Secondary Logo

Journal Logo

Original Article

Structural Position and Amount of Palmitic Acid in Infant Formulas: Effects on Fat, Fatty Acid, and Mineral Balance

Carnielli, Virgilio P.; Luijendijk, Ingrid H. T.; Van Goudoever, Johannes B.; Sulkers, Eric J.; Boerlage, Anneke A.; Degenhart, Herman J.; Sauer, Pieter J. J.

Author Information
Journal of Pediatric Gastroenterology & Nutrition: December 1996 - Volume 23 - Issue 5 - p 553-560
  • Free


Palmitic acid constitutes approximately one fourth of human milk (HM) lipids and is esterified mainly to the center sn-2 position or β-position of the milk triglycerides (TG) molecule (1,2). In cow's milk fat or in vegetable oils, commonly used in infant formulas, palmitic acid is predominantly esterified to the external sn-1 and sn-3 positions, or α-positions, with relatively little incorporation at the β-position. Because the bond between the fatty acid and the glycerol molecule at the β-position is relatively resistant to digestion by the pancreatic lipase, a large fraction of the FA at the β-position remains attached at the glycerol (as monoglycerides) during digestion and absorption (3) and probably conserves this position in plasma TG (4). Therefore, in the presence of a sufficient pancreatic lipase activity, the final products of digestion are mainly the free fatty acids deriving from the external α-position and the 2-monoglyceride with the original fatty acid still bound to the β-position (3,5,6). Most of the fatty acids are better absorbed in the form of monoglycerides than free acids, because monoglycerides form mixed micellae with bile acids and cannot form insoluble soaps with divalent cations such as calcium and magnesium. Therefore, the monoglycerides of the less well absorbed fatty acids, such as the saturated fatty acids, are better absorbed than the respective fatty acids in the free form. The absorption of palmitic acid from fat where it is esterified at the β-position, is therefore greater as compared with the absorption of other fats where it is esterified to the α-positions. This determines better coefficients of intestinal absorption for both fatty acids and calcium. Studies of the effect of the TG configuration on intestinal fat absorption were reported in the late 1960s in animals (7) as well as in healthy term infants (8). Tomarelli et al. showed that the high absorption in rats of HM fat was equaled by a fat blend of a similar fatty acid composition with a high content of palmitic acid at the β-position of the TG, in contrast to one with low palmitic at the β-position (7). In a study in term human infants, Filer et al. (8) compared fat, fatty acids, and mineral balances in 5 infants who received a formula containing natural lard (palmitic acid primarily in the β-position) and in 6 who received a formula containing randomized lard (palmitic acid equally distributed among the α- and β-positions). They noted reduced fat excretion in the lard group but no difference in mineral balance. Nonetheless, later studies of preterm infants receiving formulas with different fatty acid composition failed to demonstrate clear benefits from formulas in which palmitic acid was esterified mainly at the sn-2 position relative to formulas with palmitic acid at the sn-1,3 positions (9-11). Using novel synthetic TG in a recent study of preterm infants (12), we demonstrated that the isomeric position of fatty acids in dietary TG has significant effects on the intestinal absorption of fatty acids. The use of a formula containing TG with a isomeric structure similar to that of HM (26% palmitic acid, predominantly at the β-position) was associated with an improvement in the absorption of myristic, palmitic, and stearic acid and in the mineral balance. In that study (12), we speculated that the advantage of the β-position could be greater in term infants due to the larger contribution of pancreatic lipase (13,14) and bile salts to the digestive process. Bile salts are important for monoglyceride absorption (15-17). The novelty of our study resides in the use of synthetic TG, thus, unlike investigators in previous studies, we studied formulas of identical fatty acid composition differing only in the TG structure. In the present study, we investigated the effect of the structural position of palmitic acid on the intestinal absorption of fat and calcium in term infants.



