Gastrointestinal foreign body removal in children with intestinal failure: A single center ten-year retrospective review

Abstract

Background

The significance of foreign body (FB) ingestion in children with Intestinal Failure (IF) is unknown. We aimed to characterize differences in procedural management and clinical outcomes related to gastrointestinal FB removal in pediatric patients with IF compared to children without this condition.

Material and Methods

This ten-year retrospective review utilized electronic healthcare data from a single pediatric center. We evaluated patients who underwent post-esophageal gastrointestinal FB removal procedures between 2011 and 2020 and compared medical history, presentation, procedure type, and procedure outcomes in children with and without IF.

Results

There were 12 patients with IF and 185 controls. All FBs ingested by IF patients carried low intrinsic risk of perforation or obstruction. Esophagogastroduodenoscopy (EGD) was the most common removal procedure in both groups. Patients with IF were more likely to have FBs removed from a post-pyloric location (100 vs. 20.5 %, p < 0.0001), require lower endoscopy (41.7 vs. 9.7 %, p = 0.006), require multiple procedures for definitive removal (41.7 vs. 8.6 %, p = 0.0041), involve > 1 endoscopist or pediatric surgeon (66.7 vs. 7.0 %, p < 0.0001), and require hospital admission (83.3 vs. 28.6 %, p < 0.0001).

Conclusion

FB ingestion by children with IF is associated with increased risk of procedural complexity and hospital admission, even when the object’s intrinsic risk of gastrointestinal hazard is low. These differences may be related to altered intestinal anatomy and dysmotility. Physicians should consider involvement of an advanced proceduralist during removal. Education to prevent ingestion should be part of routine IF care.

Graphical Abstract

Introduction

Between 1995 and 2015, the annual rate of emergency department visits for foreign body (FB) ingestions in children under age six years increased by nearly two-fold [1]. Although FB ingestion in children remains a growing challenge, its significance and incidence in children with Intestinal Failure (IF) is unknown. While early intervention is mandatory for objects lodged in the esophagus, the majority of FBs that traverse the pylorus pass spontaneously and do not pose a significant risk to children with normal gastrointestinal anatomy [1], [2], [3]. In contrast, we noted multiple cases of retained intestinal FBs necessitating complex endoscopic or surgical removal among children with IF at our institution [4]. The 2015 North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition (NASPGHAN) Endoscopy Committee clinical report on FB management provides a well-validated and widely utilized set of guidelines for operative removal [5], namely when the FB is intrinsically hazardous or obstructive; however, these directives do not account for altered intestinal anatomy, gastrointestinal dysmotility, or intestinal adhesions common to IF [6], [7], [8], such that their applicability in this population is less clear.

Given these factors, we hypothesized that FB ingestion in children with IF would be associated with increased procedural complexity. Our primary aim was to characterize differences in procedural management and clinical outcomes related to gastrointestinal FB removal in pediatric patients with IF compared to children without IF. Our secondary aim was to assess the utility of clinical variables as independent predictors of FB removal related morbidity.

Material and Methods

We performed a ten-year retrospective review of patients from a single, quaternary pediatric center who underwent a post-esophageal gastrointestinal FB removal between 2011 and 2020. We obtained Institutional Review Board approval for chart review with a waiver of consent (IRB #M08–04–0163). To identify patients, we queried our institution’s secure data warehouse by Current Procedural Terminology (CPT) to extract unique patient encounters associated with “foreign body removal from the gastrointestinal tract” between 01/01/2011–12/31/2020. The start date was selected based on availability of operative records in the database. Each encounter underwent a secondary, manual review by one physician reviewer in the electronic medical record (EMR) to confirm inclusion. Duplicate records were excluded. For patients with repeat FB ingestion generating more than one removal procedure, only the first encounter was considered. Most patients with multiple FB removal encounters were adolescents with psychiatric co-morbidities and repeated, purposeful FB ingestion. Subsequent FB removal encounters were therefore excluded to limit the effect size of this specific ingestion type on the dataset.

