De-implementing the Routine Measurement of Gastric Residual Volumes in an All-Referral Newborn Intensive Care Unit

Routine gastric residual volume measurement has historically been the standard of care in many neonatal intensive care units to monitor for feeding intolerance and necrotizing enterocolitis. However, current evidence suggests that gastric residual volume is a poor indicator of these pathologies. Furthermore, the routine measurement of gastric residual volumes is associated with adverse outcomes in neonates including prolonged time to full enteral feeds, longer hospitalization, and more late-onset sepsis. In this 51-bed, all-referral neonatal intensive care unit, the clinical practice guideline recommended the routine measurement of gastric residual volumes in all infants receiving enteral feeds via enteric feeding tube. There was a need to de-implement this practice to be consistent with current evidence. Pre- and post-intervention data were compared to evaluate the success of this quality improvement project. Staff practice and opinions were assessed with surveys before and after the educational interventions. Patient outcome data were obtained by retrospective chart review prior to the commencement of the interventions and by prospective chart review following the completion of the interventions. This quality improvement project began with the removal of the clause recommending gastric residual volume measurement from the clinical practice guideline. Educational interventions including Situation-Background-Assessment-Recommendation descriptions of the change, presentations at charge nurse and leadership meetings, in-person and small group discussions with nursing staff, informational pamphlets, and reminder signs in patient rooms were used to inform staff about and garner support for the practice change. Nursing practice prior to the educational interventions was varied. Following the interventions, nursing practice changed significantly away from routine gastric residual volume measurement, with only 5.1% of nurses continuing to aspirate gastric residual volumes in all patients and 82.1% of nurses not measuring gastric residual volumes in any patients. Nursing opinion on discontinuing gastric residual volume measurement also became significantly more favorable; assessment by Likert scale demonstrated an increase from mean 3.44 to 4.05 out of 5, p=.005 (N=127). Advanced practice providers expressed high favorability of discontinuing residual volume measurement both before and after the interventions, with mean Likert scores of 4.78 and 4.90 respectively (N=28). No changes were evident before versus after the practice change in patient outcomes of time to full enteral feedings (5.5 versus 7.6 days, p=.052, N=51), weight gain (ΔZ-score/days hospitalized -0.017 vs. -0.005, p=.986, N=117), episodes of feeding interruptions for feeding intolerance (3.09 vs. 8.05 episodes, p=.156, N=117), incidence of bacteremia (1.03 vs. 2.01 episodes, p=.683, N=117), or incidence central line infections (3.08 vs. 2.01 episodes, p=.670, N=117). There were no episodes of necrotizing enterocolitis or culture-positive pneumonia in either patient group. Educational interventions for staff were effective in de-implementing routine gastric residual volume measurement in an all-referral neonatal intensive care unit, with success both in changing practice and gaining staff acceptance of the change. While benefit in patient outcomes was not observed at this institution, there was importantly no increase in adverse events for patients following the practice change.

1 De-implementing the Routine Measurement of Gastric Residual Volumes in an All-Referral Newborn Intensive Care Unit Erin B. Tucker College of Nursing: The University of Utah NURS 7703: DNP Scholarly Project III April 17, 2023 2 Abstract Background: Routine gastric residual volume measurement has historically been the standard of care in many neonatal intensive care units to monitor for feeding intolerance and necrotizing enterocolitis. However, current evidence suggests that gastric residual volume is a poor indicator of these pathologies. Furthermore, the routine measurement of gastric residual volumes is associated with adverse outcomes in neonates including prolonged time to full enteral feeds, longer hospitalization, and more late-onset sepsis. Local Problem: In this 51-bed, all-referral neonatal intensive care unit, the clinical practice guideline recommended the routine measurement of gastric residual volumes in all infants receiving enteral feeds via enteric feeding tube. There was a need to de-implement this practice to be consistent with current evidence. Methods: Pre- and post-intervention data were compared to evaluate the success of this quality improvement project. Staff practice and opinions were assessed with surveys before and after the educational interventions. Patient outcome data were obtained by retrospective chart review prior to the commencement of the interventions and by prospective chart review following the completion of the interventions. Interventions: This quality improvement project began with the removal of the clause recommending gastric residual volume measurement from the clinical practice guideline. Educational interventions including Situation-Background-Assessment-Recommendation descriptions of the change, presentations at charge nurse and leadership meetings, in-person and small group discussions with nursing staff, informational pamphlets, and reminder signs in patient rooms were used to inform staff about and garner support for the practice change. Results: Nursing practice prior to the educational interventions was varied. Following the interventions, nursing practice changed significantly away from routine gastric residual volume measurement, with only 5.1% of nurses continuing to aspirate gastric residual volumes in all patients and 82.1% of nurses not measuring gastric residual volumes in any patients. Nursing opinion on discontinuing gastric residual volume measurement also became significantly more favorable; assessment by Likert scale demonstrated 3 an increase from mean 3.44 to 4.05 out of 5, p=.005 (N=127). Advanced practice providers expressed high favorability of discontinuing residual volume measurement both before and after the interventions, with mean Likert scores of 4.78 and 4.90 respectively (N=28). No changes were evident before versus after the practice change in patient outcomes of time to full enteral feedings (5.5 versus 7.6 days, p=.052, N=51), weight gain (ΔZ-score/days hospitalized -0.017 vs. -0.005, p=.986, N=117), episodes of feeding interruptions for feeding intolerance (3.09 vs. 8.05 episodes, p=.156, N=117), incidence of bacteremia (1.03 vs. 2.01 episodes, p=.683, N=117), or incidence central line infections (3.08 vs. 2.01 episodes, p=.670, N=117). There were no episodes of necrotizing enterocolitis or culture-positive pneumonia in either patient group. Conclusion: Educational interventions for staff were effective in de-implementing routine gastric residual volume measurement in an all-referral neonatal intensive care unit, with success both in changing practice and gaining staff acceptance of the change. While benefit in patient outcomes was not observed at this institution, there was importantly no increase in adverse events for patients following the practice change. Keywords: De-implementation, quality improvement, gastric residuals, gastric aspirates, neonatal, newborn. 4 De-implementing the Routine, Universal Measurement of Gastric Residual Volumes in an AllReferral Newborn Intensive Care Unit Problem Description It is a general goal in the care of the neonatal patient to provide as much nutrition through the enteral route—as opposed to the parenteral route—as is safe and tolerated. Not limited by the formulary limitations of parenteral nutrition, enteral feeding allows for improved provision of nutrients (Wiechers, Bernhard, Goelz, Poets, & Franz, 2021). Accordingly, faster advancement to full enteral feeds is associated with improved neonatal weight gain (Krishnamurthy, Gupta, Debnath, & Gomber, 2010). With central access as a favored route for parenteral nutrition, catheter-associated infections remain a frequent complication associated with parenteral nutrition and a further incentive for its expeditious wean (Lago Rivero et al., 2013). However, the advancement of enteral feeds in the neonate is often slowed by concerns of feeding intolerance or emerging necrotizing enterocolitis. While these concerns are often valid, it is important to avoid unnecessary interruptions in feedings based on inappropriate indicators. Gastric residual volumes, measured by aspirating from an enteric feeding tube until no further output can be obtained, have traditionally been monitored on many newborn intensive care units to screen for necrotizing enterocolitis or feeding intolerance in patients receiving enteral feeds. However, a growing body of evidence discourages this practice, and consequently, the routine measurement of gastric residual volumes is being increasingly de-implemented from the standard care of premature infants. This practice change has been at odds with the inertia of tradition, as reflected by considerable variation in practice (Dorling et al., 2020; Hobson, Spence, Trivedi, & Thomas, 2019; Perumbil Pathrose et al., 2021; Tume et al., 2019; Tume et al., 2020). There is a need to translate this movement to meet the needs of the complex and surgical neonatal population— the type cared for at the all-referral newborn intensive care