Three groups, consisting of 9 healthy term male infants each, who were fed with one of the three study formulas, were studied from birth until at least age 5 weeks. All studied infants were free of manifest diseases and received no medications. Participation in the study was voluntary, and written informed consent had been obtained from the parents.

Study Feedings

The enrolled infants were fed exclusively with one of the three study formulas. The ready-made formulas (Nutricia Research-Zoetemeer, The Netherlands) were supplied in 200-ml bottles; their composition is shown in Table 1. The beta formula was designed to resemble HM fat most closely; (23.9%) of the formula fat was given as palmitic acid predominantly (66%) esterified to the sn-2 position (β-position). Part of the fat in the beta formula was constituted by synthetic TG (Betapol), which is produced by interesterifying a tripalmitin-rich palm fraction with a mixture of other oils by using the sn-1,3-specific lipase from Rhizomucor miehei (code SP-392; Novo Industries, Copenhagen, Denmark) (18). The regular formula was identical to the infant formula currently marketed by Nutricia/Cow & Gate under the brand names Nutrilon Premium and Premium, respectively. This formula contained 19.9% of palmitic acid, mainly esterified to the sn-1,3 positions. The intermediate formula was designed to represent an intermediate type of formula between the beta and the regular formula. This formula contained a lower amount of Betapol than the beta formula. The total amount of palmitic acid of the intermediate formula (24.0%) was chosen to be identical to that of the beta formula (and higher as compared with that of the regular formula) with 39% of it at the sn-2 position (or 9.4% of the formula fat) as compared with 66% (or 15.8% of the formula fat) in the beta formula. The regular formula (19.9% of total fat as palmitic acid, 2.6% of total fat as β- and 17.3% as α-palmitate) had absolute amounts of palmitic acid esterified to the α-positions relatively similar to those of the intermediate formula (24.0% of total fat as palmitic acid, 9.4% of total fat as β- and 14.7% as α-palmitate) and clearly different amounts of palmitic at the sn-2 position. The fatty acid composition of the sn-2 position of the three study formulas is shown in Table 2.

Clinical Design

The infants were randomly assigned to receive one of the three study formulas from birth and at least until age 5 weeks. Random selection was obtained by the “sealed envelope” system and the investigators were blindfolded as to the type of feeding received by the infants. The infants were studied at home where the procedures were performed by a qualified research nurse.

Anthropometric Measurements

Anthropometric measurements were performed at birth and at 2 and 4-5 weeks of age. These measurements included body weight, total length, head circumference, and skin fold thickness. Percent of total body fat was calculated according to the method of Dauncey et al. (19).

Balance Studies

Fat and mineral balance studies were performed at the postnatal age of ≈4 weeks with the infants at home with their parents. The methods we used for collection of feces and urine were the same that were used in our hospital unit. The only difference was that in the hospital a research nurse collected the urine and diapers, and the parents did so at home. However, a research nurse visited the infants home every day during the balance studies. On day 1, the nurse instructed the parents and started the collections, visiting each infant early in the morning on the subsequent days to collect the samples and to check the accuracy of balance. The nurse could always be reached by phone by the parents. The balance studies involved the measurement of formula intake, the collection of feces for a 72-h period, and the collection of urine for a 24-h period. the amount of formula taken was calculated by the nurse, the parents recorded the weight of the bottle before and after the feeding and the weight of bibs and other instruments if they had gotten wet during the feeding or because the baby regurgitated. A special scale for this purpose was given to the parents during the balance study days. Carmine red was used to bracket the fecal collections. Carmine red 100 mg was dissolved in 2 ml of distilled water and administered orally just before the 12 o'clock feeding on day 1 of the balance studies. Intestinal transit time was assumed to be equal to the time elapsed between the administration of carmine red and the production of the first red stool. Feces were collected for 72 h by use of a plastic sheet placed inside the diapers. Corrections were made for accidental losses of feces into the diapers (double weighing of the diaper) and for feces sticking to the buttocks of the infants (double weighing of the cleaning swabs). The total amount of feces collected during the 3-day balance period was weighed and homogenized, and a small sample of the homogenate was freeze-dried. Although in the present study we did not intend to assess the characteristics of the feces because of the limited number of patients to be enrolled, we noted marked differences in the first fecal samples that arrived at the laboratory. The degree of stool hardness as well as the color were then estimated by the research nurse (A.B.) using a method based on four levels of consistency (1 = watery, 2 = runny-soft, 3 = soft, and 4 = hard) and 5 colors (1 = yellow, 2 = green, 3 = light brown, 4 = dark brown, and 5 = black). Fat excretion was determined by a modification of the Jeejeebhoy method (20): twice as much hydrochloric acid was added. Ca and Mg were measured with an atomic absorption spectrophotometer (Perkin-Elmer 2380) after “acid digestion” [15 min at 100°C, 15 min at 200°C, and 15 min at 300°C, in 5 ml concentrated HNO3/concentrated H2SO4 mixture (2:1)]. The total urine volume was measured, and urinary Ca and Mg were also measured by atomic absorption spectrophotometry. Phosphate was measured by a colorimetric method (21). In all cases, matrix effects were eliminated by the standard addition method (22). Excretions were calculated by multiplying the volume of feces (or urine) produced by the concentration of the compounds of interest. Intestinal absorption was calculated by dividing the apparent amount absorbed (intake-excretion) by the intake and then multiplied by 100.

