Effect of -linolenic acid supplementation during pregnancy on maternal and neonatal polyunsaturated fatty acid status and pregnancy outcome 1 3

Transcription

1 Effect of -linolenic acid supplementation during pregnancy on maternal and neonatal polyunsaturated fatty acid status and pregnancy outcome 1 3 Renate HM de Groot, Gerard Hornstra, Adriana C van Houwelingen, and Frans Roumen ABSTRACT Background: Maternal essential fatty acid status declines during pregnancy, and as a result, neonatal concentrations of docosahexaenoic acid (DHA, 22:6n 3) and arachidonic acid (AA, 20:4n 6) may not be optimal. Objective: Our objective was to improve maternal and neonatal fatty acid status by supplementing pregnant women with a combination of -linolenic acid (ALA, 18:3n 3) and linoleic acid (LA, 18:2n 6), the ultimate dietary precursors of DHA and AA, respectively. Design: From week 14 of gestation until delivery, pregnant women consumed daily 25 g margarine supplying either 2.8 g ALA 9.0gLA(n 29) or 10.9 g LA (n 29). Venous blood was collected for plasma phospholipid fatty acid analyses at weeks 14, 26, and 36 of pregnancy, at delivery, and at 32 wk postpartum. Umbilical cord blood and vascular tissue samples were collected to study neonatal fatty acid status also. Pregnancy outcome variables were assessed. Results: ALA LA supplementation did not prevent decreases in maternal DHA and AA concentrations during pregnancy and, compared with LA supplementation, did not increase maternal and neonatal DHA concentrations but significantly increased eicosapentaenoic acid (20:5n 3) and docosapentaenoic acid (22:5n 3) concentrations. In addition, ALA LA supplementation lowered neonatal AA status. No significant differences in pregnancy outcome variables were found. Conclusions: Maternal ALA LA supplementation did not promote neonatal DHA AA status. The lower concentrations of Osbond acid (22:5n 6) in maternal plasma phospholipids and umbilical arterial wall phospholipids with ALA LA supplementation than with LA supplementation suggest only that functional DHA status improves with ALA LA supplementation. Am J Clin Nutr 2004;79: KEY WORDS Docosahexaenoic acid, arachidonic acid, -linolenic acid, linoleic acid, Osbond acid, pregnancy, neonatal outcome, essential fatty acids, pregnant women, fatty acid intakes, birth weight, gestational age INTRODUCTION It is well known that the essential fatty acid (EFA) status and long-chain polyene (LCP) status of pregnant women decrease during pregnancy (1). This particularly applies to arachidonic acid (AA, 20:4n 6) and docosahexaenoic acid (DHA, 22:6n 3), the major LCPs derived from linoleic acid (LA, 18:2n 6) and -linolenic acid (ALA, 18:3n 3), respectively. Because the developing fetus depends on its mother for LCP accretion (2, 3), neonatal LCP status may not be optimal under present dietary conditions (1, 4). AA and DHA are important building blocks in all cell membranes and are present in high concentrations in neural and retinal tissues (5 7). Infants born preterm often experience neurodevelopmental problems (8, 9), and although a causal relation with their low LCP status at birth has not been ascertained, such an association is suggested by the results of postnatal intervention studies, which generally show that early LCP supplementation improves neuro-mental development, at least temporarily (10 12). In term neonates, who have a higher LCP status than do preterm infants (13), LCP supplementation has also been shown to improve neuro-mental development, although the results are less convincing than for preterm infants (14 18). The central nervous system of a fetus undergoes a growth spurt in the last trimester of pregnancy. Therefore, adequate prenatal LCP availability can be considered of key importance for optimal brain development and function. This view is supported by the recent findings of Bakker (19), who showed that certain measures of brain maturation at 7 y of age are positively related to neonatal DHA status at birth. Maternal supplementation with fish oil has been used successfully to increase fetal DHA availability and neonatal DHA status. However, increasing the DHA status of pregnant women with fish oil lowers AA concentrations in their infants (20, 21). Because AA is the second most abundant LCP in neural tissue (22), this may not be desirable. Although endogenous DHA synthesis from dietary ALA is limited in humans (23 26), 1 From the Department of Human Biology (RHMdG and ACvH) and the Nutrition and Toxicology Research Institute Maastricht (NUTRIM) (RHMdG, GH, and ACvH), Maastricht University, Maastricht, Netherlands, and the Department of Obstetrics and Gynecology, Atrium Medical Center, Heerlen, Netherlands (FR). 2 Supported by a grant from Unilever Research and Development (Vlaardingen, Netherlands), which also donated the margarines used in the study. 3 Reprints not available. Address correspondence to RHM de Groot, Maastricht University, Department of Neuropsychology, PO Box 616, 6200MD Maastricht, Netherlands. rhm.degroot@np.unimaas.nl. Received March 25, Accepted for publication July 28, Am J Clin Nutr 2004;79: Printed in USA American Society for Clinical Nutrition 251

