Formula Supplementation and Growth

To the Editor.—Fewtrell et al1 raise interesting questions about supplementation of preterm infant formula with docosahexaenoic acid (DHA) and arachidonic acid (AA). In this short-term study—about a month during initial hospitalization—feeding premature formula with added egg lipid (a source of DHA and AA) and evening primrose oil (source of γ-linolenic acid) did not provide a significant advantage in neurodevelopment at 9 or 18 months’ corrected age (CA) and appeared to negatively impact growth. These results are in stark contrast to results of several large studies in which no negative effects on growth and development or increase in adverse events2–7 were reported. Some studies, in fact, found significant benefits in growth5,6 and/or neurodevelopment.4,6 It is important to provide an additional perspective on the differences between these studies that may help explain the disparate results, particularly to prevent undue alarm, as premature infant formulas with added long-chain polyunsaturated fatty acids (LCPUFAs) have been available commercially in many countries for a number of years and are just now becoming available in North America.First, although Fewtrell et al1 did not find statistically significant effects of LCPUFA supplementation on neurodevelopment, the authors point out that there was a small advantage in favor of the supplementation (+2.6 points on the Bayley Mental Development Index [MDI] and +2 points on the Bayley Psychomotor Development Index [PDI]) at 18 months’ CA, with greater yet statistically insignificant improvements for those infants <30 weeks’ gestational age. The authors comment that a much larger study, estimated at 256 infants per group, would be required to confirm a benefit of 2.6 points for the Bayley MDI in the overall population studied, or 60 to 70 infants <30 weeks’ gestational age to confirm a significant difference in this population.Other recent, large studies of preterm infants have found significant improvements in neurodevelopmental outcomes associated with supplementation of DHA plus AA, including differences in measures of visual acuity4 (further reviewed in the meta-analysis of SanGiovanni et al8) and language development,4 as well as the Bayley MDI6 and PDI.4,6 Key differences between the studies that found significant benefits4,6 and the Fewtrell et al study1 that could reasonably explain the difference in magnitude of benefits include the length of feeding and dose of LCPUFAs provided. In studies that demonstrated significant benefits,4,6 LCPUFA supplementation was continued through 12 months’ CA, and levels provided were significantly higher than those fed by Fewtrell et al1 (0.17% DHA and 0.31% AA), either over the entire feeding period6 (0.32–0.33% DHA and 0.64–0.67% AA) or at least during the early feeding period4 (0.26% DHA and 0.42% AA to term CA, followed by 0.16% DHA and 0.42% AA). In another short-term (∼4 weeks in-hospital) LCPUFA supplementation study, no significant difference in visual acuity of preterm infants at 2 or 4 months’ CA was found. Fewtrell et al1 cite the Cochrane Review9 as concluding that no long-term benefits have been demonstrated for LCPUFA supplementation for preterm infants. Unfortunately, this review has not been revised since 1998 (not 2000, as cited in the Fewtrell et al1 article) and thus does not include any of these more recent studies.2–8The other issue raised by the Fewtrell et al study1 that is not supported by other recent LCPUFA supplementation trials is a possible negative effect on growth. In fact, 2 preterm studies have now demonstrated significant improvements in growth of preterm infants fed DHA and AA. Preterm infants fed formula with 0.14% energy (∼0.3% of fatty acids) from DHA and 0.27% energy (∼0.6% of fatty acids) from AA had significantly higher in-hospital weight gain (+4 g/day) and greater achieved weights at 2 months’ CA (+330 g) compared with those fed control formula with LCPUFAs.5 The LCPUFA-supplemented preterm infants did not differ in weight or weight/length ratio from a reference group of breastfed term infants at 2 or 4 months’ CA. In this study, the supplemented formula was fed only during initial hospitalization, similar to the Fewtrell et al1 design. (Preliminary results from this study were published10 in 1998, although only the abstract of these results11 is cited by Fewtrell et al.1) More recently, in a large study, we6 have demonstrated that very low birth weight (<1500 g) infants fed LCPUFA-supplemented formulas had significantly improved growth compared with those fed unsupplemented control formulas. In this case increased weight and length were demonstrated at later ages. The supplemented preterm infants caught up to a reference group of breastfed term infants, in weight at 18 months’ CA and in length at 9 months’ CA.6 Our recent study7 also addresses the safety of these formulas when fed to preterm infants and, like other large studies,2–4 did not find negative effects of LCPUFAs on growth, tolerance, blood chemistry values, or adverse events. Interestingly, all of these studies2–7 fed levels of DHA and AA higher than those tested by Fewtrell et al.1 Thus perhaps the possible negative growth effect may be attributable to a deficiency of AA and DHA.Safety may be more appropriately linked to sources of LCPUFA used in the formulas. All of the studies noted above that support safety of supplemented formulas used triglyceride sources of LCPUFAs, including single cell AA oil,2–7 single cell DHA oil,2,3,5–7 high-DHA fish oil,4,6,7 and egg-derived triglyceride.4 In contrast, Fewtrell et al1 used an egg lipid source of LCPUFA presumably quite high in phospholipid (as noted for the Milupan fat source by Carnielli et al12) and also included γ-linolenic acid as well as other compositional differences in the study formulas. The bioavailability of LCPUFAs from the nontriglyceride sources is not known nor is the impact of