Secondary Logo

Journal Logo

Original Article

Significant Loss of Macronutrients During Passage Through Feeding Tube: An Observational Study

Moran-Lev, Hadar MD*,†; Azaiza, Muhammad MD; Mangel, Laurence PhD§; Mimouni, Francis B. MD; Mandel, Dror MD†,§; Lubetzky, Ronit MD*,†

Author Information
doi: 10.1097/PG9.0000000000000006
  • Open


Nasogastric tube (NGT) feeding is necessary in preterm infants who do not have yet mature suck and swallow capability. Suck–swallow coordination, cough, and gag mature only at 32–34 weeks’ gestation (1). Gastric feeding in the Neonatal Intensive Care Unit (NICU) can be given via nasal or oral tubes, either as intermittent bolus feeding or as continuous feeding (2). During the process of tube feeding, substantial losses of nutrients may occur due to refrigeration, freezing, heating, adherence to the tubing system, and/or photodegradation (2–10). Previous studies have concluded that tube infusion could decrease the fat content of human milk (HM) (3,5–10). However, little is known about the impact of passage through NGT on the formula composition.

What Is Known

  • Feeding infants born before week 34 of gestation is based mainly on providing nutrition directly to the gastrointestinal tract through a nasogastric tube.
  • Previous studies have concluded that tube infusion could decrease the fat content of human milk.
  • Little is known about the impact of passage through nasogastric tube on formula composition.

What Is New

  • Macronutrient composition of infant formula is altered after passage through a feeding tube.
  • These alterations might relate to specific formulation of each formula.

The aim of the present study was to evaluate possible changes in macronutrients content of several standard infant formulae after transfer through a feeding tube. We hypothesized that similar to HM, passage of a formula through NGT leads to a decrease in fat and energy content.


Sample Collection

We examined 11 infant formulas commonly used in the NICU at the Tel Aviv Sourasky Medical Center. The formulas that were studied included 2 preterm formulas (Nos. 1 and 2), 4 extensively hydrolyzed formulas (Nos. 3–6), 2 amino acid–based formulas (Nos. 7 and 8), and 3 standard cow’s milk–based formulas for term infants (Nos. 9–11). Ten consecutive measurements were performed for each formula by a single investigator (M.A.). Three formulas were ready to feed (RTF, Nos. 1–2, 11) and the other 6 were powder-based formulas (Nos. 3–9, 10). Table 1 depicts the formula tested according to brands and types of formulas. The preparation was done according to the manufacturer’s guidelines in a routine manner as per NICU protocol.

TABLE 1. - Formulas that Were Tested (Brand and Type)
Number Brand Protein Other Ingredients Type
1 Abbott, RTF
Similac Special Care 24
Whey, Casein LC-PUFA Preterm formula (24 Cal/flOz)
2 Nestlé, RTF
Materna Preterm infants
Whey, Casein LC-PUFA Preterm formula (24 Cal/flOz)
3 Abbott
Similac Alimentum
Casein LC-PUFA; no lactose; MCT 33% Extensively hydrolyzed formulas
4 Mead Johnson
Nutramigen stage 1
Casein LC-PUFA; no lactose Extensively hydrolyzed formulas
5 Nutricia
Pepti Junior
Whey LC-PUFA; no lactose; MCT 50% Extensively hydrolyzed formulas
6 Mead Johnson
Casein LC-PUFA; MCT 55%; no lactose Extensively hydrolyzed formulas
7 Nutricia
Amino acid LC-PUFA; no lactose Amino acid–based formulas
8 Mead Johnson
Nutramigen AA
Amino acid LC-PUFA; no lactose Amino acid–based formulas
9 Nestlé
Materna Extracare
Stage 1
Whey, Casein LC-PUFA Cow’s milk–based formulas for term infants
10 Abbott
SimilacAdvance Stage 1
Whey, Casein LC-PUFA Cow’s milk–based formulas for term infants
11 Nutricia, RTF
Nutrilon stage 1
Whey, Casein LC-PUFA Cow’s milk–based formulas for term infants
LC-PUFA indicates long-chain polyunsaturated fatty acids; MCT, medium-chain triglyceride; RTF, ready to feed.