2 252 DE GROOT ET AL evidence suggests that this synthesis may increase during pregnancy (27). Moreover, ALA supplementation has been shown to result in the accretion of ALA-derived DHA in the brains of baboon fetuses (28). Therefore, dietary ALA may be an effective alternative to fish oil for use in increasing maternal and neonatal DHA status. Because Al et al (29) showed that neonatal n 6 LCP status can be increased by supplementation with LA during pregnancy, the aim of the present study was to increase the availability of maternal DHA and AA for fetal accretion by supplementation of pregnant women with a margarine containing relatively high amounts of ALA and LA, the ultimate dietary precursors of DHA and AA, respectively. SUBJECTS AND METHODS Experimental design The study was a double-blind, randomized, controlled dietary intervention trial in pregnant women who were randomly assigned to receive daily 25 g of either an ALA-enriched, high-la margarine (experimental group) or a high-la margarine without ALA (control group) from week 14 of pregnancy until delivery. During pregnancy, the subjects were visited at their homes every 3 wk so that they could be provided with the margarines, fill out questionnaires, or have blood collected (at weeks 14, 26, and 36 of pregnancy). Immediately on delivery, another maternal blood sample was collected, together with a blood sample from the umbilical vein and a piece of the umbilical cord. Finally, a maternal blood sample was drawn 32 wk after delivery. From the blood and tissue samples collected, phospholipid fatty acid profiles were determined by gas chromatography for investigation of the effect of the intervention on the EFA and LCP status of the mothers during pregnancy and at 32 wk after delivery and of the neonates at birth. Subjects The pregnant subjects were recruited by midwives in the region around Maastricht, Heerlen, and Sittard in the southeastern part of the Netherlands and by the Departments of Obstetrics and Gynecology of hospitals in the same area (University Hospital Maastricht, Atrium Medical Center in Heerlen, and Maasland Hospital in Sittard). The selection criteria for study entry were as follows: white origin; a gestational age 14 wk; normal health, ie, not suffering from any hypertensive, metabolic, cardiovascular, renal, psychiatric, or neurologic disorder; and fish consumption 2 times/wk. Earlier studies by our group showed average plasma phospholipid DHA concentrations of 4.07% and 3.80% by wt at weeks 14 and 36 of pregnancy, respectively (30). Because the decrease in DHA concentration between weeks 14 and 36 of pregnancy was 0.27% by wt, and because we aimed at preventing this decrease by ALA supplementation, the target for the difference in DHA concentration between the experimental group and the control group at 36 wk of pregnancy was set at 0.27% by wt. At the SD of our DHA measurement (0.34% by wt) and a power of 90% (at 0.05), the number of subjects needed in each group was calculated to be 27. However, the women were oversampled because of expected withdrawals and dropouts during the study. A total of 79 women enrolled in the study, which was approved by the Medical Ethics Committees of the University Hospital Maastricht and Maasland TABLE 1 Fatty acid composition of the margarines given to the 2 groups 1 Fatty acids Control group Experimental group % of total fatty acids Saturated Monounsaturated LA (18:2n 6) Total n ALA (18:3n 3) Total n trans Unidentified LA, linoleic acid; ALA, -linolenic acid. Hospital in Sittard. Written informed consent was obtained from all participants. Supplements The margarines, which were provided by Unilever Research and Development (Vlaardingen, Netherlands), contained 79.5% fat, and the remainder consisted of water (20%), vitamins (0.04%), flavor (0.04%), lecithin (0.3%), and butylated hydroxytoluene (0.12%). The fatty acid compositions of the margarines, which were determined after lipid extraction by gas chromatographic analysis, are shown in Table 1. Inthe margarine given to the experimental group, LA and ALA constituted 45.4% and 14.2% of total fatty acids, respectively; thus, with the requested intake of 25 g margarine/d, the subjects in the experimental group consumed 9.02 g LA/d and 2.82 g ALA/d. This amount of margarine was about equal to the subjects habitual margarine or butter intake (24.9 g) as measured by using food-frequency questionnaires (FFQs). In the margarine given to the control group, LA and ALA constituted 55.02% and 0.17% of total fatty acids, respectively; thus, with the requested intake of 25 g margarine/d, the subjects in the control group consumed g LA/d and 0.03 g ALA/d. The composition of the ALA-containing margarine was based on calculations that daily consumption of 25 g of this margarine would result in a ratio of n 3 ton 6 fatty acids of 1:5 in the total diet of the experimental group. The choice of this ratio was based on guidelines for polyunsaturated fatty acid intake issued by various international authorities; these guidelines state that for optimum benefit, the ratio of n 3 ton 6 fatty acids should be between 1:4 and 1:10, preferably 1:5 (31 33). Every 3 wk the volunteers received 3 tubs each containing 250 g margarine. The leftovers were collected and weighed to estimate total consumption. The subjects were instructed to consume the margarines primarily on bread. If their consumption was lower than the required 25 g/d, the subjects were advised to put the margarine on top of potatoes or pasta (in place of the habitual sauce). The subjects were not allowed to use the margarine for baking because of possible adverse effects on the polyunsaturated fatty acid content of the margarines. The subjects were allowed to maintain their usual diets during the entire course of the study, with the exception of the use of butter or their usual brand of margarine, which were to be replaced by the experimental or control margarines.