A record was included when the following criteria were met: 1) the patient ingested a FB or had previously undergone placement of a medical device in the gastrointestinal tract, and 2) underwent a procedure to remove the FB, and 3) the FB was distal to the gastroesophageal junction (GEJ) during or after the procedure (Fig. 1). Foreign body was defined as any non-food object. Removal procedure was defined as any retrieval process performed by endoscopy, surgery, or manually under anesthesia. We included cases in which a FB removal was attempted but unsuccessful, either due to spontaneous progression of the object or technical difficulty. Cases were excluded due to FB location (i.e., mouth, esophagus, or respiratory tract) or a CPT coding error. Cases were also excluded if the FB had been inserted into the rectum. We limited the study to post-esophageal FBs, as esophageal FBs necessitate urgent removal in all cases. Patients were assigned to the IF subgroup if they had received care at the Boston Children’s Hospital intestinal rehabilitation program (the Center for Advanced Intestinal Rehabilitation) within 180 days of the FB removal and had a history of parenteral nutrition (PN) dependence for at least one period of 90 consecutive days prior to the procedure. All patients without IF were assigned to the control group.

Fig. 1.

Flow chart of patients assigned to control and IF cohorts after applying exclusion and inclusion criteria. List of abbreviations: FB: Foreign body, Intestinal Failure: IF, CPT: Current procedural terminology, PN: Parenteral nutrition, CAIR: Center for Advanced Intestinal Rehabilitation, Boston Children’s Hospital.

Data was extracted from clinical documentation directly associated with the FB removal encounter. When variables were missing from procedure documentation, EMR review was expanded to data entered within 180 days of the procedure. Anthropometric percentiles and z-scores were calculated using World Health Organization Child Growth Standards (age < 2 years) or Centers for Disease Control and Prevention Clinical Growth Charts (age >2 years) [9], [10]. The medical history was reviewed for gastrointestinal, neurologic, and psychiatric conditions considered “active” at the time of procedure per documentation of treating clinicians. Patients were classified as having a “developmental delay” if this exact term was used or if the patient was noted to have significant delays in cognitive or motor abilities after twenty-four months of age. Dysmotility diagnoses required history of an abnormal motility study or history of chronic intestinal pseudo-obstruction or other specifically defined motility disorder. Functional constipation was excluded. We recorded history of prior abdominal surgery as well as surgical technique (laparoscopy or laparotomy) if this information was available. Variables related to the FB removal included FB type, size, and indication(s) for removal. Type and size were defined and classified per the 2015 NASPGHAN Endoscopy Committee Guidelines on Foreign Body Management [5]. Indication(s) for removal were recorded per clinical notes; if the indication was not explicitly documented then it was assigned according to the 2015 NASPGHAN Endoscopy Committee guidelines. “Symptomatic” was defined as abdominal pain, vomiting, feeding intolerance, or a clinically significant change in gastrointestinal symptoms from baseline. The anatomic location of the FB prior to procedure was recorded per radiographic reports or as unknown in the instances of an incidental, intraoperative finding. Operative notes were reviewed for procedure type(s), number of attending endoscopists and/or surgeons involved in the case, and FB location at time of removal or attempted removal. Disposition was categorized as either same day, post procedure discharge, or hospital admission of at least one hospital night. Additional variables collected for IF subgroup included the primary cause of IF and whether the child had a stoma or was PN dependent at the time of FB removal.

Statistical analysis

Primary outcomes included: 1) FB characteristics including number, intrinsic hazard type, and location at removal, 2) procedure type and number of attending proceduralists involved to remove the foreign body, and 3) patient outcomes following removal procedure. Descriptive statistics were used to summarize patient characteristics and clinical factors and were presented as mean (standard deviation) for continuous variables and percentage (frequency) for categorical variables. Independent T-tests were used to compare continuous variables between IF and controls groups and Fisher’s exact tests were used for categorical variables. All statistical tests were 2-sided with p < 0.05 considered statistically significant. Data were analyzed using SAS version 9.4 (SAS Institute, Cary, NC, USA).

Results

Patient inclusion and characteristics

The initial query yielded 1760 records. Ninety-three duplicate records were removed. After manual chart review, 197 cases met inclusion criteria. Of these, 12 cases met criteria for inclusion in the IF group. The remaining 185 cases acted as controls. Fig. 1 reviews excluded records and the distribution of eligible cases to the IF and control groups, respectively. Patient characteristics are summarized in Table 1. Both control and IF groups had a male predominance and a similar mean age at the time of FB removal (7.7 ± 5.5 vs. 6.4 ± 4.6 years). The weight-for-age z-score was higher in the control group (0.3 ± 1.4 vs. −0.4 ± 0.6). Patients with IF had slightly higher rates of developmental delay and psychiatric disease, however, the presence of one or more of these conditions was common in both groups, affecting over one-quarter of all patients. In both groups, developmental delay ranged from mild impairments in speech, motor, or sensory processing to more significant difficulties including non-verbal status, severe oral pharyngeal dysphagia, and limited mobility. There were also some children in both groups with a co-diagnosis of autism spectrum disorder. Twenty-five patients (13.5 %) in the control group had had a prior FB ingestion compared to only one (8.3 %) patient with IF.

