Endoscopic Management of Pediatric Foreign Body Ingestions and Food Bolus Impactions: A Retrospective Study from a Tertiary Care Center

Abstract Background  Foreign body (FB) ingestion is a common pediatric problem with the majority of these occurring in children younger than 3 years. Management varies depending on the age of the patient, ingested object(s), its location along the digestive tract, and the available expertise. We aim to report our experience with endoscopic management of FB ingestions in children (<18 years). Materials and Methods  We retrospectively reviewed and analyzed endoscopic and medical records from our hospital database of all pediatric patients (<18 years) who presented with FB ingestion between January 2011 and December 2021. Results  Our analysis included a total of 368 patients. FB ingestions and/or food bolus impactions were noted in 242 and 11 children, respectively while 115 (31.25%) had spontaneously passed off FB from the digestive tract. Most common FB was coin (28.5%) followed by animal bones (26.2%). Endoscopic management of FBs and food bolus impaction was successful in 247 children (97.63%), while endoscopic FB retrieval failed in 6 children including 1 with fish bone and 5 with button batteries. A total of 9 out of 11 children with food bolus impaction had underlying esophageal pathology, the commonest being corrosive stricture ( n  = 7). No mortality related to endoscopic intervention was reported. Conclusions  Endoscopic retrieval of ingested FBs and food bolus impaction in children is a safe and effective approach when performed by experienced endoscopists and is associated with a high success rate and a lower incidence of complications with reduced hospital stay.


Introduction
Foreign body (FB) ingestions constitute approximately 4% of all emergency endoscopies performed. 1,2 The majority of FB ingestions occur in children because of their curiosity to explore objects orally, with peak incidence occurring between the ages of 6 months and 3 years. While the majority of the ingested FBs tend to pass off spontaneously through the digestive tract, approximately 10 to 20% of FBs require endoscopic intervention with an estimated 1% needing surgery for their extraction and/or to mitigate complications. [3][4][5][6] Impacted FBs, especially button batteries (BBs) in the esophagus, given its proximity to vital structures in the mediastinum, may result in serious complications including death if not intervened expeditiously. Western literature shows that ingested FBs account for approximately 1,500 deaths Keywords ► button battery ► coin ► fish bone ► food bolus ► foreign body ► stricture Abstract Background Foreign body (FB) ingestion is a common pediatric problem with the majority of these occurring in children younger than 3 years. Management varies depending on the age of the patient, ingested object(s), its location along the digestive tract, and the available expertise. We aim to report our experience with endoscopic management of FB ingestions in children (<18 years).

Materials and Methods
We retrospectively reviewed and analyzed endoscopic and medical records from our hospital database of all pediatric patients (<18 years) who presented with FB ingestion between January 2011 and December 2021. Results Our analysis included a total of 368 patients. FB ingestions and/or food bolus impactions were noted in 242 and 11 children, respectively while 115 (31.25%) had spontaneously passed off FB from the digestive tract. Most common FB was coin (28.5%) followed by animal bones (26.2%). Endoscopic management of FBs and food bolus impaction was successful in 247 children (97.63%), while endoscopic FB retrieval failed in 6 children including 1 with fish bone and 5 with button batteries. A total of 9 out of 11 children with food bolus impaction had underlying esophageal pathology, the commonest being corrosive stricture (n = 7). No mortality related to endoscopic intervention was reported. Conclusions Endoscopic retrieval of ingested FBs and food bolus impaction in children is a safe and effective approach when performed by experienced endoscopists and is associated with a high success rate and a lower incidence of complications with reduced hospital stay.
annually, although similar data are lacking from the Indian subcontinent. 7 While most adults with FB ingestion present  with a multitude of symptoms including dysphagia, odynophagia, retrosternal pain, FB sensation, retching, and/or  vomiting, most children given their limited vocabulary present with nonspecific symptoms such excessive crying, sialorrhea, vomiting, fever, etc., thus causing a delay in diagnosis and resulting in complications. 5,[8][9][10] The aim of this study was to report our experience in endoscopic management of ingested FBs and/or food bolus impactions of the digestive tract in children (<18 years).

Materials and Methods
We retrospectively evaluated the medical and endoscopic records of 368 children who were referred to St. John's Medical College Hospital, Bengaluru, for suspected history of FB ingestion and/or food bolus impaction between January 2011 and December 2021. Of these, 115 patients had spontaneous expulsion of FBs by the time they arrived at the emergency department (ED) or were subjected to endoscopic procedure. Of the remaining 253 patients, 242 underwent endoscopy for FBs in the digestive tract and 11 patients for food bolus impactions. All patients had a thorough and quick physical examination to evaluate for evidence of luminal obstruction and other complications, especially perforation (cervical swelling and/or crepitus in case of oropharyngeal/proximal esophageal perforation, or fever and peritonitis in case of intestinal perforation). Plain X-rays of the neck, chest, and/or abdomen were obtained for all patients upon their presentation to the ED in accordance with current recommendations and practice guidelines to assess the presence, location, size, configuration, and number of ingested objects. 3,4 Patients with FBs impacted in the esophagus, those with stridor and/or breathing difficulty, and children with sharp FBs (including pin, nails, and razor) in the upper digestive tract were treated as emergency and underwent endoscopy within 2 hours of their arrival to the ED irrespective of their fasting status. Those with asymptomatic FBs in the stomach and esophageal food bolus impactions were classified as urgent to semi-urgent procedures and subjected to endoscopy within 6 to 8 hours of their ED presentation. 11,12 All patients underwent endoscopic procedure under general anesthesia by a trained anesthetist, with close monitoring of vital signs such as blood pressure, heart rate, and oxygen saturation during the entire procedure. All endoscopic procedures were performed using flexible endoscopes; either Olympus GIF-Q150 or GIF-180 was used for esophagogastroduodenoscopy, while Olympus CF-180 was used in a sole patient with documented metallic nail in the transverse colon. A variety of accessories including FB rat tooth forceps, biopsy forceps, Roth net, FB retrieval baskets, snares, and magnet were used depending on the nature of the FB and the ease and competency of the endoscopist(s). Transparent caps, hoods, and/or over-tubes were employed when needed during retrieval of sharp FBs to protect the digestive tract during their endoscopic retrieval. Patients with documented BB ingestion irrespective of its location in the digestive tract underwent esophagogastroduodenoscopy to document any injury related to its transit along the upper digestive tract. 13,14 The on-call pediatrician or pediatric surgeon was in attendance for all the endoscopic procedures. Postprocedure, all the patients were observed and monitored closely in the hospital until discharge. Patients with impacted esophageal FBs including BBs were kept nil by mouth for 24 to 48 hours, received thin barium study to rule out any perforation, and were started on oral feeds only after barium study was reported as normal. Patients were discharged from the hospital once they tolerated oral feeds without incident.
Of note, our experience with endoscopic management of BBs has been published recently. 13 Demographic, clinical, and endoscopic data were collected and analyzed with regard to age, gender, type and location of FBs, symptoms after FB ingestion, anesthetic methods, associated upper gastrointestinal (GI) tract diseases, accessory devices utilized, and complications related to the FB impaction. Descriptive statistics including mean, median, and standard deviation were used to analyze the data. Institutional ethical committee clearance was obtained for the study.