3 -LINOLENIC ACID SUPPLEMENTATION DURING PREGNANCY 253 Questionnaires Assessment of dietary fat intake Because dietary intake can influence the plasma fatty acid profile, the subjects fat intake was measured by using a well-validated, prestructured FFQ (3). This FFQ was especially designed to collect data on fat consumption. Frequency of consumption had to be recorded, and amounts eaten had to be indicated in household units or grams. Whether fat intake changed during pregnancy was monitored by having the women complete this FFQ at weeks 14 and 36 of pregnancy. After the FFQs were returned, they were checked by an experienced dietitian and, if required, were corrected after a telephone interview. The food consumption data were converted into dietary intake data, including fatty acid intake data, by using the computer program KOMEET (34). This program is based on the database of the Dutch Nutrient Databank (35). Other questionnaires At the start of the study (week 14), the subjects filled out an additional questionnaire, which included items about age, prepregnancy weight, height of the mother, height and weight of the father, smoking habits, education, the use of supplements, and medical history, including former pregnancies and medical treatments. Education was scored on an 8-point scale, ranging from primary education to higher vocational training and university education (36). After delivery, the subjects completed a medical questionnaire about the course of the current pregnancy, blood transfusion, gestational age at delivery, course of parturition, sex of the newborn, birth weight, and Apgar score. Fatty acid analysis Plasma was separated from blood cells by centrifugation ( g, 10 min, 4 C) and collected in tubes, which were closed under nitrogen and stored at 80 C until fatty acid analysis. From the pieces of umbilical cords, umbilical arteries and veins were isolated. The vein and both arteries from each umbilical cord were frozen in liquid nitrogen, pulverized, and freeze-dried before lipid extraction (see below). Fatty acid profiles of phospholipids isolated from maternal venous plasma, umbilical venous plasma, and umbilical venous and arterial vessel walls were determined as previously described by Al et al (1) and Otto et al (37). Briefly, after addition of an internal standard (1,2-dinonadecanoyl-sn-glycero-3- phosphocholine), plasma or tissue total lipid extracts were prepared by a modified Folch extraction method (38), and phospholipid fractions were isolated from the lipid extracts by using aminopropyl (500 mg/4.0 ml) Extract-Clean columns (Alltech, Breda, Netherlands) (39). Heptadecanoic acid (17:1) was added to the samples so that any carryover of free fatty acids during the isolation of phospholipids could be detected. The phospholipid fractions were hydrolyzed, and the fatty acids were methylated with boron trifluoride in methanol (40). The fatty acid composition of the phospholipids was then determined by using capillary gas chromatography with a WCOT fused silica CP-SIL 88 fame column (50 m 0.25 mm inside diameter, film thickness of 0.2 m; Varian, Bergen op Zoom, Netherlands) and helium as the carrier gas. The injection and detection temperatures were 300 C. The starting temperature of the column was 160 C. After 10 min, the temperature was increased 3.2 C/min up to 190 C and then kept constant for 15 min. Finally, the temperature was increased up to 230 C at a rate of 5 C/min. Total amounts of phospholipid-associated fatty acids are expressed as mg/l plasma or mg/kg tissue, and relative fatty acid concentrations are expressed as percentages of the total amount of phospholipid-associated fatty acids (% by wt). Forty-two fatty acids were identified, but only the following selection will be reported (full results available on request): LA, dihomo- -linolenic acid (20:3n 6), AA, adrenic acid (22:4n 6), Osbond acid (ObA, 22:5n 6), ALA, eicosapentaenoic acid (EPA, 20:5n 3), docosapentaenoic acid (DPA, 22:5n 3), and DHA. In addition, the following fatty acid combinations and ratios are presented: sum of saturated fatty acids, sum of monounsaturated fatty acids, sum of n 7 fatty acids, sum of n 9 fatty acids, total amount of LCPs of the n 3 and n 6 families (n 3 LCPs and n 6 LCPs; LCPs were defined as fatty acids with 20 carbon atoms and 3 double bonds), EFA status index [(sum n 3 fatty acids sum n 6 fatty acids)/(sum n 7 fatty acids sum n 9 fatty acids)], DHA deficiency index (DHADI, 22:5n 6/22:4n 6), and DHA sufficiency index (DHASI, 22:6n 3/22:5n 6). Statistics The various statistical techniques used to evaluate the data are detailed in the Results. The statistical package SPSS 10.0 (SPSS Inc, Chicago) was used to perform the statistical analyses, and data are presented as means SDs. In this study, 3 fatty acids were of primary interest: DHA, AA, and ObA. The latter fatty acid is generally accepted as the deficiency indicator of DHA, because ObA synthesis increases if there is a functional DHA shortage (41). These fatty acids of primary interest were studied separately from the other fatty acids and fatty acid combinations. For these principal fatty acids, the P value required for significance was set at Analyses of the data for the other fatty acids required adaptation of this P value to P 0.01 because of multiple testing. RESULTS Subjects A total of 79 pregnant women enrolled in the study. However, 21 women were not followed up completely: 3 subjects (2 in the experimental group and 1 in the control group) had a premature delivery (before week 36 of gestation), 4 subjects (1 in the control group and 3 in the experimental group) were not motivated to complete the study because they considered it too time consuming, and 3 subjects (1 in the control group and 2 in the experimental group) were excluded for noncompliance. Two women (both in the control group) dropped out because they did not like the margarine, and 2 subjects (1 in each group) withdrew because they experienced too much morning sickness. One subject in the control group had to be removed from the study because of a stillbirth, and 1 woman in the control group was removed because she developed diabetes mellitus gravidae. Two women (1 in each group) were lost to follow-up because of long-term hospitalization during the study, 1 woman in the experimental group was lost because of a lengthy

4 254 DE GROOT ET AL stay abroad, and 2 subjects (1 in each group) were lost because of insufficient blood samples for analysis. A total of 58 women (29 in each group) completed the study until delivery. During the period after delivery, 2 subjects in the experimental group dropped out: 1 subject could not be reached in time because she moved away from the research area, and 1 subject withdrew because of postpartum depression. All the remaining mothers had uncomplicated pregnancies and delivered full-term, singleton newborns. Maternal characteristics Maternal characteristics are shown in Table 2. Comparisons between the groups were performed by using either the unpaired t test or the Mann-Whitney U test for continuous variables and the chi-square test for discrete variables. Except for the number of breastfeeding mothers, no significant differences between the groups were observed. Significantly more mothers in the experimental group than in the control group breastfed their infants (P 0.020, chi-square test). The mean duration of breastfeeding by the lactating mothers did not differ significantly between the 2 groups, although a tendency was observed for a longer duration of breastfeeding in the experimental group (P 0.095, unpaired t test). Maternal dietary fatty acid intake Daily margarine consumption was g in the control group and g in the experimental group. The difference between the 2 groups was not significant (unpaired t test). Dietary fatty acid intakes measured by using an FFQ in weeks 14 and 36 of pregnancy are shown in Table 3. At week 14, the differences between the 2 