Laboratory Methods

The preterm infants in our NICU are fed every 3 hours for a total of 8 meals per day. In the first phase of our study, we measured the average feeding time through gravity drainage of a gastric feeding tube in 22 preterm infants (gestational age ranging from 26 to 34 weeks) who were admitted to our NICU. The average feeding time for a 5-ml HM sample was 14 minutes and 30 seconds. Therefore, simulated real-life practice tube feeding of intermittent bolus was performed by using an infusion pump (Alaris; CareFusion, Becton, Dickinson and Company, NJ) connected to a feeding tube (Metric/x-ray, 40 cm CH 05; Unomedical, Denmark) to transfer 30 ml of formula at a speed of 20 ml/h into 10-ml tubes. For each formula tested, we examined 10 samples before and after each passage through a feeding tube. The NGT was changed after each simulated meal and was not washed by saline or air after the meal. In order to minimize the risk of evaporation, each formula was kept in closed bottles until the analysis. Moreover, we constructed the simulated real-life practice of tube feeding as a “closed system” by connecting the feeding tube to an infusion pump and followed that by measuring the milk composition immediately after it exited the NGT. Analysis of the samples were done using the MIRIS milk analyzer (Uppsala, Sweden), an instrument based on mid-infrared transmission spectroscopy (11), as previously described by us (12–14).

Statistical Analysis

Statistical analyses were performed using Minitab version 16 (Minitab Inc, State College, PA). Paired t tests were used to compare macronutrient concentrations in all samples and in each formula before and after tube feeding. Continuous variables were expressed as median and range when not normally distributed and as mean ± SD when normally distributed. Dichotomous variables were expressed as percent/ratio. Stepwise multiple regression analysis was performed to determine the effect between formula type and macronutrients and energy content after tube infusion. A P value ≤0.05 was considered significant.

Ethical Considerations

The study protocol was approved by the institutional review board of the medical center (TLV-0062-11).


A total of 220 measurements were performed. Twenty measurements of 1 of the amino acid–based formulas (No. 8) were excluded from further analysis due to technical error.

Variations in at least 1 macronutrient were observed in 5 out of 10 formulas (Nos. 1, 2, 3, 5, and 6). Fat and energy content were altered in 1 preterm formula (No. 1). Fat content differed in 2 other extensively hydrolyzed formulas (Nos. 3 and 5), whereas carbohydrate content was affected in another extensively hydrolyzed formula (No. 6). Protein and energy content were affected in preterm formula No. 2 (Table 2). Stepwise multiple regression analysis demonstrated that the fat, protein, and energy content after tube infusion were affected by formula type (P < 0.001 for all).

TABLE 2. - Changes in Macronutrient Content After Tube Feeding Transfer
Macronutrient Content Before Transfer* Content After Transfer* Loss % P
Formula No. 1 Fat 5.98 ± 0.09 5.85 ± 0.08 2 0.002
Protein 2.59 ± 0.03 2.55 ± 0.09 1.3 0.2
Carbohydrate 6.7 ± 0.1 6.69 ± 0.3 0.1 0.6
Energy 92.8 ± 0.9 91.4 ± 1.3 1.5 0.007
Formula No. 2 Fat 5.3 ± 0.04 5.28 ± 0.04 0.3 0.3
Protein 1.58 ± 0.04 1.49 ± 0.03 2.2 0.005
Carbohydrate 7.33 ± 0.02 7.32 ± 0.1 0.4 0.08
Energy 67.6 ± 0.69 66.8 ± 0.63 1 <0.001
Formula No. 3 Fat 4.32 ± 0.06 4.25 ± 0.08 1.6 0.04
Protein 1.62 ± 0.05 1.61 ± 0.05 0.6 0.08
Carbohydrate 6.26 ± 0.1 6.21 ± 0.07 0.7 0.06
Energy 70.1 ± 0.56 69.9 ± 0.73 0.2 0.08
Formula No. 4 Fat 3.93 ± 0.04 3.92 ± 0.04 0.2 0.5
Protein 1.67 ± 0.03 1.66 ± 0.04 0.5 0.1
Carbohydrate 8.3 ± 0.06 8.28 ± 0.07 0.2 0.5
Energy 73.1 ± 0.3 72.9 ± 0.3 0.2 0.1
Formula No. 5 Fat 5.2 ± 0.04 5.1 ± 0.06 1.5 0.01
Protein 1.69 ± 0.03 1.67 ± 0.04 1.1 0.1
Carbohydrate 8.3 ± 0.06 8.28 ± 0.07 0.2 0.5
Energy 73.1 ± 0.3 72.9 ± 0.3 0.2 0.1
Formula No. 6 Fat 5.3 ± 0.04 5.3 ± 0.06 0 0.7
Protein 1.61 ± 0.03 1.59 ± 0.03 1.2 0.1
Carbohydrate 6.86 ± 0.1 6.64 ± 0.04 3.3 0.002
Energy 80.1 ± 0.7 79.6 ± 0.8 0.6 0.1
Formula No. 7 Fat 3.17 ± 0.04 3.1 ± 0.1 0 0.1
Protein 1.2 ± 0.05 1.2 ± 0.03 0 1
Carbohydrate 6.69 ± 0.1 6.68 ± 0.01 0.1 0.8
Energy 60.7 ± 1 60.1 ± 0.8 0.9 0.2
Formula No. 9 Fat 3.89 ± 0.03 3.86 ± 0.03 0.2 0. 2
Protein 1.22 ± 0.04 1.21 ± 0.03 0.1 0.2
Carbohydrate 6.98 ± 0.06 6.97 ± 0.06 0.1 0.3
Energy 67.6 ± 0.6 66.8 ± 0.6 1 0.3
Formula No. 10 Fat 3.67 ± 0.04 3.66 ± 0.05 0.2 0.1
Protein 1.1 ± 0.5 1.1 ± 0.4 0 0.4
Carbohydrate 7.75 ± 0.07 7.74 ± 0.1 0.1 0.5
Energy 57.2 ± 0.3 57 ± 0.2 0 0.2
Formula No. 11 Fat 3.16 ± 0.04 3.15 ± 0.1 0.2 0.1
Protein 1.2 ± 0.05 1.2 ± 0.03 0 1
Carbohydrate 6.71 ± 0.1 6.7 ± 0.01 0 0.8
Energy 60.2 ± 1 60.1 ± 0.8 0.1 0.2
Data are presented as mean ± SD.
*Fat, g/100 ml; carbohydrate, g/100 ml; fat, g/100 ml; energy, kcal/100 ml.
†Dependent t test.


In the present study, we have been able to demonstrate a small but significant change in the macronutrients content for some infant formulas after tube feeding. Some of these formulas were designed for very low birth weight (VLBW) infants.

It has been known for years that tube feeding may result in quantitative loss of lipids due to separation of milk fat and adherence to tubing and syringe. In theory, since lipids are the major contributor of energy content of formula, a caloric reduction due to fat losses may have deleterious effect upon growth (2). Also, since premature infants have a limited endogenous essential fatty acid pool and limited fat stores, fat losses may lead to essential fatty acid deficiency faster than in term infants or adults (15). These essential fatty acids are known to be critical for adequate brain and retina development of the VLBW infants (15,16).

The results of our study differ from previous studies conducted on this topic. Tacken et al (3) examined triglyceride, α- and β-carotene, lutein, and lycopene concentrations of 30 samples of mature HM from mothers who delivered a term infant and 10 samples of infant formula (3 different infant fluid formulas for preterm infant, no specific fat composition indicated) after refrigeration, freezing, microwave heating, and tube feeding. They found that after tube feeding HM, triglyceride, lutein, and β-carotene concentrations decreased by 33%, 35%, and 26%, respectively. However, no differences were found when using infant formulas. Igawa et al (6) compared preinfusion and postinfusion fat contents of 15 fresh and 10 thawed HM samples and 6 formula samples (brand names and specific ingredients were not mentioned) and found that only thawed and fresh HM postinfusion fat contents were significantly lower, but none of the tested formula samples. They concluded that fat content of infant formulas was not affected by tube size, tube material, or infusion duration (6). In contrast, other studies using slow infusion rates showed a loss of fat up to 40% (17,18). In the study of Greer et al (17), mean fat loss had an inverse relationship with the rate of infusion. Mehta et al (19) measured fat residue in the feeding sets by gravimetry, and individual fatty acids were characterized by gas–liquid chromatography. They found that a major portion of medium-chain triglycerides were lost. This finding suggests that differences in formulation might change the tendency of the fat of a particular formula to adhere to the feeding tube. In our study, we used an intermittent bolus feeding model, and nevertheless, variations in at least 1 macronutrient were observed in 5 out of 10 formulas that were tested in our study. We documented fat loss in 2 extensively hydrolyzed formula and in 1 preterm formula; protein loss in standard cow’s milk formula; and carbohydrate loss in 1 extensively hydrolyzed formula. The formulas we tested differed in many aspects, in particular in triglyceride fatty acid composition, in long-chain fatty acid percentage, and in protein type or size (Table 1). Continued study that will examine identical formulas with change in 1 component at a time will help identify the exact component that is responsible for the losses over time.

A strength of our study is its relatively large sample size (the largest that was documented in the literature until now) of formulas that were tested. However, our study has several limitations. First, we do not have longitudinal information on weight gain and development of these infants in order to understand the implication of our result. Another limitation of our study is the fact that we used the near-infrared spectroscopy method of macronutrient evaluation rather than direct biochemistry methods. However, near-infrared spectroscopy method has been shown to correlate quite well with biochemical analyses of milk (20). Moreover, the results of formula compositions at baseline were within the ranges of the manufacturer’s published data.

In conclusion, we have shown that the macronutrient composition of infant formula is altered after passage through a feeding tube even at a rate simulating bolus feeding. Larger studies that employ direct biochemistry methods and compare different formulation and feeding practices are needed to better understand the biological significance of our findings as they relate to the VLBW infant, as the preferred practice for tube feeding in preterm infant in order to minimize nutrient loss during administration. If the current results will be verified in larger studies and in different feeding practice, it may be worthwhile to concentrate the formulas or add more volume when they are given to preterm babies through NGTs.


This study was conducted according to the guidelines laid down in the Declaration of Helsinki and all procedures involving research study participants were approved by the Helsinki ethics committee. Written informed consent was obtained from mothers of all participants.


1. McGuire W, Henderson G, Fowlie PW. Feeding the preterm infant. BMJ. 2004; 3291227–1230
2. Goswami I, Alshaikh B. Gavage feeding practices in VLBW infants: physiological aspects and clinical implications. Neoreviews. 2017; 18106–115
3. Tacken KJ, Vogelsang A, van Lingen RA, et al. Loss of triglycerides and carotenoids in human milk after processing. Arch Dis Child Fetal Neonatal Ed. 2009; 94447–450
4. Buss IH, McGill F, Darlow BA. Vitamin C is reduced in human milk after storage. Acta Paediatr. 2001; 90813–815
5. Ovesen L, Jakobsen J, Leth T. the effect of microwave heating on vitamin B1 and E, linoleic and linolenic acids and immunoglobulins in human milk. Int J Food Sci Nutr. 1996; 47427–436
6. Igawa M, Murase M, Mizuno K, et al. Is fat content of human milk decreased by infusion? Pediatr Int. 2014; 56230–233
7. Narayanan I, Singh B, Harvey D. Fat loss during feeding of human milk. Arch Dis Child. 1984; 59475–477
8. Van Zoeren-Grobben D, Moison RM, Ester WM, et al. Lipid peroxidation in human milk and infant formula: effect of storage, tube feeding and exposure to phototherapy. Acta Paediatrica. 1993; 82645–649
9. Rogers SP, Hicks PD, Hamzo M, et al. Continuous feeding of fortified human milk lead to nutrient losses of fat, calcium and phosphorous. Nutrients. 2010; 2230–240
10. Stocks RJ, Davies DP, Allen F, et al. Loss of breast milk nutrients during tube feeding. Arch Dis Child. 1985; 60164–166
11. Casadio YS, Williams TM, Lai CT, et al. Evaluation of a mid-infrared analyzer for the determination of the macronutrient composition of human milk. J Hum Lact. 2010; 26376–383
12. Lev HM, Ovental A, Mandel D, et al. Major losses of fat, carbohydrates and energy content of preterm human milk frozen at −80°C. J Perinatol. 2014; 34396–398
13. Moran-Lev H, Mimouni FB, Ovental A, et al. Circadian macronutrients variations over the first 7 weeks of human milk feeding of preterm infants. Breastfeed Med. 2015; 10366–370
14. Lubetzky R, Sever O, Mimouni FB, et al. Human milk macronutrients content: effect of advanced maternal age. Breastfeed Med. 2015; 10433–436
15. Decsi T, Koletzko B. Polyunsturated fatty acid in infant nutrition. Acta Paediatr Suppl. 1994; 8331–37
16. Makrides M, Neumann MA, Byrad RW, et al. Fatty Acid composition of brain, retina and erythrocytes in breast and formula fed infant. Am J Clin Nutr. 1994; 60189–194
17. Greer FR, McCormick A, Loker J. Changes in fat concentration of human milk during delivery by intermittent bolus and continuous mechanical pump infusion. J Pediatr. 1984; 105745–749
18. Brooke OG, Barely J. loses of energy during continuous infusion of breast milk. Arch Dis Child. 1978; 53344–345
19. Mehta NR, Hamosh M, Bitman J, et al. Adherence of medium chain fatty acid to feeding tubes during gavage feeding of human milk fortified with medium chain triglycerides. J Pediatr. 1988; 112474–476
20. Mimouni FB, Lubetzky R, Yochpaz S, et al. Preterm human milk macronutrient and energy composition: a systematic review and meta-analysis. Clin Perinatol. 2017; 44165–172

preterm infant; human milk; tube feeding; preterm formula; macronutrient

Copyright © 2020 The Author(s). Published by Wolters Kluwer on behalf of European Society for Pediatric Gastroenterology, Hepatology, and Nutrition and North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition.