Abstract

Background and objective: Regurgitation symptoms in many GERD infants might affect their nutrition status. This review compares thickeners and hydrolysis formula with cow’s milk in the evolution of regurgitation, weight and stool consistency changes in infants with GERD based on currently available evidence, as the benefits of using AR-F in infants with GERD is still contradictory. There is also evidence that hydrolysed formula plays a crucial role in regulating reflux. Methods: The search terms ‘thickener formula/pre-thickener’ OR ‘cow’s milk allergy/hydrolysis formula’ AND ‘gastro-oesophageal reflux disease (GERD)/regurgitations’ were used to search the electronic databases of NUsearch, Google Scholar, ScienceDirect and PubMed for randomised controlled trials (RCTs) published between 2005 and 2020 for thickened formula and 2010-2020 for hydrolysed formula that involved infants (<13 months) of both sexes who were diagnosed or at risk of regurgitation. Results: Seven eligible RCTs met the inclusion criteria. Thickened whey protein formula significantly reduced daily regurgitations by up to 7.7 episodes, while the results were insignificant in thickened or normal extensive hydrolysed casein milk (-4.2 vs -3, respectively). However, rice thickener significantly reduced regurgitation and weight gain by 1.261 kg in two months. An insignificant difference in stool consistency was found in most trials (P >0.05), except for the thickened casein formula as 64% of participants had normal stool density (P= 0.45). Conclusion: This study identified that using either thickening formula or extensively hydrolysed formula significantly reduced the number of regurgitations per day (mean range: -1.2 to -7.1; P<0.05). It does, however, seem that hydrolysed whey formula thickened with hydrolysed rice can support weight gain in failure to thrive infants.

1. Introduction

Gastroesophageal reflux (GER) is defined as the normal phenomenon of passing gastric contents from the stomach to the oesophagus ¹. However, if this reflex does not disappear after the first year and affects the intake of infants or causes health problems, it is classified as gastroesophageal reflux disease (GERD) ². GERD's reported prevalence in patients of all ages worldwide is increasing, but it is nevertheless far less common than GER ³. Population-based studies suggest reflux disorders are not as common in Eastern Asia, where the prevalence is 8.5%, compared with Western Europe and North America (10% to 20%) ^{4, 5}. Some patients are highly likely to have GERD, such as those with neurologic impairment, obesity, congenital oesophageal disease, cystic fibrosis and some genetic diseases, especially down syndrome ². This is because of the potential weakness of the oesophageal sphincter ⁶ or highly contracting abdominal muscles ⁷. However, while infants with frequent episodes of emesis do not necessarily suffer from GERD, they may have an allergy to cow’s milk or have been overfed ². Also, half of prematurely born infants have reported at least one regurgitation per day ^{2, 8}.

Symptoms in infants with GERD can include vomiting or regurgitation, feeding refusal, poor weight gain, dysphagia, painful swallowing and arching of the back during feeding ⁹. GERD symptoms can also be extraoesophageal; for example, in the respiratory system resulting in coughing and wheezing ¹⁰ or in the mouth cavity resulting in dental erosion pharyngitis, sinusitis and recurrent otitis media. Extensive and prolonged regurgitation can lead to insufficient nutrient intake, failure to thrive and an increased risk of health problems such as esophagitis, peptic stricture, Barrett oesophagus and adenocarcinoma ^{11, 12, 13}. Therefore, it was vital to search for the most appropriate and non-invasive management tools to prevent the possibility of these health complications.

GERD management: Rosen et al. ² argue that infants with apparent GERD symptoms can be treated with both pharmacological and non-pharmacological intervention. Non-pharmacological treatment involves a combination of modifying positions during and after feeding, feeding volume, frequency and type of milk ⁸. In contrast, there is a lack of evidence for GERD drug treatment ⁸ as there is a lack of diagnostic tests ^{2, 14}. Thus, non-pharmacological strategies are still the first-line therapy to manage GERD ¹⁵ as it is highly possible to reduce the use of unnecessary medications and save more money through utilising anti- regurgitation feeding (AR-F) ¹⁶.

The impact of milk on GERD infants: The fundamental mechanism in GERD feeding intervention is to thicken the meal to restrain stomach content emission after a feed ¹⁶. Therefore, thickened formula has been the subject of research since the 1980s ^{17, 18}. There is also another mechanism theory to reduce controlled emesis episodes, which is fasting the empty gastric period ¹⁹. However, each technique has positive and negative sides. The AR-F is higher in calorie than the standard formula as one tablespoon of rice cereal per ounce increases the energy density by 170% ⁸. Hydrolysed formula has similar calories to cow’s milk formula, but the protein is degrading ²⁰. Although this is more suitable when infants present with failure to thrive ²¹, restricting feed volumes and persistence may be used to manage undesirable weight gain (ibid). AR-F might decrease the absorption of nutrients, such as fat and carbohydrates and the bioavailability of iron, calcium, copper and zinc is also reduced ^{21, 22}. Some indigestible thickener agents, such as carob bean gum, can lead to diarrhoea and allergy, while rice-thickened feeds may increase coughing ²¹.

Generally, the evidence regarding the benefits of AR-F is still contradictory ²³ but there is evidence that hydrolysed formula plays a crucial role in regulating reflux in allergy patients ²⁴. This review aims to compare the use of thickeners and hydrolysis formula with regular milk in the evolution of regurgitation, weight and stool consistency changes in infants with GERD based on currently available evidence. However, one of the most significant challenges is the significant difference in GERD classification with several regurgitations per day and the variety of reflux aetiology. Additionally, it was not easy to compare the variety of thickener agents with the concentration of the formula.

Rationale: According to the latest updates from the North American Society for Paediatric Gastroenterology, Hepatology and Nutrition (NASPGHAN) and the European Society for Paediatric Gastroenterology, Hepatology and Nutrition (ESPGHAN) guidelines (2018), thickening the feed is the first-line approach to treat GERD in infants and young children but the evidence is weak ². However, the National Institute for Health and Care Excellence ²⁵ recommend a hydrolysis formula if the patient regurgitates or vomits without knowing the cause or has been diagnosed with a cow’s milk allergy.

Many reviews studied the effect of thickening/pre-thickening feeding in regurgitation ^{26, 27}. Huang et al. (2002) did not find inference; therefore, their research did not meet the requirement. Kwok et al. ²⁷ produced moderate evidence that feed thickeners must be considered in infants with GERD to reduce regurgitation by two episodes per day. However, they were unable to assess the superior AR-F for the preterm infant with gastric emission. More recently, Duncan et al. ²⁸ evaluated a variety of thickener risks and benefits in paediatric patients with GERD. They concluded that AR-F has some side effects but with continuous follow-up, patients will tolerate the formula and symptoms will improve sufficiently. Regarding hydrolysed formula, only one systemic review has compared various types of protein and grades of hydrolysation in gastric emptying and regurgitation in children ²⁴. This review is different from other reviews in that it analysed and compared both thickening and hydrolysed protein to find out which one was more superior in treating regurgitation.

2. Research Methodology

Moher et al. ²⁹ recommended using the PRISMA flow chart to point out evidence-based search guidelines in the permeated context of a systematic review meta-analysis. Therefore, the official PRISMA flow chart was used in this systematic review to enable reliable, focused and transparent reporting, interpretation and synthesis.

Review question: This review compares the evolution of regurgitation, weight and stool consistency changes when feeding thickener or hydrolysis formula to infants. The review uses currently available RCT studies as evidence.

Search strategy: Universal and exceptional systemic reviews were conducted on NUsearch, Google Scholar, PubMed and ScienceDirect using the search terms ‘thickener formula/pre-thickener’ OR ‘cow’s milk allergy/hydrolysis formula’ AND ‘gastroesophageal reflux disease (GERD)/regurgitations.’ Publication dates were restricted to 2005-2020 for thickener formula and 2010-2020 for hydrolysed formula. The difference in years results from not having enough available research to conduct a fair comparison between the two formulas. The results were filtered using open access to English language articles. Furthermore, references in other studies were reviewed to identify any additional studies. The last search on thickener formula was published by January 2008 and April 2013 for hydrolysis formula.

Inclusion criteria and study selection: The primary criterion for searching the selected database was a randomised control trial involving infants of both sexes who were less than one year old. They mainly assessed the evolution of regurgitations as well as collecting weight and stool consistency changes within the study analyses. The publication dates and titles of studies were double-checked to ensure they met the research criteria. Subsequently, the abstract sections were screened against inclusion and exclusion criteria. Finally, the full-text versions of papers that met the inclusion criteria were reviewed. Appendix 1 shows the study's acceptance criteria in more detail.

Quality and evidence-level assessment: The Cochrane risk bias tool was used to evaluate the quality of eligible RCTs ³⁰. This tool assessed the eligible trials for sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data, selective outcome reporting and other sources of bias. The evaluation was conducted twice for each study. The risk of bias graph envisaged the risk bias quality of the eligible trials into three levels: high, low and unclear risk of bias.

Data analysis and subgroup analysis: The studies were selected depending on the amount of extract data. The summary table used to collect the outcome data from the source was observed using therapeutic milk (thickened or hydrolysing formula) in infants with GERD or regurgitation. The evolution of daily regurgitations was measured. Secondary data gathered three possible outcomes: (1) the gathering weight evolution date; (2) the mean percentage of participants with formed to soft stools (reasonable constancy). Standard deviation (±SD) was difficult to conduct in the summary table, as most of the evolution dates were manually calculated. All trials were compared by ANCOVA test to the mean of the outcomes between the groups, then the result was compared with the confidence interval (95%).

3. Results

Search results: In the first stage of PRISMA, the result was eight-seven. The titles and publication dates were double-checked to ensure they met the initial eligibility criteria (n=36). Subsequently, the abstracts were screened to check if they were relevant to the research questions. A total of 16 available articles were identified and after careful review, four RCTs for thickening formula and three RCTs for hydrolysed formula were used. Figure 1 shows the excluded studies in the PRISMA flow diagram in various stages.

Quality assessment of eligible trials: Figure 2 shows the result of the Cochrane risk bias tool in the eight eligible studies. Four trials were classified as having a low risk of bias, while two were classified as having an unclear bias and one was classified high risk. The rationale underpinning the risk of the bias assessment process is reported in Appendix 2.

Included trials: Trials were conducted in America ³¹, Taiwan ³² and Japan ³³. Vandenplas et al. ³⁵ conducted the only trial across countries (Belgium, Greece, Kuwait, Lebanon, and Slovenia). The remaining trials were carried out in France ³⁴ and Italy ^{36, 37}.

Population characteristics: A total of 541 infants participated in the included RCTs and the sample sizes ranged from 20 to 115. The study duration was from two to eight weeks and the mean age of the participants was 2.6 months. Four RCTs had equal gender distribution except for Vandenplas et al. ³⁴, who had significant variations concerning the number of boys and girls (70 and 45, respectively). Three trials did not present the gender ratio, but the differences were insignificant ^{31, 32, 37}. Additionally, the baseline weight for infants between groups had nonsignificant differences (P<0.05). More information about the study characteristics is provided in Table 1.

Five trials included a thickening formula that was given to full-term infants presenting with GERD symptoms or abnormal oesophageal pH examination (n = 397). Two studies involved patients diagnosed or at risk of cow’s milk allergy who had excessive regurgitation (n = 144) ^{35, 37}. It was unclear in two studies if premature infants were included ^{32, 35}, while two trials were ambiguous about whether breast-fed babies were included or excluded ^{31, 33}. Generally, infants with complicated regurgitations (e.g., haematemesis, melena), other conditions associated with vomiting or previous treatment with therapeutic milk were excluded. In some trials, the small-for-gestational-age at birth, GERD patients with severe complications or those using antireflux medication were excluded. For more details, see Table 2.

Depending on the records from parents, the criteria of the frequency of regurgitation or vomiting symptoms were as follows.

• ≥ 3 episodes of regurgitation and/or vomiting per day: Chao and Vandenplas ³², Miyazawa et al. ³³.

• ≥ 5 episodes per day and lasting for >1 week: Vandenplas et al. ³⁴.

• ≥ 5 episodes of regurgitation and/or vomiting per day: Vandenplas et al. ³⁵.

• ≥ 6 episodes per day: Ostrom et al. ³¹.

• Symptoms defined according to the Questionnaire on Paediatric Gastrointestinal Symptoms—Rome III Version [QPGS-RIII], with a positive symptom score (≥ 7) or without specifying a score in their including criteria: Indrio, et al., ³⁷, Ummarino et al. ³⁶.

Intervention: Table 3 summarises the interventions, which converged into three categories.

1) Thickening formula versus standard formula groups (n = 153 vs. 149)

- Six grams of soy-fibre agent per 100 ml in soymilk as a feed thickener, ³¹.

- Pre-prepared AR-F, regular milk mixed with rice-cereal by one scoop for every two scoops of powder formula, ³².

- Non-hydrolysed thickening formula with 0.35 g/100 mL locust bean gum agent ³³.