groups were not significant (Mann-Whitney U test). The same holds true for week 36, with the exception of the ALA intake, which was significantly higher in the experimental group than in the control group (P 0.001, Mann-Whitney U test) During the intervention period, the total intake of polyunsaturated fatty acids and the intake of LA increased significantly in both groups, whereas total fat intake increased significantly only in the experimental group (Wilcoxon s signed-ranks test). The increase in LA intake was significantly higher in the control group than in the experimental group (Mann-Whitney U test). ALA intake increased significantly in the experimental group and decreased significantly in the control group (Wilcoxon s signedranks test), which made the difference in response between the groups significant also (Mann-Whitney U test). The subjects in the experimental group had higher DHA consumption at week 14 of pregnancy than did those in the control group, and this difference showed a trend for significance (P 0.074, Mann-Whitney U test). The changes in DHA and EPA intakes during pregnancy did not differ significantly between the 2 groups (Mann-Whitney U test). However, the decreases in DHA and EPA intakes in the experimental group showed a trend for significance (P and 0.076, respectively). Maternal plasma phospholipid fatty acids To analyze the potential differences between the control group and the experimental group in maternal fatty acid concentrations in plasma phospholipids during pregnancy (weeks TABLE 2 Subject characteristics Characteristic Control group (n 29) Experimental group (n 29) Maternal Age (y) Education Height (cm) Prepregnancy weight (kg) Parity (n) Smoking at week 14 (n) Yes 10 4 No Alcohol consumption at week 14 (n) Yes 8 11 No Medication during pregnancy (n) Yes 7 6 No Delivery (n) Location Home Hospital Method Spontaneous Forceps or vacuum extraction 2 3 Cesarean 4 3 Breastfeeding (n) Duration (d) Neonatal Birth weight (g) Gestational age (d) Sex (n) Male Female Apgar score x SD. 2 Measured on an 8-point scale. 3 Significantly different from the control group, P (chi-square test). 5,6 Significantly different from the control group after correction for gestational age, P (one-way ANOVA). 5 Maximum within 10 min ), the general linear model for repeated measures was used with correction for fatty acid concentrations at the start of the study (week 14), smoking (42), and parity (43). The between-subjects factor had 2 levels (control group and experimental group), whereas the within-subjects factor, time, had 3 levels (week 26, week 36, and partus, which was set at week 40). Missing values during pregnancy were imputed by using missing value analyses. Only one missing value per subject was allowed. Mean total amounts of phospholipid-associated fatty acids

5 -LINOLENIC ACID SUPPLEMENTATION DURING PREGNANCY 255 TABLE 3 Dietary fat intake during pregnancy in the control group (C, n 26) and the experimental group (E, n 29) as measured by using a foodfrequency questionnaire 1 Intake Week 14 Week 36 Change Total fat (g/d) C E Total SFAs (g/d) C E Total MUFAs (g/d) C E Total PUFAs (g/d) C E LA, 18:2n 6 (g/d) C E ,5 ALA, 18:3n 3 (g/d) C E ,6 EPA, 20:5n 3 (mg/d) C E DHA, 22:6n 3 (mg/d) C E x SD. SFAs, saturated fatty acids; MUFAs, monounsaturated fatty acids; PUFAs, polyunsaturated fatty acids; LA, linoleic acid; ALA, -linolenic acid; EPA, eicosapentaenoic acid; DHA, docosahexaenoic acid. 2 P 0.05 (Wilcoxon s signed-ranks test). 3 P (Wilcoxon s signed-ranks test). 4 P 0.01 (Wilcoxon s signed-ranks test). 5,6 Significantly different from C (Mann-Whitney U test): 5 P 0.05, 6 P did not differ significantly between the 2 groups. Therefore, only the relative fatty acid concentrations (% by wt) are reported in Table 4. Compared with consumption of the control margarine, consumption of the experimental margarine resulted in significantly higher ALA concentrations (P 0.001). In both groups, DHA concentrations decreased significantly during the intervention period (P 0.028). Because these decreases were not significantly different between the groups, the overall DHA concentrations were not significantly different between the groups. Changes in the same direction were observed for AA, although these were not significant. ObA concentrations increased significantly during pregnancy (P 0.026). However, overall ObA concentrations were significantly lower in the experimental group than in the control group (P 0.002). Adrenic acid concentrations were significantly lower in the experimental group than in the control group (P 0.001), whereas EPA and DPA concentrations were significantly higher (P and 0.001, respectively). Differences between the groups during pregnancy in the other fatty acids and fatty acid combinations listed in Table 4 were not significant, although the lower dihomo- -linolenic acid concentrations, lower total n 6 LCP concentrations, and higher DHASI values in the experimental group than in the control group had a probability 95% (P required for significance 0.01). Univariate analysis of variance with group as the betweensubjects factor (control group and experimental group) and with maternal fatty acid concentration at the start of the study (week 14), smoking, parity, and duration of breastfeeding (27) as covariables was used to investigate potential differences in maternal plasma fatty acid concentrations between the 2 groups at 32 wk after delivery (week 72). At this point in time, only ALA concentrations were still significantly different between the 2 groups, with higher concentrations in the experimental group than in the control group (P 0.004). Neonatal fatty acids Mean total amounts of phospholipid-associated fatty acids in umbilical plasma phospholipids and in phospholipids from umbilical vessel walls (both venous and arterial) did not differ significantly between the 2 groups (unpaired t tests). Therefore, only the relative amounts of fatty acids (% by wt) are reported in Table 5. The differences in neonatal fatty acid concentrations between the 2 groups were analyzed by univariate analysis of variance with correction for gestational age, parity, maternal smoking, and maternal fatty acid concentrations at the start of the study (week 14) if necessary after log transformation to ascertain normality. Umbilical venous plasma Umbilical venous plasma samples were available for 26 newborns in the control group and 28 newborns in the experimental group (Table 5). No significant difference in umbilical plasma DHA concentrations in phospholipids was found between the 2 groups. In contrast, the AA concentrations were significantly lower in the experimental group than in the control group (P 0.004). After log transformation and correction for potential confounders, ObA concentrations did not differ significantly between the 2 groups (P 0.095). The average EPA concentration in the experimental group was 2 times that in the control group (P 0.001), whereas total n 6 LCP concentrations were significantly lower in the experimental group than in the control group (P 0.004). Values for other fatty acids and fatty acid combinations were not significantly different between the 2 groups. Umbilical vein walls From each group, 28 samples of umbilical veins and 28 samples of umbilical arteries were available for analysis. Results are shown in Table 5. After correction for potential confounders, neither DHA (P 0.183) nor ObA (P 0.085) concentrations differed significantly between the 2 groups. AA concentrations also did not differ significantly between the 2 groups. DPA concentrations were significantly higher in the experimental group than in the control group (P 0.001). There were trends for higher DHASI and n 3 LCP values (P and 0.038, respectively) and lower DHADI values (P 0.034) in the experimental group than in the control group. Other fatty acids and fatty acid combinations were not significantly different between the 2 groups.

