the DHAGF group, than in the DHAF group. We also found a time effect (p < 0.001) caused by a decrease with time in both groups (paired Student'st-test; DHAGF p = 0.04, DHAF p = 0.014).

The fatty acid compositions of the breast-milk and formulas are shown inTable 2. The DHA levels in milk from the two groups of breast-feeding mothers were 0.53 and 0.38 wt%, respectively. There was no statistical difference in milk LCPUFA composition between the two breast-fed groups.

Back to Top | Article Outline

Visual Acuity

Because no difference was found in RBC DHA between the DHAF and DHAGF groups, data from the two groups were pooled in the analysis on visual acuity. In the following, this group is named DHA ± GF.

Of the 37 randomized formula-fed infants, 26 were successfully tested with SWEEP VEP at 4 months of age (DHA ± GF, 18; STF, 8). Eight infants were 4 months old before our equipment was available and in three of the remaining 29, the testing was not successful. All 25 breast-fed infants were successfully tested.

The visual acuity measured with SWEEP VEP at 4 months of age was significantly different among the three feeding groups (BF, DHA ± GF, and STF; one-way ANOVA; p = 0.05; Fig. 3). BF infants had significantly better visual acuity than did STF infants (Student'st-test p = 0.02). However, there were no statistical differences between the DHA ± GF and BF or between the DHA ± GF and STF groups (Student's t-test; p = 0.15 and p= 0.25, respectively).

We did not find a significant correlation between visual acuity at 4 months and RBC DHA concentrations, regardless of age, within each of the DHA ± GF and STF groups or when all formula-fed infants were combined in the analysis (all p values ≥0.19). Nor did we find a correlation between visual acuity and AA or EPA. It is interesting to note that we recorded a negative correlation between linoleic acid in the PC fraction at 4 months and visual acuity (r = 0.45, p = 0.03).

There were no significant differences among feeding groups in length, weight, and head circumference at 1, 2, or 4 months of age, or growth velocity during the intervention (two-way ANOVA) with repeated measurements.

Back to Top | Article Outline

Visual Acuity and Anthropometry

Within the formula groups, we found a positive, significant correlation between visual acuity and weight and length at birth, 1, 2, and 4 months of age (r between 0.4 and 0.66; all p ≤ 0.05); The correlations between head circumference at the tested ages and visual acuity were positive (r ≥ 0.40), albeit only significant at 1 month. We observed no significant correlation between weight gain or length velocity from birth to 4 months of age and visual acuity. We recorded a significant correlation, however, between duration of breast-feeding and visual acuity with a better visual acuity in infants breast-fed for a longer time. Regression coefficients and p-values are listed inTable 6.

In a multiple regression analysis including only formula-fed infants, influence of duration of breast-feeding, gestational age, formula type, weight, and length at birth on visual acuity was examined(Table 7). Weight at delivery but not type of formula was significantly associated with visual acuity (p = 0.002), whereas the association with duration of breast-feeding was of borderline significance(p = 0.09). Including RBC DHA or RBC linoleic acid in the analysis did not affect the results.

Within the BF group, we saw no significant correlations between visual acuity and anthropometric data or gestational age (data not shown).

Back to Top | Article Outline


Supplementation of formula with 0.3 wt% DHA increased the levels of RBC DHA to levels almost comparable with those in breast-fed infants. Furthermore, infants receiving a standard formula devoid of DHA were gradually depleted in RBC DHA with time. These findings are in accordance with those in randomized intervention studies of term (9,10,24) and preterm infants (5,12), and with those in observational studies of breast-fed and formula-fed term infants(1,2,8). At all ages, the DHA formula-fed infants had RBC DHA levels slightly lower than those of the breast-fed infants. This could be because of the short period (interquartile range, 6-17 days) in the first month of life, during which the DHA formula-fed infants received a standard formula; the higher level of DHA in the breast milk(0.4-0.5 wt%) compared with that in the formula (0.3 wt%); or factors in breast-milk that may improve absorption of LCPUFA from breast milk-e.g., bile salt-stimulated lipase (25) or nucleotides(26), which are present in breast-milk but not in the formulas used.

