| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Research & Development and Scientific Affairs, Ross Products Division, Abbott Laboratories, Columbus, Ohio (K.M.O., M.W.B.)
Section of Nutrition, Department of Pediatrics, University of Colorado Health Sciences Center, Denver, Colorado (J.E.W., K.S.R., N.F.K.)
Address reprint requests to: Marlene Borschel, PhD, 105215 RP32, Ross Products Division, 625 Cleveland Avenue, Columbus, OH 43215. E-mail: marlene.borschel{at}abbott.com
 |
ABSTRACT |
|---|
 TOPFOOTNOTES ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSACKNOWLEDGMENTSREFERENCES |
|---|
Methods: Two randomized, blinded, crossover balance studies were conducted in normal term infants using a three-day home balance method. One study evaluated 10 infants fed commercially available CHF with or without PO, and the other study evaluated 12 infants fed commercially available SPF with or without PO. Fat and calcium absorption were determined based on the weight of formula intake, weight of stools, and measured calcium and fat in formula and stools.
Results: Fat and calcium intake did not differ between the groups fed CHF. However, infant’s calcium and fat absorption was less, 41 ± 6% (Mean ± SEM) and 92.0 ± 0.8%, respectively, when fed CHF with PO compared to 66 ± 5% and 96.6 ± 1.1%, respectively, when fed CHF without PO, (p < 0.01). For infants fed SPF, fat and calcium intake did not differ between the feeding groups. Mean calcium absorption was also significantly less when infants were fed SPF with PO, 22 ± 3%, than when fed SPF with no PO, 37 ± 4% (p < 0.05). Fat absorption did not differ between the two SPFs.
Conclusion: This study demonstrates that PO, as the predominant fat, is associated with significantly lower absorption of calcium from infant formulas in which calcium salts are the source of calcium. These findings corroborate previous reports of this negative effect of PO in cow milk-based infant formulas in which most of the calcium is a component of the cow milk protein source.
Key words: palm olein, infant formula, fat absorption, calcium absorption
|
INTRODUCTION |
|---|
|
TOPFOOTNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSACKNOWLEDGMENTSREFERENCES |
|---|
Recently, in two separate studies, Nelson et al. reported significantly lower fat and calcium absorption in healthy infants fed cow milk protein-based infant formulas where the fat blend contained either 53% [1] or 45% [2] of total fat as palm olein (PO) compared to infants fed similar formulas without added PO. Studies by both Lucas et al. [3] and Carnielli et al. [4] have shown this is due to palmitic acid (PA) positioning in PO triglycerides and that when PO fatty acids are redistributed on the triglyceride, fat and calcium absorption are improved.
In formulas based on cow milk proteins, most of the calcium in the formula is inherent in the protein source, and only small amounts of added calcium salts are required. This is in contrast to formulas based on casein hydrolysate (CHF) or soy protein (SPF) where no appreciable calcium is supplied by the protein source and the formula must be fortified with additional calcium and phosphorus sources to meet the needs of the infant.
Since the previous studies [1–4] were conducted with milk-based formulas with primarily inherent calcium, the present studies were conducted to compare fat and calcium absorption in healthy infants fed commercially available formulas with primarily added calcium salts, CHF and SPF, with or without PO.
|
MATERIALS AND METHODS |
|---|
|
TOPFOOTNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSACKNOWLEDGMENTSREFERENCES |
|---|
Feedings
All study formulas were Ready To Feed, providing 20 kcal/fl oz, and packaged in clinically labeled 32 fl oz cans. For the CHF study, the formulas were Nutramigen® Hypoallergenic Protein Hydrolysate Formula With Iron (NUTR) [Mead Johnson Nutritionals, Evansville, IN] and Alimentum® Protein Hydrolysate Formula With Iron (ALIM) [Ross Products Division, Abbott Laboratories, Columbus, OH]. For the SPF study, the formulas were ProSobee® Soy Protein Formula With Iron (PRO) [Mead Johnson] and Isomil® Soy Protein Formula With Iron (ISO) [Ross Products Division]. The source of protein was the same for both CHFs and for both SPFs, but carbohydrate sources and oil blends differed. NUTR and PRO contained PO, and all formulas contained added calcium salts. All formulas were fortified with minerals and vitamins. The composition of the study formulas are presented in Table 1.
|
Casein Hydrolysate Study.
Sixteen healthy term infants (nine males) were enrolled between July 1995 and July 1996. All infants had a gestational age of 
37 weeks and a birth weight 2500 g. Their mean age at the beginning of the first crossover period was 84 days (median, 94 days) for those completing the study fed NUTR and 75 days (median, 83 days) for those fed ALIM. Ten infants successfully completed both periods of the balance study. Six infants did not complete the study, five (4 NUTR, 1 ALIM) did not complete one of the balance periods, and one infant completed both periods but had incomplete stool collections so the balance data were excluded.
Soy Protein Study.
