| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Commentary |
Departments of Pediatrics, Obstetrics and Gynecology, Wayne State University, Detroit, Michigan
Address correspondence to: Dr. Winston Koo, Department of Pediatrics, Hutzel Hospital, 4707 St Antoine Blvd, Detroit, MI 48201. E-mail: wkoo{at}wayne.edu
ABSTRACT
Long chain polyunsaturated fatty acids (LCPUFA) namely arachidonic acid (ARA, 20:4n-6) and docosahexaenoic acid (DHA, 22:6n-3) are highly concentrated in the phospholipid bilayer of biologically active brain and retinal neural membranes and are important in phototransduction and neuronal function. The rationale for adding these LCPUFA to infant formula (IF) was primarily because of their presence in large quantities in the retina and brain and in human milk. In addition, infants fed IF containing LCPUFA and breastfed infants have comparable ARA and DHA levels in red cell and plasma, in contrast to the lower ARA and DHA levels in those fed IF containing only the essential fatty acids: linoleic (LA, 18:2n-6) and linolenic (LNA, 18:3n-3), the precursors to ARA and DHA, respectively. However, functional benefits in particular visual or neural development from IF containing LCPUFA remains controversial. Potential for excessive and/or imbalanced intake of n-6 and n-3 fatty acids exists with increasing fortification of LCPUFA to infant foods other than IF.
Key words: infant formula, docosahexaenoic acid, arachidonic acid, long chain polyunsaturated fatty acid
Introduction
Long chain polyunsaturated fatty acids (LCPUFA) namely arachidonic acid (ARA, 20:4n-6) and docosahexaenoic acid (DHA, 22:6n-3) are selectively incorporated, retained and highly concentrated in the phospholipid bilayer of biologically active brain and retinal neural membranes and are accumulated in large amounts during infancy [1–3]. The structural properties of LCPUFA may have a beneficial effect on photoreceptor membrane permeability, fluidity and activation of membrane-bound proteins important in phototransduction and neuronal function [4,5]. In addition, LCPUFA are precursors of eicosanoids, important modulators and mediators of a variety of essential biological processes [6].
In February 2002, infant formulas (IF) with added LCPUFA for infants born at term became commercially available in the US, and now IF with added LCPUFA for preterm infants is available. The rationale for adding these LCPUFA to IF was primarily because of their presence in large quantities in the retina and brain [1–3] and in human milk [7,8]. In addition, infants fed IF containing only the essential fatty acids: linoleic (LA, 18:2n-6) and linolenic (LNA, 18:3n-3), the precursors to ARA and DHA, respectively, had lower ARA and DHA levels in red cell and plasma, and lower tissue content of LCPUFA in particular DHA [2–3] compared to breastfed infants. In 1998, an expert panel for the Food and Drug Administration (FDA) in the US [9] and a working group for the Canadian authorities [10] did not recommend the addition of LCPUFA to IF because of uncertainties over the efficacy and safety data. There are approximately four million infants born in the US each year, and it is estimated that >35% received IF from birth and >70% between 6 to 12 months [11,12]. Thus the efficacy and safety of any additive to the human milk substitute is of paramount importance and a review of these issues concerning IF with LCPUFA is appropriate.
Factors Affecting the Efficacy of LCPUFA in Infant Formula
The addition of DHA and subsequently ARA to IF has been studied for several decades in an attempt to make IF resemble the LCPUFA content of human milk with the presumption that their addition might also provide potential visual and neurodevelopment benefits. However, human milk content of fatty acids especially the n-3 series [7,8] is highly dependent on maternal diet and is extremely variable. In addition, the extremely low intake of n-3 fatty acids in healthy populations with minimal or no dietary marine animal products, for example, vegans, Jain Hindus and Seventh Day Adventists, have added to the difficulty in determining the optimal amount and ratio of n-6 and n-3 fatty acids added to IF.
Many studies have insufficient sample size to determine the true functional benefit and safety profile for IF with added LCPUFA. Also comparison of different reports is made difficult due to confounding factors among different studies including the source, amount and ratio of ARA:DHA, the amount and ratio of LA and LNA and the varied duration of consumption of the IF containing these fatty acids.