Regarding the total amount of LCPUFA n-3, the observed difference among groups was caused by the slightly lower EPA levels and the dramatically lower DHA levels in the STF group. In the STF infants, the content of 22:5n-3, an elongation product of EPA, was lower in the PC fraction. In fact, we noted a considerable increase in 22:5n-3 in the PE fraction in all feeding groups(mean increase from 1.2 wt% to 3.5 wt%), which could be explained by the increase in capacity with advancing age. This indicates that infants fed a standard formula can synthesize 22:5n-3 to levels almost comparable to those in BF infants who receive 22:5n-3 in breast milk. The rate-limiting step in infants who are not fed DHA may be the elongation-desaturation,β-oxidation steps between 22:5n-3 and 22:6n-3 (DHA). This is supported by results in the study by Farquharson et al. (27), which suggest that the enzymes included in this pathway are insufficiently active during the first 4 months of life. The lower level of RBC EPA in the STF group could be related to a drain of this fatty acid because of increased synthesis of 22:5n-3.

γ-Linolenic acid had a slight effect on EPA PE concentrations in RBCs, which decreased, and on AA concentrations, which increased, compared with those in the DHAF group. Furthermore, we showed an increased concentration of dihomo-γ-linolenic acid in the group supplemented with GLA. These observations were confirmed by an increase in the total amount of LCPUFA n-6. In spite of this, there were higher levels of EPA and lower levels of AA PE in RBCs in the two fish-oil supplemented groups, compared with those levels in breast-fed infants or in infants fed the standard formula, which is in accordance with the findings of Makrides et al. (9) Similar to their observations (9), differences in AA levels in RBCs did not result in differences in growth pattern, indicating that decreased RBC AA in term infants may not influence growth, as was suggested in preterm infants (14).

In the results of the current study, we did not note a significant difference in visual acuity between infants supplemented with DHA (DHA± GF) and infants receiving a standard formula without LCPUFA (STF). However, we did record a difference in visual acuity among all three feeding groups, with the best visual acuity in BF infants, followed by the DHA-supplemented infants (DHA ± G), with the STF infants having the poorest visual acuity. We therefore could not confirm the observations made in two previous studies in term infants (9,10) in which results showed that DHA supplementation of formula had a positive effect on visual acuity. There are several possible explanations for this discrepancy. Firstly, dietary DHA may not be necessary for optimal visual development in term infants. This would be in agreement with results from another study in term infants in which data showed no effect of DHA supplementation was noted (28). However, it should be noted that the DHA supplementation (0.1 wt%) was lower than that in the current study. Secondly, we noted a trend for better visual acuity in DHA-supplemented infants, and it is possible that if we had examined a larger number of infants, we would have found a significant difference (type II error). Thirdly, the intervention with DHA started after a median period of 10 days postpartum, during which the infants were fed a formula without DHA. This period may be particularly vulnerable in terms of visual development.

Within the combined groups of formula-fed infants, weight at delivery was a strong predictor of visual acuity. In contrast, within the breast-fed group, size at birth did not correlate with visual acuity. Reanalyzing data from a previous study in breast-fed infants and infants fed a standard formula devoid of DHA (2), we determined a positive, but not significant, correlation between visual acuity measured with Teller Acuity Cards and weight and length at delivery within the formula-fed group(r = 0.47, p = 0.09; r = 0.44, p = 0.11, respectively). We speculate that the DHA status of the mother during pregnancy, which has been shown to have a positive influence on birth weight(29), is a predictor of visual acuity at 4 months of age in infants who receive a suboptimal intake of DHA during the first postpartum month.

The trend for a positive effect of duration of breast-feeding on visual acuity in the multiple regression analysis supports the role of human milk, and thereby dietary DHA, in visual development in term infants. Furthermore, the fact that the median duration was only 14 days suggests that breastfeeding during the first weeks of life are of particular importance.

The mean concentration of DHA in human milk in this study and in previous studies of Danish mothers (30) was higher than that in most other Western countries (31), suggesting a higher average fish intake in Danish mothers. According to a recent nationwide study on dietary habits in Denmark, almost 25% of 19- to 34-year-old women stated that they never ate fish, suggesting a wide range of fish intake. If visual acuity in the formula-fed infants at 4 months is indeed influenced by the DHA status of the mother through DHA accretion in utero, it could blur an effect of DHA supplementation during the first months of life. If so, this might be an additional reason why we did not find a significant effect of DHA supplementation, especially because we examined only a small number of infants. It is tempting to speculate that in populations with lower fish intake, the effect of DHA would be stronger.