unit which was the site of this project—and to persuade a change in culture to more quickly and effectively onboard evidence-based recommendations. Available Knowledge Two databases were utilized to search for applicable research: PubMed and Cumulative Index for 5 Nursing and Allied Health Literature. The following phrases were employed to search both databases: “(Infant* OR Neonat* OR Newborn*) AND (Gastric OR enteral) AND (Residual OR Aspirat*).” Results were limited to studies published in the last 10 years. The search returned 302 articles, and an additional five articles were identified through snowballing. Title and abstract review found that 44 of the articles were applicable. Of these, seven were duplicates between databases, three were inaccessible, two were inaccessible in English, and two were commentaries on other included articles. Non-systematic reviews were excluded, also, resulting in the exclusion of two more articles. In total, 30 articles were considered in this synthesis from the database searches. This process is illustrated in Figure 1. Four case-control studies identified in the literature sought to determine if gastric residual volume measurements are valuable screening tools for necrotizing enterocolitis in human infants (Cobb, Carlo, & Ambalavanan, 2004; Bertino et al., 2009; Gephart, Fleiner, & Kijewski, 2017; Purohit, Mehkarkar, Athalye-Jape, Nathan, & Patole, 2021). Two studies followed gastric residual volumes from birth to the recognition of necrotizing enterocolitis (Bertino et al., 2009; Purohit, Mehkarkar, Athalye-Jape, Nathan, & Patole, 2021). These reported higher mean or median maximal residual volumes between birth and the occurrence of necrotizing enterocolitis in cases compared to controls. However, the timings of the maximal residuals were multiple days—as many as 17 on average (Bertino et al., 2009)— prior to diagnosis. When focusing on comparisons of residuals at timepoints proximal to the onset of necrotizing enterocolitis—72, 48, 24 hours prior in one study and 36, 24, 12, 6, and 0 hours prior in another—one of two relevant studies found a difference in residual volumes at 36 hours prior to onset in the most severe classification of necrotizing enterocolitis, but there were no differences at other severities or other timepoints (Gephart, Fleiner, & Kijewski, 2017; Purohit et al., 2021). One study from 2004 did show an increase in gastric residual volumes in cases in the three and six days prior to diagnosis, but its authors cautioned that there was considerable overlap between cases’ and controls’ residual volumes, making it difficult to differentiate normal from abnormal values (Cobb, Carlo, & Ambalavanan, 2004). This overlap is not surprising, as factors like patient body position prior to aspiration can affect gastric residual 6 volume by as much as 50% (Sangers et al., 2013). In summary, there appears to be some association between gastric residual volumes and necrotizing enterocolitis, but the clinical utility of this association is limited by a delay between the occurrence of the maximal residual volumes and the onset of necrotizing enterocolitis and by insufficient separation between normal and abnormal volumes to make a clinical distinction. These findings agree with those from animal models, which show residual volumes to be of low predictive value for small bowel necrotizing enterocolitis, the type most affecting human preterm infants (Cao et al., 2016; Kappel et al., 2021). The most common purpose of articles reviewed, accounting for 11 of the articles, was to compare patient outcomes with and without routine gastric residual volume measurements. Four randomized controlled trials (Kaur et al., 2015; Parker et al., 2019, 2020; Singh et al., 2018) and three pre-post studies (Akar & Turgut, 2020; Elia, Ciarcià, Miselli, Bertini, & Dani, 2022; Riskin et al., 2017) compared patient outcomes with routine gastric residual volume measurement versus only performing selective residual volume measurement in response to concerning clinical signs, such as abnormal abdominal exam or increasing abdominal girth. Two randomized controlled trials compared patient outcomes with routine gastric residual volume measurement versus absolutely no residual volume measurements (Thomas et al., 2018; Torrazza et al., 2015). The most common patient outcome analyzed was time to full enteral feeding, the definition of this ranging from 120 mL/kg/d to 180 mL/kg/d. Five out of the eight studies examining this outcome found a statistically significant reduction in time to full enteral feedings (Akar & Turgut, 2020; Elia et al., 2022; Kaur et al., 2015; Riskin et al., 2017; Thomas et al., 2018). Two studies found a non-significant reduction (Singh et al., 2018; Torrazza et al., 2015). Two additional pre-post studies compared patient outcomes with routine versus no gastric residual volume measurement, but they included the change in residual volume measurement practice as one of many changes associated with the introduction of new feeding protocols (Morton et al., 2018; Staub, Van Vuuren, Priest, McNaught, & Paradisis, 2019). Both reported exclusively favorable or neutral results, but their manipulation of multiple variables limits the interpretation of these results as they relate to residual volume practice. Notably, however, one of these studies included complex neonates and took 7 place in an all-referral newborn intensive care unit (Morton et al., 2018). This demonstrates feasibility for the discontinuation of gastric residual volumes in a population and setting similar to that of this project. Two systematic reviews with meta-analysis of randomized controlled trials and one without metaanalysis have been conducted comparing patient outcomes with and without routine gastric residual volume measurement (Abiramalatha, Thanigainathan, & Ninan, 2019; Kumar et al., 2021; Lin et al., 2022). The most recent meta-analysis, published in 2021, identified and synthesized the five aforementioned randomized controlled trials as well as one additional abstract (Kumar et al., 2021). By meta-analysis, they found that avoiding routine gastric residual volume measurements was associated with shorter time to full enteral feeds, shorter duration of hospitalization, and lower incidence of late onset sepsis. While the earliest systematic review had insufficient evidence to make a judgement, the two most recent systematic reviews concluded that routine residuals should not be recommended. The presumed causal link between avoiding gastric residual volume measurements and shorter time to full enteral feeds is that residual volumes perceived as abnormal prompt iatrogenic interruptions in enteral feeding which would not occur if feeding intolerance was monitored by other signs. Supporting this hypothesis, definitions of feeding intolerance in the literature which include consideration of gastric residual volumes tend to be associated with the highest reported prevalences of feeding intolerance compared to definitions based on other signs (Weeks, Marino, & Johnson, 2021). While it is a theoretical concern that routine aspiration could cause traumatic injury to the gastric mucosa, this concern has not been confirmed. Markers of gastrointestinal function, intestinal inflammation, and gastrointestinal mucosal bleeding were not increased with routine residual volume measurement in a randomized controlled trial (Parker et al., 2020). While there are multiple patient outcomes deleteriously affected by gastric residual volume measurement, there is inadequate evidence to definitively include mucosal damage on that list. Existing literature supports the cessation of routine gastric residual volume measurements in neonates. Because most relevant research excludes medically and surgically complex neonates, the application of this research to the tertiary neonatal intensive care unit is understandably hesitant. 8 However, the burden of proof should be in justifying performing an intervention rather than discontinuing it. It is unreasonable to continue gastric residual volume measurement in these populations with reason to suspect harm and no evidence of benefit. Rationale The theoretical framework underpinning this project is the Stevens Star Model, developed by Kathleen Stevens at the University of Texas’ Academic Center for Evidence-Based Practice (2015). This model describes the transformation knowledge must undergo to inform clinical decisions. It divides the process into five steps, visualized as the five points of a star and included in Appendix A. The first point of the star, “discovery research,” refers to the generation of knowledge through research projects. Once multiple projects have been completed regarding a question, they are considered together in the second point of the star, the “evidence summary.” This point was completed with the literature review at the start of this project. In the third point of the star, “translation to guidelines,” the summary of evidence is used, in combination with expert opinion where evidence is lacking, to produce an evidence-based recommendation. Point four of the star is focused on the execution of the clinical practice guideline, called “practice integration.” The goal of this point is to make practice agree with the new guidelines. Finally, the fifth point of the star evaluates whether the change produced the desired impact in clinical practice and patient outcomes. Specific Aims The objectives of this project were 1) to de-implement the routine, universal aspiration of gastric residual volumes through educational interventions, 2) to evaluate the success of these interventions in producing real change in practice and staff acceptance of the change, and 3) to evaluate the success of the practice change in its effects on relevant patient outcomes. Methods Context This project was conducted at a 51-bed, all-referral level IV newborn intensive care unit in the urban, Intermountain West. Its average 35 admissions per month are neonatal patients requiring specialist 9 or surgical services in addition to neonatology care. It employs approximately 206 newborn intensive care unit nurses, 55 advanced practice providers, and 23 neonatologists and fellows. Nursing coverage often includes some registered nurses “floated” from other units within the children’s hospital or from the central resource staffing pool, which staffs all pediatric units. The clinicians called first by nurses with patient concerns are advanced practice providers, a group composed of neonatal nurse practitioners and a minority of physician assistants. Prior to this project, the decision to aspirate, measure, and consider gastric residual volumes in the assessment of a patient was variable, largely depending upon the provider or nurse caring for a patient that day. The unit’s clinical practice guideline recommended that nursing staff measure gastric residual volumes in every patient prior to each enteral feeding, with instructions to notify providers when >25% of the previous feed’s volume or >1 hour’s feeding volume were obtained. Some limited attempts to discontinue this practice had been attempted in previous years, but they were of marginal success due to word-of-mouth communication strategies and the continued presence of the clause to report high residuals in the institution’s enteral tube feeding clinical practice guideline document. Interventions Following a literature review, the third point of Stevens’ Star is translation to guidelines. For this step, a practice change description and educational materials were prepared in collaboration with the neonatal unit’s medical director, the neonatal nurse practitioner quality improvement leader, as well as other neonatologist and nurse practitioner stakeholders. The practice change proposal was presented to the pediatric surgical team, who voiced no objections. The unit policy and the educational materials discontinued gastric residual volume measurement but continued to recommend small-volume aspiration of <1mL of aspirate to assess gastric residual color and to provide a secondary indicator of oro- or nasogastric tube placement. It was also permitted to therapeutically aspirate for the purpose of relieving gastric pressure related to aerophagia. The decision to measure volumes selectively in response to other observed, concerning abdominal signs was left to provider discretion. 10 The de-implementation of gastric residual volume measurement, and the fourth point of Stevens’ Star, began with the removal of reference to gastric residual volumes in evaluating feeding tolerance from the institution’s neonatal enteral tube feeding clinical practice guideline. Multiple interventions were performed to educate staff on the change. First, Situation-Background-Assessment-Recommendation (SBAR) descriptions of the practice change were disseminated through email in the electronic newsletters sent to nursing and providers. A brief presentation was provided in meetings of charge nurses and the leadership team. At these meetings, charge nurses and management in attendance were encouraged to voice questions and concerns, and those were addressed in real-time. As nursing buy-in was crucial to the success of the practice change, in-person, one-on-one or small group educational in-services were performed for nursing staff. These were performed on the nurses’ shifts and consisted of a verbal explanation of the change, distribution of an educational pamphlet describing the practice change and its supporting evidence, and opportunities to voice and have addressed any questions or concerns. Upon completion of the in-service, nurses were asked to initial that they understood that routinely measuring gastric residual volumes is now contrary to unit policy. It was targeted to achieve 80% participation in the in-services. Additionally, as a reminder to the unit’s nurses and to inform nurses floating in, laminated signs were posted on the milk warmers for two months; they included a QR code to the nursing SBAR for more information. The SBARs for nursing and providers, informational pamphlet, initialed attestation, and milk warmer signs are included as Appendices B-F. All educational interventions were performed by the project lead, a neonatal Doctor of Nursing Practice student. Study of the Interventions For the fifth point of Stevens’ star, the success of the interventions in producing real change in practice and achieving staff acceptance of the change were evaluated through pre- and post-intervention electronic surveys in the Research Electronic Database Capture (REDCap) tool, hosted at the University of Utah. The surveys were distributed by email to neonatal nurses and advanced practice providers employed on this unit. Because practice was being assessed by self-report through these surveys, respondent anonymity was imperative to ensure that responses were not affected by fear of disciplinary 11 action. For this reason, the surveys were distributed by anonymous links. Post-intervention surveys were distributed four weeks after the completion of the educational interventions. They remained open for two weeks. Data was imported from REDCap to Statistical Package for Social Sciences (SPSS) software for analysis. The success of the interventions in producing improvements in relevant patient outcomes was evaluated by pre-post comparison, obtained through retrospective chart review for pre-intervention data and prospective chart review for post-intervention data. Data was collected for all infants receiving enteral feeds via enteric feeding tube admitted in the two-months before the initiation of and two-months after the completion of the project’s educational interventions until the ends of those two-month periods. Data was entered into SPSS software for analysis. Measures The pre- and post-intervention surveys for nurses consisted of five questions, included as appendices G and H. The first two questions were to establish demographic information, asking about years of experience in nursing and additional roles on the unit, such as charge nurse or tele-critical care nurse. The next question asked about nurses’ practice regarding gastric residual volume measurement, prompting them to select if they measured residual volumes routinely in every patient; only in some patients, depending on risk factors and history; or not in any patients. The pre-intervention survey asked about their practice prior to and at the time of completion of the survey, and the post-intervention survey asked about their practice in the one-month post-intervention. These responses were compared to estimate change in nursing practice following the educational interventions. The fourth question asked nurses to rate the favorability of their opinions on discontinuing routine residual volume measurement on a Likert scale of 1 to 5. It was compared to estimate nursing’s acceptance of the practice change. The fifth question provided nurses the opportunity to write any specific reservations they held about the change before intervention or problems encountered with the change after the intervention. It was qualitatively synthesized to determine any additional needs for education or revisions in the guidelines. 12 A limitation of the nursing surveys is that the estimations of practice change are based on selfreported data. This limitation is attenuated by the wholly anonymous nature of the surveys. However, the anonymous link distribution imposes the secondary limitation that the pre- and post-responses could not be matched, yielding susceptibility to bias from demographic differences in what would optimally be identical samples. The final two questions of the nursing survey, asking for their opinions on, reservations about, or problems encountered with the discontinuation of routine gastric residual volume measurement were distributed to advanced practice providers. The earlier questions were omitted in this group because aspiration of residuals is a bedside nursing task not performed by advanced practice providers. Some patient outcome measures were based upon the outcomes determined to be of importance in the stakeholder consensus meeting reported for the ongoing GASTRIC feasibility study, a trial to discontinue gastric residual volume monitoring in multiple sites in the UK (Lyvonne N. Tume et al., 2021; L.N. Tume et al., 2020). These outcomes included change in weight between birth and discharge and incidence of iatrogenic infections, catheter-associated bloodstream infections, and aspiration pneumonia. Because this unit’s patients are admitted at varied postnatal and gestational ages and have extremely variable lengths of stay, weight gain was assessed by comparison of the average change in the z-score of patient weight per day of hospitalization. Infective incidences were compared as episodes of blood and tracheal aspirate culture positivity per patient-days. This has the limitation of excluding culture-negative or unrecognized sepsis, but this was determined necessary to maintain objectivity. Additional patient outcome measures evaluated were time to full enteral feedings and episodes of interruptions in feeding