Introduction
CHDs are the most common types of birth defects, impacting nearly 1% of births per year in the United States.1 While the term CHD refers to structural disease that forms in utero, children can also have acquired cardiovascular disease that develops after birth. Both congenital and acquired heart disease can contribute to significant morbidity and mortality.Reference Leong, Kane and Joy2 Approximately 25% of infants born with CHD will have a critical defect, generally requiring surgery or catheter-based intervention in the first year of life.1 These interventions often correspond to the time period in typical infant development during which oral feeding skills are established and can contribute to heightened risk for feeding and swallowing difficulties.Reference Maurer, Latal, Geissmann, Knirsch, Bauersfeld and Balmer3,Reference Norman, Zühlke, Murray and Morrow4 During the surgical and/or hospital course, infants with congenital and acquired heart disease often experience multiple disruptions to feeding development and may have reduced opportunities to feed by mouth in comparison to infants without heart disease.Reference Jones, Desai and Fogel5 Early opportunity for oral feeding skill development may be crucial for infants with CHD requiring surgical intervention, as some studies have reported an association between delayed oral feeding initiation and poor oral feeding outcomes.Reference Sables-Baus, Kaufman, Cook and da Cruz6–Reference Milligan, Mills and Ge8
Mechanical ventilation and an ongoing need for respiratory support can interfere with oral feeding opportunities. Noninvasive ventilation therapy including high flow nasal cannula and nasal continuous positive airway pressure are often used after postoperative extubation from mechanical ventilation. An analysis of the Pediatric Cardiac Critical Care Consortium clinical registry found that 72% of neonates and 43% of infants who underwent cardiothoracic surgery required non-invasive ventilation, with a median duration of 6 days.Reference Romans, Schwartz and Costello9 While many patients benefit from non-invasive ventilation support to optimise gas exchange and cardiac output,Reference Beshish, Hupp and Dryer10 positive pressure and high flow rates may be considered added risks for aspiration while feeding due to potential impact on airway protective mechanisms.
Currently, evidence on oral feeding practices for infants on non-invasive ventilation is inconsistent.Reference Canning, Clarke, Thorning, Chauhan and Weir11 It is difficult to establish safe feeding guidelines due to heterogeneity in the populations studied, the types of non-invasive ventilation used, and the level of support applied.Reference Barnes, Herbert and Bonilha12 To our knowledge, only one study,Reference Bassi, Morris, Rao, Rao, Nigro and Brigger13 available as a conference presentation, has focused on oral feeding while on non-invasive ventilation for infants with CHD. The authors described an institutional guideline that supported safe progression of oral feeding for infants with single ventricle physiology while on non-invasive ventilation, using a phased approach. Most infants (67%) demonstrated adequate weight gain and ability to wean respiratory support, with the majority of patients fully feed by mouth by two weeks (53%) or one month post discharge (67%). However, little is known about current oral feeding practices for infants with CHD who require nasal continuous positive airway pressure or high flow nasal cannula. Therefore, the aim of this study was to describe current US institutional feeding practices for infants with CHD while on non-invasive ventilation.
Materials and method
This was a cross-sectional survey of clinicians from primarily US institutions that were members of the Cardiac Neurodevelopmental Outcomes Collaborative. Survey questions were developed in Microsoft Forms by a group of clinicians working in acute care and/or intensive care paediatric cardiac units, including speech-language pathologists, occupational therapists, a cardiac care intensivist, and a pulmonologist. The survey was piloted at 8 institutions, with questions revised and finalised based on pilot feedback. A survey team member then contacted each Cardiac Neurodevelopmental Outcomes Collaborative member institution and requested contact information for a clinician involved in cardiac ICU feeding practices (e.g., a feeding therapist, advanced practice provider, or physician). Individual centres were instructed to determine the most appropriate professional in their cardiac ICUs to complete the survey (i.e., open to all disciplines). The final survey focused on feeding practices for hospitalized infants with CHD <6 months old (see Supplementary Material S1) and was sent directly to clinicians in June 2023, with a three-week response window. Analysis included descriptive statistics (e.g., n, %) and data visualisation.
Results
The survey was sent to clinicians at 35 institutions, of which 20 (57%) responded. The majority of respondents were speech-language pathologists (n = 14, 70%), with occupational therapists (5, 25%) and physical therapists (1, 5%) also represented. Institutions were located in all regions of the US, including the South (7, 35%), West (6, 30%) Midwest (4, 20%), and Northeast (2, 10%), and in one international site. All respondents were from academic centres. Respondents’ cardiac ICUs most commonly had 11–20 (9, 45%) or 21–30 beds (7, 35%), with three institutions including >31 beds.