In conclusion, the addition of 0.3 wt% DHA in fish oil to infant formula raised the relative DHA concentrations in RBCs to levels comparable to those in breast-fed infants. The further addition of 0.5 wt% GLA in borage oil to the formula also increased the concentration of AA but could not fully compensate for the lack of AA in the formula. However, it is still not known whether lower AA levels have any functional consequences in term infants. Although, we did not note a significant effect on visual development, our results do not exclude the possibility that dietary DHA has a role in visual development in term infants. Other investigators have shown that intervention with DHA in formula has an effect on developmental scores in preterm infants at 12 months of age (32,33) and in term infants at 4 months of age (34), suggesting that the effect of n-3 fatty acids may not be limited to visual development, but may affect neurodevelopment in general. However, it still remains to be verified that such effect has long-term consequences in term infants.

Acknowledgment: The authors thank all parents and infants who participated in the study; Dr. Anthony Norcia for his advice concerning SWEEP VEP measurements; and all laboratory technicians, dietitians, and secretaries, who took part in data collection, fatty acid analysis, and preparation of the manuscript.

The study was supported by grants from the Food Technology Research and Development Program (FØTEK), BASF Health and Nutrition, Denmark, and the Swedish Medical Research Council (19x-05708). Semper AB, Stockholm supplied the formula used in the study.