for feeding intolerance. Time to full enteral feedings was selected as it is the most investigated outcome in the literature. It was defined as attainment of 140 mL/kg/d of enteral feedings after admission or after a period of non-per-os. Incidence of interruptions in feeding for feeding intolerance per patient days was evaluated because gastric residual volume measurements tend to prompt more determinations of feeding intolerance than other metrics (Weeks et al., 2021). Non-per-os orders placed for surgery, blood transfusions, or other interventions were excluded from these counts. These 13 exclusions are especially important in the medically complex neonates making up this population; in the subpopulation of infants with congenital heart disease, only 32.2% of feeding interruptions are for gastrointestinal concerns (Qi, Li, Cun, & Li, 2013). Incidence of necrotizing enterocolitis was included as a balancing measure. It was measured as episodes of Bell’s stage II necrotizing enterocolitis per patient-days, excluding episodes in which necrotizing enterocolitis was the admitting diagnosis. However, the effect size would need to be profound for a project of this power to accurately detect changes in necrotizing enterocolitis rates. Analysis Descriptive statistics were applied to demographic data and the significance of differences between those groups was evaluated by Pearson X2. The author intended to compare means by independent samples t-test for parametric data or Mann-Whitney U-test for non-parametric data. While surveys were collected from the same population of staff, it was uncertain how many participants were present in both the pre- and post-intervention surveys due to low participation rates, so it was reasonable to treat them as independent samples. Rare-event frequencies were compared using the MedCalc rate comparison calculator, which utilizes the Exact Poisson method to compare incidence rate ratios (MedCalc Software Ltd., 2023). Free-text survey questions on reservations regarding and problems encountered with the practice change were analyzed by content analysis, with grouping of responses by themes. Ethical Considerations This project was reviewed by a single panelist and determined to be exempt from full review by the Intermountain Healthcare Institutional Review Board. It was also exempt from human subjects review by the University of Utah as a quality improvement project and did not require evaluation. The author has no conflicts of interest to disclose. Results The post-education attestation was initialed by 79.6% (n=164) of 206 nurses. Eighty-eight (42.7%) responded to the nursing pre-intervention survey and 39 (18.9%) responded to the nursing post- 14 intervention survey. The demographic compositions of the responding groups, in terms of years of experience and roles on the unit, were not significantly different by Pearson X2 analysis. They are presented in Table 1. Prior to the educational interventions, nursing practice regarding gastric residual volumes was very mixed, with about half (n=42, 47.7%) measuring gastric residual volumes in some of their patients based on their personal assessment of risk factors and history, one quarter (n=22, 25.0%) measuring residual volumes in every patient, and the remaining quarter (n=23, 27.7%) never measuring residual volumes. They were most often (n=32, 36.6%) “somewhat in favor” of discontinuing gastric residual volume measurements, with a mean of 3.44 (SD = 1.25) out of 5 on a Likert scale. The most common (n=6) concern described was that discontinuing residual volume measurements would remove a valuable tool in screening for necrotizing enterocolitis. The next most common concerns, in order of decreasing frequency, were that residuals were important to verify feeding tube placement (n=4), screen for feeding intolerance or poor gastric motility (n=5), relieve gastric pressure in the context of aerophagia (n=3), and prevent aspiration by limiting emesis (n=2). These concerns were anticipated, so they were already addressed in the educational pamphlets and nursing SBAR communications and no modifications of those materials were necessary. The charge nurse meeting presentation was designed around these concerns, including a section in which common concerns were quoted and then evidence provided to address them. The slides are included as Appendix I. The presentation was designed this way because those in attendance may have shared those concerns and, as charge nurses, they should be prepared to respond to them if asked. The nursing post-intervention survey demonstrated a statistically significant change in nursing practice away from gastric residual volume measurement by Pearson X2 test (2, N=127) =32.934, p <.001. Shown in Table 2, 82.1% (n=32) of nurses reported that they do not check gastric residual volumes in any patients, and only 5.1% (n=2) reported continuing to check in every patient. The nurses were also significantly more in favor of discontinuing gastric residual volume measurement by Mann-Whitney Utest, with a post-intervention mean Likert scale rating of 4.05 out of 5 (SD = 1.19, p = .005). Most— 15 79.5% (n=31)—were either somewhat in favor or entirely in favor of discontinuing gastric residual volume measurement. This is illustrated in Figure 2. Their free-text responses about issues encountered were most often about verifying tube placement (n=3) and therapeutic aspiration to relieve gastric air (n=3). Two nurses expressed that they had issues with parents adjusting to the new routine, and one expressed confusion on what to chart in the electronic health record following the change. These responses indicated a need for clarifying education, so an additional communication was posted in the nursing newsletter, included as Appendix J. Eighteen (32.7%) of advanced practice providers responded to the pre-intervention survey and 10 (18.1%) responded to the post-intervention survey. They were generally in favor of discontinuing gastric residual volume measurements both before and after their educational SBAR, though there was a nonsignificant shift towards increased favorability (Likert 4.78 out of 5, SD=0.55 pre-education; 4.90, SD=0.32 post-education, p= .76). Before education, they expressed some of the same concerns as nursing, that residuals may be helpful to assess placement (n=2), screen for feeding intolerance (n=2), avoid aspiration related to emesis (n=1), or relieve gastric air from aerophagia (n=1). Following the educational SBAR, the only concern raised was, “nurses seem uncomfortable not checking residuals, but it is getting better.” The demographic information for the patients admitted in the 2-months pre- and 2-months postintervention are included in Table 3. They did not significantly differ by sex, admitting diagnosis, gestational age at birth, or postnatal age on admission, but the gestational age at admission of the postimplementation group was significantly younger than the pre-intervention group by Mann-Whitney U-test (36.74 weeks, SD=4.36 versus 38.81, SD = 3.54, p = .031). Regarding the social determinants of health, most patients in both groups did not have a race identified in the electronic health record. Most patient families were English speaking, and language spoken was not significantly different between groups. Patient outcome data is provided in Table 4. There were no significant differences between the groups on time to full enteral feeds, weight gain, interruptions in feedings for feeding intolerance, number of episodes of culture-positive bacteremia, or number of episodes of necrotizing enterocolitis. There were 16 no episodes of central line-associated bloodstream infections or culture-positive pneumonia in either group. Only 26 patients and 25 patients in the pre- and post-intervention groups respectively could be included in the time to full enteral feeds analysis because many patients either did not reach full enteral feeds within the study period (n=26 pre-intervention, n=24 post-intervention), were never non-per-os (n=9 pre-intervention, n=5 post-intervention), or the exact timing could not be determined because they were primarily breast-fed around that point (n=1 pre-intervention, n=1 post-intervention). These reasons for exclusion did not differ between groups by Pearson X2 (2, N=117) = 1.026, p = .59. There was a nonsignificant increase in time to full enteral feeds pre- to post-intervention. Discussion Summary The educational interventions in this project were successful in effecting change in nursing practice and opinion. The fraction of nurses reporting to routinely measure gastric residual volumes significantly decreased, and nursing opinion of discontinuing gastric residual volumes significantly improved by Likert scale rating. Advanced practice providers were generally in favor of the practice change both before and after its introduction. The practice change produced no demonstrable changes in patient weight gain, time to full enteral feeds, episodes feeding interruptions for feeding intolerance, episodes of bacteremia, or episodes of necrotizing enterocolitis during the study period. Interpretation De-implementation of ineffective healthcare interventions is important in the minimization of patient harm and the provision of efficient care (Norton & Chambers, 2020). This project demonstrates the feasibility of de-implementation of gastric residual volume measurement in an all-referral newborn intensive care unit. While a vital branch of implementation science, de-implementation carries unique challenges, particularly in persuading staff acceptance of the change. Among health professionals, common barriers to de-implementation include personal experience with negative events, cognitive dissonance, and fear of 17 malpractice accusations (Norton & Chambers, 2020). In dialoguing with nurses in one-on-one educational in-services and receiving nurse and advance-practice provider written concerns in surveys, these barriers were evident. Past experience with negative events generally refers to anecdotes in which appropriate de-implementation led to a poor outcome or poor reception (Norton & Chambers, 2020). It could also be extended to anecdotes in which someone perceived the ineffective intervention to avert a poor outcome. A very experienced charge nurse, in their one-on-one educational session, described events in their career in which a high residual volume contributed to the identification of necrotizing enterocolitis. While other signs were present in those cases, they described a reluctance to forgo residual volume measurements because of their experiences. Cognitive dissonance may result from conflict between the belief a caregiver holds that he or she is providing high-quality care with an intervention and emerging evidence that the intervention was not actually valuable or caused harm; this unpleasant dissonance can be mitigated by rejecting the emerging evidence (Norton & Chambers, 2020). Another experienced nurse, in their one-on-one in-service, communicated that they had contributed to writing the original policy requiring gastric residual volume measurement many years ago. They were extremely skeptical of the more recent evidence presented. No caregivers specifically referenced a fear of malpractice accusations verbally or in writing. However, the survey comments concerned about explaining the change to parents and documenting information related to gastric residuals may be related to this underlying concern. Despite these barriers, the pre- and post-intervention surveys demonstrated strong staff acceptance of the change and adherence to the new practice expectations. There was high agreement between nurses’ self-reported favorability of opinion on the change and actual practice; 79.5% of nurses reported somewhat or entirely positive opinions on discontinuing the practice and 82.1% reported that they now avoid routine residual volume measurements except by provider order. This illustrates the importance of persuading healthcare professionals’ acceptance of a de-implementation practice change in order to achieve translation to practice. Despite the changes achieved in nursing practice, this project failed to reproduce the 18 improvements observed in the literature in time to full enteral feeds, weight gain, episodes feeding interruptions for feeding intolerance, or episodes bacteremia or central line infections during the study period. This may be due to multiple factors. First, the advanced practice providers expressed an enthusiasm to discontinue gastric residual volume measurement even in their pre-intervention survey. As nursing would report large residuals to them, advanced practice providers’ skepticism of the value of residual volumes may have prevented residual volumes from affecting the plan of care. This buffering of effect would not have been present at institutions where providers either valued residual volumes or based decisions on them by protocol. Second, as many nurses had independently decided to stop residual volume measurement even before the change in protocol, this data was not a true reflection of universal residual volume measurement versus no measurement and should have a correspondingly smaller effect size. Third, there were differences in patient demographics pre- and post-intervention. The younger gestational age in the post-implementation group could be a confounding factor due to generally slower feeding advancement in more premature infants. Finally, the relatively short data collection window may have been inadequate to demonstrate changes in the frequency of rare events. It is encouraging that there was not an increase in necrotizing enterocolitis or pneumonia rates, but interpretation of these balancing measures is similarly affected by that limitation. Limitations There were a couple of necessary limitations in data measurement. First, nursing practice before and after the change was assessed by self-report. This was necessary because a nurse’s gastric residual management practice could not be obtained with certainty from the electronic health record. Nurses inconsistently charted in a “gastric residual volume” field, but if they charted 1-2 mL, a chart reviewer would not be able to differentiate if that was an intentionally small volumes aspirated just to assess placement and color versus the entirety of the gastric contents aspirated to assess volume. However, selfreport may introduce bias; a nurse could over-report their compliance with guidelines out of fear of retaliation. This risk was combatted by the wholly anonymous nature of the surveys. Though, opting not to collect identifiers from the staff and distributing the surveys through universal rather than personalized 19 links prohibited responses from being paired pre- to post-intervention. This was another limitation. There were relatively low response rates to the surveys. In order to avoid the impression of compelling participation, the surveys were clearly identified as voluntary at the recommendation of the Institutional Review Board, and no requests to take the survey were issued during in-person communications. Incomplete participation was a necessary consequence of these ethical safeguards. A limited window for data collection in this student-led project meant that relatively small samples of patient data could be collected. The samples may have been too small to demonstrate changes in patient outcomes, particularly for rare events. Furthermore, sepsis and pneumonia were only counted if they were culture-positive, and necrotizing enterocolitis was only counted if it was Bell’s stage II or greater. These limitations may lead to under-recognition of these events, but they were necessary to maintain objectivity. The practice recommendations in this project left the decision to aspirate selectively in response to other concerning signs to provider discretion. About half of studies in the literature permitted selective gastric residuals after de-implementing routine residuals, while the other half no longer performed residuals at all. No studies were identified which compared selective versus no residuals. This gap in the literature meant the authors were uncomfortable writing a definitive recommendation regarding selective residuals for this project. This recommendation may require revision as more evidence becomes available. Additionally, the practice recommendations disseminated in this project recommended routine aspiration of small volumes of gastric contents to assess enteric tube placement. Gastric aspirates are unreliable indicators of enteric feeding tube placement. The presence of straw-colored aspirate does not ensure that the tube does not terminate in the respiratory system as respiratory secretions may appear similar (Parker et al., 2015). The absence of aspirate does not necessarily indicate malposition, as 38% of aspiration attempts will not yield aspirate (Nyqvist, Sorell, & Ewald, 2005). While use of this tube placement verification technique is not supported by evidence, there were also no satisfactory indicators to replace it. Gastric aspirate pH testing had been previously trialed at this institution, but frequent local 20 use of acid suppressive medication limited its utility. Routine radiographic verification would result in unacceptable radiation exposure for patients. Auscultation with injection of air may produce a “whoosh” even if the feeding tube terminates in the esophagus. The Nose-Ear-Midpoint between the xiphoid process and the Umbilicus (NEMU) measurement technique is recommended by the Neonatal Resuscitation Program but results in accurate placement in only 90% of cases (Parker, Withers, & Talaga, 2018). Accuracy improves when multiple placement confirmation methods are combined (Parker et al., 2018). Therefore, this project’s guidelines continued to recommend aspiration as a secondary measure of placement, to be used in combination with the NEMU method and confirmation on incidental radiographs. This recommendation may also require revision in the future should more reliable methods of placement verification emerge. Conclusions The routine measurement of gastric residual volumes in the neonatal intensive care unit has not been demonstrated to provide benefit, and increasing evidence suggests it may cause harm. Deimplementation of existing healthcare interventions is an important variety of quality improvement, and it carries its own unique challenges, particularly in obtaining buy-in of healthcare professionals. This project demonstrated the feasibility of de-implementing gastric residual volume measurement by educational interventions in an all-referral newborn intensive care unit, with success in both changing practice and gaining staff acceptance of the change. While the project failed to produce demonstrable changes in patient outcomes within the study period, this may be due to the limited data collection period and multiple confounders. Further research is necessary to fine-tune the nursing management of enteric feeding tubes in neonates; this should include investigation into the value of selective gastric residual volume measurements and exploration to determine reliable ways to verify enteric tube placement. 21 Acknowledgements The author would like to acknowledge John Marshall Smith, DNP, FNP-BC, CCRN, who provided his expertise as content expert to this project and whose Doctor of Nursing Practice Scholarly Project in an adult intensive care unit provided inspiration for multiple components of this project (Smith, 2020; Smith, Smith, & Robinson, 2022). She would also like to thank Con Yee Ling, MD and Kim Friddle, PhD, APRN, NNP-BC for their involvement in developing the practice change recommendations. Jenny Hearne, APRN, NNP-BC provided guidance as project chair. 22 References Abiramalatha, T., Thanigainathan, S., & Ninan, B. (2019). Routine monitoring of gastric residual for prevention of necrotising enterocolitis in preterm infants. Cochrane Database of Systematic Reviews(7). doi:10.1002/14651858.CD012937.pub2 Akar, S., & Turgut, M. (2020). Do we control gastric residuals unnecessarily in premature newborns? AGRA study: a voidance of g astric r esidual a spiration. World Journal of Pediatric Surgery, 3(1), e000056. doi:10.1136/wjps-2019-000056 Bertino, E., Giuliani, F., Prandi, G., Coscia, A., Martano, C., & Fabris, C. (2009). Necrotizing enterocolitis: risk factor analysis and role of gastric residuals in very low birth weight infants. Journal of Pediatric Gastroenterology and Nutrition, 48(4). doi: 10.1097/MPG.0b013e31817b6dbe Cao, M., Andersen, A. D., Li, Y., Thymann, T., Jing, J., & Sangild, P. T. (2016). Physical activity and gastric residuals as biomarkers for region-specific nec lesions in preterm neonates. Neonatology, 110(4), 241-247. doi:10.1159/000445707 Cobb, B. A., Carlo, W. A., & Ambalavanan, N. (2004). Gastric residuals and their relationship to necrotizing enterocolitis in very low birth weight infants. Pediatrics, 113(1), 50-53. doi:10.1542/peds.113.1.50 Dorling, J., Tume, L., Arch, B., Woolfall, K., Latten, L., Roper, L., . . . Gale, C. (2020). Gastric residual volume measurement in British neonatal intensive care units: a survey of practice. BMJ Paediatr Open, 4(1), e000601-e000601. doi:10.1136/bmjpo-2019-000601 Elia, S., Ciarcià, M., Miselli, F., Bertini, G., & Dani, C. (2022). Effect of selective gastric residual monitoring on enteral intake in preterm infants. Italian Journal of Pediatrics, 48(1), 30. doi:10.1186/s13052-022-01208-7 Gephart, S. M., Fleiner, M., & Kijewski, A. (2017). The conNECction between abdominal signs and necrotizing enterocolitis in infants 501 to 1500 g. Advances in Neonatal Care, 17(1). doi: 10.1097/ANC.0000000000000345 23 Hobson, D., Spence, K., Trivedi, A., & Thomas, G. (2019). Differences in attitudes to feeding post repair of Gastroschisis and development of a standardized feeding protocol. BMC Pediatrics, 19(1), 475. doi:10.1186/s12887-019-1858-z Kappel, S. S., Sangild, P. T., Hilsted, L., Hartmann, B., Thymann, T., & Aunsholt, L. (2021). Gastric residual to predict necrotizing enterocolitis in preterm piglets as models for infants. Journal of Parenteral and Enteral Nutrition, 45(1), 87-93. doi: 10.1002/jpen.1814 Kaur, A., Kler, N., Saluja, S., Modi, M., Soni, A., Thakur, A., & Garg, P. (2015). Abdominal circumference or gastric residual volume as measure of feed intolerance in VLBW infants. Journal of Pediatric Gastroenterology and Nutrition, 60(2), 259-263. doi:10.1097/MPG.0000000000000576 Krishnamurthy, S., Gupta, P., Debnath, S., & Gomber, S. (2010). Slow versus rapid enteral feeding advancement in preterm newborn infants 1000–1499g: a randomized controlled trial. Acta Paediatrica, 99(1), 42-46. doi:https://doi.org/10.1111/j.1651-2227.2009.01519.x Kumar, J., Meena, J., Mittal, P., Shankar, J., Kumar, P., & Shenoi, A. (2021). Routine prefeed gastric aspiration in preterm infants: a systematic review and meta-analysis. European Journal of Pediatrics, 180(8), 2367-2377. doi:10.1007/s00431-021-04122-y Lago Rivero, N., Rodríguez Rodríguez, M., Piñeiro Corrales, G., Paradela Carreiro, A., Gayoso Rey, M., & Arroyo Conde, C. (2013). PP246-MON Incidence of catheter-related infections in preterm infants with parenteral nutrition. Clinical nutrition (Edinburgh, Scotland), 32, S214-S214. doi:10.1016/S0261-5614(13)60556-7 Lin, T., Hu, J., Zhang, L., Qin, X., Liu, X., Lan, Y., . . . You, T. (2022). Promoting enteral tube feeding safety and performance in preterm infants: A systematic review. International Journal of Nursing Studies, 128, 104188. doi: 10.1016/j.ijnurstu.2022.104188 MedCalc Software Ltd. (2023). Comparison of two rates. MedCal: Easy to Use Statistical Software. https://www.medcalc.org/calc/rate_comparison.php 24 Morton, S. U., Belfort, M. B., Kahlon, P. S., Hajizadeh Barfjani, S., Rudie, C., Hashim, E., . . . Huh, S. Y. (2018). Reducing time to initiation and advancement of enteral feeding in an all-referral neonatal intensive care unit. Journal of Perinatoly, 38(7), 936-943. doi:10.1038/s41372-018-0110-2 Norton, W. E., & Chambers, D. A. (2020). Unpacking the complexities of de-implementing inappropriate health interventions. Implementation Science, 15(1), 2. doi:10.1186/s13012-019-0960-9 Nyqvist, K. H., Sorell, A., & Ewald, U. (2005). Litmus tests for verification of feeding tube location in infants: evaluation of their clinical use. Journal of Clinical Nursing, 14(4), 486-495. doi: 10.1111/j.1365-2702.2004.01074.x Parker, L. A., Torrazza, R. M., Li, Y., Talaga, E., Shuster, J., & Neu, J. (2015). Aspiration and Evaluation of Gastric Residuals in the Neonatal Intensive Care Unit: State of the Science. Journal of Perinatal and Neonatal Nursing, 29(1), 51-59. doi:10.1097/JPN.0000000000000080 Parker, L. A., Weaver, M., Murgas Torrazza, R. J., Shuster, J., Li, N., Krueger, C., & Neu, J. (2019). Effect of gastric residual evaluation on enteral intake in extremely preterm infants: a randomized clinical trial. JAMA Pediatrics, 173(6), 534-543. doi:10.1001/jamapediatrics.2019.0800 Parker, L. A., Weaver, M., Murgas Torrazza, R. J., Shuster, J., Li, N., Krueger, C., & Neu, J. (2020). Effect of aspiration and evaluation of gastric residuals on intestinal inflammation, bleeding, and gastrointestinal peptide level. Journal of Pediatrics, 217, 165-171.e162. doi:10.1016/j.jpeds.2019.10.036 Parker, L. A., Withers, J. H., & Talaga, E. (2018). comparison of neonatal nursing practices for determining feeding tube insertion length and verifying gastric placement with current best evidence. Advances in Neonatal Care, 18(4), 307-317. doi:10.1097/ANC.0000000000000526 Perumbil Pathrose, S., Spence, K., Taylor, C., Psalia, K., Schmied, V., Dahlen, H., . . . Foster, J. (2021). A cross-sectional survey of enteral feeding tube placement and gastric residual aspiration practices: need for an evidence-based clinical practice guideline. Advances in Neonatal Care, 21(5), 418-424. doi:10.1097/ANC.0000000000000822 25 Purohit, G., Mehkarkar, P., Athalye-Jape, G., Nathan, E., & Patole, S. (2021). Association of gastric residual volumes with necrotising enterocolitis in extremely preterm infants—a case–control study. European Journal of Pediatrics, 181(1), 253-260. doi:10.1007/s00431-021-04193-x Qi, J., Li, Z., Cun, Y., & Li, X. (2013). Causes of interruptions in postoperative enteral nutrition in children with congenital heart disease. Asia Pacific Journal of Clinical Nutrition, 26(3), 402-405. doi:10.6133/apjcn.032016.05 Riskin, A. M. D. M. H. A., Cohen, K. M. D., Kugelman, A. M. D., Toropine, A. M. D., Said, W. M. D., & Bader, D. M. D. M. H. A. (2017). The impact of routine evaluation of gastric residual volumes on the time to achieve full enteral feeding in preterm infants. Journal of Pediatrics, 189, 128-134. doi:10.1016/j.jpeds.2017.05.054 Sangers, H., de Jong, P. M., Mulder, S. E., Stigter, G. D., van den Berg, C. M., te Pas, A. B., & Walther, F. J. (2013). Outcomes of gastric residuals whilst feeding preterm infants in various body positions. Journal of neonatal nursing : JNN, 19(6), 337-341. doi:10.1016/j.jnn.2012.12.003 Singh, B., Rochow, N., Chessell, L., Wilson, J., Cunningham, K., Fusch, C., . . . Thomas, S. (2018). Gastric residual volume in feeding advancement in preterm infants (grip study): a randomized trial. Journal of Pediatrics, 200, 79-83.e71. doi:10.1016/j.jpeds.2018.04.072 Smith, J. M. (2020). Eliminating routine gastric residual volume assessments in the intensive care setting. Georgia College Doctor of Nursing Practice (DNP) Translational and Clinical Research Projects, 42. Retrieved from https://kb.gcsu.edu/dnp/42 Smith, J. M., Smith, M., & Robinson, K. N. (2022). Using nurse-driven protocols to eliminate routine gastric residual volume measurements: a retrospective study. Critical care nurse, 42(4), e1-e10. doi:10.4037/ccn2022584 Staub, E., Van Vuuren, A., Priest, C., McNaught, E., & Paradisis, M. (2019). Introduction of a standardized feeding protocol without gastric aspirates improves enteral nutrition in vlbw infants. Journal of Paediatrics and Child Health, 55(S1), 102-102. doi:1111/jpc.14410_143 Stevens, K. (2015). STAR Model. University of Texas Health, San Antonio, School of Nursing. 