Effects of long-chain polyunsaturated fatty acid supplementation on fatty acid status and visual function in treated children with hyperphenylalaninemia

Effects of long-chain polyunsaturated fatty acid supplementation on fatty acid status and visual function in treated children with hyperphenylalaninemia

Effects of long-chain polyunsaturated fatty acid supplementation on fatty acid status and visual function in treated children with hyperphenylalaninem...

84KB Sizes 2 Downloads 61 Views

Effects of long-chain polyunsaturated fatty acid supplementation on fatty acid status and visual function in treated children with hyperphenylalaninemia Carlo Agostoni, MD, Nicoletta Massetto, MD, Giacomo Biasucci, MD, Amilcare Rottoli, MD, Milena Bonvissuto, MD, MariaGrazia Bruzzese, PhD, Marcello Giovannini, MD, and Enrica Riva, MD

Background: Children with phenylalanine-hydroxylase deficiency (type-I hyperphenylalaninemia, HPA) follow a low-phenylalanine diet, severely restricted in animal foods and long-chain polyunsaturated fatty acids (LCPUFA). Consequently, they have a poor LCPUFA status, particularly for docosahexaenoic acid (DHA). DHA is relevant to visual and neural development. Objective: To investigate the effects of a 12-month supplementation with LCPUFA in a double-blind, placebo-controlled trial in treated children with HPA. Study design: Twenty children with well-controlled HPA were randomly allocated to receive either a fat supplement (supplying 26% as fatty acids including DHA, 8%) or a placebo. The fatty acid composition of erythrocyte lipids and the visual evoked potentials were measured at baseline and after 12 months of supplementation. Reference data were obtained from healthy children of comparable age. Results: At baseline children with HPA had a poorer DHA status and prolonged P100 wave latencies than the reference group. At the end of the trial the LCPUFA group showed a significant increase in DHA levels of erythrocyte lipids. In the LCPUFA group P100 wave latency decreased and was negatively associated with the DHA changes. Conclusions: A balanced dietary supplementation with LCPUFA in children with HPA is associated with an increase of the DHA pool and improved visual function. (J Pediatr 2000;137:504-9)

From the Department of Pediatrics and the Department of Neurology, San Paolo Hospital, University of Milan, Milan, Italy.

The study supplements were provided by SHS International Ltd. Submitted for publication July 29, 1999; revisions received Dec 10, 1999, and Mar 22, 2000; accepted Apr 28, 2000. Reprint requests: Carlo Agostoni, MD, Department of Pediatrics, San Paolo Hospital, 8 Via A. di Rudinì, 20142 Milan, Italy. Copyright © 2000 by Mosby, Inc 0022-3476/2000/$12.00 + 0 9/21/108398 doi:10.1067/mpd.2000.108398

504

The diet for children with phenylalanine-hydroxylase deficiency (hyperphenylalaninemia) is based mainly on Phe-free protein substitutes and vegetable foods with a low Phe content.1 The resulting dietary regimens provide a high-carbohydrate, low-saturated fat and low-cholesterol intake because of the very low intake of Phe-containing animal foods. Long-chain polyunsaturated fatty acids such as arachidonic AA DHA EPA FA HPA LCPUFA PC Phe VEP

Arachidonic acid Docosahexaenoic acid Eicosapentaenoic acid Fatty acids Hyperphenylalaninemia Long-chain polyunsaturated fatty acids Phosphatidylcholine Phenylalanine Visual evoked potentials

acid (20:4n-6), eicosapentaenoic acid (20:5n-3), and docosahexaenoic acid (22:6n-3) are found preformed only in animal foods. In case of low or no intake of these fatty acids, human subjects rely almost exclusively on the endogenous synthesis from the 18-carbon precursors, linoleic acid (18:2n-6) for the n-6 series and α-linolenic acid (18:3n-3) for the n-3 series, respectively, through elongating and desaturating reactions.2 LCPUFA are structural components of cell membranes. Dietary treated children with HPA have low levels of DHA in plasma and erythrocyte lipids when matched with members of control groups, whereas