Determination of FAs in the Formulas and in the Feces

The individual FA content of the feeding and of the feces was determined by gas chromatography, and analyses were performed in duplicate. Fresh fecal samples of 10-20 mg each were subjected to trans-esterification by HCl methanol after the addition of C7, C9, C11, C15, C17, and C23 as internal standards. Fatty acid methyl esters were separated and identified by a gas chromatograph (GC) (Hewlett Packard 5890 II) equipped with a fused silica column (Supelcowas 10, 60 × 0.20 mm ID, 0.20-μm film thickness), a flame ionization detector (280°C), and a split-splitless injector used in splitless mode (280°C). The GC was operated with the following temperature program: 60°C initially for 5 min, with the oven temperature then raised at 20°C per min to 205°C and held at that temperature for 15 min. The temperature was then increased again at 0.2°C per min to 222°C. Helium was used as a carrier gas (2 ml/min), and peak areas were calculated by HP-Chem station software. Fatty acids were identified by comparison of the retention times with known standards (Nu Chek Prep, Elysian, MN, U.S.A.). All reagents were of analytical grade.

Determination of the Fatty Acid Composition of the sn-2 Position of the Formulas

The percentage of the individual fatty acids esterified to the sn-2 position of formula TG was calculated from the fatty acid composition of the two monoglycerides obtained after the digestion with pancreatic lipase (23).

Statistical Analysis

Data are group mean ± SD. Means were compared by analysis of variance. Multiple post hoc comparisons were analyzed by the Tukey test. The degrees of consistency as well as the color of the feces of the different feeding groups were compared by the Pearson chi-square test. Correlation between the data were calculated by simple linear regression. All calculations were performed with the statistical package Systat, version 5.2.

Ethical Consideration

The project was approved by the local ethics and scientific committees. Written informed consent was obtained from both parents of each infant in the study. Only infants whose mothers decided not to breast feed were approached and requested to participate in the study. We performed “home balances” and not “in-hospital balances” because we did not consider it ethical to admit normal full-term infants to the hospital for 4-5 consecutive days and impose such a burden on a family just a few weeks after the delivery of a normal infant. The parents received free formula for the duration of the study period (≅5-6 weeks).


The anthropometric characteristics of the infants at birth and at the time of the metabolic balance studies are shown in Table 3. No statistically significant differences were detected among the three groups either at birth or at the time of the metabolic balance studies (at ≈30 days of postnatal age). Mean formula intakes of the three groups during the 72-h balance period are shown in Table 4 together with the data on the intestinal transit time, fecal and urine productions, and composition of feces. Infants fed the beta formula produced a smaller amount of feces than did infants fed the regular formula, and their stool contained significantly less fat than the stool of the other two groups (Table 4). Fat balance data are shown in Table 5. Fat excretion of the infants fed the beta formula was significantly lower than that of the other two groups. Coefficients of intestinal fat absorption were also significantly higher in the infants receiving the beta formula (97.5 ± 0.90) than in the infants receiving the intermediate formula (93.2 ± 1.8) or the regular formula (90.09 ± 4.59). The intestinal absorption of the major saturated fatty acids, i.e., C12 (lauric acid), C14 (myristic acid), C16 (palmitic acid), and C18 (stearic acid) was significantly greater in the beta group than in the other two groups. No significant differences were noted for the major mono- and polyunsaturated FA. The results of the mineral balances are shown in Table 6. The fecal calcium excretion of the infants fed the beta formula was significantly lower than that of the infants fed one of the other two formulas, resulting in significantly improved calcium absorption and better mean retention values. On the average calcium retention ≅16 mg · kg-1 · day-1 greater in infants fed the beta formula than in infants fed the intermediate and regular formulas, although this difference did not reach statistical significance (p = 0.10).

Phosphorus and magnesium intake, excretion, absorption, and retention values were not significantly different among the three groups. Palmitic acid was the major fecal fatty acid in most of the infants. We noted significant correlation between fecal calcium excretion and the excretions of fat and of the major fatty acids. Palmitic acid showed the highest correlation coefficient (r = 0.84), followed by oleic acid (r = 0.60) and linoleic acid (r = 0.51) (Fig. 1). Although in the present study design a systematic scoring of the degree of stool hardness and color was not included, differences were observed in the characteristics of the stools. Color and consistency of the “fresh” feces could be obtained from 8, 7, and 8 infants of the beta, intermediate, and regular groups, respectively. Consistency was significantly different among the groups (p = 0.003). In the beta group, 2 infants had soft feces, 6 had runny-soft feces, and none had hard stools. In the regular group, 4 infants had hard feces and 4 had soft feces. Infants fed the intermediate formula were in an intermediate condition; no infant had hard stools, 5 had soft feces, and only 2 had runnysoft feces. None of the infants had watery stools. The color of the feces was also statistically different (p = 0.026): All infants in the beta group had yellow feces (n = 8, 100%); the infants in the intermediate and regular groups had more brown and green feces (4 and 1 vs. 2 and 3, respectively).


Several studies have focused on the effect of the TG configuration on intestinal fat absorption in preterm infants (9-11). Only one group of investigators has studied the influence of the TG structure on the absorption of fatty acids in term infants (8); Filer et al. compared a formula based on natural lard (palmitic primarily at the β-position) with one based on a randomized lard. The infants fed the natural lard showed an improved absorption of all fatty acids, most markedly of palmitic and stearic acids, but calcium absorption was not improved. In our study using synthetic TG we compared fat in which palmitic is predominantly in β-position with fat in which palmitic is predominantly in α-position, whereas in the second diet used by Filer et al. (8), the fatty acids were randomized and thus equally distributed among the α- and β-positions. Our results confirm the finding of Filer et al. (8) of improved fat absorption in infants receiving the palmitic acid predominantly at the β-position, but we were also able to demonstrate a reduction in the fecal excretion of calcium. In term infants, absorption of calcium is clearly related to the absorption of fat (24-27). A possible explanation for the lack of effect on calcium absorption in the study of Filer et al. (8) might be the difference in calcium concentration between the formulas they used (69 and 72 mg/100 ml) and those used in our study (52.5-54.0 mg/100 ml) (Table 1). In our study, the fecal excretion of calcium exceeded the excretion of palmitic acid of ≈0.5 mmol · kg-1 · day-1(Fig. 1). If the formulas had had a lower calcium content, the fecal calcium concentration would have been less and the palmitic acid calcium ratio higher (nearer 1:1). In such case, the effect of fecal palmitic acid (and of sn-2 palmitate) on calcium excretion might be more pronounced because of the larger percentage of calcium available for calcium palmitate formation and thus for fecal losses. Conversely, when the calcium concentration in formulas is more than double the amount in breast milk, the effect of fecal palmitate and of sn-2 palmitate on calcium absorption percentage would no longer be measurable.

Although we did not plan a systematic scoring of the degree of stool hardness and color differences, we decided to include such an evaluation when differences were noted in the stool samples of the first patients: The samples from the infants fed the beta formula were softer than the stool samples from infants fed the intermediate and the regular formulas. However, unlike in the recent study of Quinlan et al. (28) in which hardness of stools of breast-fed and formula-fed infants was compared, the variation in stool hardness in our study was not due to a significant difference in the water content of the infants' feces. Because this observation requires confirmation in future studies involving a larger number of infants, we may hypothesize that the lipid content of the stools, including the calcium soaps, has a much more pronounced effect on stool hardness than does its water content. The differences in total fecal output among the three feeding groups are difficult to explain. We know of no report in the literature describing a clearly lower fecal output in breast-fed infants than in formula-fed infants or in infants fed formulas of a composition relatively similar to those we used in our study. Although calcium absorption was greater in the beta group, we could not demonstrate a significant improvement in calcium and phosphate retention. Nonetheless, on the average, the infants fed the beta formula retained 16 mg · kg-1 · day-1 more calcium than the other two groups. This difference should be considered biologically significant and, if confirmed in a larger group of infants, certainly is further evidence of the importance of the structure of dietary TG. In term infants we showed that the isomeric position of fatty acids in dietary TG has a significant effect on the intestinal metabolism of the dietary FA. The use of beta formula containing TG with a structure more similar to those of HM fats than the currently available infant formulas is associated with an improvement in the intestinal absorption of the major saturated fatty acids and of total fat. Calcium excretion and absorption were also significantly improved. Formulas containing these novel TG similar in structure to HM lipids offer an advantage as compared with conventional formulas.

Acknowledgment: This work was supported by Nutricia Research (Zoetermeer, The Netherlands).

FIG. 1.
FIG. 1.:
Correlation between the excretion of calcium and of fatty acids C16:0 (squares), C18:1ω9 + ω7 (circles), and C18:2ω6 (triangles). All regressions were statistically significant.


1. Jensen RG. Lipids in human milk-composition and fatsoluble vitamins. In: Lebenthal E, ed. Textbook of gastroenterology and nutrition in infancy, 2nd ed. New York: Raven Press, 1989:157-208.
2. Breckenridge WC. Stereospecific analysis of triacylglycerols. In: Kuksis A, ed. Handbook of lipid research 1-fatty acids and glycerides. New York: Plenum Press, 1978:197-232.
3. Mattson FH, Volpenhein RA. Rearrangement of glyceride fatty acid during digestion and absorption. J Biol Chem 1962;237:53-61.
4. Innis SM, Dyer R, Nelson CM. Evidence that palmitic acid is absorbed as sn-2 monoacylglycerol from human milk by breast-fed infants. Lipids 1994;29:541-5.
5. Thomson AB, Keelan M, Garg ML, Clandinin MT. Intestinal aspects of lipid absorption: in review. Can J Physiol Pharmacol 1989;67:179-91.
6. Bernback S, Blackberg L, Hernell O. The complete digestion of human milk triacylglycerol in vitro requires gastric lipase, pancreatic colipase-dependent lipase, and bile salt-stimulated lipase. J Clin Invest 1990;85:1221-6.
7. Tomarelli RM, Meyer BJ, Weaber JR, Bernhart FW. Effect of positional distribution on the absorption of the fatty acids of human milk and infant formulas. J Nutr 1968;95:583-90.
8. Filer LJ, Jr, Mattson FH, Fomon SJ. Triglyceride configuration and fat absorption by the human infant. J Nutr 1969;99:293-8.
9. Brooke OG. Absorption of lard by infants. Hum Nutr Appl Nutr 1985;39A:221-3.
10. Verkade HJ, van Asselt WA, Vonk RJ, et al. Fat absorption in premature infants: the effect of lard and antibiotics. Eur J Pediatr 1989;149:126-9.
11. Verkade HJ, Hoving EB, Muskiet FAJ, et al. Fat absorption in neonates: comparison of long-chain-fatty acid and triglyceride compositions of formula, feces, and blood. Am J Clin Nutr 1991;53:643-51.
12. Carnielli VP, Luijendijk IHT, Van Goudoever JB, et al. Feeding premature newborn infants palmitic acid in amounts and stereo isomeric position similar to human milk: effects on fat and mineral balance. Am J Clin Nutr 1995;61:1037-42.
13. Zoppi G, Andreotti G, Pajno-Ferrara F, Njai DM, Gaburro D. Exocrine pancreas function in premature and full term neonates. Pediatr Res 1972;6:880-6.
14. Fredrikzon B, Olivecrona T. Decrease of lipase and esterase activities in intestinal contents of newborn infants during test meals. Pediatr Res 1978;12:631-4.
15. Brueton MJ, Berger HM, Brown GA, Ablitt L, Iyngkaran N, Wharton BA. Duodenal bile acid conjugation and dietary sulphur amino acids in the newborn. Gut 1978;19:95-8.
16. Murphy GM, Singer E. Bile acid metabolism in infants and children. Gut 1974;15:151-63.
17. Watkins JB, Szczepanic P, Gould JB, Klein P, Lester R. Bile salt metabolism in human premature infant. Gastroenterology 1975;69:706-13.
18. Quinlan P, Moore S. Modification of triglycerides by lipases: process technology and its application to the production of nutritionally improved fats. Inform 1993;4:580-5.
19. Dauncey MJ Gandy G, Gairdner D. Assessment of total body fat in infancy from skinfold thickness measurements. Arch Dis Child 1977;52:223-7.
20. Jeejeebhoy KN, Ahmad S, Kozak G. Determination of fecal fats containing both medium and long chain triglycerides and fatty acids. Clin Biochem 1970;3:157-63.
21. Bartels PC, Roijers AFM. A kinetic study on the influence of the parameters in the determination of inorganic phosphate by the molybdenum blue reaction. Clin Chim Acta 1975;61:135-44.
22. Willard HH, Merritt LL Jr, Dean JA. Instrumental methods of analysis, 4th ed. Princeton, NJ: D. Van Nostrand, 1965:342-4.
23. International Organization for Standardization. Animal and vegetable oils. Determination of the composition of fatty acids in the 2-position. Geneva: International Organization for Standardization, (Publication No. ISO 6800), 1985:00-00.
24. Hanna FM, Navarrete DA, Hsu FA. Calcium-fatty acid absorption in term infants fed human milk and prepared formulas simulating human milk. Pediatrics 1970;45:216-24.
25. Barness LA, Morrow GI, Silverio J, Finnegan LP, Heitman SE. Calcium and fat absorption from infant formulas with different fat blends. Pediatrics 1974;54:217-21.
26. Williams ML, Rose CS, Morrow GD, Sloan SE, Barness LA. Calcium and fat absorption in neonatal period. Am J Clin Nutr 1970;23:1322-30.
27. Widdowson EM. Absorption and excretion of fat, nitrogen and minerals from filled milks by babies one week old. Lancet 1965;2:1099-105.
28. Quinlan PT, Lockton S, Irwin J, Lucas AL. The relationship between stool hardness and stool composition in breast-and formula-fed infants. J Pediatr Gastroenterol Nutr 1995;20:81-90.

Triglyceride structure; Fat absorption; Infant formula; Fatty acids; Calcium; Mineral balance

© Lippincott-Raven Publishers