The accuracy and reliability of tests for determining the visual and neurodevelopment effect of LCPUFA during the first two years after birth are controversial. The tests of visual resolution acuity in normal infants fed LCPUFA containing IF may be statistically significant but clinically irrelevant. For example, a meta-analysis of 12 studies on the role of LCPUFA supplemented IF for infants born at term has reported that the combined visual resolution acuity measured with behaviorally based methods shows an average improvement of 0.32 octaves at two months compared to infants fed standard non-supplemented IF. On the other hand, breastfed infants show an average improvement of 0.49 octaves at two months and 0.18 octaves at four months compared to those fed standard IF [13]. The average improvement in electrophysiologic based measure of visual acuity was 0.37 octaves between the latter two groups at four months of age [13]. The significance of any study must be interpreted in conjunction with documented reliability among the testers; thus, the significance of the combined improvement in visual acuity resolution documented in the meta-analysis [13] should be judged against the documented agreement of between 0.5 to 1 octaves among different testers for behaviorally based measurements of acuity [14,15].
Electrophysiologic studies of visual acuity also have limitations in particular if each infant’s entire data is reduced to a single extrapolated threshold [16]. The above tests of visual resolution acuity also are suboptimal for the determination of the functional role of DHA in visual development [16]. Furthermore, conversion of an octave difference in visual resolution acuity testing to "line equivalent" on the Snellen eye chart [15,17–19] is an oversimplification of a complex relationship between these measurements and may exaggerate the true clinical significance of the difference in visual acuity obtained from LCPUFA supplemented IF.
Many neurodevelopment tests were never designed to test normal infants. Neurodevelopment tests, particularly for healthy infants, lack predictive ability for long-term neurodevelopment outcome [16]. Inadequate or absent documentation of tester- and subject-based bias of visual and neurodevelopment testing for each study further hampers the comparison among different studies [13,20–22].
Efficacy of LCPUFA in Infant Formula
There is no doubt that DHA or ARA supplementation to IF can raise their respective levels in red cell and plasma of infants born at term [14,17–19,23] or preterm [15,24–26]. However, the existence of functional benefits in particular visual or neural development from IF containing DHA with or without ARA remains controversial. It is well known that human milk content of nutrients are highly variable especially with respect to LCPUFA [7,8] and there are many socio-cultural and intellectual reasons that lead to the decision to breast feed, which confound the functional outcome of breastfed infants. Thus it is inappropriate to consider that the benefit of LCPUFA addition to IF must be documented against the breastfed infants. Instead, the most appropriate means to determine the benefit of LCPUFA supplemented IF is from randomized masked trials of infants fed formulas with or without LCPUFA. The use of breastfed infants as a reference group is recommended but not critical to the determination of the benefit of LCPUFA supplemented IF.
For infants born at term, one meta-analysis of 12 studies of visual resolution acuity has reported that the combined data showed a benefit at two months and possibly four months from breast feeding or feeding LCPUFA supplemented IF compared to those infants fed the same IF but without LCPUFA [13]. There was no demonstrated benefit beyond these ages. The improvement in visual resolution acuity from human milk or LCPUFA supplemented IF is relatively small. Based on the overall gains, the authors conclude that it would require about 30 subjects per group at two months and almost 500 subjects per group at four months to have adequate power to detect any difference between groups [13]. Another meta-analysis of 10 randomized studies of term infants fed IF with or without LCPUFA shows little evidence to support the hypothesis that LCPUFA supplementation confers a benefit for visual development [21]. The relative benefit of visual development from LCPUFA supplementation also may be exaggerated by the subnormal performance of some infants fed control formula [27].
Dietary intake and age have independent and interactive effects on sweep visual evoked potential acuity. After six weeks of breast-feeding, the use of LCPUFA supplemented IF resulted in an average improvement of about 0.16 log of the minimum angle of resolution at one year of age when compared to those fed the same IF without LCPUFA. The authors speculated that this finding reflects the greater maturation of cortical function from LCPUFA [19]. Reports with much greater numbers of term infants have demonstrated no significant difference among infants fed IF with or without DHA/ARA and breastfed reference group on serial assessment of visual resolution acuity based on the behavioral [14,17,28] or electrophysiologic [17] methods for up to 12 months. This holds true regardless of the content of LNA (1.1% to 2.6%), LA (12.4% to 22.4%), DHA (0.12% to 0.60%) and ARA (0.43% to 0.60%), the source of LCPUFA (egg derived triacylglycerol, fish or fungal oils) or the duration (6 or 12 months) of feeding these LCPUFA containing IF (Table 1).