Table 1.

Characteristics of patients with and without Intestinal Failure who underwent a gastrointestinal foreign body removal procedure.

	mean (± SD) or n (%)
	Control Patients n = 185	Patients with IF n = 12
Age at date of procedure --- years	7.7 (5.5)	6.4 (4.6)
Male sex	98 (53.0)	7 (58.3)
Weight for age --- z-score	0.3 (1.4)	−0.4 (0.6)
Neurologic and psychiatric conditions1
Developmental delay, autism spectrum	21 (11.4)	2 (16.7)
Psychiatric disease2	32 (17.3)	3 (25.0)
ADHD	7 (3.8)	0 (0.0)
Pica	11 (5.9)	0 (0.0)
Gastrointestinal conditions and history1
Intestinal motility disorder	22 (11.9)	4 (33.3)
Inflammatory bowel disease	2 (1.1)	1 (8.3)
Presence of enteral tube	7 (3.8)	10 (83.3)
Prior foreign body ingestion	25 (13.5)	1 (8.3)
Prior stricture or obstruction	0 (0.0)	6 (50.0)
Abdominal surgical history1
None	170 (91.9)	0 (0.0)
Prior laparoscopy only	9 (4.9)	0 (0.0)
Prior laparotomy	5 (2.7)	12 (100.0)
Intestinal failure diagnosis1
Volvulus	---	4 (33.3)
Hirschsprung’s Disease	---	3 (25.0)
Gastroschisis	---	2 (16.7)
Intestinal atresia	---	2 (16.7)
Necrotizing enterocolitis	---	2 (16.7)
Chronic intestinal pseudo-obstruction	---	1 (8.3)
Meconium ileus	---	1 (8.3)
Intestinal failure characteristics at time of FB removal
PN dependent	---	9 (75.0)
Presence of stoma	---	6 (50.0)
History of STEP	---	2 (16.7)

IF: Intestinal Failure, ADHD: Attention deficit hyperactivity disorder, FB: Foreign body, PN: Parenteral nutrition, STEP: Serial Transverse Enteroplasty Procedure. 1) A patient may have had multiple medical diagnoses and/or surgery types within categories; percentages add to greater than 100 %. 2) Psychiatric diseases included anxiety, depression, bipolar disorder, schizophrenia, and psychosis.

The most common cause of IF in this cohort was volvulus, followed by Hirschsprung’s disease. As expected, all patients with IF had a history of more than one major abdominal surgery, six (50 %) had a history of prior intestinal stricture or obstruction, and two (16.7 %) had undergone a Serial Transverse Enteroplasty Procedure. At the time of the FB removal procedure, nine patients with IF (75 %) were PN dependent and six (50 %) had an intestinal stoma with intestinal discontinuity. In contrast, only 15 (8.1 %) control patients had undergone prior abdominal surgery, the majority of which were laparoscopy for appendicitis or enteral tube placement. No control patients had an ostomy at the time FB removal. Control patients also had lower rates of gastrointestinal motility disorders (11.9 vs 33.3 %).

Foreign bodies and removal procedures

Table 2a characterizes FB removal by presenting history, location, and procedure type. Caregivers of control patients were more likely than those of patients with IF to report witnessing or suspecting FB ingestion at presentation (88.1 vs. 33.3 %, p < 0.0001). There were similar rates of acute abdominal pain or other, non-baseline symptomatology at presentation (45.9 vs. 58.3 %, p = 0.55). Table 2b summarizes types of ingested FBs, classified by urgency of removal (emergent/urgent versus clinician discretion) and defined by the object’s intrinsic risk of perforation, obstruction, or caustic injury [5]. While the most common ingestion in both groups was that of a single, small, blunt object (i.e. objects with intrinsically low-risk object of gastrointestinal hazard), control patients were significantly more likely to ingest intrinsically hazardous objects that necessitated urgent or emergent removal (45.9 vs. 0.0 %, p = 0.0014). Hazardous objects included glass shards, multiple magnets, and button batteries. No intrinsically hazardous objects were ingested by patients with IF.

