Secondary Logo

Journal Logo

Original Articles: Gastroenterology

Through Thick and Thin: The In Vitro Effects of Thickeners on Infant Feed Viscosity

Koo, Jenny K.; Narvasa, Averyl; Bode, Lars∗,†; Kim, Jae H.∗,†

Author Information
Journal of Pediatric Gastroenterology and Nutrition: November 2019 - Volume 69 - Issue 5 - p e122-e128
doi: 10.1097/MPG.0000000000002470

Abstract

What Is Known/What Is New

What Is Known

  • Thickening infant feeds is an effective treatment for gastroesophageal reflux.
  • Human milk is difficult to thicken because of the presence of digestive enzymes.
  • Gum-based thickeners can thicken human milk, but the properties are not well understood.

What Is New

  • Gum-based thickeners overly thicken human milk, and continue to rise in viscosity for up to 2 hours.
  • Acid added to gum-based thickened human milk causes precipitation.
  • Objective measurements of human milk viscosities after the addition of thickening agents.

Gastroesophageal reflux (GER) is highly prevalent in preterm infants, where it occurs up to 5 times per hour (1). GER is physiological, but is classified as gastroesophageal reflux disease (GERD) if it results in symptoms of pain, respiratory complications, or nutritional compromise (2). Treatment for reflux begins with upright positioning, followed by thickening of feeds, pharmacologic intervention, and surgery (3). A meta-analysis of randomized controlled trials demonstrated effective reduction of GERD symptoms as well as better weight gain in infants fed thickened feeds (4).

Products marked for infants with GERD include milk thickeners and formulas premixed with thickening starches. There are starch-based thickeners (SBT) and gum-based thickeners (GBT) available as additives. Human milk cannot be thickened by SBT because of innate digestive enzymes, such as amylase in human milk (5,6). Conversely, GBT effectively thickens human milk but has been linked to serious adverse effects (7–9). It is important to determine effective ways to thicken human milk, as there are countless benefits to feeding human milk to the premature population, including: reducing the risk of necrotizing enterocolitis, reducing the risk of infections, promoting better nutrient absorption, and many more (10).

It has been described that thickened feeds likely ameliorates symptoms of reflux via the added weight and thickness of the feeds (11). Measuring the viscosity of thickened feeds is the most appropriate in vitro, objective and numerical measurement that best describes this potential mechanism of action. The American Academy of Pediatrics (AAP) recommends caution with use of thickeners with preterm infants, but this practice continues to be widespread across neonatal intensive care units (12). Much is still unknown regarding the effects of thickeners on human milk, and it remains unclear exactly what the mechanism of action of thickened feeds is on reflux.

To understand the mechanism of how thickened feeds affect infants with reflux and aspiration, we aim to address the most fundamental question: What are the basic objective, measurable characteristics of thickened feeds? In this study, we explored various thickeners and their effects on the viscosity of human milk. The primary aim of this study is to obtain objective measurements of viscosity of formula and human milk after adding thickeners. We assessed other variables that could affect the viscosity of milk, including time, temperature, acidity, and presence of human milk fortifier.

METHODS

Human milk and formula were thickened using Thick It (TI-SBT), rice cereal (RC-SBT), SimplyThick xanthum gum (ST-GBT), Thicken Up Clear xanthan gum (TUC-GBT), or GelMix carob bean (GM-GBT). Each sample of milk or formula was thickened per manufacturer recommendations for each of the products (Supplemental Table 1, Supplemental Digital Content, https://links.lww.com/MPG/B704). The viscosity of each sample was measured at 0, 15, 30, and 60 minutes after addition of thickeners. Each thickener was tested in triplicates (n = 3), and 3 separate measurements of each sample were taken and the average value was calculated. GM-GBT requires the liquid to be heated to 100oF before adding the thickener and allowed to rest 5 minutes for thickening to occur; for that reason, GM-GBT was prepared in a heated liquid per manufacturer recommendations, and the time “0 minutes” measurement was made 5 minutes after introduction of the thickener to the liquid. All samples were 120 mL.