feeding high levels of phospholipid (and phosphorus?) in the diet well-known. The publication by Fewtrell et al1 did not provide measures in plasma for AA and DHA, so it is not possible to know whether the AA and DHA provided were absorbed. The authors1 appropriately note that “the precise nature, dosage, and duration of LCPUFA supplementation used might influence any effects on growth.” They continue, “Clearly, additional work is needed to exclude the possibility that this is a chance finding.”We believe that, despite the results of the Fewtrell et al study,1 the results of recent clinical trials indicate that feeding AA and DHA are necessary13–19 to normalize the levels20 of these essential fatty acids in the plasma and provide important additional support indicating benefits of LCPUFAs for both growth5,6 and neurodevelopment,4,6 immune response of preterm infants21 solidifying the safety2–7 of specific sources of LCPUFAs. The publication by Fewtrell et al1 clearly indicates that alternate sources need to be evaluated with similar rigor. We view the addition of appropriate levels of these fatty acids to preterm formulas as an important step forward in providing the best nutrition to formula-fed infants.In Reply.—Recently we published the results of a large trial of long-chain polyunsaturated fatty acid (LCPUFA) supplementation in preterm infants, showing reduced linear growth to 18-month follow-up in the group fed LCPUFAs.1 We are pleased to see the letter of Clandinin and VanAerde2 placing this trial in the broader context of some recent publications.In contrast to our findings, Clandinin and VanAerde2 cite 2 studies showing LCPUFAs actually promoted growth in preterm infants.3,4 Both were published (one in abstract form) while our own manuscript was in press and therefore not cited by us. Moreover, we recently completed a further trial (unpublished) showing growth (and neurodevelopmental) benefits up to 18-month follow-up for LCPUFA supplementation (using fish and plant oils), notably in preterm males.Interestingly, however, another article (Vanderhoof et al5) cited by Clandinin and VanAerde as showing no negative effect on growth actually reports, in line with our current article in Pediatrics, significantly lower length gain between term and 1-year postterm in preterm infants supplemented with LCPUFAs (from single cell oils) compared with those fed unsupplemented formula or human milk. We suspect this difference could have been even greater if the authors had adjusted for the sex imbalance between randomized groups. However, although growth was the principal clinical (as opposed to biochemical) efficacy outcome, this finding was not mentioned in the abstract or discussion, both of which conclude that the findings support the efficacy and safety of LCPUFA-supplemented formula! Previous studies6,7 showed long-term reduction of growth (beyond the neonatal period) with LCPUFA supplementation when docosahexenoic acid ([DHA] from fish oil) was used without arachidonic acid (AA). Now our study,1 using egg lipids, and the one cited5 by Clandinin and VanAerde, using single cell oils, indicate that reduction in linear growth, apparent only beyond the neonatal period, may occur with formulas supplemented with DHA plus AA. This reduction in postneonatal linear growth requires future study.Clandinin and VanAerde also cite 2 references in support of positive effects of LCPUFAs on neurodevelopment. One of these was the abstract (above) published while our own article was in press.4 The other was a multisite study,5 in fact cited by us, in which our own center played a significant part. The study used both fish-fungal oil and a fish-egg oil preparation to provide LCPUFA supplementation, and significant benefits were seen within the first year (either in the whole study group or subpopulations) for several short-term cognitive and visual outcomes. However, our own experience in 4 large LCPUFA studies involving over 1200 preterm or term infants (unpublished, see above1,8,9—the latter with 18-month follow-up published but 4- to 6-year follow-up unpublished), indicates paradoxically inconsistent effects on neurodevelopment.Earlier LCPUFA intervention trials showed negative, neutral, or positive effects on growth and neurodevelopmental outcomes. This inconsistency might have been attributable to small study samples and flawed design. But it would seem, based on our own experience (see above) and that of others, that even larger studies that use different strategies for LCPUFA supplementation may show inconsistent findings. This could be attributable to chance—but it might also reflect a genuine difference in outcome according to the strategy for LCPUFA supplementation used, as discussed by Clandinin and VanAerde. The latter possibility would not be surprising considering the complexity of LCPUFA sources and the broad, complex, and active biological roles of n3 and n6 fatty acids and their derivatives. If so, it might be prudent not to take a generic view on the value of particular intakes of LCPUFAs in preterm infants. Rather, it might be more prudent to select those products with proven efficacy and safety.We agree with Clandinin and VanAerde that, collectively, current data indicate that it is possible (at least with selected products) to improve short-term outcomes in preterm infants with LCPUFA supplementation—two of our own trials confirm this (see above). Nevertheless, it must be born in mind that current studies, including our own, have focused on a narrow range of outcomes (growth and neurodevelopment) and there are no published long-term efficacy and safety data beyond 2 years. We would agree that while more outcome data is accumulating, the use of approved LCPUFA-supplemented formulas, with established efficacy and safety, is appropriate. However, investigators, health professionals, and regulators should keep an open mind.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.