6 256 DE GROOT ET AL TABLE 4 Plasma phospholipid-associated fatty acid concentrations over time in mothers in the control group (C) and the experimental group (E) 1 Fatty acids Week 14 (n 29/29) 2 Week 26 (n 29/29) Week 36 (n 28/29) Week 40 (n 26/27) Week 72 (n 28/28) Total (mg/l) 3 C E :2n 6, LA (% by wt) C E :3n 6, DGLA (% by wt) 4 C E :4n 6, AA (% by wt) C E :4n 6, AdrA (% by wt) 3,5 C E :5n 6, ObA (% by wt) 6,7 C E n 6 LCPs (% by wt) 4 C E :3n 3, ALA (% by wt) 5 C E :5n 3, EPA (% by wt) 6 C E :5n 3, DPA (% by wt) 5 C E :6n 3, DHA (% by wt) 7 C E n 3 LCPs (% by wt) 7 C E Sum of n 7 FAs (% by wt) C E Sum of n 9 FAs (% by wt) C E EFA index C E SFAs (% by wt) C E PUFAs (% by wt) C E MUFAs (% by wt) C E DHADI C E DHASI 4 C E x SD. LA, linoleic acid; DGLA, dihomo- -linolenic acid; AA, arachidonic acid; AdrA, adrenic acid; ObA, Osbond acid; LCPs, long-chain polyenes (fatty acids with 20 carbon atoms and 3 double bonds); ALA, -linolenic acid; EPA, eicosapentaenoic acid; DPA, docosapentaenoic acid; DHA, docosahexaenoic acid; FAs, fatty acids; EFA index, essential fatty acid index [(sum of n 3 FAs sum of n 6 FAs)/(sum of n 7 FAs sum of n 9 FAs)]; SFAs, sum of saturated fatty acids; PUFAs, sum of polyunsaturated fatty acids; MUFAs, sum of monounsaturated fatty acids; DHADI, DHA deficiency index (22:5n 6/22:4n 6); DHASI, DHA sufficiency index (22:6n 3/22:5n 6). No significant time by group interactions were found. 2 Control group/experimental group. 3,7 Significant change over time during pregnancy (weeks 14 40) (general linear model for repeated measures with correction for several potential confounders): 3 P 0.01, 7 P Significant difference between the 2 groups during pregnancy (weeks 14 40) (general linear model for repeated measures with correction for several potential confounders): 4 P 0.05, 5 P 0.001, 6 P Significantly different from C, P 0.01 (univariate ANOVA).

7 -LINOLENIC ACID SUPPLEMENTATION DURING PREGNANCY 257 TABLE 5 Neonatal phospholipid-associated fatty acid concentrations in the control group (C) and the experimental group (E) 1 Fatty acids Plasma Venous wall (n 26/28) 2 (n 28/28) Arterial wall (n 28/28) Total (mg/l plasma or mg/kg tissue) C E :2n 6, LA (% by wt) C E :3n 6, DGLA (% by wt) C E :4n 6, AA (% by wt) C E :4n 6, AdrA (% by wt) C E :5n 6, ObA (% by wt) C E n 6 LCPs (% by wt) C E :3n 3, ALA (% by wt) C ND E ND :5n 3, EPA (% by wt) C ND ND E ND ND 22:5n 3, DPA (% by wt) C E :6n 3, DHA (% by wt) C E n 3 LCPs (% by wt) C E Sum of n 7 FAs (% by wt) C E Sum of n 9 FAs (% by wt) C E EFA index C E SFAs (% by wt) C E PUFAs (% by wt) C E MUFAs (% by wt) C E DHADI C E DHASI C E x SD. LA, linoleic acid; DGLA, dihomo- -linolenic acid; AA, arachidonic acid; AdrA, adrenic acid; ObA, Osband acid; LCPs, long-chain polyenes (fatty acids with 20 carbon atoms and 3 double bonds); ALA, -linolenic acid; EPA, eicosapentaenoic acid; DPA, docosapentaenoic acid; DHA, docosahexaenoic acid; FAs, fatty acids; EFA index, essential fatty acid index [(sum of n 3 FAs sum of n 6 FAs)/(sum of n 7 FAs sum of n 9 FAs)]; SFAs, sum of saturated fatty acids; PUFAs, sum of polyunsaturated fatty acids; MUFAs, sum of monounsaturated fatty acids; DHADI, DHA deficiency index (22:5n 6/22:4n 6); DHASI, DHA sufficiency index (22:6n 3/22:5 n 6); ND, not reliably detectable (concentrations 0.05% by wt). GLM was used to evaluate the arteriovenous differences (data not shown). Significant within-subjects effects (venous wall and arterial wall) were found for total FAs, LA, ObA, sum of n 9 FAs, MUFAs, DHADI, and DHASI, whereas significant between-subjects effects (control group and experimental group) were found for ObA, DPA, DHADI, and DHASI. No significant interaction effects were found. 2 Control group/experimental group. 3 5 Significantly different from C [univariate ANOVA with correction for gestational age, parity, maternal smoking, and maternal fatty acid concentrations at the start of the study (week 14)]: 3 P 0.05, 4 P 0.01, 5 P

8 258 DE GROOT ET AL Umbilical arterial walls After correction for potential confounders, phospholipid ObA concentrations in the experimental group were significantly lower than those in the control group (P 0.037). The differences in DHA and AA concentrations between the 2 groups were not significant. DPA concentrations were significantly higher in the experimental group than in the control group (P 0.001). Differences between the 2 groups in the other fatty acids and fatty acid combinations were not significant (Table 5). Arteriovenous differences As shown by the general linear model, significant withinsubjects effects for arteriovenous differences in many of the fatty acids listed in Table 5 (total fatty acids, LA, ObA, sum of n 9 fatty acids, monounsaturated fatty acids, and DHADI) were found (data not shown), whereas the group arteriovenous interactions were not significant. Consequently, the arteriovenous differences in the experimental group were not significantly different from those in the control group. However, significant between-subjects (control group and experimental group) effects were found for ObA, DPA, DHADI, and DHASI, which indicated that the average amount of phospholipid fatty acids in arterial and venous vessel walls was lower for DPA and DHASI and higher for ObA and DHADI in the control group than in the experimental group. Neonatal outcome variables With the use of one-way analysis of variance, the newborns in the experimental group had a significantly higher mean birth weight than did those in the control group (P 0.043; Table 2). Mean pregnancy duration was 4.5 d longer in the experimental group than in the control group (Table 2). This difference tended to be significant as well (P 0.091, Kruskal- Wallis test). No significant differences between the groups in Apgar score (P 0.05, Mann-Whitney U test) or sex (P 0.05, chi-square test) were observed. DISCUSSION It is known from longitudinal studies that under present dietary conditions, relative AA and DHA concentrations in maternal plasma phospholipids decrease during the second and third trimester of pregnancy (1, 37, 44, 45). Supplementation with a combination of n 6 and n 3 fatty acids was suggested as being required for optimal fatty acid status (46). Although endogenous DHA production from dietary ALA is known to be low (23 26), Otto et al (27) obtained some indications that LCP synthesis from EFA precursors may be enhanced during pregnancy. Therefore, in the present study, margarine high in ALA and LA was tested for its efficacy in increasing maternal and neonatal DHA concentrations without reducing AA concentrations. Some neonatal outcome variables were also measured in this study. Some of the results obtained require special attention and are discussed here. Maternal dietary fat intake Analysis of the FFQs used in the present study showed that the participants total fat intake increased during pregnancy, although this increment was significant for the experimental group only, and the difference in intake between the 2 groups was not significant. Using the same FFQ, Al et al (3) did not observe a change in habitual fat consumption during the second and third trimesters of pregnancy. Therefore, the difference that we observed probably resulted from the study regimen. This was confirmed by an analysis of the items in the FFQs, which showed a rather frequent habitual use of light margarines at week 14, which were replaced in the study by the full-fat experimental and control intervention products. The higher fat intake provided about 108 kcal (0.026 MJ)/d. This partly satisfied the recommendation for higher energy intake during pregnancy (47). Maternal plasma fatty acid concentrations during pregnancy In the present study, supplementation with margarines rich in ALA LA or in LA only did not prevent the well-known reductions in DHA and AA concentrations during the second and third trimesters of pregnancy (Table 4). In addition, plasma DHA concentrations after supplementation with the ALA LA rich margarine were not significantly higher than those after supplementation with a comparable margarine without ALA. Although the possibility that a higher ALA dose may increase plasma DHA concentrations cannot be excluded, this seems rather unlikely because Francois et al (48) showed that even an ALA dosage of 10.7 g/d does not increase plasma DHA concentrations in lactating women. Interestingly, the lower ObA concentrations in the experimental group of our study than in the control group indicate that functional DHA status may have been slightly higher with ALA supplementation, because ObA is generally accepted as a functional shortage marker for DHA (41). This suggests that any additional DHA that may have been produced from the supplemented ALA was transferred directly to certain (fetal) target tissues and, therefore, did not increase the DHA concentration of maternal plasma phospholipids. Maternal fatty acids 32 wk postpartum In the present study, the effect of ALA supplementation during pregnancy on maternal plasma fatty acid concentrations was also studied in blood collected 32 wk postpartum. The choice of this time was made on the basis of an earlier study by Otto et al (27), which showed that maternal EFA and LCP concentrations return to prepregnancy values 32 wk postpartum, irrespective of whether or not the mother breastfeeds. We confirmed that the concentrations of most fatty acids at 32 wk postpartum were either equal to or higher than the corresponding concentrations during early pregnancy. Moreover, the concentrations were not significantly different between the 2 groups. The only exception was ALA, which was still higher in the experimental group. Although this may indicate a longterm effect of the dietary intervention, a difference due to selection bias cannot be ruled out. Neonatal phospholipid fatty acids The significantly lower ObA concentrations in umbilical arterial vessel walls in the experimental group than in the control group suggests a higher functional DHA status in infants of ALA-supplemented mothers, although DHA concentrations themselves were not significantly higher in the exper-

9 -LINOLENIC ACID SUPPLEMENTATION DURING PREGNANCY 259 imental group. This suggestion is supported by the significantly higher DHASI values and significantly lower DHADI values in umbilical venous wall phospholipids in the neonates in the experimental group. Moreover, maternal ALA supplementation during pregnancy resulted in significantly higher neonatal concentrations of EPA (plasma only) and DPA (umbilical vessel walls) than did supplementation with LA without ALA. Maternal ALA supplementation during pregnancy was also associated with lower neonatal AA concentrations and lower neonatal total concentrations of n 6 LCPs (significant in plasma only). These results show that maternal ALA consumption during pregnancy hardly increases neonatal DHA status and does not prevent a reduction in n 6 LCPs. This reduction in n 6 LCPs during pregnancy was also not prevented by fish-oil supplementation during pregnancy (20, 21). Neonatal outcome variables High birth weight has been associated with a low risk