Formula with added starches that are marketed for reflux, including Similac Spit Up (SSU), Similac Spit Up ready-to-feed (SSU-RTF) and Enfamil AR (EAR), were evaluated for their ability to thicken as per manufacturer claims. These formulas were intended to have the consistency of thin-liquid before consumption, and upon reaching the warm and acidic environment of the stomach, the formula thickens to aid with reflux symptoms. Because some hospitals treat reflux in infants by combining human milk with added-starch formulas, we evaluated the viscosity of a 1 : 1 mixture of human milk with added-starch formula. Sixty milliliters of DHM was mixed with 60 mL of added-starch formulas. The viscosities of SSU, EAR, SSU-RTF, DHM + SSU, DHM + EAR, and DHM + SSU-RTF were measured before and after the addition of HCl to mimic the acidic environment of the stomach. It is known that stomach parietal cells secrete HCl at an approximate concentration of 100 to 160 mmol/L (13,14). Ten millilitres of 100 mmol/L HCl was added to the thickened solution after it reached a steady peak viscosity at 2 hours. The addition of 10 mL 100 mmol/L HCl to 120 mL sample volume (<10% of feed volume) is a approximation of the ratio of an infant's innate gastric juice volume to the volume of an average feed (15–17). The acidity (pH) of each sample was also measured using pH paper.

At baseline, all samples were kept at room temperature. We explored the effects of high heat on human milk digestive enzymes. We autoclaved donor human milk at 120oC for 15 minutes to inactivate any digestive enzymes in the human milk in order to further establish that bioactive digestive enzymes in DHM are responsible for its inability to be thickened by SBT.

Human milk fortifiers (HMF) are commonly used for preterm infants because they are often times fed DBM (obtained from mothers of term infants) that lack the extra fat, protein, and calories that preterm infants require. Enfamil HMF has 1 g of fat per 4 packets (70% MCT oil, 30% soy oil), 1.1 g of protein per 4 packets (whey: casein ratio is 60 : 40), <0.4 g of carbohydrates per 4 packets (low CHO content to keep the osmolality low), and various vitamins, minerals, and electrolytes. We tested human milk samples with and without HMF to determine if HMF alters the end viscosity of human milk.

Human milk (HM) used in this study was raw human milk donated for research purposes. Samples were pumped by mothers, delivered to the research laboratory where it was frozen and stored at −20°C. The milk was de-identified so no demographic information was available for these samples. The milk did not undergo any pasteurization process. To minimize variation between each batch of donor milk, the individual aliquots of donor milk were thawed and combined into large volumes enough to supply all duplicate trials in each experiment (ie, the experiments looking at added-starch formulas required a total volume of 1620 mL for triplicates in all conditions so enough thawed human milk was combined to achieve that volume).

The viscosity of the milk was measured using the NDJ-1 rotary viscometer (detection range, 10–100,000 mPa seconds, measurement error +/−5%). The rate of rotation, as well as the size of rotors, were changed to achieve specific reading ranges per manufacturer recommendations. Measurements made by the machine were converted to units of viscosity mPa second) using coefficients unique to the rotor size and speed of rotation (Supplemental Table 2, Supplemental Digital Content, https://links.lww.com/MPG/B704). Viscosities were categorized according to the National Dysphagia Diet Task Force (NDDTF) into the following thicknesses: thin (1–50 mPa seconds), nectar-like (51–350 mPa seconds), honey-like (351–1,750 mPa seconds), and spoon-thick (>1751 mPa seconds).

RESULTS

Starch-based Thickeners Versus Gum-based Thickeners in Various Conditions

TI-SBT was unable to effectively thicken HM (Fig. 1). On the contrary, formula was rapidly thickened by TI-SBT and the viscosity of the solution continued to rise over time. When RC-SBT was mixed with formula, the resulting mixture was nonhomogenous with clumps and particles randomly dispersed throughout the mixture. This rendered viscosity readings imprecise, given the vast range of viscosities within a single fluid sample. The difference, however, between formula and HM and their interaction with RC-SBT was striking: HM fully dissolves the rice cereal, leaving no solid particles but at the same time did not result in a notable viscosity change from baseline following dissolution of the rice cereal particles.