Table 2a.

Characteristics of foreign body removal by history, location, and procedure type in patients with and without Intestinal Failure.

	n (%)
	Control patients n = 185	Patients with IF n = 12	p-value1
Witnessed or suspected FB ingestion by caregiver	163 (88.1)	4 (33.3) 7	< 0.0001 0
Acute abdominal symptoms and/or change from symptom baseline	85 (45.9)	7 (58.3)	0.55
Ingestion of > 1 FB	53 (28.7)	2 (16.7)	0.19
Ingestion of large or long FB2	71 (38.4)	3 (25.0)	0.54
FB location prior to removal3
Stomach	158 (85.4)	0 (0.0)	< 0.0001
Distal to stomach	23 (12.4)	8 (66.7)	< 0.0001
Not identified4	7 (3.8)	4 (33.3)	0.0021
FB location at removal3
Stomach	133 (71.9)	0 (0.0)	< 0.0001
Distal to stomach	38 (20.5)	12 (100.0)	< 0.0001
Not identified5	25 (13.5)	0 (0.0)	0.37
FB removal procedures6
EGD or upper endoscopy	175 (94.6)	8 (66.6)	0.0058
Colonoscopy, sigmoidoscopy, or anoscopy	18 (9.7)	5 (41.7)	0.0064
Laparoscopy	13 (7.0)	1 (8.3)	0.60
Laparotomy	19 (10.3)	3 (25.0)	0.14
Patient required surgery	17 (9.2)	3 (25.0)	0.11
Patient required > 1 unique procedure	16 (8.6)	5 (41.7)	0.0041
Patient required > 1 endoscopist or surgeon during removal	13 (7.0)	8 (66.7)	< 0.0001

IF: Intestinal Failure, FB: Foreign body, EGD: Esophagogastroduodenoscopy 1) p-values are from Fisher’s Exact test. 2) FB ≥ 2.5 cm wide or ≥ 5 cm long. In cases of multiple FB ingestion, size was classified based on the largest FB. 3) When FBs were identified at or removed from multiple locations, each unique location was recorded. If multiple FBs were identified at or removed from the same location, only a single location was recorded. Percentages add to greater than 100 %. 4) FB was radiopaque, pre-procedure imaging was not performed, or the FB was found incidentally. 5) Proceduralist was unable to locate the FB during attempted removal procedure. 6) All procedures related to FB removal were recorded for each patient. Upper endoscopy included ileostomy and jejunostomy. Percentages add to greater than 100 %.

Table 2b.

Types of ingested foreign bodies, classified by urgency of removal per North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition 2015 Endoscopy Committee Guidelines in patients with and without Intestinal Failure.

			n (%)
	Control patients n = 185		Patients with IF n = 12	p-value1
Emergent/urgent removal always indicated				0.0014
Pointed Button battery > 1 magnet Toxin container Superabsorbent Total		23 (12.4) 21 (11.4) 37 (20.0) 2 (1.1) 2 (1.1) 85 (45.9)	-- -- -- -- -- 0 (0.0)
Indication for removal based on clinician assessment
Coin/blunt Bezoar Rubber/compressible Single magnet Retained medical device Total		77 (41.6) 9 (4.8) 2 (1.1) 10 (5.4) 2 (1.1) 100 (54.1)	7 (58.3) 3 (25.0) 0 (0.0) 0 (0.0) 2 (16.7) 12 (100.0)

IF: Intestinal Failure 1) p-value is from Fisher’s Exact test.

One-hundred percent of FBs in the IF group were removed from a post-pyloric location, compared to only 20.5 % of objects in the control group (p < 0.0001). While the majority of children in both groups underwent removal by esophagogastroduodenoscopy (EGD) only, children with IF were significantly more likely to require lower endoscopy (41.7 vs. 9.7 %, p = 0.006). Children with IF were also more likely to undergo multiple procedures for definitive removal (41.7 vs. 8.6 %, p = 0.001) and require more than one endoscopist or pediatric surgeon during the removal due to procedural complexity (66.7 vs. 7.0 %, p < 0.0001). The requirement for exploratory laparotomy was also higher in patients with IF compared to controls, however this did not reach significance (25.0vs. 10.3 %, p = 0.14).

Patient outcomes and disposition

Table 3 summarizes patient outcomes and disposition following the FB removal procedure. IF patients were significantly more likely to be require hospital admission after the procedure (83.3 vs. 28.6 %, p < 0.0001) and if admitted, were more likely to have a hospital stay greater than 1 night (58.3 vs. 12.4 %, p < 0.0001).