|
Recent reports with larger numbers of preterm infants show inconsistent benefits from LCPUFA supplemented IF. One report of preterm infants fed 0.3% DHA and 0.6% ARA supplemented IF during the initial hospitalization after birth show no improvement in visual resolution acuity at eight weeks and 17 weeks corrected age compared to preterm infants fed the same IF but without LCPUFA [26]. Another report documented no improvement in serial behavioral assessment of visual resolution acuity at two, four and six months corrected age among preterm infants fed IF with or without DHA/ARA, or fed their own mother’s milk, although there was a statistically significant improvement in sweep visual evoked potential measurements at six months corrected age in infants fed LCPUFA supplemented IF or fed their own mother’s milk when compared to those fed unsupplemented IF [15]. These infants were fed the study formulas throughout hospitalization and up to 12 months corrected age after hospital discharge. The DHA content of the supplemented in-hospital formulation ranged from 0.24% to 0.27%, and the post-discharge IF had 0.16%, but the ARA content was similar for the in-hospital and post-discharge IF and ranged from 0.41% to 0.43%. There were three different sources of LCPUFA, which included egg derived triacylglycerol, fish and fungal oils. It therefore appears that amongst the levels of LCPUFA supplementation studied, the longer duration of feeding is more important in determining the beneficial outcome (Table 2). In any case, the visual benefit beyond six months remains poorly defined.
|
Maternal supplementation of DHA in the form of cod liver oil from 18 weeks of pregnancy until three months after delivery was reported to result in an average increase of 4.1 points in cognitive function of the offspring at four years of age [35]. The breast milk content of DHA averaged 1.26% and ARA content averaged 0.33% when measured at three months of lactation. However, the number of infants studied at four years was approximately 14% of the original cohort of mothers enrolled to the study and was 32% of the cohort of infants studied at nine months [35]. In another study on raising the DHA content of breast milk to as high as 1.7% of total fatty acids from maternal supplementation during the first 12 weeks of lactation, there was no effect on the infants’ visual evoked potential at 12 and 16 weeks and no effect on Bayley scale of mental development index at two years [36]. In both studies, plasma DHA [35,36] and red cell DHA [36] were elevated with increased breast milk DHA content, although no further increase in plasma or red cell DHA was noted beyond a breast milk DHA content of about 0.8% of total fatty acids [36]. Thus, it would appear that neither the source and content of DHA and ARA nor the duration of feeding LCPUFA containing IF has any influence on the outcome in term infants (Table 1). The use of higher levels of DHA in IF is unlikely to improve visual or neurodevelopment outcome significantly.
The reports that some standard IF without LCPUFA can (Table 1) whereas other standard IF cannot [27,34] support visual and neurodevelopment comparable to those fed IF with LCPUFA or breastfed infants are intriguing. It is theoretically possible that different sources of fats, for example, presence or absence of palm or palm olein oil as the dominant fat blend, or the different content or ratio of LA and LNA content in different infant formulas may account for some of the visual or neurodevelopmental differences, since different fat blends have documented influence on other biologic systems including gut absorption of calcium [37–41] and bone mineralization [42].
Meta-analysis of eight randomized trials of preterm infants fed IF with or without LCPUFA shows no long-term benefit for neurodevelopment [22]. Of the reports with greater numbers of preterm infants [15,26,33], only one study shows infants with birth weights 
1250g and who received >80% of the IF with DHA/ARA from the fish/fungal oils have higher Bayley motor index scores at 12 months corrected age compared to infants fed the same IF without LCPUFA [15]. The neurodevelopment benefit from LCPUFA supplemented IF appears to be dependent on prolonged use from in-hospital period until 12 months corrected age rather than the DHA or ARA contents (Table 2). The relative importance of different sources of LCPUFA to the functional outcome remains ill defined.
The consistency of results in all reports suggests that any functional benefit in visual or neurodevelopment from LCPUFA supplemented IF is likely to be of minor clinical significance, at least for the term infant. It is possible that preterm infants with greater needs may have greater benefit from LCPUFA supplementation although the relative importance of the duration of feeding with these IF and the source of LCPUFA remain ill defined. Some multi-center studies with larger sample size suffer from additional confounders in the interpretation of the data, for example, inadequate or no documentation of interobserver reliability in the performance of tests.
Safety of LCPUFA in Infant Formula
There have been remarkably few reports of adverse events in infants fed LCPUFA containing IF despite the numerous biological functions of LCPUFA. Most of the adverse events seem to be associated with the use of IF containing DHA only. Preterm infants fed IF with DHA but without ARA have lower absolute or Z scores in weight [24,25,43], length [24,43] or head circumference [24,43], and a decreased fat free mass [25]. The growth delay may be associated with the relative high eicosapentaenoic acid (EPA) content of fish oil source of DHA or with a relatively inadequate intake of ARA [44]. EPA is another n-3 fatty acid that may have contributed to the inhibition of n-6 fatty acid metabolism. In contrast to the variations in fatty acids in fish oil, DHA as oil from the microalgal Crypthecodinium cohnii and ARA as oil from the soil fungus Mortierella alpina (Martek Biosciences, Columbia, MD) have received the "generally recognized as safe" status from the United States Food and Drug Administration (FDA) and are approved for use in infant formulas and baby foods [45].