F1
FIGURE 1:
Donor human milk and formula thickened with carob bean gum-based thickeners (GelMix), xanthan gum-based thickeners (SimplyThick and Thicken Up Clear), and starch-based thickeners (Thick It) using manufacturer-recommended quantities to achieve a target of nectar-thick viscosities. (A) Carob bean GBT thickened both formula and DHM and achieved the target nectar-thick viscosity by following the manufacturer recipe. (B and C) Xanthan GBT also thickened both formula and DHM, but achieved much higher viscosities than carob bean GBT. Thicken Up Clear notably achieved viscosities 4-fold higher than carob bean GBT, and is in the range of honey-thick. (D) SBT briefly thickened DHM at 5 minutes, but the solution quickly thinned out by 15 minutes. On the contrary, formula is effectivelythickened by SBT and the viscosity continued to rise over time. DHM = donor human milk; GBT = gum-based thickeners; SBT = starch-based thickeners.

All GBTs successfully thickened both HM and formula. The 2 xanthan GBTs, ST-GBT and TUC-GBT, achieved notably higher viscosities than the carob GBT, GM-GBT (Fig. 1). Despite following the manufacturer recipes for achieving nectar-thick solutions, both xanthan GBT thickened HM to honey-thick viscosities.

The addition of HMF was evaluated for its potential effects on how thickeners interact with HM. Adding HMF to HM did not cause statistically significant changes to the viscosities of the milk when thickened by GM-GBT, ST-GBT, and TUC-GBT (Fig. 2 and Supplemental Table 3, Supplemental Digital Content, https://links.lww.com/MPG/B704). Adding HMF to HM causes TI-SBT to achieve higher starting viscosity at 0 minutes, but the viscosity quickly drops and does not differ from the viscosity of unfortified HM thickened with TI-SBT.

F2
FIGURE 2:
Donor human milk with added human milk fortifiers, and autoclaved donor human milk, thickened with carob bean gum-based thickeners (GelMix), xanthan gum-based thickeners (SimplyThick and Thicken Up Clear), and starch-based thickeners (Thick It) using manufacturer-recommended quantities to achieve a target of nectar-thick viscosities. P-values are listed in Supplemental Table 3 (https://links.lww.com/MPG/B704). (A and B) The addition of HMF to DHM and autoclaving DHM did not cause statistically significant changes to the viscosities of the milk when thickened by GelMix and SimplyThick. At 1 hour, Thicken Up Clear achieved significantly higher viscosities in the autoclaved DHM group than in the DHM group. (D) Adding HMF to DHM allows for SBT to achieve higher viscosity at 5 minutes, but the viscosity quickly drops by 15 minutes. Autoclaving DHM at 120°C for 15 minutes allows for the human milk to be successfully thickened by SBT. DHM = donor human milk; GBT = gum-based thickeners; HMF = human milk fortifiers; SBT = starch-based thickeners.

When HM was autoclaved, TUC-GBT achieved significantly higher viscosities in the autoclaved HM group than in the control HM group. Autoclaving HM enabled TI-SBT (Fig. 2 ) to thicken human milk.

Formula With “Added Rice”

Per the formula manufacturers, the formula has “normal” thin consistency before feeding, and upon ingestion into the warmer, acidic environment of the stomach, the starches will “activate” and increase the viscosity of the feed, thus reducing GERD symptoms. After adding 10 mL 0.1 mol/L HCl to 120 mL SSU, SSU ready-to-feed (RTF) and EAR, the resulting viscosities were 495, 323, and 435 mPa seconds, respectively. The addition of HM in a 1 : 1 ratio to formula with “added rice” renders the solution thin and incapable of thickening even in an acidic environment (Table 1).

T1
TABLE 1:
Formula with “added rice”

Gum-based Thickeners Over 2 Hours

The viscosities of xanthan GBT-thickened solutions continued to rise at the 60 minutes. The viscosities of HM thickened with GBT were further evaluated over a longer duration, up to 120 minutes. All GBT appears to plateau in viscosity prior to 120 minutes.

When HM is thickened per manufacturer recommendations to target honey-thick viscosities, the resulting viscosities were in the spoon-thick range for all GBT by 15 minutes. Doubling the quantity of GBT thickeners GM-GBT, ST-GBT, and TUC-GBT, resulted in 25×, 3.9×, and 3.1× increase in viscosity, respectively. This demonstrates that all GBT do not have a linear relationship between the quantity of thickener added, and the resulting viscosity (Fig. 3).

F3
FIGURE 3:
Donor human milk thickened with gum-based thickeners, followed over 2 hours. All GBT appears to plateau in viscosity before 120 minutes. DHM thickened per manufacturer recommendations to nectar-thick viscosity (light shade) attained honey-thick viscosity with xanthan gum-based thickeners. When DHM was thickened to honey-thick viscosity (medium shade), the resulting viscosities were in the spoon-thick range (dark shade) for all GBT by 15 minutes. Doubling the quantity of GBT thickeners GelMix, SimplyThick, and Thicken Up Clear, resulted in 25×, 3.9×, and 3.1× increase in viscosity, respectively. This demonstrates that all GBT do not have a linear relationship between the quantity of thickener added, and the resulting viscosity. DHM = donor human milk; GBT = gum-based thickeners.

Addition of Acid to Gum-based thickeners-thickened Solution

Adding acid (0.1 mol/L HCl) after HM has been thickened by xanthan GBT and has reached peak viscosity, leads to precipitation at a pH of 5 (Fig. 4). This phenomenon is not seen with carob bean GBT. There is less precipitation with formula. HM was also first acidified with 10 mL 0.1 mol/L HCl to a pH of 5 before the addition of GBT, and as per manufacturer claims, preacidifying HM did not affect the final viscosities (data not shown).

F4
FIGURE 4:
Adding 0.1mol/L HCl to donor human milk thickened with gum-based thickeners after 2 hours. DHM thickened with xanthan GBT precipitate when acid is added ([A] SimplyThick, [B] Thicken Up Clear). (C) Formula thickened with xanthan GBT initially forms smaller amounts of precipitant, but stirring and agitation of the solution dissolves it. DHM = donor human milk; GBT = gum-based thickeners.

DISCUSSION

Thickening feeds remains an active strategy in the management of GERD and aspiration risk in infants. There are numerous strategies that are used currently to thicken infant feeding but without an objective measure to compare the resulting viscosities. We conducted this study to fill this comparison gap and characterize the challenges that are present in the process of thickening infant feeding. Thickening of human milk still remains complicated and unresolved. We demonstrated that SBT rapidly loses its thickening effect on human milk. Inactivating digestive enzyme activity based on published heat inactivation data required autoclave temperatures (120oC for 15 min), which unfavorably alter many of the macronutrients, micronutrients, and bioactive factors present in human milk (18). The treatment required is far higher than typical donor human milk pasteurization temperatures (62.5oC for 30 min). Some practices have used thickening formulas that are marketed for reflux (with added starches) and add this to human milk with hopes of thickening it, but we demonstrated that an equal mixture of added-starch formula and DHM results in a thin liquid, even in the presence of acid.

On the other hand, all forms of GBT tested provided immediate thickening of both HM and formula. However, the dynamic nature of the thickening of HM with GBT was surprising. There were time-dependent increases in viscosity. Both xanthan GBT tested (ST-GBT and TUC-GBT) notably achieved higher viscosities than the manufacturers had intended based on their recommended recipes. The most striking finding was the precipitation of the thickened solution when acid was added to HM thickened with xanthan GBT (Fig. 4). The tendency to precipitate could result in slower intestinal transit that may lead to severe ileus or small-bowel obstruction, and possibly necrotizing enterocolitis. Interestingly, carob bean GBT did not exhibit this precipitation in the presence of acid.