Table 3.

Outcome and patient disposition following foreign body removal procedure in patients with and without Intestinal Failure.

		n (%)
	Control patients n = 185	Patients with IF n = 12	p value1	Risk Difference (95 % CI)
Outcome
≥1 FB retained after procedure2	26 (14.1)	0 (0.0)	0.37	14.1 % (9.1 %, 19.1 %)
Disposition
Same day discharge	132 (71.4)	2 (16.7)	<0.0001	54.7 % (11.3 %, 32.6 %)
Inpatient admission > 1 hospital night	23 (12.4)	7 (58.3)	< 0.0001	−45.9 % (−74.2 %, −17.6 %)

IF: Intestinal Failure, FB: Foreign body, CI: Confidence Interval 1) All p-values are from Fisher’s exact test. 2) Proceduralist was unable to locate the FB during attempted removal procedure or the removal was unsuccessful due to spontaneous progression or technical difficulty.

Discussion

To our knowledge, this is the first study to characterize FB removal procedures in children with IF and to compare their outcomes to children without this condition. Average age and slight male predominance was consistent with prior studies which show that young, school age males tend to be at highest risk for FB ingestion [1], [2], [11], [12]. The transition to elementary school naturally coincides with a decrease in adult supervision compared to toddlerhood and this may partially account for increased risk of ingestion at this age.

The rate of developmental delay was notable in both IF (16.7 %) and control (11 %) groups and was higher than the general pediatric population [12]. This aligns with previous studies that report neurologic impairment and intellectual disability as significant risk factors for FB ingestion [13], [14]. There are numerous mechanisms by which neurologic impairment may increase the risk of ingestion; a prolonged oral phase, dysphagia with limited control over objects placed in the oral cavity, and communication impairment may all play a role [13]. Prior studies have also shown similarly increased rates of autism spectrum in patients presenting with FB ingestion [14], [15], [16]. Autism is associated with sensory processing disorders, which can lead to a variety of self-soothing behaviors that involve biting or sucking on non-nutritive objects. Given that neurologic and intellectual impairments are well-described comorbid conditions for children with IF, affecting up to 30 % of this population [17], [18], [19], it is especially important to educate caregivers of children with IF on ways to prevent FB ingestion. Notably, three children with IF ingested medical equipment (adhesive bandages and plastic tubing from a central line extension). Routine education on safe storage of medical supplies and medications should also be implemented for these children.

The need for increased supervision in children with IF is further underscored by the finding that FB ingestions in this group were significantly more likely to be unwitnessed or unsuspected by caregivers. Five out of eight children with IF had an unwitnessed ingestion and the FB was discovered on a radiograph or during an endoscopic or surgical procedure performed for another indication. Interestingly, in three of these cases, there was a documented history of increased obstructive symptoms (e.g., emesis, increased ostomy output) or gastrointestinal bleeding preceding the incidental FB finding. Given the high incidence of strictures and dysmotility in these patients, it is difficult to know exactly what role the FB played in the development of symptoms. However, in two cases associated with gastrointestinal bleeding, mucosal changes including erythema and friability or ulceration were seen surrounding the incidental FBs. In the third case, the patient underwent a laparotomy for persistent obstructive symptoms. A 2 cm Lego® block was identified proximal to a strictured area at the site of a prior anastomosis. It is also notable that 75 % of the IF cohort were PN dependent at the time of ingestion. It is possible that caregivers and gastroenterologists may be less inclined to consider FB ingestion in these patients due to oral aversion or nil per os status, as well as the co-existence of chronic gastrointestinal diseases to also explain symptoms.

While it is well described that FB ingestion can portend substantial morbidity for any child, regardless of preexisting gastrointestinal disease [2], [20], [21], our findings suggest that the risk of an adverse outcome after ingestion substantially increases for children with IF. Children with IF commonly develop profound anatomical and motility changes [6], [7], which may predispose them to both FB retention and increased procedural complexity during removal. Extensive surgical interventions lead to adhesions, causing abnormal bowel fixation and blind loops, while strictures, commonly at anastomosis sites, may impede object passage, raising obstruction risk. Chronic intestinal dysfunction and neuropathy result in dysmotility and bowel dilatation, slowing transit and further limiting spontaneous foreign body progression. Among this cohort of IF patients, it is notable that FBs were found in close proximity to a stricture in three patients, and in a blind ending bowel loop in a fourth.

These factors also contribute to increased procedural complexity, highlighting the need for multidisciplinary planning for FB removal in this population. In our study, children with IF were significantly more likely to require multiple procedures for definitive removal. Similar factors may have contributed to this finding; pre-procedural radiographs are less able to precisely identify the location of a FB within a surgically altered intestine, and distorted anatomy, adhesions, and strictures further limit the endoscopist’s ability to discover and safely remove an object.

It is also notable that the increased morbidity associated with FB removal in children with IF could not be attributed to the intrinsic properties of the ingested object itself. One-hundred percent of objects ingested by the IF group were classified as intrinsically low-risk (blunt objects, bezoars, or coins) [5]. This was in stark contrast to control patients, one-half of whom ingested objects carrying a risk of perforation or caustic injury. The decision to pursue multiple, and in some cases highly invasive, removal procedures for children in the IF group was not a reflection of the objects themselves, but rather of the increased surgical and medical complexity of the patients who ingested them. We observed that once a FB removal procedure was initiated for a patient with IF, proceduralists were more inclined to complete the removal, even when the object was difficult to locate or extract. In documentation associated with these cases, proceduralists noted both the increased risk of FB-related bowel obstruction and low likelihood of spontaneous passage as primary indications for pursuing more invasive removal procedures. In contrast, when similarly low-risk objects could not be identified by upper endoscopy in a child in the control group, the object was left in place and further procedures were not attempted. In most cases the child was discharged home on the same day with the assumption that the object would pass spontaneously.

Strengths of this study include the ten-year timespan and single center experience. A further strength is the size and composition of the control cohort. Comprised of generally healthy, school-age children without significant gastrointestinal disease or prior gastrointestinal surgery, the control group reflects the majority of children who undergo FB removal procedures at our institution and broadly. Conversely, our study is limited by the small IF sample size. The high rate of PN dependence in this group also suggests that the included patients with IF were relatively sicker and more complex, which may limit the applicability of this study to the broader IF population. Lastly, given that the study was limited to patients who underwent removal procedures, our findings are most applicable to referral centers or those with pediatric endoscopy and surgery capacity.

Conclusion

In summary, FB ingestion was associated with increased number of removal procedures, procedural complexity, and hospital admission in children with IF compared to controls. When evaluating acute changes in gastrointestinal symptoms in children with IF, occult FB ingestion should be considered alongside other important causes of gastrointestinal symptoms. In the case of a known FB ingestion, clinicians should review the patient’s anatomy and make a plan for removal. Future guidelines on the management of FBs in children could include special considerations for children IF as well as for other children with a history of major abdominal surgery. Recommendations could include a lower threshold to remove benign objects from the stomach to mitigate the risk of the object becoming lodged in the intestine, and to involve an interdisciplinary team including a pediatric surgeon and advanced endoscopist with experience navigating surgically altered gastrointestinal anatomy and complex FB removal, if available. Education to prevent FB ingestion should be implemented as part of routine care for children with IF.

Ethical clearance

Not required.

Funding

This work was supported in part by National Institute of Health grants P30 DK040561 (to CPD) and T32DK007477-41 (to KLC).

CRediT authorship contribution statement

Alexandra N. Carey: Writing – review & editing, Supervision, Data curation, Conceptualization. Kieley L. Chapman: Writing – original draft, Formal analysis, Data curation, Conceptualization. Paul Crowley: Writing – review & editing, Methodology, Data curation. Evan Brociner: Writing – review & editing, Data curation. Enju Liu: Writing – review & editing, Methodology, Formal analysis. Brenna Fullerton: Writing – review & editing, Data curation, Conceptualization. Biren P. Modi: Writing – review & editing, Data curation. Tom Jaksic: Writing – review & editing, Data curation. Lissette Jimenez: Writing – review & editing, Data curation. Christopher P. Duggan: Writing – review & editing, Supervision, Resources, Data curation, Conceptualization.

Patient's/ Guardian's consent

Not applicable.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Glossary

FB

Foreign body,

IF

Intestinal Failure

NASPGHAN

North American Society for Pediatric Gastroenterology Hepatology, and Nutrition,

CPT

Current procedural terminology, PN: Parenteral nutrition

CAIR

Center for Advanced Intestinal Rehabilitation, Boston Children’s Hospital.

EGD

Esophagogastroduodenoscopy