For infants fed DHA and ARA containing IF there has been no documented adverse growth except for one report in preterm infants [33]. It showed preterm infants fed IF with DHA and ARA during their hospital stay had lower weight and length upon follow up at 18 month corrected age compared to those fed IF containing only the essential fatty acid precursors, although the Z score was lowered only for length [33]. The reports of diminished growth in preterm infants many months beyond feeding the LCPUFA supplemented IF [24,33,43] indicate prolonged follow up is needed.
Negative neurodevelopment effect was reported in one study of term infants fed IF with DHA supplementation. These infants had lower scores in selected subscales of language development, vocabulary comprehension and production [31], but there was no difference in the results from formal testing of intelligence quotient and vocabulary at 39 months among infants fed IF with DHA, DHA and ARA or no LCPUFA supplementation [30].
DHA supplemented IF raises DHA level in red cell and plasma, but there was an associated decreased in the level of ARA in term [17,35,36] and preterm [15,24,25] infants. This presumably reflects the imbalanced intake of n-6 and n-3 fatty acids [46] although the functional significance of this alteration is not known. In one report of preterm infants receiving IF with approximately 0.3% DHA from marine oil, prolonged bleeding time was noted although the values were within the normal limits [47], and there was no report of increased frequency of clinically significant bleeding. Theoretical risks of increased infection, and susceptibility to oxidant injury, from increased LCPUFA in the diet have not been reflected in increased clinical sepsis, intracranial hemorrhage, chronic lung disease or necrotizing entercolitis [14,15,26,28,32,33,48].
Recent in vitro data showing DHA at a concentration of 0.8 mM can affect apolipoprotein B and A-I secretion as well as triacylglycerol and phospholipid synthesis and secretion in newborn swine enterocytes [49] and that DHA (50 µM) and ARA (20 µM) can impair human osteoblast proliferation [50] suggest that one should remain vigilant for potential additional in vivo side effects.
Increasing fortification of DHA of commercial baby food becomes an added safety concern both from the excessive intake and/or imbalanced ratio of n-6 and n-3 fatty acids. The FDA requirement for continued monitoring through scientific studies and post market surveillance of infants who consume IF with added LCPUFA [45] is a prudent measure associated with the introduction of any nutrient product although the success of this endeavor remains to be determined.
Conclusion
The use of LCPUFA containing IF undoubtedly raises the plasma and red cell concentrations of DHA and ARA, thus achieving one more physiological response of formula fed infants that compares favorably with breastfed infants. The IF industry should be commended for this effort.
After much fanfare in the lay press [51] and aggressive marketing to the families and professionals by appealing at the emotional level that what is in the brain, eyes and human milk must be good, the infant formulas with added LCPUFA are rapidly gaining acceptance. However, it comes with a dramatic increase in the cost of IF feeding by up to 25%, although there is little evidence that LCPUFA containing IF provides clinically significant improved vision and intelligence in healthy infants born at term. The added cost is a significant burden on the family budget and also to public nutrition funding such as the Women, Infants and Children supplemental food program which provides milk feedings to 47% of the infants in USA [52]. One can only hope that the price hike will decrease as the market for LCPUFA supplemented IF becomes saturated.
Preterm infants, particularly those with birth weights 1250g, have greater needs and may benefit from the prolonged use of LCPUFA containing IF. However, documented benefits from LCPUFA supplemented preterm IF remain limited. Much remains to be learned [53,54], and one needs to remain vigilant for potential adverse effects of excessive intake of LCPUFA or imbalanced intake of n-6 and n-3 fatty acids as LCPUFA is added to (and also increases the cost of) infant foods other than IF.
Received September 12, 2002. Accepted January 23, 2003.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  W. W. Koo and E. M Hockman Posthospital discharge feeding for preterm infants: effects of standard compared with enriched milk formula on growth, bone mass, and body composition Am. J. Clinical Nutrition, December 1, 2006; 84(6): 1357 - 1364. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Oh Practical Applications of Fish Oil ({Omega}-3 Fatty Acids) in Primary Care J Am Board Fam Med, January 1, 2005; 18(1): 28 - 36. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