Nineteen healthy term infants (eight males) were enrolled between July 1996 and April 1997. Mean age at study entry was 89 days (median, 81 days) for those completing the study fed PRO initially and 89 days (median, 96 days) for those fed ISO. Thirteen infants successfully completed both periods of the balance study, but one was not included in the analyses due to incomplete formula intake data. Of the remaining six infants who did not complete, five infants had perceived intolerance to one of the study formulas (3 PRO, 2 ISO), and one infant exited for reasons unrelated to the formula.
Procedures
A dose of 1 mg brilliant blue per kg body weight in 2 fl oz formula was given immediately before the first feeding of each 72-hour balance period and before the feeding 72 hours later. Assessment of intake during each balance period began with the first feeding after the brilliant blue feeding up to, and including, the second brilliant blue feeding. The first marker-containing feeding (2 fl oz) was not included when calculating the intake of formula during the balance period. To determine formula intake, parents weighed the feeding bottle immediately before and after each feeding using an Ohaus Model CT600 Scale (Florham, New Jersey). Spit-up and vomiting losses were assessed by using pre-weighed cloths and wipes. For the SPF study, parent(s) recorded the characteristics of their infant’s stools and the incidence of spitting up during the last four days of each study feeding period.
At the start of the stool collection period, infants were washed with Ivory® soap (Proctor & Gamble, Cincinnati, Ohio) and fitted with a net pant that was open in the back. Throughout the balance study, infants wore a T-shirt and the net pant when positioned in the collection seat. Urine and fecal matter were kept separate by using a folded washcloth inside the net pant. The parents were instructed to use no lotions on themselves or their infant and to wash their hands frequently.
The infant was seated in a specially designed seat with a center hole containing a pre-weighed fecal collection bag [5]. A tongue depressor or ashless filter paper was used to wipe/collect any residual stool left on the infant that did not appear in the plastic bag and was added to the total collection. Alternately, the infant was fitted with a reverse disposable diaper or plastic chux containing an ashless filter paper when the infant was held, fed or slept. Stools were collected on the filter paper and blotted off the plastic with additional filter paper when necessary. Bagged stools were stored on ice for up to two days in the home and frozen in the laboratory at -70°C until analysis.
Sample Analyses
Fecal collections obtained during each 72-hour balance period were pooled for analyses, excluding the first stool colored with brilliant blue, but including the second blue stool marking the end of the balance period. The frozen fecal matter was transferred into a pre-weighed blender jar along with an equal volume of water used to rinse the collection bag and depressors. After homogenization with equal parts deionized water, a 20 g aliquot was analyzed for fecal fat according to the method of Jeejeebhoy et al. [6]. A 20 to 50 g aliquot, after microwave digestion, was assessed for stool calcium by flame atomic absorption spectrometry using a Perkin Elmer atomic absorption spectrophotometer, Model PE 2380 (Norwalk, Connecticut) (CEM Corporation Model MDS2000, Charlotte, North Carolina). Formula was analyzed by the same methods. Fat and calcium intakes were calculated using the weight of the formula consumed, corrected for spit-up and vomiting losses, using measured fat and calcium concentrations of the formula.
Data Analysis
Absorption of nutrients was calculated as intake minus fecal excretion adjusted to daily rate between brilliant blue markers. Percent absorption was determined by dividing amount absorbed by intake amount and multiplying by 100. Fat and calcium absorptions were compared using statistical crossover analysis of variance techniques. No carryover or feeding-order effects were detected and thus were eliminated from the statistical models. The arcsine variance stabilizing transformation was applied to percentage absorptions. Observed differences between formulas were considered statistically significant at the 5% level.
|
RESULTS |
|---|
|
TOPFOOTNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSACKNOWLEDGMENTSREFERENCES |
|---|
|
Soy Protein Study
Fat and calcium intake did not differ between feeding groups. Mean (± SEM) calcium absorption as a percent of intake was significantly less when infants were fed PRO than when fed ISO, 22 ± 3% and 37 ± 4% (p < 0.05), respectively. Nine of the 12 infants had higher percentage calcium absorption when fed ISO than when fed PRO with eight having markedly better calcium absorption (difference ranged from 19% to 33%). Two infants had similar values on both feedings, and one had less calcium absorption when fed ISO. Mean fat absorption did not differ when infants were fed either formula, and was greater than 90% for both. One young infant had low fat absorption on either formula. No medical problem was detected, and the malabsorption was regarded as a developmental condition. Median absorption was 96.6% for PRO and 97.6% for ISO. If this subject is eliminated from the analysis, mean fat absorption is 96% for both PRO and ISO.
Infants averaged one to two stools per day when fed PRO or ISO. Mean rank stool consistency was 3.4 ± 0.2 (1 = watery, 2 = loose, 3 = soft, 4 = formed, 5 = hard) for PRO and 3.2 ± 0.2 for ISO. The percentage of stools that were formed were significantly greater when infants were fed PRO, 57%, than when fed ISO, 28% (p < 0.05). Percent of feedings with spit-up and vomit for infants fed PRO, 32.2 ± 7.7% did not differ significantly from infants fed ISO, 39.5 ± 6.2%.
|
DISCUSSION |
|---|
Several studies have demonstrated that this phenomenon significantly reduces calcium absorption in term [1,2,4] and preterm [3,13] infants fed milk-based, lactose-containing formulas with primarily inherent calcium sources. Both NUTR and PRO have the same PO-containing fat blend as the milk-based PO-containing formula studied by Nelson et al. [2]. The present studies demonstrate that PO also reduces calcium absorption in formulas with mineral salts as the primary source of calcium.
Soy formulas are generally fortified with higher levels of calcium than milk-based formulas to account for the calcium-binding effect of the phytate inherent in soy protein isolate. The calcium absorption from ISO in the current study [37 ± 4% (SEM)] is similar to that of 135 infants [34 ± 18% (SD)] of the same age (8–122 days) as described in Fomon’s text on infant nutrition [14]. The absorption of calcium from PRO, however, [22 ± 3% (SEM)] is considerably lower.
There are physiological consequences to the digestion characteristics of PO by infants. The formation of calcium soaps has been reported to increase hard stools in infants fed PO-containing formula [3,4] and in breast-fed infants weaned to a PO-containing formula [15]. But of more concern are the reports that suggest a reduction in bone mineral accretion as measured by dual energy x-ray absorptiometry (DXA) in healthy term infants fed PO-containing formulas [16–18].
Kennedy et al. [17] fed infants formulas that were identical except for the positional distribution of PA on triglycerides. An experimental formula containing an enzymatically interesterified version of PO in which 50% of the PA was in the sn-2 position was compared to one with native PO (12% of PA in sn-2). By 12 weeks of age, infants fed the experimental formula had significantly higher BMC than those fed the PO-containing formula. Infants fed the PO-containing formula also had significantly lower BMC than infants exclusively breast-fed for 12 weeks. This cross-sectional study specifically tested the effect of PA triglyceride positioning.
In a study designed to test the effects of low vs. moderate levels of calcium and phosphorus levels in formula on skeletal development, Specker et al. [16] reported that infants fed the low mineral formula GS (Good StartTM, Carnation Nutritional Products, Glendale, CA] had significantly lower bone mineral content (BMC) at three and six months of age than infants fed the moderate mineral formula SWI (Similac with IronTM, Ross Products Division, Abbott Laboratories). However, GS also contains 47% of its fat blend as PO, whereas there was no PO in SWI. Given the emerging data on the effects of PO on calcium absorption, one cannot rule out that the differences in BMC could be attributed, at least in part, to the differences in fat blend rather than mineral levels.
Most recently, Koo et al. [18] demonstrated that the difference in calcium absorption reported in the formulas studied by Nelson et al. led to a significant difference in BMC in infants fed the same formulas at three and six months of age. This longitudinal, randomized, blinded study, supports the likelihood that the differences in bone mineralization reported by Kennedy [17] and Specker [16] were a result of poor calcium absorption associated with PO.
There are no longitudinal randomized studies of bone mineralization in infants fed soy protein-based formulas using DEXA. A study by Mimouni et al. [19] reported no differences in bone mineralization of infants fed human milk, soy protein- and milk protein-based formulas through the first year of life using single-photon absorptiometry. Interestingly, the two soy-protein based formulas in that study were ISO and PRO, but at that time PRO did not contain PO in its fat blend. There are no published studies of infants fed the current PO-containing version of PRO.
The relationship between calcium absorption and calcium availability for skeletal development is important for this study. As proposed by Heaney et al. [20], calcium is a threshold nutrient, and an individual’s calcium requirement is the lowest intake resulting in maximal retention. They recommend a calcium intake at or above the threshold level from childhood through young adulthood to order to achieve one’s genetic potential for peak bone mass because, although " ... it’s possible that ‘catch-up’ consolidation may occur to make up for periods of nutritional insufficiency, data to support this concept are lacking in humans."
In conclusion, the current study adds to the growing body of evidence that using palm or PO in infant formulas at levels needed to match the PA content of human milk significantly reduces calcium absorption and that this holds true in both hydrolyzed protein- and soy protein-based infant formulas that use mineral salts as a primary source of calcium. Designing infant formula based on human milk composition rather than clinical outcome may have untoward physiological consequences.
|
ACKNOWLEDGMENTS |
|---|
|
TOPFOOTNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSACKNOWLEDGMENTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
|
TOPFOOTNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSACKNOWLEDGMENTSREFERENCES |
|---|
Received March 1, 2002. Accepted August 14, 2002.
|
REFERENCES |
|---|
|
TOPFOOTNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSACKNOWLEDGMENTSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  W. W.K. Koo, E. M. Hockman, and M. Dow Palm olein in the fat blend of infant formulas: effect on the intestinal absorption of calcium and fat, and bone mineralization. J. Am. Coll. Nutr., April 1, 2006; 25(2): 117 - 122. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. W. K. Koo Efficacy and Safety of Docosahexaenoic Acid and Arachidonic Acid Addition to Infant Formulas: Can One Buy Better Vision and Intelligence? J. Am. Coll. Nutr., April 1, 2003; 22(2): 101 - 107. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