Xanthan GBT can readily thicken human milk, but there have been several reports using ST-GBT in preterm infants later in their hospital stay that, after discharge, resulted in severe complications including necrotizing enterocolitis and death (8,9). This resulted in an Food and Drug Administration (FDA) warning to avoid using this product in infants born preterm, less than 37 weeks’ gestation (19).

Concern for the cause of bowel morbidities from increased viscosity in infant feeding has been longstanding in growing preterm infants (20). The increased risk of NEC with ST-GBT is not fully known. It may be because of increased intraluminal water, sugars, hydrogen, short-chain fatty acid, and bile acid resulting from bacterial fermentation. In vitro experiments showed that xanthan GBT activates gut lymphocytes and macrophages to trigger excessive inflammatory cascade including an increase in TLR4, TNFa, and IL-12 (21). Other in vitro studies suggest the opposite effect, that xanthan GBT may be a positive immunomodulator on macrophages (22,23). The possibility of bacterial contaminants, such as toxins in the powder during manufacturing cannot be ruled out fully either. Furthermore, it is not clear at this point if the bowel complications seen with 1 form of xanthan GBT crosses over to other xanthan GBT products. All similar GBT share similar characteristics of adsorbing water, being fermentable by bacteria, and subject to manufacturing caution because of the nonsterilizable nature of the powder. Carob bean GMT have also been reported in association with NEC but this was introduced very early in the diet (days of life 12 and 24) (24). Caution in the use of any GBT in preterm infants is prudent (25).

A recent publication identified significant osmolality shifts in feeds when adjusted with thickening agents (26). Infant feeds are typically aimed to be under 450 mOsm/kg to avoid excessive intraluminal osmolar load that could affect intestinal health, microbiome, and stool consistency. Contrary to the study conducted by Levy et al, our study focuses on the physical viscosity of feeds given the assumption that thickened feeds prevent reflux via its increased weight in the stomach (11). This study demonstrates the physical burden of excessively thick feeds rather than the role of osmolality. Preterm infants are frequently fed via feeding tube, and the slow introduction of feeds through narrow tubes can lend itself to clogging issues. For infants who are bottle fed, excessively thickened feeds can create difficulty in extracting feeds from a nipple. Moreover, our study demonstrated a continuous rise in viscosity over 2 hours after the thickening agents are added to feeds. This raises the concerns of obstruction not only in the feeding tubes and apparatus but also the risk of intestinal obstruction.

Moving forward, more studies will need to be done to determine why human milk thickened by xanthan GBT precipitates in the presence of acid. Xanthan gum thickens solutions by forming a pseudoplastic matrix, and whey-proteins interact with xanthan gum in protein-polysaccharide interactions that are affected by pH and presence of ions (27). One theory for explaining this observed phenomenon is that the addition of acid disrupts the pseudoplastic matrix, causing a phase separation. The use of carob bean GBT may be preferable as the characteristics appear more stable with infant feeding tested (28). Wenzl et al (10) demonstrated that human milk can be effectively thickened with carob bean GBT to reduce the symptoms of GERD, but the safety of using carob bean GBT in the neonatal population, particularly preterm infants, has not been fully explored.

Practical use of thickeners varies widely between centers (29). To date, there are no specific recommendations on what viscosities need to be achieved in formula before it is deemed effective in reducing GERD symptoms. We observed that formula specifically marketed for reflux achieves viscosities in the honey-thick range in an acidic environment, suggesting that nectar-thick solutions may not be sufficient in reducing symptoms of reflux. Moreover, the thickening potential of these formulas will vary by individual differences in gastric pH as well as concurrent treatment with proton-pump inhibitors or H2-blockers.

We also noted that there is large potential for significant variability in the amount of thickener added to the solutions if users follow manufacturer recommendations without weighing out the content with a scale. For example, 4 teaspoons (∼8 g) of TI-SBT in 120 mL should yield nectar-thick viscosities, but the measurement tools provided with the product can yield up to 10% variability in the actual mass of product used. Moreover, all GBTs did not exhibit a linear relationship between the quantities of thickener added and the resulting viscosities. Using twice the amount of thickener resulted in much greater than 2-fold increases in viscosity. Not only is the viscosity of fluid difficult to predict based on the quantity of thickener added, operator differences, such as degree of agitation of solutions can significantly alter the final viscosity of the solution. Although this was controlled for in this experiment by limiting the number of people performing the experiment, using consistent mixing tools, and timing the duration of mixing, in clinical settings milk viscosities have been reported to be vastly different depending on the person preparing the milk.

Also not clearly understood is the effect of thickening agents on the nutritional properties of infant feeds. It is unclear what the final caloric and macronutrient shifts on the feeds are, and there is a lack of data demonstrating the effects of thickened feeds on infant growth and body composition.

There continues to be serious reservations regarding the use of GBT in infants given the recent cases of necrotizing enterocolitis associated with xanthan GBT. Carob bean GBT is a promising option for thickening human milk, but additional studies need to be done to define its safety profile in the preterm infant population. Thickened feeds should be treated as a dose-dependent therapy, as infants likely respond to different viscosities to reduce GERD symptoms. Moreover, thickened feeds should be limited to a defined “therapeutic window” to minimize the possibility of complications secondary to excessively thickened feeds.

Thickening feeds to address GERD is considered to be a conservative approach when compared with pharmacologic and surgical interventions. However, there may be additional clinical implications that are not yet understood. Thickeners add to infant feeding a host of complex metabolites, increase protein and carbohydrate content, increase caloric density, potentially alter the microbiome, and make significant changes the physical properties of milk. Moreover, use of additives to human milk precludes an infant from being able to directly breast-feed. This study allows us to conclude that thickeners make significant changes to the properties of milk. Objectively quantifying the viscosity properties with thickeners in practical scenarios is an early step to improving our application of thickeners in infants. Caution must be taken when using thickening agents to treat reflux, as thickened feeds may not be such a safe and conservative regimen after all.