of development of adult heart disease, type 2 diabetes mellitus, and related risk factors such as hypertension and dyslipidemia (49 52). Consequently, in the present study, the significantly higher birth weight of the neonates born to ALA-supplemented mothers than of those born to mothers supplemented with LA only may be relevant indeed. Therefore, we performed a post hoc stepwise multiple regression analysis to detect possible confounding variables. On the basis of the literature, gestational age, alcohol consumption, neonatal sex, maternal prepregnancy weight, maternal height, height and weight of the father, smoking, and drug usage during pregnancy were included in the model. Gestational age, maternal prepregnancy weight, smoking, and height of the father contributed significantly to the model (P 0.05), and after correction for these possible confounders, no significant difference in birth weight remained between the experimental group and the control group. Conclusion Supplementation with ALA LA during pregnancy does not prevent a decrease in maternal LCPs during pregnancy, whereas the small improvement in neonatal DHA status is accompanied by a reduction in neonatal n 6 LCP status. Because supplementation with DHA is more efficient than is supplementation with ALA in promoting DHA status (53), we suggest that a mixture of DHA and AA may be a more efficient option to optimize maternal and neonatal LCP status. We thank all the mothers and their infants for participating in this study. The participation of the midwives from the region of Southern Limburg and of the medical staff of the Departments of Obstetrics and Gynecology of the hospitals in the same region is greatly appreciated. We thank H Aydeniz for technical assistance; A Kester for statistical advice; A Badart- Smook, J Breedveld (NutriScience, Maastricht, Netherlands), and I Verkooijen (NutriScience) for their help in the analyses of the FFQs; and E de Deckere (Unilever Research and Development, Vlaardingen, Netherlands) for critically reading the manuscript. RHMdG was the principal investigator. GH was the supervisor of the study. RHMdG and GH wrote the manuscript. ACvH was involved in development of the study design and in management of the study and critically reviewed the manuscript. FR assisted in subject recruitment and reviewed the manuscript. None of the authors had any conflicts of interest. REFERENCES 1. Al MD, van Houwelingen AC, Kester AD, Hasaart TH, de Jong AE, Hornstra G. Maternal essential fatty acid patterns during normal pregnancy and their relation to the neonatal essential fatty acid status. Br J Nutr 1995;74: Crawford M. Placental delivery of arachidonic and docosahexaenoic acids: implications for the lipid nutrition of preterm infants. Am J Clin Nutr 2000;71(suppl):275S 84S. 3. Al MD, Badart-Smook A, van Houwelingen AC, Hasaart TH, Hornstra G. Fat intake of women during normal pregnancy: relation with maternal and neonatal essential fatty acid status J Am Coll Nutr 1996;15: Hornstra G. Essential fatty acids in mothers and their neonates. Am J Clin Nutr 2000;71(suppl):1262S 9S. 5. Fliesler SJ, Anderson RE. Chemistry and metabolism of lipids in the vertebrate retina. Prog Lipid Res 1983;22: Sastry PS. Lipids of nervous tissue: composition and metabolism. Prog Lipid Res 1985;24: Svennerholm L. Distribution and fatty acid composition of phosphoglycerides in normal human brain. J Lipid Res 1968;9: Wolke D, Meyer R. Cognitive status, language attainment, and prereading skills of 6-year-old very preterm children and their peers: the Bavarian Longitudinal Study. Dev Med Child Neurol 1999;41: Wolke D, Ratschinski G, Ohrt B, Riegel K. The cognitive outcome of very preterm infants may be poorer than often reported: an empirical investigation of how methodologic issues make a big difference. Eur J Pediatr 1994;153: Simmer K. Longchain polyunsaturated fatty acid supplementation in preterm infants. Cochrane Database Syst Rev 2000;2:CD SanGiovanni JP, Parra-Cabrera S, Colditz GA, Berkey CS, Dwyer JT. Meta-analysis of dietary essential fatty acids and long-chain polyunsaturated fatty acids as they relate to visual resolution acuity in healthy preterm infants. Pediatrics 2000;105: O Connor DL, Hall R, Adamkin D, et al. Growth and development in preterm infants fed long-chain polyunsaturated fatty acids: a prospective, randomized controlled trial. Pediatrics 2001;108: Foreman-van Drongelen MM, Al MD, van Houwelingen AC, Blanco CE, Hornstra G. Comparison between the essential fatty acid status of preterm and full-term infants, measured in umbilical vessel walls. Early Hum Dev 1995;42: Simmer K. Longchain polyunsaturated fatty acid supplementation in infants born at term. Cochrane Database Syst Rev 2001;4:CD SanGiovanni JP, Berkey CS, Dwyer JT, Colditz GA. Dietary essential fatty acids, long-chain polyunsaturated fatty acids, and visual resolution acuity in healthy fullterm infants: a systematic review. Early Hum Dev 2000;57: Agostoni C, Trojan S, Bellu R, Riva E, Giovannini M. Neurodevelopmental quotient of healthy term infants at 4 mo and feeding practice: the role of long-chain polyunsaturated fatty acids. Pediatr Res 1995; 38: Willatts P, Forsyth JS, DiModugno MK, Varma S, Colvin M. Effect of long-chain polyunsaturated fatty acids in infant formula on problem solving at 10 mo of age. Lancet 1998;352: Birch EE, Garfield S, Hoffman DR, Uauy R, Birch DG. A randomized controlled trial of early dietary supply of long-chain polyunsaturated fatty acids and mental development in term infants. Dev Med Child Neurol 2000;42: Bakker EC. Long-chain polyunsaturated fatty acids and child development. PhD thesis. Maastricht University, Maastricht, Netherlands, 2002: Helland IB, Saugstad OD, Smith L, et al. Similar effects on infants of n 3 and n 6 fatty acids supplementation to pregnant and lactating women. Pediatrics [serial online] 2001;108:e82.Internet: (accessed 5 November 2003). 