Results
Of the total 368 patients, endoscopic intervention was required in 242 patients with FB ingestions and 11 with food bolus impactions (n ¼ 253). In the remaining 115 patients (31.25%), the FB had passed off beyond the reach of the upper GI scope spontaneously from the digestive tract by the time the patient arrived at the ED (or endoscopy was undertaken). The mean age of our study population was 2.8 years (range: 3 months-17 years), with 67% (n ¼ 169) being less than 3 years of age.
The mean age of patient, sex, and type and location of FBs are depicted in ►Table 1.
Coins were the most common FBs followed by BBs. Although the most common site of FB location was the stomach (64.97%); given its narrow diameter, the esophagus was the most common site of FB impaction. Only one patient underwent colonoscopy for extraction of a metallic nail from the transverse colon as the nail would not move or change its position even after 3 days of monitoring in the hospital (►Fig. 1). The average "door to endoscopy time" in those with esophageal BBs was 1.3 and 6.2 hours in those with other FBs (including BBs) beyond the gastroesophageal junction.
While a majority of our patient had no symptoms (n ¼ 200); in the remainder of the study population, the most common symptoms (either alone or in combination) were dysphagia noted in 20 patients followed by odynophagia in 12, sialorrhea in 8, retrosternal chest discomfort in 8, and dyspnea and cough in 5 patients.
A total of 11 patients presented with a history of food bolus impaction, of which 9 had an underlying esophageal pathology, with corrosive stricture being the most common (n ¼ 7) followed by anastomotic stricture at the site of previous tracheoesophageal fistula surgery (n ¼ 2), while 2 patients had no obvious cause for impaction.
Food bolus impactions were treated by either "push technique" wherein the food bolus was pushed into the stomach carefully under endoscopic guidance or retrieved piecemeal using either a snare or Roth net (►Fig. 2).
Endoscopic management of FB and food bolus impactions was successful in 247 (97.63%) children but failed in 6 including 5 with BBs and 1 with fish bone ingestion who developed retropharyngeal abscess, all of whom underwent surgical intervention. Two of the five children who failed BB extraction had esophageal perforation with resultant peritonitis needing surgical intervention. 13 No complications related to the endoscopic procedure or general anesthesia were observed in our study.  in diameter) may get lodged anywhere along the GI tract; however, the esophagus, due to its inherent low motility and anatomical narrowing, is a common site of FB and food bolus impactions. 2 True FB ingestion is mostly encountered in pediatric populations, with 75% of cases occurring in children younger than 10 years; BBs, coins, and metallic pins are the common culprits. 1 In our present study, FB and food bolus impactions were found in 242 and 11 patients, respectively, out of 368 patients (68.75%), which differs from western studies that reported FBs in the range of 80%. 2,[17][18][19][20][21] This inconsistency could be related to delay in referral from the outlying hospitals for endoscopic procedure resulting in a high likelihood of spontaneous passage of FB through the digestive tract with time. Of the total 253  patients with FB ingestion in our study, 67% were aged less than 3 years. Endoscopic intervention was successful in 97.63% (n ¼ 247) of our patients but failed in 6 patients, which is comparable to other studies. 21,22 Patients in whom endoscopic retrieval was unsuccessful included five children with impacted BBs and one child with fish bone who developed retropharyngeal abscess needing surgical drainage and intravenous antibiotics. While the stomach was the most common site of FB, the most common site of FB and food bolus impactions was the esophagus given its narrow lumen. The need for and timing of an endoscopic intervention are dependent on multiple factors; these include patient age and clinical condition, the location and characteristics of the ingested FB, time since ingestion, the technical capabilities of the endoscopist, and availability of endoscopic accessories and on-call anesthetists. 23 Patients unable to effectively manage their secretions secondary to complete esophageal obstruction from the ingested FB and/or those with sharp objects or BB require emergent endoscopic intervention, preferably within 2 hours, and at the least within 6 hours. 3 Both North American Society for Pediatric Gastroenterology, Hepatology and Nutrition (NASPGHAN) and European Society for Paediatric Gastroenterology, Hepatology and Nutrition (ESPGHAN) recommend performing emergent (<2 hours) endoscopic removal of BBs impacted in the esophagus regardless of the patient's symptoms. 11,12 This protocol was followed in our patients and all those with impacted esophageal BBs were taken up for endoscopy within 2 hours of their arrival to the ED. Also, emergency endoscopic retrieval was undertaken in those presenting with symptoms of complete esophageal obstruction or dyspnea and those with sharp pointed objects in the upper digestive tract in accordance with other studies. 3,21 Patients with BBs and other FBs (except sharp objects) in the stomach were subjected to endoscopy on an average of 6.2 hours from the time of their arrival to the ED (range: 2-8 hours).

Discussion
Our study included a total of 11 children with food bolus impaction, with vegetable and fruit seeds being the commonest culprits followed by animal meat. Corrosive stricture was noted in seven of these patients and another two had anastomotic stricture developing after surgery for tracheoesophageal fistula. All these patients underwent stricture dilatation once the FB was extracted. Studies have demonstrated that around 75 to 100% of adult patients with food bolus impaction have an underlying esophageal pathology, with the most common abnormalities being hiatus hernia, eosinophilic esophagitis, strictures, surgery, and esophageal motility disorders, similar studies are lacking in pediatric patients. [24][25][26][27] Endoscopic "push technique," which has been advocated by several authors as the primary endoscopic method to treat esophageal food bolus impaction, was successful in the remaining eight patients in our study, while others underwent piecemeal extraction using either a snare or Roth net. 28,29 Although radiographic evaluation is not always required, plain radiographic evaluation of the neck, chest, and abdomen is recommended to assess the presence, number, location, size, and shape of the radiopaque FB. X-rays can also provide useful information regarding possible aspiration, presence of free mediastinal or peritoneal air, or subcutaneous emphysema. 30 The major limitation of our study is the retrospective nature of the study, and being a single-center study, the findings of our study cannot be generalized. There was no follow-up in the majority of our subjects after the initial endoscopy as most of them were referred from other outlying hospitals.
In general, any FB along the esophagus should be considered for extraction on an emergency/urgent basis. Any sharp objects, multiple magnets, and BBs in the stomach should be removed expeditiously to avoid complications, while blunt objects such as coins in the stomach and duodenum in an asymptomatic patient warrant a "wait and watch" approach and such patients need to be closely monitored radiologically and clinically till expulsion of the FB is confirmed.

Conclusion
In conclusion, FB ingestion and impacted food boluses in the digestive tract are a common pediatric problem needing endoscopic intervention. At presentation, a quick history and physical examination along with plain X-rays of the chest and abdomen should be performed to assess the type and location of the ingested FBs. While a proportion of digestive tract FBs are expelled without incidence, flexible endoscopic retrieval under general anesthesia is safe and effective in those presenting with obstructive symptoms when performed by a skilled endoscopist with a high success rate.

Funding
None.