26 https://www.uthscsa.edu/academics/nursing/star-model. Thomas, S., Nesargi, S., Roshan, P., Raju, R., Mathew, S., P, S., & Rao, S. (2018). Gastric residual volumes versus abdominal girth measurement in assessment of feed tolerance in preterm neonates: A Randomized Controlled Trial. Advances in Neonatal Care, 18(4). doi: 10.1097/ANC.0000000000000532 Torrazza, R. M., Parker, L. A., Li, Y., Talaga, E., Shuster, J., & Neu, J. (2015). The value of routine evaluation of gastric residuals in very low birth weight infants. Journal of Perinatology, 35(1), 57-60. doi:10.1038/jp.2014.147 Tume, L. N., Arch, B., Woolfall, K., Roper, L., Deja, E., Jones, A. P., . . . Dorling, J. (2021). Determining optimal outcome measures in a trial investigating no routine gastric residual volume measurement in critically ill children. JPEN Journal of Parenteral and Enteral Nutrition, 45(1), 79-86. doi:10.1002/jpen.1817 Tume, L. N., Valla, F. V., Dorling, J., Gale, C., Pathan, N., Arch, B., . . . Ecclestone, H. (2019). MONPO586: Gastric residual volume to guide enteral feeding in UK PICUs and neonatal units. Clinical Nutrition (Edinburgh, Scotland), 38, S276-S276. doi:10.1016/S0261-5614(19)32419-7 Tume, L. N., Woolfall, K., Arch, B., Roper, L., Deja, E., Jones, A. P., . . . Dorling, J. (2020). Routine gastric residual volume measurement to guide enteral feeding in mechanically ventilated infants and children: the GASTRIC feasibility study. Health Technology Assessment, 24(23), 1-120. doi:10.3310/hta24230 Weeks, C. L., Marino, L. V., & Johnson, M. J. (2021). A systematic review of the definitions and prevalence of feeding intolerance in preterm infants. Clinical Nutrition, 40(11), 5576-5586. doi:10.1016/j.clnu.2021.09.010 Wiechers, C., Bernhard, W., Goelz, R., Poets, C. F., & Franz, A. R. (2021). Optimizing early neonatal nutrition and dietary pattern in premature infants. International Journal of Environmental Research and Public Health, 18(14). doi:10.3390/ijerph18147544 27 Tables and Figures Table 1 Nursing Survey Demographics Pre-intervention (n=88) n % Nursing experience Less than 1 year 1 to less than 3 years 3 to less than 5 years 5 to less than 10 years 10 or more years Roles Bedside nurse Charge nurse Tele-critical care nurse 13 17 9 14.8% 19.3% 10.2% Post-intervention (n=39) n % ` 6 15.4% 5 12.8% 4 10.3% 19 30 21.6% 34.1% 7 17 17.9% 43.6% 87 14 15 98.9% 15.9% 17.0% 39 4 8 100% 10.3% 20.5% Х2 (df) 1.499 (4) p .827 0.447 (1) 0.710 (1) 0.219 (1) .504 .399 .640 28 Table 2 Nursing Practice Survey Results Pre-intervention (n=88) n % Post-intervention (n=39) n % 23 27.3% 32 82.1% “Only in some patients, 42 depending on risk factors and history” 47.7% 5 12.8% “Yes, in every patient” 25.0% 2 5.1% Do you routinely measure gastric residual volumes in your patients? “No, I do not measure residual volumes in any patients” 22 Х2 (df) 32.934 (2) p <.001 29 Table 3 Patient Demographics PMAa at birth, weeks PMAa on admission, weeks Day of life on admission Sex Male Admitting diagnosis Gastrointestinal Cardiac Renal Respiratory Infectious Neurologic Genetic Endocrine/Metabolic Other Language spoken by family English Spanish American sign language Race White Black Hispanic Asian Pacific Islander No data a PMA= post-menstrual age. Pre-intervention (n=62) M SD 36.35 4.15 38.81 3.54 17.24 33.9 Post-intervention (n=55) M SD 35.89 4.46 36.74 4.36 5.91 7.96 n % n % 41 66.1% 31 56.4% 17 10 1 13 1 9 7 1 3 27.4% 16.1% 1.6% 21.0% 1.6% 14.5% 11.3% 1.6% 4.8% 17 12 3 5 0 10 1 2 5 p .672 .031 .420 X2 (df) 1.17 (1) p .279 13.45 (9) .217 6.33 (3) .096 6.78 (5) .238 30.9% 21.8% 5.5% 9.1% 0.0% 18.2% 1.8% 3.6% 9.1% 60 1 1 96.8% 1.6% 1.6% 51 4 0 92.7% 7.3% 0.0% 21 0 0 4 0 39 33.9% 0.0% 0.0% 3.2% 0.0% 62.9% 21 2 1 0 1 30 38.2% 3.6% 1.8% 0.0% 1.8% 50.4% 30 Table 4 Patient Outcome Data Time to full enteral feedingsc, days Post-intervention (n=25) M SD/CIa M SD/CIa IRRb p 5.5 3.1 7.6 3.9 - .052 IRRb p Pre-intervention (n=62) Post-intervention (n=55) M M SD/CIa SD/CIa Weight gain, ΔZscore/days hospitalized -0.017 0.079 -0.005 0.181 - .986 Episodes feeding interruption for intolerance, per 1000 patient-days 3.09 0.64-9.02 8.05 0.35-15.90 2.61 .156 Episodes culturepositive bacteremia, per 1000 patientdays 1.03 0.03-5.73 2.01 0.24-7.27 1.96 .638 Episodes necrotizing enterocolitis, per 1000 patient-days 3.08 0.64-9.02 2.01 0.24-7.27 0.651 .670 a 95% confidence interval b IRR=Incidence c Pre-intervention (n=26) rate ratio Full enteral feedings= 140 mL/kg/d. 31 Figure 1 Literature Review 32 Figure 2 Nursing Survey: Opinion on Discontinuing Gastric Residual Volume Measurement Pre-implementation Post-implementation Note. Values were defined as: 1= Entirely opposed, gastric residual volume measurement should continue. 2= Somewhat opposed 3= Neutral 4= Somewhat in favor 5= Entirely in favor, gastric residual volume measurement should stop. 33 Appendix A Stevens’ Star Model 34 Appendix B SBAR for Nursing Situation: [Hospital] Newborn Intensive Care Unit is officially recommending against the routine, universal aspiration of gastric residual volumes in patients receiving enteral feedings via enteric feeding tubes. This refers to the regular aspiration of gastric contents from an enteric feeding tube until no more liquid can be obtained with the intention of evaluating the volume obtained. Background: Nursing practice in [Hospital’s] NICU has been inconsistent, with the decision to aspirate and measure gastric residual volume measurements depending largely on the nurse and provider caring for a given patient. Other [Hospital System] Newborn Intensive Care Units no longer practice gastric residual volume aspiration. It has been removed from the “Enteral gastric tube feedings, neonatal” clinical practice guideline. An increasing body of evidence is recommending against the routine, universal measurement of gastric residual volumes in neonates.  Utilizing gastric residual volumes as a component in assessing for feeding intolerance results in more frequent determinations of feeding intolerance and secondary interruptions in enteral feeding. In the last 10 years, 5 randomized controlled trials and 3 pre-post studies examined patient outcomes after discontinuing routine gastric residual measurements. In randomized controlled trials, discontinuing residual checks is associated with shorter time to full enteral feeds, shorter hospitalization, and lower incidence of late onset sepsis (Kumar et al., 2021). The prepost studies similarly found a reduction in time to full enteral feeds (Akar & Turgut, 2020; Elia et al., 2022; Riskin et al., 2017).  Gastric residual volume is very dependent on factors unrelated to the patient’s physiologic gastric motility, including the diameter of the feeding tubing, patient positioning, and viscosity of the aspirate (Parker et al., 2015; Sangers et al., 2013)  Gastric residual volume is of limited clinical utility in predicting necrotizing enterocolitis. While mean maximal residuals in the days preceding necrotizing enterocolitis tend to be elevated in children with NEC versus healthy children in case-control trials, there are multiple days between the timing of the elevated residual and the onset of necrotizing enterocolitis, and high overlap between residuals found in healthy versus NEC patients makes it impractical to define normal values (Bertino et al., 2009; Gephart, Fleiner, & Kijewski, 2017; Purohit, Mehkarkar, AthalyeJape, Nathan, & Patole, 2021). Assessment: Gastric residual volume measurement is not evidenced to provide benefit to patients, and there is growing evidence that it may cause harm. Recommendation: Do NOT check gastric residual volumes routinely. This refers to the prefeed aspiration of all gastric contents with the intention of measuring the volume obtained. Gastric residual volumes may be requested in some patients by provider order. Continue to assess for feeding intolerance. Notify physician or licensed independent practitioner for any of the following (stop feeding until consultation is completed):  Rapid increase in abdominal girth  Apnea/bradycardia/desaturation  Aspiration, decreased or adventitious breath sounds  Projectile emesis / bilious emesis/ bloody emesis 35       Bilious or bloody residual Emesis greater than 50% of previous feeding on two separate occasions Diarrhea / constipation Blood in stool (occult or frank blood) Visible bowel loops Absence of bowel sounds Aspiration may still be appropriate  In small volumes to assess color and placement  Therapeutically to relieve gastric pressure  As instructed by a provider You can chart your small volume aspiration used to check placement under "tube placement verification" as "aspiration." Remember, it is recommended to use two methods to verify placement, such as both aspiration and measurement. Reserve the "residual amount field" for situations in which you and provider decide a residual is prudent, as in response to symptoms or in a patient with ordered residuals. References are available here: " 36 Appendix C SBAR for Providers Situation: [Hospital’s] Newborn Intensive Care Unit is officially recommending against the routine, universal aspiration of gastric residual volumes by nursing in patients receiving enteral feedings via enteric feeding tubes. This refers to the regular aspiration of gastric contents from an enteric feeding tube until no more liquid can be obtained with the intention of evaluating the volume obtained. Background: Nursing practice in [Hospital’s] NICU has been historically inconsistent, with the decision to aspirate and measure gastric residual volume measurements depending largely on the nurse and provider caring for a given patient. Other [Hospital System] Newborn Intensive Care Units already no longer practice gastric residual volume aspiration. As of this summer, it was removed from [Hospital System’s] “Enteral gastric tube feedings, neonatal” clinical practice guideline. An increasing body of evidence is recommending against the routine, universal measurement of gastric residual volumes in neonates.  Utilizing gastric residual volumes as a component in assessing for feeding intolerance results in more frequent determinations of feeding intolerance and secondary interruptions in enteral feeding. In the last 10 years, 5 randomized controlled trials and 3 pre-post studies examined patient outcomes after discontinuing routine gastric residual measurements. In randomized controlled trials, discontinuing residual checks is associated with shorter time to full enteral feeds, shorter hospitalization, and lower incidence of late onset sepsis (Kumar et al., 2021). Pre-poststudies similarly found a significant reduction in time to full enteral feeds (Akar & Turgut, 2020; Elia et al., 2022; Riskin et al., 2017).  Gastric residual volume is very dependent on factors unrelated to the patient’s physiologic gastric motility, including the diameter of the feeding tubing, patient positioning, and viscosity of the aspirate (Parker et al.,2015; Sangers et al., 2013)  Gastric residual volume is of limited clinical utility in predicting necrotizing enterocolitis. While mean maximal residuals in the days preceding necrotizing enterocolitis tend to be elevated in children with NEC versus healthy children in case-control trials, there are multiple days between the timing of the elevated residual and onset of necrotizing enterocolitis and high overlap between residuals found in healthy versus NEC patients makes it impractical to define normal values (Bertino et al., 2009; Gephart, Fleiner, & Kijewski,2017; Purohit, Mehkarkar, Athalye-Jape, Nathan, & Patole, 2021). Free full-text of most recent systematic review and meta-analysis regarding gastric residual volume measurement: https://sci-hub.ru/10.1007/s00431-021-04122-y 37 Assessment: Routine gastric residual volume measurement is not shown to provide benefit to patients, and there is growing evidence that it may cause harm. Recommendation: Nursing will no longer routinely aspirate gastric residual volumes prior to feeds. This refers to the prefeed aspiration of all gastric contents with the intention of measuring the volume obtained. You may still request routine gastric residual volume measurement in certain exceptional patients by communication order. Nursing will continue to monitor and notify for  Rapid increase in abdominal girth  Apnea/bradycardia/desaturation  Aspiration, decreased or adventitious breath sounds  Projectile, bilious, or bloody emesis  Bilious or bloody residuals  Emesis greater than 50% of previous feeding on two separate occasions  Diarrhea / constipation  Blood in stool (occult or frank blood)  Visible bowel loops  Absence of bowel sounds They will also continue to aspirate small volumes (<1mL) to check placement and color. References are available here: https://docs.google.com/document/d/1ADSgDra97vh6BkvHSnBi26y8ZB_LvzumttXxzkQAHo/mobilebasic 38 Appendix D Informational Pamphlet 39 Appendix E Initialed Attestation By signing my initials below, I certify my understanding that    The expectation at [Hospital’s] NICU is not to routinely aspirate pre-feed, whole gastric residual volumes in any infants receiving enteral feeds via enteric feeding tube except by provider order Aspiration may still be appropriate o In small volumes to assess color and placement o Therapeutically to relieve gastric pressure o As instructed by a provider Monitoring for feeding intolerance should be completed in accordance with the “Enteral gastric tube feedings, neonatal” clinical practice guideline 40 Appendix F Laminated Signs Placed on Milk Warmers 41 Appendix G Pre-Implementation Survey for Nurses Please complete the following survey regarding gastric residual volume measurement. Your responses are anonymous. They will be used for my Doctor of Nursing Practice scholarly project. Thank you for your time. 1. How many years of experience do you have as a nurse? (select 1) a. Less than 1 year b. 1 to less than 3 years c. 3 to less than 5 years d. 5 to less than 10 years e. 10 or more years 2. Select your role(s) at [Hospital’s] NICU: (select all that apply) a. Bedside RN b. Charge RN c. Tele-critical care RN 3. Do you routinely measure gastric residual volumes in your patients? Take “routinely” to mean at regular intervals independent of the presence or absence of concerning clinical findings. Take “measure gastric residual volumes” to mean aspiration of all gastric contents by aspirating until no more liquid volume can be obtained with the intention of assessing the volume obtained. a. Yes, in every patient b. Only in some patients, depending on risk factors and history c. No, I do not measure residual volumes in any patients 4. On a scale of 1 to 5, how do you feel about not routinely measuring gastric residual volumes in [Hospital’s] Newborn ICU patients? (Likert scale) 1= Entirely opposed, gastric residual volume measurement should continue. 2= Somewhat opposed 3= Neutral 4= Somewhat in favor 5= Entirely in favor, gastric residual volume measurement should stop. 5. If you have specific reservations about not routinely measuring gastric residual volumes at [Hospital’s] NICU, please describe them here: (free text) 42 Appendix H Post-Implementation Survey for Nurses [Hospital] NICU has recently undergone practice change regarding the routine aspiration of gastric residual volumes. The following survey will be used to compare staff practice and perception following implementation of this change. Your responses are anonymous. Please answer honestly. Thank you for your time. 1. How many years of experience do you have as a nurse? (select 1) a. Less than 1 year b. 1-3 years c. 3 to 5 years d. 5 to 10 years e. 10 or more years 2. Select your role(s) at [Hospital] NICU: (select all that apply) a. Bedside RN b. Charge RN c. Tele-critical care RN 3. In the past 1 month, have you routinely measured gastric residual volumes in your patients? Take “routinely” to mean at regular intervals independent of the presence or absence of concerning clinical findings. Take “measure gastric residual volumes” to mean aspiration of all gastric contents by aspirating until no more liquid volume can be obtained with the intention of assessing the volume obtained. This excludes patients with a written provider order for gastric residual volume measurements. a. Yes, in every patient b. Only in some patients, depending on risk factors and history c. No, I have not routinely measured residual volumes in any patients 4. On a scale of 1 to 5, how do you feel about not routinely measuring gastric residual volumes in [Hospital] Newborn ICU patients? (Likert scale) 1= Entirely opposed 2= Somewhat opposed 3= Neutral 4= Somewhat in favor 5= Entirely in favor 5. Have you encountered any problems related to the new expectation to not routinely measure gastric residual volumes at [Hospital] NICU? If so, please describe them here: (free text) 43 Appendix I Charge Nurse Meeting Presentation Slides [Hospital] 44 45 [ H os pi tal ] 46 47 48 49 50 51 Appendix J Follow-Up Communication to Nurses Thank you to all who participated in the surveys regarding gastric residual volume measurements. I would like to clarify some common concerns raised in the post-implementation survey. First, aspirating to check placement: It is still expected to aspirate small volumes (0.5-1mL) of gastric fluid prior to feedings as one indicator of correct enteric tube placement. Aspiration alone is not a reliable indicator of placement-as straw-colored liquid may be aspirated from the lungs, and many correctly placed tubes will not return aspirate-- so this should be combined with a second indicator, such as the NEMU method (below) or tube position on incidental x-rays. The cm depth you pass on at shift change should be based on at least one of those secondary methods. Aspirating small volumes is also helpful so that you can assess color; bilious or hemorrhagic residuals should still be reported. Second, aspirating to remove air: When you have a patient whose stomach contains large amounts of air, like may happen with any type of positive pressure respiratory support, it is still appropriate to aspirate to pull off that air. While you will likely obtain all the liquid contents of the stomach while you aspirate the air, the important distinction is that you are not continuing to draw back from the tube with the purpose of measuring how much liquid volume was there. The act of aspirating has not been shown to be harmful. We are only trying to avoid routine measurement of volume because has not been demonstrated to be helpful in informing clinical decisions in the absence of other concerning signs. Third, charting: Under “Tube placement verification,” you may select “aspiration” to indicate that you aspirated a small volume to check placement. You should select a second method of verification, also, usually “measurement.” “Residual Volume” should usually be left blank. You will use this field in the exceptional cases when you and your provider determine that a residual volume would be valuable to complete a clinical picture. 52 Finally, how to explain the change to parents: Prior to this practice standardization, practice regarding residual aspiration varied considerably nurse-tonurse. Optimally, getting everyone on the same page regarding practice expectations will lead to more consistent modeling for parents and, by extension, more trust in us as caregivers. Some parents who have been here since before the practice change have been taught to expect aspiration of whole gastric volumes. We can explain to these parents that practice is changing in order to be consistent with current evidence. If you have parents with persistent concerns, please do not hesitate to contact me, and I will be happy to speak with them and provide literature.