21. van Houwelingen AC, Sorensen JD, Hornstra G, et al. Essential fatty acid status in neonates after fish-oil supplementation during late pregnancy. Br J Nutr 1995;74: Lauritzen L, Hansen HS, Jorgensen MH, Michaelsen KF. The essentiality of long chain n 3 fatty acids in relation to development and function of the brain and retina. Prog Lipid Res 2001;40: Burdge GC, Jones AE, Wootton SA. Eicosapentaenoic and docosapentaenoic acids are the principal products of -linolenic acid metabolism in young men. Br J Nutr 2002;88:

10 260 DE GROOT ET AL 24. Burdge GC, Wootton SA. Conversion of -linolenic acid to eicosapentaenoic, docosapentaenoic and docosahexaenoic acids in young women. Br J Nutr 2002;88: Gerster H. Can adults adequately convert -linolenic acid (18:3n 3) to eicosapentaenoic acid (20:5n 3) and docosahexaenoic acid (22: 6n 3)? Int J Vitam Nutr Res 1998;68: Vermunt SH, Mensink RP, Simonis MM, Hornstra G. Effects of dietary -linolenic acid on the conversion and oxidation of 13 C- linolenic acid. Lipids 2000;35: Otto SJ, van Houwelingen AC, Badart-Smook A, Hornstra G. The postpartum docosahexaenoic acid status of lactating and nonlactating mothers. Lipids 1999;34(suppl):S Greiner RC, Winter J, Nathanielsz PW, Brenna JT. Brain docosahexaenoate accretion in fetal baboons: bioequivalence of dietary -linolenic and docosahexaenoic acids. Pediatr Res 1997;42: Al MD, van Houwelingen AC, Badart-Smook A, Hornstra G. Some aspects of neonatal essential fatty acid status are altered by linoleic acid supplementation of women during pregnancy. J Nutr 1995;125: Al MDM. Essential fatty acids, pregnancy, and pregnancy outcome: relation between mother and child. PhD thesis. Maastricht University, Maastricht, Netherlands, 1994: European Commission. Scientific Committee for Food. Nutrient and energy intakes for the European community. Luxembourg: Office for Official Publications of the European Communities, 1993: Scientific Review committee. Nutrition recommendations. Ottawa: Minister of National Health and Welfare, Canada, FAO/WHO. Fats and oils in human nutrition: report of a joint expert consultation. Rome: FAO, Scholte BA. KOMEET 1996, version d ed. Wageningen, Netherlands: Wageningen University, Netherlands Bureau of Nutrition Education. Dutch food composition table The Hague: NEVO Foundation, de Bie SE. Standaardvragen 1987: voorstellen voor uniformering van vraagstellingen naar achtergrondkenmerken en interviews. [Standard questions 1987: proposal for standardization of questions regarding background variables and interviews.] 2nd ed. Leiden, Netherlands: Leiden University Press, 1987 (in Dutch). 37. Otto SJ, van Houwelingen AC, Antal M, et al. Maternal and neonatal essential fatty acid status in phospholipids: an international comparative study. Eur J Clin Nutr 1997;51: Bligh EG, Dyer WJ. A rapid method of total lipid extraction and purification. Can J Biochem Physiol 1959;37: Kaluzny MA, Duncan LA, Merritt MV, Epps DE. Rapid separation of lipid classes in high yield and purity using bonded phase columns. J Lipid Res 1985;26: Morrison WR, Smith LM. Preparation of fatty acid methyl esters and dimethylacetals from lipids with boron fluoride-methanol. J Lipid Res 1964;5: Neuringer M, Connor WE, Lin DS, Barstad L, Luck S. Biochemical and functional effects of prenatal and postnatal omega 3 fatty acid deficiency on retina and brain in rhesus monkeys. Proc Natl Acad Sci U S A 1986;83: Rise P, Marangoni F, Galli C. Regulation of PUFA metabolism: pharmacologic and toxicological aspects. Prostaglandins Leukot Essent Fatty Acids 2002;67: Al MD, van Houwelingen AC, Hornstra G. Long-chain polyunsaturated fatty acids, pregnancy, and pregnancy outcome. Am J Clin Nutr 2000;71(suppl):285S 91S. 44. Matorras R, Ruiz JI, Perteagudo L, et al. Longitudinal study of fatty acids in plasma and erythrocyte phospholipids during pregnancy. J Perinat Med 2001;29: Sanjurjo P, Matorras R, Ingunza N, Alonso M, Rodriguez-Alarcon J, Perteagudo L. Cross-sectional study of percentual changes in total plasmatic fatty acids during pregnancy. Horm Metab Res 1993;25: Hornstra G, Al MD, van Houwelingen AC, Foreman-van Drongelen MM. Essential fatty acids in pregnancy and early human development. Eur J Obstet Gynecol Reprod Biol 1995;61: Pitkin RM. Nutritional requirements in normal pregnancy. Diabetes Care 1980;3: Francois CA, Connor SL, Bolewicz LC, Connor WE. Supplementing lactating women with flaxseed oil does not increase docosahexaenoic acid in their milk. Am J Clin Nutr 2003;77: Barker DJ, Winter PD, Osmond C, Margetts B, Simmonds SJ. Weight in infancy and death from ischaemic heart disease. Lancet 1989;2: Hales CN, Barker DJ, Clark PM, et al. Fetal and infant growth and impaired glucose tolerance at age 64. BMJ 1991;303: Osmond C, Barker DJ, Winter PD, Fall CH, Simmonds SJ. Early growth and death from cardiovascular disease in women. BMJ 1993; 307: Barker DJ, Hales CN, Fall CH, Osmond C, Phipps K, Clark PM. Type 2 (non-insulin-dependent) diabetes mellitus, hypertension and hyperlipidaemia (syndrome X): relation to reduced fetal growth. Diabetologia 1993;36: Brenna JT. Efficiency of conversion of -linolenic acid to long chain n 3 fatty acids in man. Curr Opin Clin Nutr Metab Care 2002;5: