Abstract
Background
Feeding practices for infants with gastroschisis are often based on institutional protocols, clinician judgement, and usual practice. The purpose of this study was to describe the range of feeding practices used in neonatal intensive care units across Canada.
Methods
We developed an 18-item survey with open-ended questions focused on feeding and nutrition among infants with gastroschisis. Clinicians from neonatal intensive care units that manage infants with gastroschisis were invited to participate. Interviews were completed via virtual meeting or email. Data were analyzed using descriptive statistics.
Results
Semi-structured interviews were conducted with 14 of 17 centers (82% response rate). Participants included neonatologists (12/14 centers) and dieticians (10/14 centers). Pediatric surgeons from 3/14 centers participated at the request of neonatologists. None of the centers reported using a gastroschisis feeding protocol routinely, although two centers noted occasional use of a protocol from another hospital, and another indicated that a gastroschisis feeding protocol is in development. All centers reported that patients with gastroschisis qualify for donor human milk if maternal breast milk is unavailable. Routine use of sham feeding, contrast studies, rectal irrigations, or motility agents was not reported. There was variation between centers regarding feeding advancement, type of formula used if breast milk is unavailable, use of probiotics, and strategies to introduce breastfeeding.
Conclusion
There is variation in some but not all aspects of enteral feeding among infants with gastroschisis at neonatal intensive care units in Canada. The development of institutional feeding protocols and clinical practice guidelines may help standardize practice.
Introduction
Gastroschisis is a congenital condition characterized by a full-thickness defect in the abdominal wall. 1 This causes the intestines and other organs to protrude outside the abdomen. During fetal development, the intestines undergo prolonged exposure to amniotic fluid. This causes inflammation and bowel injury, which leads to poor motility and slow return of bowel function in the postnatal period. Some infants with gastroschisis have complex disease, defined as the presence of volvulus, intestinal necrosis, perforation, and/or atresia. Patients with complex gastroschisis may require multiple surgeries to restore intestinal continuity.2,3
Gastroschisis affects approximately 1 in 2400 newborns in the United States each year. 4 In Canada, the incidence of gastroschisis has ranged from as low as 1 in 4,300 live births in the year 2020 to as high as 1 in 2,400 in the year 2007. 5 This equates to more than 1,800 cases in the last 15 years, surpassing the number of infants born with congenital diaphragmatic hernia and many other congenital anomalies. 5 Survival rates in developed countries exceed 90%, but many of these patients experience prolonged stays in hospital and other morbidities. 6
Infants with gastroschisis are supported with parenteral nutrition following delivery and abdominal closure. Introducing and advancing enteral feeds can be challenging, as many of these patients struggle with feeding intolerance and poor intestinal motility. Prolonged dependence on parenteral nutrition increases the risk of cholestasis and central-line associated bloodstream infection. Early introduction of enteral feeds appears to be associated with a shorter duration of parenteral nutrition, decreased time to achieve enteral autonomy, shorter length of stay, and reduced risk of infectious complications.7,8 A growing body of evidence also suggests that exclusive breast milk is associated with better patient outcomes, including decreased duration on parenteral nutrition, time to full enteral feeds, and length of stay in hospital.9–11
Despite emerging evidence in favor of exclusive breast milk and early enteral nutrition, there appears to be variation in enteral feeding practices during the neonatal period. 12 A survey of neonatal clinicians in Australia and New Zealand revealed significant variation in healthcare providers’ opinions regarding the initiation and advancement of enteral feeds. Many of these differences appeared to stem from varying definitions of feeding intolerance. This prompted the authors to develop an institutional feeding protocol based on input from bedside clinicians and the best available literature. 12 This underscores the importance of surveying practitioners to better understand practice patterns and identify areas of uncertainty. In addition, a systematic review by Raduma et al. found that implementing a standardized feeding protocol for infants with gastroschisis improves consistency in practice, reduces time to first enteral feeds, and decreases complications. 13
Many children’s hospitals have developed institutional pathways to facilitate postnatal management. The Quality and Safety Committee from the American Pediatric Surgical Association has collected and shared many of these documents as part of a “toolkit” for the management of gastroschisis. 14 These documents serve as an excellent foundation for other centers to develop institutional pathways, but they do not represent a consensus statement on best practice.
The Outcomes and Evidence Based Practice Committee from American Pediatric Surgical Association recently published clinical practice guidelines for infants with uncomplicated gastroschisis. This consisted of two systematic reviews focused on the use of feeding protocols and postnatal management strategies. 15 The authors concluded that institutional feeding protocols and care pathways may standardize feeding and decrease the duration of parenteral nutrition. They also noted that advancing feeds 12–30 mL/kg every 24–48 hours may promote feeding tolerance and reduce the risk of necrotizing enterocolitis. Implementation of feeding protocols also appears to decrease the time to full enteral feeds among patients who undergo delayed abdominal closure. Recommendations related to sham feeding, probiotics, maternal breast milk, donor human milk, introduction of breastfeeding, motility agents, and other feeding practices were not addressed.
Another recent initiative was the development of management guidelines by the European Reference Network for Inherited Congenital Anomalies. 16 This group conducted a systematic review of studies published from 2010 to 2022. The areas of focus included prenatal care, abdominal closure, feeding, and organization of care. As part of this review, the authors recommended starting enteral feeds within 14 days for patients with uncomplicated gastroschisis and to consider starting feeds before 7 days for stable infants with low gastric aspirates. The authors also recommended the implementation of standardized feeding protocols to guide the introduction of enteral feeds. This review did not address other enteral feeding practices. The level of evidence for each recommendation was assessed using the Grading of Recommendations, Assessment, Development and Evaluation approach, 17 but no consensus exercise was conducted to indicate the authors degree of agreement with each recommendation.
The purpose of the current study was to better understand enteral feeding practices among infants with gastroschisis as reported by clinicians from neonatal intensive care units across Canada. To do this, we conducted semi-structured interviews with representatives from each center to gain insights into the practices of neonatal clinicians across Canada. This will help us identify areas of variation in enteral feeding practices and may form the basis for the development of consensus-based clinical practice guidelines.
Methods
Survey development
We created a survey consisting of 18 open-ended questions across four phases of care: (1) post-closure practices while awaiting return of bowel function; (2) introduction of enteral feeds; (3) advancement of enteral feeds; and (4) use of feeding adjuncts. We did not define the clinical criteria for each of these phases because we wanted respondents to speak honestly about their center’s usual practices. The questions were developed by our core working group. Feedback was sought from clinicians from our center, including pediatric surgeons, neonatologists, registered dieticians, and neonatal nurse practitioners. This resulted in further revisions. The finalized list of interview questions can be found in Appendix 1.
The section focused on the introduction of enteral feeds included items related to the type of feeds (e.g., maternal breast milk vs formula), type of formula used if donor or maternal breast milk is unavailable (e.g., partially hydrolyzed, elemental, etc.), route of administration (i.e., oral or nasogastric), and rate of administration (i.e., bolus or continuous). We also asked questions related to which clinicians are involved in the decision to start enteral feeds and whether each center has a written feeding protocol for infants with gastroschisis.
Study design and participants
We planned to describe enteral feeding practices as reported by clinicians from the 17 neonatal intensive care units in Canada that manage infants with gastroschisis. To increase awareness, the study protocol and survey questions were presented to the Gut Health subgroup from the Evidence-Based Practice of Improving Quality program. This initiative is a national collaboration of multidisciplinary clinicians focused on improving the quality of care in neonatal intensive care units across Canada. Each neonatal intensive care unit in Canada that manages infants with gastroschisis identified an attending neonatologist and registered dietician to serve as possible study participants. Invitations were sent to these individuals via email along with the interview questions. We encouraged participants to discuss the interview questions with their colleagues to ensure they could accurately report their center’s “usual practice” rather than their personal opinions. If the neonatologist or dietician indicated that they lacked sufficient experience in treating infants with gastroschisis, they were asked to identify an alternative representative from their center to respond.