Oral feeding practices on non-invasive ventilation
Only four (20%) respondents indicated that their institution utilised a formal guideline for introduction of oral feeding on nasal continuous positive airway pressure or high flow nasal cannula. Further details of institutional practices for oral feeding on non-invasive ventilation can be seen in Figure 1a–d. Most institutions did not allow full oral nutritive feeds while infants were on nasal continuous positive airway pressure, with only one allowing full oral feeds (Figure 1a). Four institutions allowed small volume oral feeds for practice and/or maintenance of feeding skills on nasal continuous positive airway pressure and two centres stated in the comments section that there are requirements for a specific slow flow nipple (Figure 1b).
Survey responses regarding institutional practices for oral feeding on respiratory support.
Figure 1. Long description
Panel A: A bar graph shows the number of institutions allowing full oral nutritive feeds on nCPAP. The x-axis represents the number of institutions, and the y-axis represents the response categories: Yes, No, and Other. One institution (5 percent) allows it, sixteen institutions (80 percent) do not, and three institutions (15 percent) have other practices. Panel B: A bar graph shows the number of institutions allowing small volume oral feeds on nCPAP. The x-axis represents the number of institutions, and the y-axis represents the response categories: Yes, No, and Other. Four institutions (20 percent) allow it, twelve institutions (60 percent) do not, and four institutions (20 percent) have other practices. Panel C: A bar graph shows the number of institutions allowing full oral nutritive feeds on HFNC. The x-axis represents the number of institutions, and the y-axis represents the response categories: Yes, Depends on age, and Other. Fourteen institutions (70 percent) allow it, three institutions (15 percent) depend on age, and three institutions (15 percent) have other practices. Panel D: A bar graph shows the number of institutions allowing small volume oral feeds on HFNC. The x-axis represents the number of institutions, and the y-axis represents the response categories: Yes, No, and Other. Seventeen institutions (85 percent) allow it, one institution (5 percent) does not, and two institutions (10 percent) have other practices.
The majority (70%) of centres allowed full oral nutrition on high flow nasal cannula (Figure 1c). Practice at three additional centres varied based on the infant’s age as indicated in comments. For example, one respondent wrote: “Older babies that have been previously fed successfully may be fed on more respiratory support at times.” Nearly all institutions (85%) allowed small volume oral feeding on high flow nasal cannula (Figure 1d).
Postoperative oral feeding practices and assessments
Most institutions (65%) did not have a formal guideline specifying the timing of post extubation oral feeding (Figure 2a). Once extubated to nasal continuous positive airway pressure or high flow nasal cannula, infants were typically fed by nasogastric (n = 11, 55%), or nasoduodenal (2, 10%) tube, with additional written options including “combination/variety, usually nasojejunal initially,” “both nasojejunal and nasogastric,” “varies based on provider and infant,” and “total parenteral nutrition.” Respondents were asked about indications for an instrumental swallowing evaluation, prior to allowing full oral feeds postoperatively (Figure 2b). Nearly all (95%) reported that instrumental swallow evaluations are completed when clinical signs of aspiration or swallowing dysfunction are observed. All institutions specified that facilitative feeding strategies were first implemented (e.g., changing nipple flow rate, positioning, and external bottle pacing) prior to making recommendations for instrumental assessment. Approximately a third of institutions performed postoperative instrumental swallow evaluations in cases involving the aortic arch. Institutions did not specify when these instrumental swallow evaluations were completed during the postoperative course. Most institutions (60%) allowed infants with single ventricle and two ventricle physiology similar opportunities to feed while on nasal continuous positive airway pressure or high flow nasal cannula (Figure 2c). Two additional respondents indicated that infants with single ventricle physiology may be allowed to feed on non-invasive ventilation, but that the healthcare team is generally more cautious in their feeding approach.
Survey responses regarding oral feeding practices postoperative and feeding assessments.
While most respondents (65%) reported completing a feeding assessment on all infants in the cardiac ICU both pre and postoperatively (Figure 2e), only 4 (30%) institutions utilised a formal cue-based assessment (Figure 2d). Other indications for feeding assessment included all infants with single ventricle or other high-risk cardiac physiology, all infants postoperatively, or all neonates postoperatively.
Discussion
The aim of this study was to describe current US institutional feeding practices for infants with CHD while on non-invasive ventilation therapy, including high flow nasal cannula and nasal continuous positive airway pressure . Survey responses indicated that practices related to oral feeding on non-invasive ventilation vary across centres, with only 20% having formal guidelines. Most institutions did not allow full oral feeds on nasal continuous positive airway pressure, although some allowed small volumes or limited frequencies of oral nutrition. It was more common for institutions to allow full or restricted small-volume oral feeding on high flow nasal cannula. These findings of greater institutional comfort with oral feeding on high flow nasal cannula, compared to nasal continuous positive airway pressure, are consistent with Canning et al.’sReference Canning, Fairhurst, Chauhan and Weir14 2020 survey of feeding practices on nasal continuous positive airway pressure and high flow nasal cannula in neonatal/paediatric ICUs in Australia and New Zealand, in which far fewer institutions offered full oral feeding on nasal continuous positive airway pressure than on high flow nasal cannula.
The variability in cardiac ICU feeding practices identified in the current study is likely due to limited research on the topic specifically focused on infants with CHD. This limited evidence may prompt cardiac ICU clinicians to base feeding practices on studies conducted in populations of similar age or level of care.
Given the lack of conclusive evidence to guide practice for infants with CHD, however, the low rates of full oral feeding on nasal continuous positive airway pressure identified in the current study suggest that providers likely err on the side of caution in cardiac populations. Most institutions in the current study appear to view the transition from positive pressure support to high flow as a threshold for safe initiation of oral feeding, perhaps based on limited evidence suggesting increased risk of silent aspiration.Reference Ferrara, Bidiwala and Sher18
The current lack of formal guidelines for introduction of oral feeding on non-invasive ventilation in the cardiac ICU is concerning. A conservative approach may prevent infants from having an opportunity to demonstrate compensatory efforts to maintain airway protection, while aligning breathing and swallowing skills, throughout their recovery process. A collaborative team approach to feeding includes attention to feeding cues, gas exchange, and haemodynamic stability at the forefront of oral feeding progression plans. Optimising opportunities for oral feeding progression during the rapidly changing recovery process after infant cardiac surgery may positively impact infant health outcomes, as initiation of early postoperative feeding has been shown to support optimal growth and nutrition,Reference Du, Cui, Xie, Yin, Gong and Chen24 and it is critical to capitaliseReference Harjpal, Kovela and Qureshi25 on an infant’s primitive reflexes to feed orally. Furthermore, infants with CHD benefit from the use of facilitative feeding strategiesReference Harjpal, Kovela and Qureshi25–Reference Thoyre, Holditch-Davis, Schwartz, Melendez Roman and Nix27 to support feeding safety and progression of feeding and swallowing skill (e.g., breastfeeding opportunities, cue-based feeding approaches, use of slow-flow nipples, attention to suck:swallow ratio to support minute ventilation, and external bottle pacing). The current evidence, including critically ill infants without heart disease, has not adequately examined the effect of these facilitative feeding strategies on infants’ physiologic stability while feeding on non-invasive ventilation and thus presents opportunity for investigation.
Limitations
While this study included institutions from across the US, it does not necessarily represent the practices of all US paediatric cardiac centres and included only institutions that were Cardiac Neurodevelopmental Outcomes Collaborative members. The results may have been impacted by self-selection bias, as the majority of respondents were feeding and swallowing therapists (SLP/OT) with an interest in the topic and may have been more likely to complete the survey. Additionally, as one clinician from each institution completed the survey, there could be individual variation in practice within institutions that was not captured in this study. While the survey was open to Advanced Nurse Practitioners and Cardiac Care Unit intensivists, these individuals were not chosen by any centre as respondents to the survey. Given that only therapists responded, the survey may overrepresent the perspectives, clinical reasoning, and practice patterns of feeding therapists and cannot be assumed to represent the full perspectives of the interdisciplinary team and may impact the generalisability of the survey results. Survey questions elicited responses targeted at team decision-making for oral feeding based on positive pressure support and high flow nasal cannula. Specific questions for oral feeding decision-making in relation to oxygen delivery vs. flow and considerations of L/kg/min for high flow nasal cannula were not included in the survey as respondents were presumed to interpret the question accordingly. However, the lack of explicit unit definition introduces the potential for variability in interpretation. Additionally, for responses indicating whether a centre “allows small volume feeds” at a given level of respiratory support, specific volume parameters were not defined. As a result, variability in respondent interpretation of “small volume feeds” represents a limitation in this survey.
Conclusion
The aim of this study was to describe current institutional feeding practices for infants with CHD while on non-invasive ventilation. The results indicate that the majority of institutions do not allow oral feeding until a threshold of transition to high flow nasal cannula. Most centres do not utilise formal guidelines or assessments to determine infant readiness to initiate or progress oral feeding while on non-invasive ventilation, and formal cue-based feeding approaches were infrequent. Future research and quality improvement initiatives are needed to examine the effects of feeding infants with CHD while on non-invasive ventilation and to develop centre-specific guidelines to direct feeding progression, with potential to improve oral feeding outcomes while supporting multisystem stability.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S1047951126113705.
Acknowledgements
A special thanks to the survey respondents who took the time to report on institution practices and to the professionals who contributed their expertise and time in formulation of this survey and study.
Financial support
No reported financial support.
Competing interests
No reported conflicts of interest.
Ethical standard
The authors assert that all procedures contributing to this work comply with ethical standards of the United States Federal Policy for the Protection of Human Subjects and with the Helsinki Declaration of 1975, as revised in 2008, and have been approved by the University of Minnesota Institutional Review Board.