REFERENCES

1. Poets CF. Gastroesophageal reflux: a critical review of its role in preterm infants. Pediatrics 2004; 113:e128–e132.
2. Hyman PE, Milla PJ, Benninga MA, et al. Childhood functional gastrointestinal disorders: neonate/toddler. Gastroenterology 2006; 130:1519–1526.
3. Corvaglia L, Martini S, Aceti A, et al. Nonpharmacological management of gastroesophageal reflux in preterm infants. Biomed Res Int 2013; 2013:141967.
4. Horvath A, Dziechciarz P, Szajewska H. The effect of thickened-feed interventions on gastroesophageal reflux in infants: systematic review and meta-analysis of randomized, controlled trials. Pediatrics 2008; 122:e1268–e1277.
5. Garcia JM, Chambers E 4th, Matta Z, et al. Serving temperature viscosity measurements of nectar- and honey-thick liquids. Dysphagia 2008; 23:65–75.
6. Almeida MB, Almeida JA, Moreira ME, et al. Adequacy of human milk viscosity to respond to infants with dysphagia: experimental study. J Appl Oral Sci 2011; 19:554–559.
7. Erick M. Breast milk is conditionally perfect. Med Hypotheses 2018; 111:82–89.
8. Beal J, Silverman B, Bellant J, et al. Late onset necrotizing enterocolitis in infants following use of a xanthan gum-containing thickening agent. J Pediatr 2012; 161:354–356.
9. Woods CW, Oliver T, Lewis K, et al. Development of necrotizing enterocolitis in premature infants receiving thickened feeds using SimplyThick((R)). J Perinatol 2012; 32:150–152.
10. Wenzl TG, Schneider S, Scheele F, et al. Effects of thickened feeding on gastroesophageal reflux in infants: a placebo-controlled crossover study using intraluminal impedance. Pediatrics 2003; 111 (4 Pt 1):e355–e359.
11. Orenstein SR, Magill HL, Brooks P. Thickening of infant feedings for therapy of gastroesophageal reflux. J Pediatr 1987; 110:181–186.
12. Abrams SA. Be cautious in using thickening agents for preemies. AAP News. Chicago, IL: AAP; 2011.
13. Forte JG, Zhu L. Apical recycling of the gastric parietal cell H,K-ATPase. Annu Rev Physiol 2010; 72:273–296.
14. Yao X, Forte JG. Cell biology of acid secretion by the parietal cell. Annu Rev Physiol 2003; 65:103–131.
15. Mason S. Some aspects of gastric function in the newborn. Arch Dis Child 1962; 37:387–391.
16. Harada T, Hyman PE, Everett S, et al. Meal-stimulated gastric acid secretion in infants. J Pediatr 1984; 104:534–538.
17. Widstrom AM, Christensson K, Ransjo-Arvidson AB, et al. Gastric aspirates of newborn infants: pH, volume and levels of gastrin- and somatostatin-like immunoreactivity. Acta Paediatr Scand 1988; 77:502–508.
18. Griffiths MW. Use of millk enzymes as indices of heat treatment. J Food Prot 1986; 49:696–705.
19. FDA. FDA expands caution about SimplyThick. Available at: https://wayback.archive-it.org/7993/20171101144041/https://www.fda.gov/ForConsumers/ConsumerUpdates/ucm256250.htm. Accessed August 23, 2019.
20. Quinlan PT, Lockton S, Irwin J, et al. The relationship between stool hardness and stool composition in breast- and formula-fed infants. J Pediatr Gastroenterol Nutr 1995; 20:81–90.
21. Ishizaka S, Sugawara I, Hasuma T, et al. Immune responses to xanthan gum. I. The characteristics of lymphocyte activation by xanthan gum. Eur J Immunol 1983; 13:225–231.
22. Takeuchi A, Kamiryou Y, Yamada H, et al. Oral administration of xanthan gum enhances antitumor activity through Toll-like receptor 4. Int Immunopharmacol 2009; 9:1562–1567.
23. Liu F, Zhang X, Ling P, et al. Immunomodulatory effects of xanthan gum in LPS-stimulated RAW 264.7 macrophages. Carbohydr Polym 2017; 169:65–74.
24. Clarke P, Robinson MJ. Thickening milk feeds may cause necrotising enterocolitis. Arch Dis Child Fetal Neonatal Ed 2004; 89:F280.
25. Elhassan NO. Similarity of in vitro activity of different gum thickeners on the gastrointestinal tract. J Pediatr 2015; 166:1096.
26. Levy DS, Osborn E, Hasenstab KA, et al. The effect of additives for reflux or dysphagia management on osmolality in ready-to-feed preterm formula: practice implications. JPEN J Parenter Enteral Nutr 2019; 43:290–297.
27. Bertrand ME, Turgeon SL. Improved gelling properties of whey protein isolate by addition of xanthan gum. Food Hydrocolloids 2007; 21:159–166.
28. Meunier L, Garthoff JA, Schaafsma A, et al. Locust bean gum safety in neonates and young infants: an integrated review of the toxicological database and clinical evidence. Regul Toxicol Pharmacol 2014; 70:155–169.
29. Madhoun LL, Siler-Wurst KK, Sitaram S, et al. Feed-thickening practices in NICUs in the current era: variability in prescription and implementation patterns. J Neonatal Nurs 2015; 21:255–262.
Keywords:

carob bean; gastroesophageal reflux; human milk; xanthan gum

Copyright © 2019 by European Society for Pediatric Gastroenterology, Hepatology, and Nutrition and North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition