Protocol for Feeding Intolerance in Preterm Infants

Globally, approximately 15 million premature infants are born every year, with this number increasing progressively. Due to the immature digestive, absorptive, and immunologic functions, preterm infants are particularly susceptible to mucosal inflammation and bacterial overgrowth, which can lead to feeding intolerance (FI). FI frequently occurs in premature infants, especially in those with a gestational age of <32 weeks or a birth weight of <1500 g. FI is defined as the inability to digest enteral feeding, and is characterized by increased gastric residuals, abdominal distension, vomiting, or both. This delays the establishment of full enteral nutrition and extends the duration of parenteral nutrition, thus increasing the risk of infections, prolonging the length of hospital stay, and increasing economic costs At present, there are some prevention and treatment measures for FI including optimization of enteral nutrition, modification of feeding methods, use of probiotics or medicine, and nursing interventions, but these measures are not fully effective. Feeding strategy for FI is an important clinical challenge for neonatologists. Associated with less feeding intolerance, human milk (HM) is recommended by The World Health Organization (WHO) as the first-choice milk for preterm. However, HM is not always available because of the lack of breast milk banks, diseases of mother and geographic factors, and therefore, formula feeding is required.

Currently, the alternative formulas include Preterm Formula (PF), Partially Hydrolyzed Formula (PHF), Extensively Hydrolyzed Formula (EHF) and Amino Acid-Based Formula (AAF), et al. PF is used in preterm infants when human milk is not available. Clinically, PHF, EHF and AAF are commonly used in treatment for moderate to severe cow's milk protein allergy and prevention for patients at high risk for allergy. PF containing intact protein may not be appropriate for infants with FI. Recently, a guideline mentioned that PHF, EHF and AAF could be considered for use in severe feeding intolerance.

Mihatsch et al. reported that EHF improved the feeding tolerance and enabled a more rapid establishment of full enteral feeding compared with standard PF in preterm infants. The use of EHF could reduce acid gastro-esophageal reflux in preterm infants with FI. A recent Cochrane meta-analysis found that existing data did not support conclusions that feeding PHF or EHF affected the risk of FI or necrotizing enterocolitis (NEC), but the data that could be abstracted from published studies for analysis were limited. Raimondi et al. reported that preterm infants with severe feeding intolerance significantly and rapidly reduced the gastric residual volume after AAF introduction. Jang et al. found the fecal calprotectin levels in AAF-fed infants with FI were significantly lower than those in the HM- or PF-fed infants with FI and showed improvement in the symptoms and signs of FI. However, these studies have some defects, such as small sample sizes, and no randomization. Which formula is more suitable for preterm infants with FI? No scientific evidence is available at present.

Based on previous research, investigators design a randomized, prospective, clinical trial of AAF vs EHF in feeding intolerance. investigators hypothesize that AAF can improve feeding intolerance and establish full enteral feeding more rapidly compared with EHF.

The aim of the present study is to investigate whether AAF enables a more rapid establishment of full enteral feeding in preterm infants with FI compared with EHF and help establish improved guidelines and feeding practices.

This study is designed as a single-center, randomized, prospective, blinded clinical trial. A total of 190 preterm infants with gestational age < 32 weeks or birth weight < 1500g with a diagnosis of FI will be included in the study. Participants will be randomized to the AAF group or the EHF group.

All subjects will be recruited in Children's Hospital of Chongqing Medical University (Chongqing, China), a tertiary university hospital that has a 250-bed neonatal unit with an annual admission rate of around 8000 neonates in the past two years.

Termination criteria: discharge in a stable condition with advice.

Exit criteria:

1. Death or discharge before total enteral nutrition.
2. Guardian request to withdraw from the study.
3. Sepsis, NEC, use of ventilatory support, or critical illness is not suitable for continuing to participate in the study.

This is a prospective randomized controlled study. Patients meeting the inclusion and exclusion criteria are randomly divided into two groups, group A (AAF) and group B (EHF). Time to reach full enteral feeding is the primary outcome measure; the same research is unreported at present. According to Raimondi's study and experience in clinical practice, the mean time to full enteral feeding in group A is 23.6±15.6 days, accounting for 30% disparity, the time to full enteral feeding between two groups differ by approximately 7.08 days. Hypothesis tests are two-sided with a significance level of 5%, while the statistical power is set at 80%. After calculation, a sample size of 76 per group is required, accounting for a 20% miss rate. 95 patients per group need to be included, the planned sample size of this study is at least 190 patients in total.

Randomization will be performed in SPSS software package (version 22.0) using a random number generator in a 1:1 ratio. A statistician who is not involved in recruitment and subsequent data analysis will generate the randomization list, and the list will be concealed. After random allocation lists have been generated, the allocation group (AAF or EHF) will be stored in sequentially numbered, sealed, opaque envelopes. These envelopes will be opened by a third-party individual at each study enrolment, after baseline measures have been obtained.

According to the guideline for nutrition support, enteral feeding is started with in the first 24 hours of life. Human milk is encouraged, and preterm formula is fed when human milk is not available due to mother's or family's condition. Patients with FI are temporarily fed with AAF (NeocateⓇ; Nutricia, London, UK) or EHF (AlfareⓇ; Nestle, Netherlands) instead of HM or PF. Nutritional feeds start at 15-20 mL/kg/day, are increased by 20-30 mL/kg/day. The gastric residual volume (GRV) is checked before each feeding. If GRV is ≥ 50% of the previous feeding, then cease enteral feeding for 1 hours and patients are reassessed. If GRV is not improved, complete blood count, C-reactive protein (CRP) and abdominal X-rays are performed and assessed. If these results can rule out NEC and severe infection, patients are fed in the same amounts for 2-3 days. Enteral feeding is increased at the same rate until full enteral feeding (150 mL/kg/day) every 3 hours is reached. Enteral feedings are withheld: in the presence of bloody or biliary gastric residuals, in cases of abnormal abdominal examination, and/or in cases of abnormal abdominal X-ray. As soon as the residuals, the abdominal examination, and/or X-ray return to normal, feeding is resumed with AAF or EHF at the same speed. Once the patients with FI show clinical improvement, AAF or EHF feeding is discontinued, and the previous feeding, either HM or PF is resumed. When full enteral nutrition is insufficient, all patients receive parenteral nutrition solutions, which is gradually decreased with increasing enteral feeding. Other routine treatments for FI are given in the two groups according to the guidelines.

The investigator who is responsible for recruitment will perform all the evaluations and remain blinded to the status of the patients with regard to the intervention throughout the study.

The following variables: age, sex, birth weight, birth length, delivery mode, gestational age, multiple gestation, family history, birth history, history of present condition.

The statistical analysis is performed using the SPSS software package, version 22.0. The measurement data obeying the normal distribution are expressed as x±s, and the t-test is used for comparison between groups. Non-normally distributed continuous variables are expressed as median with interquartile range and compared using the Mann-Whitney U test. The enumeration data are expressed by rate (%) and analyzed by chi-square test or Fisher exact test. P < 0.05 is considered significant statistical differences.

First, persons in charge of recruitment and quality control are identified. Patients who met all inclusion criteria and none of the exclusion criteria are recruited in this study through the electronic medical record system. Then, their legal guardians are informed of the content of this study, sign a written informed consent, and fill in a registration form. The quality controller chooses randomly from guardians of patients over the course of recruitment and asks if they have read and understood the written informed consent form, to ensure that the rights of patients and guardians are safeguarded.

Data management and monitoring will be performed by using ResMan Research Manager (http://www.medresman.org), and all data sets will be password protected. Only the investigators directly involved with the study will have access to the account number and password. Investigators correctly, completely, clearly, and timely record data in the case report forms (CRFs) according to original observation. The CRFs after inspection need to be transmitted to the data administrator of the clinical research in time. Data will be entered into the ResMan by authorized researchers. Data entry will be by double entry, and matching will be conducted after inconsistent data has been reviewed. Original CRFs are stored in numerical order and kept in locked cabinets after the completion of data entry and review.