AGOSTONI ET AL

THE JOURNAL OF PEDIATRICS VOLUME 137, NUMBER 4 AA has been reported as normal3 or even increased in membrane lipids.4 This result may be due to a high intake of AA 18-C precursor, linoleic acid, and its consequent facilitated synthesis within the common n-6 and n-3 LCPUFA enzymatic pathway even with a good intake of the DHA 18-C precursor, α-linolenic acid. Because DHA is relevant to normal brain and visual development,5 children with HPA might have untoward effects from their restrictive dietary treatment. Brain structural changes6 and abnormalities of the visual function7 have been shown in otherwise well-treated patients with HPA, raising questions on the adequacy of the dietary intervention.8 The visual effects of different LCPUFA intakes have been investigated in newborns.9 Most observations with the visual evoked potential technique indicate a positive effect of dietary LCPUFA, particularly DHA, on the visual function of both preterm10 and term11 infants. It is still unknown whether treated subjects with HPA may have direct clinical consequences from the dietary lack of LCPUFA and, if so, whether the supplementation with LCPUFA could restore their body pools and related functional activities. The aim of our study was to look at the effects of a balanced LCPUFA supplement on the biochemical fatty acid status and the visual function in treated children with HPA.

METHODS In a double-blind, placebo-controlled trial, children monitored for HPA in our department and attending the primary school were randomly allocated to receive either LCPUFA or placebo (olive oil) supplements for 12 months. Entrants tested positive at the neonatal screening and were given the diagnosis of type-I HPA (classical phenylketonuria) on the basis of pretreatment plasma Phe levels >1200 µmol/L, an assessment of the plasma

Table I. Fatty acid content in test capsules and placebo

Fatty acid Myristic Palmitic Palmitoleic Stearic Oleic Linoleic acid α-Linolenic acid Gamma-linolenic Eicosatrienoic (n-3) AA EPA Docosapentaenoic (n-3) DHA

Test (mg/0.5 g) (%)

Placebo (mg/0.5g) (%)

7.3 (1.4) 52.4 (10.4) 13.1 (2.6) 20.9 (4.1) 131.4 (26.2) 52.5 (10.5) 10.1 (2.0) 23.2 (4.6) 3.7 (0.7) 37.0 (7.4) 27.5 (5.5) 20.0 (0.4) 40.0 (8.0)

—— 68.5 (13.7) <5 (<1) 13.5 (2.7) 352.5 (70.5) 58.5 (11.7) <5 (<1) —— —— —— —— —— ——

Data supplied by SHS International Ltd.

Phe curve after an oral Phe tolerance test (100 mg/kg) at 1 year of age, and a dietary tolerance <20/mg per kg body weight per day at 5 years. All had started the dietary treatment by the sixth week of life. Clinical records were available for the neonatal period, and parental education, employment status (coded according to the Italian Census), and clinical and biochemical data had been included. The blood Phe before and during the trial was monitored by means of dried blood spots on card mailed monthly by families for the Guthrie test.1 Entrants had to score >70 (that is, within a range of 2SD from the average standard) at the Wechsler Intelligence Scale for Children-Revised developmental scale (Italian Version, OS, Firenze, 1986). Compliance with the prescribed supplementation was assessed monthly by interviewing parents either by direct phone calls or at clinical controls. Poorly compliant children with HPA (so defined after parental admission of more than 2 days per week without supplementation at 2 consecutive interviews) discontinued the trial. Reference dietary and biochemical data were performed on a sample of healthy subjects randomly selected among schoolchildren from the referring area of the hospital. Age- and sex-matched healthy

relatives of subjects with HPA underwent VEPs as a reference group for the visual function. The study protocol was approved by the local Ethical Committee, and parents gave written informed consent at recruitment. Dietary intakes were calculated through a weighed food diary of 3 days for children in the non-HPA reference group and by means of the individual dietary schedules (precisely indicating the daily amounts of allowed low-Phe foods and formulas) for children with HPA. The intakes of macronutrients and individual fatty acids were derived from the Italian Food composition tables.12 The LCPUFA content was not available for some foods in these tables and were therefore drawn from a source13 including data of various agriculture handbooks issued by the U.S. Department of Agriculture in the 1980-1990 period (USDA Human Nutrition Information Service, Washington, DC). The lipid supplements were provided by SHS International Ltd (Liverpool, UK). The supplements were presented in gelatin capsules containing 500 mg oil per capsule. The test capsules supplied 26% FA as LCPUFA and contained equivalent amounts (approximately 13%) of polyunsaturated FA of both the n-6 and n-3 series immediately beyond the rate limiting step (delta-6-desatura505