Study participation was voluntary, and respondents were allowed to withdraw at any time. No patient-specific data were requested. The questions were also translated into French by one of the authors (NC) and sent to the three French-speaking centers in Montreal, Sherbrooke, and Quebec City. In cases of non-response, we followed up via email and professional contacts to encourage participation.
Responses were collected through semi-structured interviews using video conferencing software (i.e., Zoom). This format allowed participants to express uncertainty, clarify areas of ambiguity, and respond to follow-up questions. The first and senior authors (HS and MHL) were present for all interviews. Some respondents declined to meet virtually and provided written responses via email instead. The study protocol adheres to the guidelines outlined in the Declaration of Helsinki and was approved by the Hamilton Integrated Research Ethics Board (HiREB: 16036).
Statistical analysis
All responses were transcribed and stored in a secure, online database designed using Research Electronic Data Capture software. Data were exported to Microsoft Excel and R (Version 4.4.1) for descriptive analysis. Categorical variables were reported as frequencies and percentages.
Results
Participants
Characteristics of participants.
Enteral feeding practices reported by clinicians from neonatal intensive care units in Canada.
a2/14 “occasionally” use a protocol from another institution and 1/14 is developing a protocol.
bDonor human milk is used until full enteral feeds (10/14), up to 4 weeks of duration (2/14), selectively due to prematurity (1/14), or weaned prior to discharge (1/14).
Stimulation of bowel function post-abdominal closure
None of the centers reported routine use of sham feeding, suppositories, rectal irrigations, or contrast studies to stimulate bowel function following abdominal closure. As expected, some centers mentioned that contrast studies are used in a delayed fashion to rule out a stricture or atresia among infants who are unable to tolerate enteral feeds.
All centers reported that they provide oral immune therapy to infants with gastroschisis. This treatment strategy involves inserting a few drops (or a swab) of expressed breast milk (ideally colostrum) to the infants mouth several times per day. 18 Two of 14 (14%) centers reported that they would consider providing donor human milk as oral immune therapy in cases where maternal breast milk is unavailable. There was also variation in the timing of the commencement of oral immune therapy. Six of 14 centers specifically discussed the timing of starting oral immune therapy. Three centers indicated that they try to start oral immune therapy as soon as possible (even while a silo is still in place). Another three centers indicated that they only consider starting oral immune therapy once the abdominal defect is closed.
Introduction of enteral feeds
None of the participants reported routinely using a gastroschisis feeding protocol. Two centers (14%) reported occasionally using a gastroschisis feeding protocol developed by another children’s hospital. Another center indicated that a feeding protocol for infants with gastroschisis is in development. The decision to start enteral feeds was reported to be collaborative between the pediatric surgery and neonatal teams at 8 centers (57%), driven mainly by pediatric surgeons at 5 centers (36%), and largely up to the neonatal team at another.
All centers reported a preference for maternal breast milk as the initial source of enteral nutrition. In cases where maternal milk is unavailable, the eligibility criteria for donor human milk varied. One center reported that infants with gastroschisis are given donor human milk only in cases of prematurity. Others reported that donor human milk is provided to all infants with gastroschisis until they reach full enteral feeds (n = 10), for a duration of up to 4 weeks (n = 2), or until they are weaned prior to discharge from hospital (n = 1).
In Canada, donor human milk is pasteurized, stored, and distributed to hospitals by four licensed, not-for-profit human milk banks. These include the British Columbia Women’s Provincial Milk Bank, the NorthernStar Mothers Milk Bank in Alberta, the Rogers Hixon Ontario Human Milk Bank, and Héma-Québec’s Public Mothers’ Milk Bank. 19 All four donor human milk banks are compliant with Health Canada regulations and three are members of the Human Milk Banking Association of North America (HMBANA).
Participants were also asked about the type of formula used for babies with uncomplicated gastroschisis when maternal or donor human milk is not available. The majority (n = 10) reported a preference for standard infant formula. Others reported starting with formula that is partially hydrolyzed (n = 1), extensively hydrolyzed (n = 1), or elemental (n = 1). One center had no preference (n = 1). Examples of partially hydrolyzed formulas include Gentlease®, Total Comfort®, and Good Start®. Examples of extensively hydrolyzed formulas include Nutramigen® and Alimentum®. Examples of elemental formulas include Neocate® and PurAmino A+®. We did not ask about the use or endorsement of specific commercial products, since the brand of formula used is usually dictated by contracts between hospitals and formula companies.
Almost all centers (n = 13) indicated that enteral nutrition is introduced as bolus, trophic feeds unless there are signs of feeding intolerance, necessitating the transition to a continuous feeding regimen. The preferred mode of administration for bolus feeds was by mouth (n = 12) except in cases where infants require respiratory support.
Advancement of enteral feeds
The rate of increasing enteral feeds varied among respondents. The majority of centers reported a preference towards small to moderate increments (i.e., less than 20 mL/kg/day). Three centers advanced feeds at 20 mL/kg/day or more, and two centers indicated that they do not advance feedings using a standardized approach. The distribution of reported feeding advancements is shown in Figure 1. Several factors were considered in determining the rate of increase, including gestational age, duration of post-operative ileus, and weight gain. Most centers reported that the rate of feeding advancement increased over time as infants showed improved ability to tolerate greater volumes. Initial advancement rate of enteral feeds reported by clinicians from neonatal intensive care units in Canada.
The decision to advance enteral feeds was reported as a collaborative decision between pediatric surgeons and the neonatal team at most centers (n = 12). Two centers indicated that it was driven primarily by the neonatal team. Participants from all centers reported stopping enteral feeds or slowing feeding advancements among infants who show signs of necrotizing enterocolitis and/or feeding intolerance, including vomiting, abdominal distention, and/or bilious aspirates.
Use of feeding adjuncts
Participants indicated a preference towards increasing caloric intake with higher enteral volume rather than fortification (defined as the addition of infant formula fortifiers or human milk fortifiers). Most centers reported rarely using fortifiers, and one center reported never using fortifiers for infants with gastroschisis due to concerns about milk bezoar formation. Some centers reported that infants with gastroschisis are considered for human milk fortifiers (rather than infant formula fortifiers) if they are premature, close to tolerating full enteral feeds, or show signs of poor weight gain.
There were no concerns from respondents about starting enteral nutrition with bottle feeding. As for the introduction of breastfeeding, some participants indicated that their center prefers to start with non-nutritive breastfeeding and transition to ad lib breastfeeding after full enteral feeds are achieved.
Motility agents such as domperidone or metoclopramide were reported to be used on a selective basis only. Similarly, anti-reflux medications such as omeprazole were used only in cases of clinically significant gastroesophageal reflux. Two centers reported that probiotics are part of their usual care for infants with gastroschisis, with one center indicating that they use FloraBABY from RenewLife®. Other centers reported that they only provide probiotics if the infant has been on antibiotics for a prolonged period or is premature.
All participants reported using SMOFlipid® for the lipid component of parenteral nutrition for infants with gastroschisis. No other considerations were reported in the formulation of parenteral nutrition.
Discussion
This study demonstrated variation in some but not all areas of enteral feeding practices for infants with gastroschisis managed by neonatal intensive care units in Canada. Areas of variation included the use of donor human milk, types of formula used when maternal breast milk is unavailable, and use of probiotics. There also appeared to be variation in decision-making regarding the introduction and advancement of enteral feeds, which may be partly because none of the participating centers reported routinely using a gastroschisis feeding protocol.
Respondents from all centers reported that they routinely provide oral immune therapy to infants with gastroschisis. Some participants expressed concerns regarding the use of donor milk for oral immune therapy in cases where maternal breast milk is unavailable. The administration of colostrum has been shown to confer numerous benefits to premature babies, such as reduction in the incidence of necrotizing enterocolitis, feeding intolerance, sepsis, and time to full enteral feeds. 20 To the best of our knowledge, there are no published studies describing the administration of small amounts of donor human milk akin to the administration of oral immune therapy. This is likely because all donor milk is pasteurized, which not only deactivates any contaminating microorganisms but also affects the concentrations of vital bioactive components such as immunoglobulin A, lactoferrin, and lysozyme. 19
None of the participants reported using sham feeds for cases of uncomplicated or complicated gastroschisis. Sham feeding consists of early enteral feeds and nasogastric decompression. Early enteral feeding has been documented in developing countries, such as Uganda, resulting in the first known survivors of gastroschisis in jurisdictions where parenteral nutrition is unavailable. 21 In the United States, sham feeding has been reported by at least one children’s hospital. 22 This practice appears safe but its effectiveness is uncertain.
The majority of participants noted that the decision to initiate enteral feeds is made jointly by the pediatric surgery and neonatal teams. A collaborative approach provides opportunities to share differing perspectives and the ability to challenge usual practice with the best available evidence. Indeed, a retrospective cohort study found lower rates of mortality among infants with gastroschisis in centers with multidisciplinary teams than those without. 23
The rate of feeding advancement reported by participants varied widely, ranging from 5 mL/kg/day to more than 20 mL/kg/day. Utria et al. found that infants with gastroschisis experienced statistically significant improvements in length of stay, time to achieve full enteral feeds, and days on parenteral nutrition when feeds were advanced at 20 mL/kg/day compared to 10 mL/kg/day. 24
Two of the 14 centers reported using probiotics routinely for infants with gastroschisis. Some studies have shown that probiotics result in limited changes in the gut microbiome, and no significant differences in length of stay and duration of parenteral nutrition among newborns with gastrointestinal conditions.25,26 In contrast, a randomized controlled trial in premature infants found that probiotics led to faster gastric emptying and a marked decrease in regurgitation. 27 The role of probiotics among infants with gastroschisis remains unclear.
None of the centers in our study reported using motility agents routinely. Some centers specifically avoided these medications and only reported using proton pump inhibitors such as omeprazole if there are signs of significant gastroesophageal reflux. Some clinicians also expressed a preference towards the selective use of domperidone or metoclopramide.
One of the strengths of our study was the use of a modified member-checking technique during verbal interviews to ensure that the participants’ responses were accurately interpreted before recording responses in our database. In addition, centers that manage infants with gastroschisis were systematically identified using a list obtained from a national quality improvement consortium, ensuring adequate representation across Canada. This approach likely enhances the validity of our findings.
This study has several important limitations. First, interview responses were self-reported and may have been affected by recall and social desirability bias. For example, the majority of participants in our study indicated that their center routinely provides donor human milk to infants with gastroschisis, particularly before they reach full enteral feeds. However, a national cohort study of infants with uncomplicated gastroschisis treated at centers from 2014 to 2022 found that 58 of 267 patients (22%) received at least 1 day of formula in the first 28 days of life. 28 Efforts are underway to conduct a follow-up study of infants with both uncomplicated and complicated disease to describe the use of formula, maternal breast milk, and donor human milk at neonatal intensive care units across Canada.
Another limitation is that some respondents did not complete the survey questions via virtual meeting. The mode of survey administration (i.e., virtual meeting vs email) did not appear to affect whether respondents answered all of the questions, but the responses from virtual meetings generally resulted in higher-quality data. The open-ended format allowed us to seek clarification and ask follow-up questions, leading to more detailed descriptions of usual practice.
We were also unable to provide more descriptive details about the participating clinicians. All respondents were identified through the Evidence-Based Practice of Improving Quality collaboration as the clinician from their center with the greatest interest and/or experience with managing patients with gastroschisis. To foster an honest and collaborative discussion, we intentionally avoided requesting details about clinical experience, such as the number of years in practice and number of cases managed annually. We were also unable to explore whether feeding practices varied by center volume. The Canadian Pediatric Surgery Network routinely assesses outcomes for gastroschisis by each center’s case volume (which is classified as “low,” “medium,” and “high”). However, these data are kept anonymous to encourage honest reporting and ongoing participation.
Finally, it is unclear whether the three centers who did not respond to our invitation to participate were systematically different compared to those who did. With a sample of only 17 eligible centers across Canada, the aggregate results should be interpreted with caution.
None of the participants in our study reported using a gastroschisis feeding protocol routinely. Two centers occasionally use a protocol from another hospital, and another is in the process of development. Creating enteral feeding recommendations for infants with gastroschisis is just one component of their care that could be optimized through the implementation of standardized protocols and care pathways. Haddock et al. found that implementing such protocols for infants with gastroschisis led to shorter ventilator duration, reduced surgical site infection, and lower patient costs. 29
Efforts are underway to develop clinical practice guidelines for the management of gastroschisis, similar to those established for congenital diaphragmatic hernia.30,31 In Canada, patients with these two diagnoses are routinely assessed and followed as part of the Canadian Neonatal Network and Canadian Pediatric Surgery Network. This provides researchers with a unique opportunity to evaluate the impact of clinical practice guidelines using national data.
Conclusions
This survey of clinicians from neonatal intensive care units in Canada showed variation in some but not all areas of enteral feeding among infants with gastroschisis. None of the centers reported using a gastroschisis feeding guideline routinely, so clinicians should consider implementing a written pathway for enteral feeds or adopting one from another center. Although there was strong support for the use of maternal breast milk, practices differed regarding the advancement of enteral feeds, use of donor human milk, type of formula used when donor human milk and maternal breast milk is unavailable, use of probiotics, and introduction of breastfeeding. Additionally, motility agents such as domperidone or metoclopramide do not appear to be used routinely. The development of evidence-based clinical practice guidelines for patients with gastroschisis may help standardize feeding practices and lead to superior outcomes.
Statements and declarations
Footnotes
Acknowledgments
We thank the participants for making time to meet with us and respond to our interview questions. We also want to acknowledge our colleagues at McMaster Children’s Hospital who reviewed our interview questions and provided feedback.
Conflicting interest
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Interview script and questions
These questions have to do with your usual care of infants with gastroschisis post-abdominal closure. 1. Does your team do anything to stimulate bowel function after closure of infants with gastroschisis (e.g., sham feeding, suppositories, rectal irrigations, and contrast studies)? 2. Do infants with gastroschisis receive oral immune therapy? If so, when is this started? What happens if maternal breast milk (“mother’s own milk”) is not available?
These next questions are focused on practices related to starting enteral feeds. 3. Does your team have a written feeding protocol for infants with gastroschisis? 4. Who decides when to start feeds? 5. What type of feeds are used initially? 6. If maternal breast milk (“mother’s own milk”) is not available, do infants with uncomplicated gastroschisis typically receive donor breast milk? If so, when and how do infants wean from donor breast milk to formula? 7. If formula is needed, what type is typically used (i.e., partially hydrolyzed and elemental)? 8. Are feeds typically started in a continuous fashion (i.e., ml/hour) or bolus? If continuous, when and how do infants transition to bolus feeds? 9. Are bolus feeds typically given via nasogastric tube or by mouth?
These questions are focused on practices related to advancing enteral feeds. 10. How are feeds advanced (how often and by how much)? 11. Who decides when to advance feeds to higher volumes? 12. When do you hold feeds or slow the rate of advancement?
These last few questions are related to feeding adjuncts. 13. What formulation of parenteral nutrition is typically used for infants with gastroschisis at your center? 14. When and how do you fortify feeds for infants with gastroschisis? If applicable, what type of fortifier is used? 15. When do you allow infants with gastroschisis to bottle or breastfeed? How is this introduced (e.g., non-nutritive feeds, feeding ad lib, and other)? 16. Do you typically use motility agents or other medications to help with feeding in infants with uncomplicated gastroschisis (e.g., domperidone, metoclopramide, omeprazole, and oral antibiotics)? 17. Do you typically use probiotics in infants with gastroschisis? 18. Do you have any other comments regarding feeding practices for infants with gastroschisis at your center?