Back to Top | Article Outline


1. Ponder DL, Innis SM, Benson JD, Siegman JS. Docosahexaenoic acid status of term infants fed breast milk or infant formula containing soy oil or corn oil. Pediatr Res 1992;32:683-8.
2. 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 4 months of life.Lipids 1996;31:99-105.
3. Farquharson J, Cockburn F, Patrick WA, Jamieson EC, Logan RW. Infant cerebral cortex phospholipid fatty-acid composition and diet.Lancet 1992;340:810-3.
4. Makrides M, Neumann MA, Byard RW, Simmer K, Gibson RA. Fatty acid composition of brain, retina, and erythrocytes in breastand formula-fed infants. Am J Clin Nutr 1994;60:189-94.
5. Birch EE, Birch DG, Hoffman DR, Uauy R. Dietary essential fatty acid supply and visual acuity development. Invest Ophthalmol Visual Sci 1992;33:3242-53.
6. Carlson SE, Werkman SH, Rhodes PG, Tolley EA. Visual-acuity development in healthy preterm infants: Effect of marine-oil supplementation. Am J Clin Nutr 1993;58:35-42.
7. Makrides M, Simmer K, Goggin M, Gibson RA. Erythrocyte docosahexaenoic acid correlates with the visual response of healthy, term infants. Pediatr Res 1993;33:425-7.
8. Innis SM, Nelson CM, Rioux MF, King DJ. Development of visual acuity in relation to plasma and erythrocyte omega-6 and omega-3 fatty acids in healthy term gestation infants. Am J Clin Nutr 1994;60:347-52.
9. Makrides M, Neumann M, Simmer K, Pater J, Gibson R. Are long-chain polyunsaturated fatty acids essential nutrients in infancy?Lancet 1995;345:1463-8.
10. Carlson SE, Ford AJ, Werkman SH, Peeples JM, Koo WWK. Visual acuity and fatty acid status of term infants fed human milk and formulas with and without docosahexaenoate and arachidonate from egg yolk lecithin. Pediatr Res 1996;39:882-8.
11. Bjerve KS, Thoresen L, Bonaa K, Vik T, Johnsen H, Brubakk AM. Clinical studies with alpha-linolenic acid and long-chain n-3 fatty acids. Nutrition 1992;8:130-2.
12. Carlson SE, Cooke RJ, Rhodes PG, Peeples JM, Werkman SH, Tolley EA. Long-term feeding of formulas high in linolenic acid and marine oil to very low birth weight infants: Phospholipid fatty acids. Pediatr Res 1991;30:404-12.
13. Clandinin MT, Garg ML, Parrott A, Van Aerde J, Hervada A, Lien E. Addition of long-chain polyunsaturated fatty acids to formula for very low birth weight infants. Lipids 1992;27:896-900.
14. Carlson SE, Werkman SH, Peeples JM, Cooke RJ, Tolley EA. Arachidonic acid status correlates with first year growth in preterm infants. Proc Natl Acad Sci USA 1993;90:1073-7.
15. Uauy R, Hoffman DR, Birch EE, Birch DG, Jameson DM, Tyson J. Safety and efficacy of omega-3 fatty acids in the nutrition of very low birth weight infants-soy oil and marine oil supplementation of formula.J Pediatr 1994;124:612-20.
16. Aggett PJ, Haschke F, Heine W, et al. Comment on the content and composition of lipids in infant formulas. ESPGAN Committee on Nutrition. Acta Paediatr Scand 1991;80:887-96.
17. Norcia AM, Tyler CW. Spatial frequency sweep VEP: Visual acuity during the first year of life. Vision Res 1985;25:1399-1408.
18. Norcia AM, Tyler CW. Infant VEP acuity measurements: Analysis of individual differences and measurement error.Electroencephalogr Clin Neurophysiol 1985;61:359-69.
19. Norcia AM, Clarke M, Tyler CW. Digital filtering and robust regression techniques for estimating sensory thresholds from the evoked potential. IEEE 1985;26-32.
20. Folch J, Lees M, Sloane Stanley GM. A simple method for the isolation and purification of total lipids from animal tissues. J Biol Chem 1959;226:497-509.
21. Gilfillan AM, Chu AJ, Stuart DA, Roney SA. Single plate separation of lungs phospholipids including disaturated phosphatidylcholine.J Lipid Res 1983;24:1651-6.
22. Høy CE, Hølmer G. Dietary linoleic acid and the fatty acid profiles in rats fed partially hydrogenated marine oils.Lipids 1988;23:973-80.
23. Christophersen SW, Glass RL. Preparation of milk fat methyl esters by alcoholysis in an essentially nonalcoholic solution. J Dairy Sci 1969;52:1289-90.
24. Decsi T, Koletzko B. Growth, fatty acid composition of plasma lipid class, and plasma retinol and alfa-tocopherol concentrations in full-term infants fed formula enriched with w-6 and w-3 longchain polyunsaturated fatty acids. Acta Paediatr 1995;84:725-32.
25. Hernell O, Blackberg L, Chen Q, Sternby B, Nilsson A. Does the bile salt-stimulated lipase of human milk have a role in the use of the milk long-chain polyunsaturated fatty acids? J Pediatr Gastroenterol Nutr 1993;16:426-31.
26. DeLucchi C, Pita ML, Faus MJ, Molina JA, Uauy R, Gil A. Effects of dietary nucleotides on the fatty acid composition of erythrocyte membrane lipids in term infants. J Pediatr Gastroenterol Nutr 1987;6:568-74.
27. Farquharson J, Jamieson EC, Logan RW, Patrick WA, Howatson AG, Cockburn F. Age- and dietary-related distributions of hepatic arachidonic and docosahexaenoic acid in early infancy. Pediatr Res 1995;38:361-5.
28. Auested N, Montaito MB, Hall RT, et al. Visual acuity, erythrocyte composition in term infants fed formulas with long chain polyunsaturated fatty acids for one year. Pediatr Res 1997;41:1-10.
29. Olsen SF, Sørensen JD, Secher NJ, et al. Randomised controlled trial of effect of fish-oil supplementation on pregnancy duration. Lancet 1992;339:1003-7.
30. Jørgensen MH, Lassen A, Michaelsen KF. Fatty acid composition in Danish infant formula compared with human milk.Scand J Nutr 1995;39:50-4.
31. Koletzko B, Thiel I, Abiodun PO. The fatty acid composition of human milk in Europe and Africa. J Pediatr 1992;120:S62-S70.
32. Carlson SE, Werkman SH. A randomized trial of visual attention of preterm infants fed docosahexaenoic acid until two months.Lipids 1996;31:85-90.
33. Werkman SH, Carlson SE. A randomized trial of visual attention in preterm infants fed docosahexaenoic acid until nine months.Lipids 1996;31:91-7.
34. Agostoni C, Trojan S, Bellu R, Riva E, Giovannini M. Neurodevelopmental quotient of healthy term infants at 4 months and feeding practice: The role of long-chain polyunsaturated fatty acids. Pediatr Res 1995;38:262-6.

Cited By:

This article has been cited 2 time(s).

Journal of Pediatric Gastroenterology and Nutrition
Is There a Relation between Docosahexaenoic Acid Concentration in Mothers' Milk and Visual Development in Term Infants?
Jørgensen, MH; Hernell, O; Hughes, EL; Michaelsen, KF
Journal of Pediatric Gastroenterology and Nutrition, 32(3): 293-296.

PDF (77)
Journal of Pediatric Gastroenterology and Nutrition
Effect of Type of Early Infant Feeding on Fatty Acid Composition of Plasma Lipid Classes in Full-Term Infants During the Second 6 Months of Life
Decsi, T; Kelemen, B; Minda, H; Burus, I; Kohn, G
Journal of Pediatric Gastroenterology and Nutrition, 30(5): 547-551.

Back to Top | Article Outline

Arachidonic acid; Docosahexaenoic acid; Formula feeding; Human milk; Visual development

© Lippincott-Raven Publishers