AGOSTONI ET AL

THE JOURNAL OF PEDIATRICS OCTOBER 2000

Table II. Daily nutrient intakes in reference and HPA groups (mean ± SD)

Energy (total kcal) Weight (kg) Kcal/kg Protein (%) Animal/total protein Carbohydrate (%) Lipid (%) Saturated% Monounsaturated (%) Polyunsaturated (%) Linoleic acid (%) α-Linolenic acid (%) AA (mg/d) DHA (mg/d) Cholesterol (mg/d)

Reference (n = 18)

HPA (n = 20)

P value*

2037 ± 398 28 ± 6 74 ± 13 14 ± 1 1.8 ± 0.5 56 ± 5 30 ± 4 12 ± 2 11 ± 2 4±1 3.2 ± 0.3 0.4 ± 0.1 202 ± 32 141 ± 54 307 ± 59

2000 ± 359 31 ± 7 65 ± 14 10 ± 2 0.07 ± 0.02 62 ± 3 27 ± 2 6±1 13 ± 1 6±1 5.2 ± 1.0 0.5 ± 0.05 12 ± 2 — 31 ± 6

0.88 .09 .07 <.001† <.001† .001† .04† <.001† <.001† <.001† <.001† <.001† <.001† <.001†

*Mann-Whitney test. †Statistical significance.

tion) for LCPUFA synthesis. The LCPUFA profile used in this study was specifically developed to raise DHA in circulating lipids to a physiological range without affecting AA levels to prevent possible untoward effects on growth and eicosanoid metabolism.14,15 The control group received an equivalent fat dose without LCPUFA used as placebo (Table I). The daily dosage of the supplement (approximately 1 capsule per 4 kg body weight) was calculated to provide 0.3% to 0.5% of the daily energy requirements as LCPUFA (supplemented group) according to expert advice for healthy subjects of 0.27% energy from n-3 LCPUFA per day.16 Based on these assumptions, a 30-kg 9year-old child with HPA would have taken 7 daily capsules, equivalent to 472 mg/d as n-3 LCPUFA (DHA, 280 mg, and EPA, 192 mg), equivalent to 4.3 kcal, that is, 0.26% of the lower limit of the recommended range of energy intake for age.17 After lipid extraction was performed according to Folch et al, 18 erythrocyte phosphatidylcholine and phosphatidylethanolamine were separated as described previously,19 and the composition of FA methyl esters 506

(wt%) was measured with high-resolution capillary gas chromatography. Calculations included the response factor for each individual FA. Plasma Phe levels were measured by ion-exchange liquid chromatography (amino acid analyzer supplied by Carlo Erba, Rodano, Italy) at baseline and at the end of the supplementation period. The visual function was investigated with the pattern reversal (P-) and the flash-VEPs with 3 stimuli, a stroboscopic flash, a large pattern, and a small pattern, respectively, according to standardized methods.20 In designing the study, a betweengroup difference of 100% in the change of plasma phospholipid DHA concentrations was considered to be a meaningful biochemical effect of the treatment. Based on previous data19 and allowing a type I error (α) level of .05 and a power of the test of 90% with the expected SD for DHA (30% of mean levels), at least 9 patients per group were required to detect this difference. Assuming a 25% loss to follow-up over a 12-month period, 12 children were recruited per each HPA group. This sample size would be able to detect a change of 10% in variation of P100 wave latencies according to age-

related values as described by McCulloch and Skarf 21 with the same α and power. The percentage changes of variables over a 12-month period have been calculated for the statistical evaluation. Nonparametric tests have been used for analyses of continuous variables (Kruskal Wallis test with Dunn’s posthoc analysis and Mann-Whitney test, Spearman’s R to check correlations). Categoric variables were compared with the Fisher exact test. The correlation between P100 wave latency and DHA changes was adjusted for blood Phe by multiple linear regression. Data in the text are reported as means ± SD.