- Non-hydrolysed thickening formula with rice starch 14.3 g/100 ml of milk (<6 months) and 14.2 g/100 mL (>6 months) plus keeping the patients in a supine position during sleep ³⁶.

2) Thickening hydrolysate protein formula compared to AR-F (n=58 vs. 57)

- Thickened whey hydrolysate formula (HWF) with 0.66 grams starch per 100ml along with 1.15 g/100 ml of locust bean gum ³⁴.

3) Thickening extensive hydrolysed protein formula vs. non-thickening extensive hydrolysed protein formula (n = 71 vs. 69):

- Thickening extensively casein-based hydrolysate formula (EHCF) with pectin fibres (1.2 g/100 ml) and 0.3 gram per 100ml of the starch agent ³⁵.

- Thickening extensive hydrolysate protein milk with starch, but the amount or type of starch was not stated ³⁷.

Control: Standard formula without feed thickeners was used as the control in trials that used extensive hydrolysates protein formula or soy fibre, rice cereals and carob bean gum as feed thickener. Two trials used a similar formula in the intervention group, but Vandenplas et al. ³⁴ utilised a nonhydrolyzed form of the thickening formula, while Vandenplas et al. ³⁵ did not apply any thickener agent in the EHCF.

Co-intervention: Some studied engaged in lifestyle modification where the infant was kept in a supine position while asleep, a straightforward postprandial position for 90 minutes and then fed small and frequent meal in all groups ³⁶. One study used lifestyle modification in the control group ³², while peri-positioning feedings in infants were at the parents’ discretion ³¹.

Primary outcome: A total of 541 participants across seven RCTs measured the evolution of regurgitations caused by the particular thickening and/or extensively hydrolysed formula during experiments conducted over two to eight weeks. As Table 3 shows, six trials reported the mean episodes of regurgitations or vomiting per day ^{31, 32, 33, 34, 35, 37} whereas the remaining trial reported the proportion of infants without regurgitation or vomiting at the end of the intervention ³⁶.

Three of the RCTs studied the effect of thickening hydrolyse protein milk on the evolution of gastric emission (one unclear and two low risks of bias, n = 259). Vandenplas et al. ³⁴ and Indrio et al. ³⁷ found a significant decrease in regurgitation in patients fed with thickened HWF with starch and locust bean gum or pectin fibre by -7.7 and -4.8 episodes per day, respectively. However, Vandenplas et al. ³⁴ conducted a pleasing regurgitation reduction in the control group fed with standard formula supplemented with locust bean gum (7.7 vs. 7.1; P<0.013). The non-thickening standard formula used by Indrio et al. ³⁷ showed a little significant improvement by 2.2 regurgitations per day. In the last study on hydrolysed protein feeding, Vandenplas et al. ³⁵ found a nonsignificant difference between groups (P=0.24), whether the EHCF was thickened (with pectin fibre and starch) or not; infants were controlled by three to four gastric emissions per day.

On the other hand, four trials examined the effect of different thickener agents in the evolution of regurgitations in infants with a high GERD risk (one high, one unclear and two low risks of bias, n = 282). In two trials, the standard formula decreased regurgitation comparable to the soy-fibre agent (6 grams per 100 mL; P=0.029) and light locust bean gum (0.35 grams per 100 mL; P= 0.00048) thickening formula ^{31, 33}. The other two studies discovered a significant decrease in daily regurgitation mean episodes if the evaluation position was compounded with thickened formula regardless of the type of the thickener agent used ^{32, 36}. Furthermore, Ummarino et al. ³⁶ discovered that 43.5% of participants recovered from regurgitations within eight weeks.

1) Weight evolution

Six trials explored weight changes (n= 477) during the regurgitation nutrition intervention journey.

Four of trials (two unclear and two low-risk bias trials) observed a nonsignificant difference in weight change between the group that had added thickening in the standard formula or EHCF ^{32, 33, 34, 35}. The remaining studies found that by adding a starch agent to thicken the milk, even hydrolysed or nonhydrolyzed formula can significantly increase weight (P <0.05) ^{32, 37}. However, Chao and Vandenplas ³² had the highest significant weight gain in the six eligible trials at almost 140 grams per 8 weeks, by adding hydrolysed rice cereal to regular milk at a ratio of 1:2 scoops. More details about the results can be found in Table 3.

2) Stool consistency

Three RCTs with a total of 322 participants studied the changes in stool consistency ^{31, 34, 35}. Ostrom et al. ³¹ and Vandenplas et al. ³⁴ observed a nonsignificant alteration between the two groups, by adding six grams of soy-fibre into soymilk or adding 0.66 grams of starch into 100mL of HWF (low and unclear risk of bias trials, respectively; P<0.05). Ostrom et al. ³¹, also discovered a nonsignificant improvement in stool density within the soymilk and regular milk groups (96.6 vs. 91%, respectively; P<0.05). Conversely, a low-risk bias trial monitored a significant increase in participants with formed or soft stools when fed with EHCF (26.5 vs. 8.6 %; P= 0.45) thickened mainly with pectin fibre (1.2 g/100mL) and only 0.3g/100mL of starch ³⁵.

4. Discussion

In 2018, NASPGHAN and ESPGHAN released clinical practice guidelines for paediatric gastroesophageal reflux. It noted that all nonpharmacological intervention in infants with GERD was weak, except for the position modification strategy ². With paradoxical opinions on the benefits of AR-F in recent times ^{23, 26, 28}, they concluded that thickening formula, as empirical treatment for infants and young children, was dependent on the type and consistency of the thickener as well as the follow-up adherence with the dietitian ²⁸.

While our findings regarding hydrolysed protein formula concern the work of Meyer et al. ²⁴, it has a role in regulating reflux in patients allergic to cow’s milk but not in other pathophysiology ²⁴. However, on reviewing Meyer et al. ²⁴, it was observed that 100% whey protein formula emptied the stomach faster than the whole casein formula. Therefore, examining the effectiveness of thickened hydrolysed formula compared with therapeutic singular formula in infants with GERD had three outcomes, regurgitation evolution, weight and changes in stool consistency.

Interestingly, the evolution of regurgitation episodes significantly decreased within groups, regardless of the type of intervention. This result may be explained by the fact that the tips on lifestyle modification might control regurgitation but only for a limited amount of time. This finding was also reported by Gonzalez Ayerbe et al. ¹. Lifestyle modification can be evaluated by a patient’s peri-time feeding (around 60 minutes) and being fed with small and frequent meals ². However, lifestyle modification combined with suitable therapeutic milk may advance the improvement of shrinking reflux. As observed in work by Ummarino et al. ³⁶, the non-pharmacological intervention group had three to seven times regurgitation-free participants than the modified lifestyle group.

Two studies found decreased regurgitation with standard formula versus thickened formula ^{31, 33}. However, the study by Miyazawa et al. ³³ did have a significantly small sample size and short duration, which could have affected the result; furthermore, the concentration of thickener (locust bean gum 0.35 g/ml) was less than that of Vandenplas et al. 34(2013), who used a higher dose (1.5 g/ml).

In contrast, two trials compared daily GERD reflux between the group fed rice-cereal thickened milk (intervention group) and the group fed with standard formula and modification of their body position ^{32, 36}. Regardless of how the cereal was added (prepared or ready to use), regurgitation was significantly depressed compared to conventional milk.

Of the other formula intervention type in this review, only three studies, with a total of 259 infants, that examined hydrolysed protein formula on thickened vs. non-thickened hydrolysed protein milk or regular milk were suitable for inclusion. Vandenplas et al. ³⁵ concluded that thickened EHCF slightly lowered regurgitation episodes than non-thickened feeds. This was supported by Indrio et al. ³⁷, who found that thickened hydrolysed protein formula trial decreased the frequency of regurgitation, so it could assess functional regurgitation in infants. The study by Vandenplas et al. ³⁴ provided the most significant clusters of infants with GERD but the ANOVA test showed that the evolution of regurgitation results was not statistically difference between thickened HWF and the non-thickened version. This data must be interpreted with caution because patients crossed-over the formula after two weeks of feeding. This method of intervention possibly led to bias as infants needed at least two weeks to tolerate the new formula ². Conceptually, this superior regurgitation pattern with HWF or EHCF can demonstrate faster gastric emptying properties ^{38, 39}. Therefore, results seem to indicate that hydrolysed protein formula suppresses reflux better if it was thickened.

The second question in this research was if the infants’ difference in weight was caused by whether the hydrolysed protein formula was thickened or not? One interesting finding is that the HWF used by Vandenplas et al. ³⁴ and Indrio et al. ³⁷ increased the weight gain more than the EHCF used by Vandenplas et al. ³⁵. The main discrepancy could be attributed to the palpability of the formula. This finding supports evidence from previous observations by Miraglia et al. ⁴⁰, which found better palatability of HWF than EHCF (p < 0.05). Surprisingly, the weight gain was significantly higher in thickened HWF than conventional milk (P<0.05). This discrepancy could be attributed to the higher HWF being better absorbed than whole protein in conventional milk. This result seems to be consistent with other research that found that breast milk rich with whey protein was faster in gastric emptying and absorption than regular formula ²⁴.

With the various types of thickener and concentration, bodyweight results were varied. The most prominent finding to emerge from the analysis was that thickening regular milk with hydrolysed rice-cereal showed a significant increase in weight, provided that it was added in high concentration (one scoop for every two scoops of the regular formula) and for a long duration, more than four weeks ³². This outcome is contrary to that of Hegar et al. ⁴¹, who found that weight gain was significantly higher when using bean gum rather than rice cereal with healthy infants. This brings us back to the importance of the concentration of thickener applied in respect to weight gain; Hegar et al. ⁴¹ used only 5 grams of cereal thickener in 100 mL of the standard formula, which was less than the amount used by Chao and Vandenplas ³².

The type of thickener is a factor that may impact weight change. Starch agent adds more calories depending on the amount used ⁴². Therefore, Chao and Vandenplas ³² and Indrio et al. ³⁷ had a significant increase in weight compared with the similar non-thickening formula. The other insoluble fibre agents, such as locust bean gum and pectin fibre, did not significantly enhance weight gain, regardless if it was used in standard or hydrolysed protein formula ^{33, 34, 35}. The result may be explained by the fact that starch-thickener agents can be digested and absorbed in gastrointestinal full-term infants, but insoluble fibre cannot ⁴³. This result agrees with the findings of Salvatore et al. ¹⁶, which supported that enhanced calories do not exist in locust bean gum as it is not absorbed.

Apart from normalised stool consistency, one study reported a significant change in the EHCF group ³⁵ because EHCF is known to cause soft, liquid stools ⁴⁴. The other two studies reported no significant differences in side effects between the control, feed thickener and large thickened hydrolysed casein groups. However, the studies were mainly short term, with follow-up periods of up to four weeks. Hence, small but real differences in adverse effects may not be found without a large-scale RCT or observational study. Besides, some reports excluded infants who had side effects from the analysis.

The methodological quality of the included studies was varied (Figure 2). The included studies depended on parental records of regurgitation symptoms. Despite attempts to blind the parents from the formula type, it was likely that parents might have noted the higher viscosity of the thickened formula and the odour of the hydrolysed protein milk compared to the previous regular milk. It is essential to bear in mind that it could potentially lead to overestimation bias in their responses. However, this weakness of blinding may be unavoidable in such studies. Also, the findings of this review may not apply to infants with some neurological impairment or other medical causes of regurgitation symptoms. Further limitations of this systematic review, which reduces its ability to draw firm conclusions, are that most studies had small sample sizes, varying thickening composition (i.e., varying type, concentration and osmolality), type of hydrolysis, different patient groups (GERD versus cow’s milk allergy with regurgitating) and the variety of comparable groups (standard formula, similar intervention milk but without thickener agent or similar intervention milk but non-hydrolysed protein milk).

5. Conclusion

This systematic review aimed to compare the impact of thickening versus thickening hydrolysis protein formula on the reduction of regurgitation, improved weight gain and change in stool consistency. This study has identified that both thickening formula and extensively hydrolysed formula significantly reduced the number of regurgitations per day (mean range: -1.2 to -7.1; P<0.05). It does, however, seem that thickened HWF better suppresses gastric reflux in comparison with whole thickened nonhydrolyzed protein formula and regular milk. However, this may be affected by the study duration and underlying diagnosis. It was challenging to find the impact of thickened hydrolysis protein formula on regurgitation due to the varying study designs and the thickener agent type and concentration. In general, it seems that adding a high concentration of starch to HWF is suitable for some GERD patients, particularly those suffering from failure to thrive. Only a small number of RCT studies have identified the changes in stool consistency and, therefore, limited conclusions can be drawn from this number of studies.

This research has thrown up many questions in need of further investigation. One of the questions is related to weight gain using starchy whey milk: because of the improvement in the hydrolysed test's pliability, what encourages drinking more milk? Or does the weight gain result from only calories that came from starch? Future research is needed to assess this explanation. Additionally, large randomised controlled trials that included premature infants could provide more definitive evidence.

References

Appendix 1.

Appendix 2.