RESULTS Of 24 children recruited in the trial over a 6-month period, 20 subjects (10 per group) completed the 12-month supplementation period. During the follow-up 2 children from each group were excluded for poor compliance. Age (10.7 years ± 2.4 vs 10.5 ± 2.8 years), weight (32 kg ± 8 vs 30.5 kg ± 6), sex distribution (5 boys and 5 girls vs 6 boys and 4 girls), maternal education, family income, and blood Phe (532 ± 232 vs 472 ± 130 µmol/L at baseline; 468 ± 219 vs 421 ± 220 µmol/L through the trial period; 464 ± 208 vs 484 ± 238 µmol/L at the end of the trial) were comparable in the unsupplemented and the supplemented subjects with HPA, respectively. All subjects scored >70 at the Weschler Intelligence Score for Children-Revised scale (92 ± 16 in unsupplemented vs 96 ± 14 in supplemented subjects). Eighteen children, 9 boys and 9 girls with a mean age of 10.8 ± 1.2 years, formed the reference group for dietary and biochemical data, and 12 healthy relatives of children with HPA, 6 boys and 6 girls with a mean age of 10.7 ± 0.9 years, were considered to be the reference group for the visual function. Despite energy intakes comparable to reference values, subjects with HPA had clearly different macronutrient

AGOSTONI ET AL

THE JOURNAL OF PEDIATRICS VOLUME 137, NUMBER 4 Table III. LCPUFA (weight%) in erythrocyte lipids (mean ± SD, minimum-maximum)

Baseline Reference (n = 18)

Unsupplemented HPA (n = 10)

3.4 ± 0.6, 2.0-4.4 0.1 ± 0.1, 0.07-0.5 0.6 ± 0.2, 0.2-1.0

3.4 ± 1.04, 1.9-5.3 0.1 ± 0.05, 0.04-0.2 0.4 ± 0.1, 0.1-0.7

End of treatment Supplemented Unsupplemented HPA P HPA (n = 10) value* (n = 10)

Supplemented P (n = 10) value†

Erythrocyte PC 20:4n-6 20:5n-3 22:6n-3

3.6 ± 1.0, 1.9-4.7 .78 0.1 ± 0.05, 0.03-0.2 .30 0.4 ± 0.2, 0.1-0.6 .08

3.2 ± 1.4, 1.6-5.5 3.2 ± 1.5, 1.0-6.0 0.1 ± 0.04, 0.01-0.16 0.1 ± 0.07, 0.06-0.3 0.4 ± 0.2, 0.1-0.8 0.9 ± 0.3, 0.5-1.3

.93 .76 .02‡

Erythrocyte phosphatidylethanolamine 20:4n-6 20:5n-3 22:6n-3

18.2 ± 2.8, 12.2-23.1 0.3 ± 0.1, 0.1-0.5§ 3.2 ± 1.2, 1.6-6.1§

15.7 ± 3.0, 9.4-19.9 18.3 ± 6.0, 5.4-24.7 0.2 ± 0.07, 0.09-0.3 0.2 ± 0.1, 0.05-0.4 1.8 ± 0.7, 0.5-2.7 2.2 ± 1.0, 0.3-4.3

.15 14.5 ± 7.3, 5.9-23.3 .001‡ 0.2 ± 0.1, 0.07-0.5 .003‡ 1.3 ± 0.9, 0.3-2.8

16.1 ± 5.2, 6.7-23.9 0.3 ± 0.1, 0.1-0.6 3.7 ± 1.7, 1.5-6.1

.54 .82 .04‡

*Comparison among basal values, Kruskal-Wallis’ test. †Comparison of percentage variations at the end of treatment between supplemented and unsupplemented children with HPA, Mann-Whitney test. ‡Statistical significance. §Significant between-group differences. Significant between-group differences.

Table IV. Latencies (msec) of P100 wave (mean ± SD, minimum-maximum)

Baseline Reference (n = 18)

Unsupplemented HPA (n = 10)

End of treatment Supplemented Unsupplemented HPA P HPA (n = 10) value* (n = 10)

Supplemented HPA P (n = 10) value†

Pattern-reversal VEP 60’ 15’

106 ± 6, 96-117 108 ± 5, 98-118‡

109 ± 6, 99-119 116 ± 8, 100-133§

106 ± 3, 100-112 113 ± 8, 96-128

.49 .03

109 ± 9, 96-125 118 ± 11, 101-132

104 ± 4, 100-112 107 ± 6, 98-115

.30 .02

114 ± 5, 106-121‡ 112 ± 6, 104-124‡

123 ± 8, 112-136§ 124 ± 11, 105-142‡

122 ± 9, 111-142 121 ± 11, 104-138

.04 .03

114 ± 8, 104-130 121 ± 8, 110-138

113 ± 10, 100-128 111 ± 12, 97-131

.85 .04

Flash-VEP 1Hz-2J 2Hz-1J

*Comparison among basal values, Kruskal-Wallis’ test. †Comparison of percentage variations at the end of treatment between supplemented and unsupplemented children with HPA, Mann-Whitney test. ‡Significant between-group differences. §Significant between-group differences. Statistical significance.

distribution and lipid intake (Table II). As far as dietary LCPUFA, HPA showed a minimal intake of AA derived from whole cow’s milk, whose intake is allowed in controlled quantities (ranging from 100 to 200 mL/d) as part of the Phe-restricted diet. No DHA-containing food was allowed. According to our dietary protocols, all subjects with HPA were supplemented daily with one tablet of a multivitamin, multimineral preparation (Supradyn Roche, Basel, Switzerland).

On the whole, fat capsules provided approximately 1.5% of the daily energy intakes. The supplemented children with HPA were supplied with 7.5 capsules on average per day (range, 6 to 10 capsules) versus 8 capsules (range, 6 to 12 capsules) in the placebo group. Their calculated total and per kilogram body weight intakes of LCPUFA in the trial period were 273 ± 52 and 9 ± 0.4 mg, respectively, for AA, 185 ± 35 and 6 ± 0.3 mg, respectively, for EPA, and 296 ± 57 and 10 ± 0.4 mg, respectively, for

DHA. The amount of supplementary DHA was higher than that found in the members of the reference group because of their limited consumption of fish. Oleic acid from capsules in the placebo group contributed approximately 1% the daily caloric intake. Children with HPA increased their DHA status by approximately 100% in erythrocyte PC and phosphatidylethanolamine through the supplementation period, whereas AA and EPA remained almost stable (Table III). 507

AGOSTONI ET AL

THE JOURNAL OF PEDIATRICS OCTOBER 2000

Figure. Correlation between 12-month change in DHA % of erythrocyte PC, x axis, and change in P100 wave latency time, y axis (R2 = 0.21; outer lines represent 95% CI).

Compared with the reference group, the FA status of children with HPA was characterized by lower levels of saturated FA and linoleic acid and higher monounsaturated FA (data not shown). Indexes of visual function in the reference and HPA groups were different, the P100 wave being delayed in the latter group, as an indicator of subclinical impairment. At the end of the trial a decrease of latency times was found in the supplemented HPA group (Table IV). Considering the subjects with HPA altogether, the 12-month changes of P100 wave latency at 15’ check size showed negative associations with the changes in DHA % of erythrocyte PC (r = –0.45, P = .04) (Figure). After adjustment was done for blood Phe values (at baseline and after 12 months), the correlation between P100 wave latency and PC DHA changes was still significant (β = –0.49, P = 0.05). Within the supplemented group the major visual effects were evident in subjects with the lowest DHA status at baseline, but we cannot exclude a poor compliance in cases of minor DHA changes. Also, indexes of P100 wave amplitude im508

proved in the LCPUFA group, particularly P-VEP at 60’ (milliseconds, mean ± SD: 5.3 ± 2.7 at baseline and 6.0 ± 4.3 at 12 months in the control group vs 4.0 ± 3.0 at baseline and 7.2 ± 2.5 at 12 months in the supplemented group, P = .03 for the difference in percentage variation). A high degree of interindividual variability was found in all amplitude indexes (within-group values ranging up to threefold per time-point) in agreement with the wide range of age-related changes reported in patients with PKU.22

DISCUSSION Although supplying children with HPA with just n-3 LCPUFA leads to major biochemical unbalances,23 the supplementation with a fat blend balanced in both n-3 and n-6 LCPUFA restored DHA levels in erythrocyte lipids within a physiological range while it maintained unaltered the AA status. These results are consistent with experimental observations showing an increase of DHA in neural phospholipids of piglets supplemented

with both DHA and AA.24 Thus added AA, even if not directly incorporated into membranes, might regulate the availability of the other n-3 LCPUFA.2 Visual function improved in children with HPA supplemented with LCPUFA. P100 wave latency times, whose basal values were found prolonged in agreement with previous reports,22,25 improved at the end of the trial. The association with the DHA changes in erythrocyte lipids suggests a causeeffect relationship, whereas the lack of effect of blood Phe could be explained by the relative homogeneity of the two groups as far as dietary control. A 10% difference in the P100 wave latency after supplementation, even if negligible in the context of clinical ophthalmology, could reflect the nutritional requirements of the developing neural tissues. In healthy term infants the direct supply of DHA results in differences of visual indexes11 and brain composition 26 and developmental scores27 in a period of sustained nervous tissue accretion. Nevertheless, a beneficial effect of n-3 LCPUFA supplementation on vision and brain function also has been shown in children older than 10 years affected by neurodegenerative disorders with expected poor DHA status.28 For children with HPA we hypothesize an increased response to dietary supplementations with LCPUFA because of the longlasting dietary deprivation, whose effects on brain structure and function, separated from those of high blood Phe levels, are still unknown. Because in the disk membrane of retinal photo receptors DHA contributes 50% of the total PL FA content and is believed to have a pivotal role in the conformational change of rhodopsin,29 higher DHA concentrations in the membrane bilayer might favorably influence the cytoplasmic domain of the rhodopsin molecule through protein-related signal transducing systems.30 Larger studies beginning in infants with HPA (in progress at this time) and followed for longer would be nec-

AGOSTONI ET AL

THE JOURNAL OF PEDIATRICS VOLUME 137, NUMBER 4 essary to confirm our observations and assess the persistency of the functional effects or the strict dependence on the dietary supplementation of LCPUFA.

11.

REFERENCES 1. Scriver CR, Kaufman S, Eisensmith RC, Woo SLC. The hyperphenylalaninemias. In: Scriver RC, Beaudet AL, Sly WS, Valle D, editors. The metabolic and molecular bases of inherited disease. 7th ed. New York: McGraw-Hill; p. 1015-75. 2. Lands EM. Renewed questions about polyunsaturated fatty acids. Nutr Rev 1986;44:189-95. 3. Poge AP, Baumann K, Muller E, Leichsenring M, Schmidt H, Bremer HJ. Long-chain polyunsaturated fatty acids in plasma and erythrocyte membrane lipids of children with phenylketonuria after controlled linoleic acid intake. J Inherit Metab Dis 1998;21: 373-81. 4. Sanjurjo P, Perteagudo L, Rodriguez Soriano J, Vilaseca A, Campistol J. Polyunsaturated fatty acid status in patients with phenylketonuria. J Inherit Metab Dis 1994;17:704-9. 5. Neuringer M, Connor WE. N-3 fatty acids in the brain and retina: evidence for their essentiality. Nutr Rev 1986; 44:285-94. 6. Bick U, Fahrendorf G, Ludolph AG, Vasallo P, Weglage J, Ullrich K. Disturbed myelination in patients with treated hyperphenylalaninemias: evaluation with magnetic resonance imaging. Eur J Pediatr 1991;150:185-9. 7. Korinthenberg R, Ullrich K, Fullenkemper F. Evoked potentials and electroencefalography in adolescents with phenylketonuria. Neuropediatrics 1988;19:175-8. 8. Medical Research Council Working Party on Phenylketonuria. Phenylketonuria due to phenylalanine hydroxylase deficiency: an unfolding story. Br Med J 1993;306:115-9. 9. Heird WC, Prager TC, Anderson RE. Docosahexaenoic acid and the development and function of infant retina. Curr Opin Lipidol 1997;8:12-6. 10. Uauy R, Birch D, Birch E, Tyson J,

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

Hoffman D. Effect of dietary w-3 fatty acids on retinal function of very low birth weight neonates. Pediatr Res 1990;28:485-92. Birch EE, Hoffman DR, Uauy R, Birch DG, Prestidge C. Visual acuity and the essentiality of docosahexaenoic acid and arachidonic acid in the diet of term infants. Pediatr Res 1998; 44:201-9. Carnovale E, Marletta L, Miuccio F. Tabelle di composizione degli alimenti. (Italian Food Composition Tables). National Institute of Nutrition, Rome, 1995. Chow CK, editor. Fatty acids in foods and their health implications. New York: Marcel Dekker Inc; 1992. Carlson SE, Cooke RJ, Werkman SH, Tolley EA. First year growth of preterm infants fed standard compared to marine oil n-3 supplemented formula. Lipids 1992;27:901-7. Sinclair AJ, Mann NJ. Short-term diets rich in arachidonic acid influence plasma phospholipid polyunsaturated fatty acid levels and prostacyclin and thromboxane production in humans. J Nutr 1996;126:1110S-4S. General Recommendations on dietary fats for human consumption. In: Galli C, Simopoulos A, editors. Dietary omega3 and omega6 fatty acids: biological effects and nutritional essentiality. New York: Plenum Publishing; 1989. p. 403. Livelli di Assunzione Raccomandati di Energia e Nutrienti per la Popolazione Italiana. LARN. (Italian Recommended Dietary Allowances). 1996 Revision. Italian Society of Human Nutrition, Rome, 1996. Folch J, Lees M, Stanley GHS. A simple method for the isolation and purification of total lipids from animal tissues. J Biol Chem 1957;226:497-509. Agostoni C, Trojan S, Bellù R, Bruzzese MG, Giovannini M. Developmental quotient at 24 months and fatty acid composition of diet in early infancy: a follow up study. Arch Dis Child 1997;76:421-4. McCulloch DL, Taylor M, Whyte H. Visual evoked potentials and visual prognosis following perinatal asphyxia. Arch Ophthalmol 1991;109:229-33. McCulloch DL, Skarf B. Develop-

22.

23.

24.

25.

26.

27.

28.

29.

30.

ment of the human visual system: monocular and binocular pattern VEP latency. Invest Ophthalmol Vis Sci 1991;32:2372-81. Henderson RM, McCulloch DL, Herbert AM, Robinson PH, Tajlor MJ. Visual event-related potentials in children with phenylketonuria. Acta Paediatr 2000;89:52-7. Agostoni C, Riva E, Biasucci G, Luotti D, Bruzzese MG, Marangoni F, Giovannini. M. The effects of n-3 and n-6 polyunsaturated fatty acids on plasma lipids and fatty acids of treated phenylketonuric children. Prostaglandins Leukot Essent Fatty Acids 1995;53: 401-4. Craig-Schmidt MC, Stieh KE, Lien EL. Retinal fatty acids of piglets fed docosahexaenoic and arachidonic acids from microbial sources. Lipids 1996;31:53-9. Leuzzi V, Rinalduzzi S, Chiarotti F, Garzia P, Trasimeni G, Accornero N. Subclinical visual impairment in phenylketonuria. A neurophysiological study (VEP-P) with clinical, biochemical and neuroradiological (MRI) correlations. J Inher Metab Dis 1998;21: 351-64. Makrides M, Neumann MA, Byard RW, Simmer K, Gibson RA. Fatty acid composition of brain, retina, and erythrocytes in breast- and formula-fed infants. Am J Clin Nutr 1994;60:189-94. Agostoni C, Trojan S, Bellù 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. Bennett MJ, Gayton AR, Rittey CDC, Hosking JP. Juvenile neuronal ceroidlipofuscinosis: developmental progress after supplementation with polyunsaturated fatty acids. Dev Med Child Neurol 1994;36:630-8. Kurlak LO, Stephenson TJ. Plausible explanations for the effects of longchain polyunsaturated fatty acids in neonates. Arch Dis Child Fetal Neonatal Ed 1999;80:F148-54. Slater SJ, Kelley MB, Yeager MD, Larkin J, Ho C, Stubbs CD. Polyunsaturation in cell membranes and lipid bilayers and its effects on membrane proteins. Lipids 1996;31:S189-92.

509