Increase in foreign body and harmful substance ingestion and associated complications in children: a retrospective study of 1199 cases from 2005 to 2017

Accidental caustic and foreign body ingestions (FBI) are frequently reported in children world-wide and affect up to 75% of children under 6 years of age, since infants evaluate objects by tasting and swallowing [1‐7]. A variety of objects and toxic bodies are ingested, among which coins appear to be the most commonly ingested world-wide, followed by fish bones, toys, jewellery, button batteries, magnets, household items, household cleaning substances, caustic sodas, and many others, and food impaction is also reported frequently in this cohort [1, 2, 5, 8‐12]. The initial assessment includes a careful history-taking of symptoms, timing of the presentation, the type of foreign body or chemical agent, and associated conditions. This assessment is complemented by a physical examination assessing patient status, vital signs, and airway evaluation and emergency conditions. Management depends on the individual presentation of the patient, his or her underlying oesophageal pathology, the ingested object or liquid and its localization as well as on the physician’s clinical experience [7, 9]. Most objects will pass through the gastrointestinal tract (GIT) spontaneously, especially if they have already reached the small intestine and even up to 50% of large foreign bodies (FBs) with a diameter over 30 mm may pass the GIT without intervention [13, 14]. In contrast, FBs lodged in the oesophagus or sharp and relatively large FBs will require an intervention. In such cases, rigid or flexible endoscopy are typically used, where Foley catheter extraction, oesophageal bougienage, McGill forceps and magnetic catheters are less frequently implemented [5, 6, 15, 16]. Button batteries, multiple magnets and sharp objects need to be removed immediately to avoid serious complications [5, 7, 17‐22]. Caustic agent ingestion is associated with high morbidity depending on the extent of injury on initial endoscopic evaluation [3, 9]. Food bolus impaction in children and adolescents may be a clinical feature of an underlying eosinophilic oesophagitis and not only requires a therapeutic but also a diagnostic endoscopy [23]. Depending on the objects and agents involved, a variety of associated complications are possible, ranging from mucosal abrasions, lacerations, perforations, mediastinitis, peritonitis, and strictures to death [5, 24, 25]. Complications are more likely if the FB remains impacted for an extended period of time [5, 26]. In addition, unobserved ingestion of objects or chemical substances may delay and complicate diagnosis and treatment, as infants may not inform their parents that they have ingested something [26]. This study aims to elucidate the clinical presentation, diagnostic procedures, and complications associated with the ingestion of different object categories over a 13-year time period. In particular, we evaluated the effect of implementing a diagnostic and therapeutic standard algorithm on the percentage of hospitalizations and complications.

We performed a structured retrospective review of medical records collected over 13 years (from January 2005 to December 2017) from paediatric patients who presented with food bolus impaction and ingestion of foreign objects or chemical substances, at the Department of Paediatric and Adolescent Medicine, University Medical Centre Ulm, Germany. It is the only paediatric hospital of a catchment area in Germany that counts approximately 108,000 children. This study adheres to the ethical principles of the Declaration of Helsinki. It was approved by the Ethics Committee of Ulm University (No.399/17). For the presented data, the requirement to obtain informed consent was waived.

In total, 2427 records of patients aged 0 to 17 years were extracted from the hospital database using codes corresponding to FBI, chemical agents and food impaction from the International Statistical Classification of Diseases and Related Health Problems (ICD10-GM) codes including K22.1–3, T18.0–5, T28.5–9, T30.4, T52.0, T55, T60.8, T62.1–2, and T65.8–9 (Fig. 1). Episodes not corresponding to an accidental ingestion, e.g., alcoholic intoxication, iatrogenic injuries, and erroneous coding, were excluded and multiple presentations for the same event were merged.

Demographic patient information, associate conditions, type of foreign body or chemical substance, clinical symptoms, diagnostic and therapeutic management, and complications were collected from the electronic patient file and electronic database. The miscellaneous ingested foreign bodies, food boli and chemical substances were classified into 30 categories according to their characteristics (object type (solid, liquid and others), object form and size (below or over 2 cm*3 cm for infants, and below or over 3 cm*5 cm for children > 1 year of age, and sharp) [27], organic or inorganic, radiopacity, and object class (metal, plastic, wood, glass, food, drug, button battery, other batteries, coins, magnet, caustic solution, acid solution, oil, paper, plants, body parts, surfactant/cleanser, stone, and cigarette).

All authors had access to the study data and reviewed and approved the final manuscript.

The analysis of 1199 accidental paediatric foreign body and chemical substance ingestions over 13 years (2005–2017) at a German University Medical Centre revealed a significant annual increase of 80% from 6.1 in 2005 to approximately 11 per 10,000 children in 2017 in the catchment area.

Overall, the annual rate of complications also significantly increased, which was probably restricted by the implementation of a diagnostic and therapeutic algorithm at the end of 2010.

Ingestion of foreign bodies or chemical substances is frequent among children below 6 years of age around the world [1‐5]. The majority of patients who presented to our hospital were toddlers (median 2.2 years) and predominantly male (53.3%). In addition, we registered comorbidities such as psychiatric disorders, intellectual disability, postsurgical oesophageal atresia and eosinophilic oesophagitis which may also predispose patients to FBI or food bolus impaction [28, 29].

A variety of foreign bodies were ingested, of which coins were the most common in our cohort, as in many other studies [1, 5, 8, 11, 12]. However, the proportion was lower in our study (18%) than that reported by others (49 to 88%), which may reflect changes over time or differences in patient referral, selection and inclusion, in object classification and localization, or in the diets and habits of the patient populations [4, 5, 12, 30, 31] (supplemental Table). The ingestion patterns also differed by sex, age and season (e.g., Christmas decorations) [8, 32]. We found, that girls were at risk for harmful ingestion of sharp objects in our population. Another study showed that it is 2.5 times more likely that girls rather than boys ingest jewellery or hair products compared to boys [8]. We also found relevant liquid ingestions of potentially harmful cleansers (12.3%), acids (2.7%) or bases (0.6%). Caustic ingestions in children are mostly accidental, and the severity depends on the type and quantity of the ingested substance [33].

We observed an alarming trend concerning the number of accidental ingestions in our population. This has also been reported in the United States, with an annual increase of 91.5%, from 9.5 in 1995 to 18 per 10,000 children with FBI in 2015 [8]. In contrast, considerably fewer FBIs have been reported at Chiang Mai University in Thailand (only 194 cases from 2006 to 2017), with a population comparable to Ulm. This cannot be explained by different age limits (< 15 versus < 18 years) and the inclusion of ingestion of chemical substances in our study. Thus, other factors, e.g., difficult access to medical care or fewer harmful items in the household, may influence the observed lower frequency of FBIs. The increase in our study affected all categories of objects and substances, many of which are increasingly used in German households. In particular, we observed an alarming increase in lithium button battery ingestions in infants, which was associated with hospitalisation and major complications. This trend had also been previously reported in comparable studies from the U.S. [34, 35].

In our cohort, the presenting symptoms varied depending on the type and localization of the ingested FB; e.g., gagging, pain and coughing were frequently observed after coin ingestion. In contrast, a primary association with vomiting and drooling had been reported in other studies after coin ingestion [5]. Patients with oesophageal FB mainly present with drooling, vomiting and dysphagia, especially if the FB was located in the first narrowing of the oesophagus [36]. Nevertheless, most patients had a normal physical examination [31]. In children, the diagnosis of FB ingestion may be complicated if the ingestion is not observed or the child is asymptomatic. As in other studies, approximately half of the patients were asymptomatic [2].

A carefully obtained history, the type of ingestion and the level of suspicion will determine the course of action to avoid severe and life-threatening complications. In this study, approximately 16% of all patients required hospitalization, which is slightly more than in similar studies that have reporting approximately 10% [8]. Twenty-two percent of patients who ingested coins and more girls than boys were hospitalized. In fact, the hospitalization rate for coin ingestions was considerably lower in our cohort than the 55% reported in a European study [4]. The highest rate of complications was observed for button battery, glass and food ingestions. Others have reported an increased prevalence of complications after sharp FB ingestion and a 4- to 8-fold increase in complications if the endoscopic retrieval of the oesophageal FB was performed beyond 24 to 48 h after ingestion [37]. Serious complications and fatal outcomes have been reported for button battery ingestions [18, 19, 25]. Although we observed an annual increase in total patients and harmful ingestions, the number of patients admitted to the hospital remained constant.

In 2010, we implemented an algorithm for the diagnostic and therapeutic management of FBI at our hospital that takes into account the age of the patient, symptoms, and the size, type and location of the radiopaque and radiolucent FB based on an interdisciplinary consensus from the departments of radiology, paediatric and adolescent medicine (including paediatric intensive care and paediatric gastroenterology), and ENT [27]. This may potentially influence the practice of admissions at our hospital and partially explain the constant number of hospitalized patients. In addition, the complication rate was reduced by 29%.

The European Society of Paediatric Gastroenterology and Nutrition (ESPGHAN) and European Society of Gastrointestinal Endoscopy (ESGE) recommend X-ray examination for all patients with suspected FB ingestion even without symptoms [22]. In our study, medical imaging was performed for half of the patients, as nearly half of the objects were radiopaque. Imaging is important to confirm the presence, type, number and localization of FBs as well as to detect complications and guide further management and follow-up if indicated [38]. X-ray of the chest and abdomen was frequently performed and detected and localized the FB in 86% of the patients. In fact, we observed an annual increase in the number of detected FBs, particularly through X-ray investigations. Although hand-held metal detectors (HHMD) have been shown to detect and localize the majority of metallic FBs, we did not use them in our department at that time [39]. It has been shown that HHMDs have a high sensitivity in the detection of coins and seems to be a good early screening tool for faster triage in the emergency room setting, potentially reducing radiation exposure [40, 41].

The primary management was awaiting spontaneous passage for most FBIs, and interventional removal was performed for 126 patients. The indication and timing of medical intervention to remove a foreign body was based on the location, size and type of the FB, the duration of impaction, and patient symptoms according to our in-house standard with special attention towards button battery and magnet ingestions [27]. The initial localization and size of the FB are determining factors of the likelihood of spontaneous passage [5, 13].

Removal of the FB was mainly performed by rigid or flexible endoscopy according to the localization of the FB and symptoms of the patient and was successful in 81% of rigid and 90% of flexible endoscopy. High success rates for rigid and flexible endoscopy have been reported in retrospective studies on oesophageal FBs among children and adults with low rates of complications (supplemental Table) [15, 36, 42, 43]. The management of foreign body, food and toxic substance ingestions needs to be adapted for infants, e.g., smaller endoscopes in children below ten kilos, and paediatric-trained endoscopists are required [22, 44]. Oesophageal food impaction is a frequent finding and requires special attention [45]. Removal is performed either en bloc or by a piecemeal approach using various grasping devices and after examining the oesophagus distal to the bolus; the push technique is also sometimes used [20]. Food impaction is frequently associated with oesophageal pathology, e.g., postsurgical oesophageal atresia and eosinophilic oesophagitis [28, 46, 47]. Magnet ingestion is rarely registered but frequently requires surgery, especially for children with neurological or psychiatric diseases who have an increased risk of ingesting multiple magnets [48, 49]. There has been an alarming increase in magnet ingestions in emergency departments according to data from the National Electronic Injury Surveillance System (NEISS) [21].

We found a very low need to perform surgical interventions (0.3% of all, 2% of hospitalized cases, 3% of removed FB) in our cohort, unlike the 18% reported by others for hospitalized patients, despite providing paediatric surgery twenty-four-seven [50]. Cohorts that once required surgical interventions could be treated by experienced endoscopic removal today [26]. In children, mild oesophageal lesions (88%) have been predominantly identified following caustic ingestions, and severe oesophageal lesions have been associated with the presence of signs and symptoms (e.g., oral lesions, vomiting, dyspnoea, drooling, dysphagia, and haematemesis) [51]. Thus, endoscopy could be avoided in the absence of signs and symptoms [51], as in our study cohort.

In our study, 9% of the children presented to our hospital more than 24 h after ingestion versus the 22% reported in a study from Pittsburgh on children who underwent oesophagoscopy for suspected FB [31]. The lack of awareness of harmful situations requiring emergency care and prompt intervention may lead to serious complications, e.g., as observed with button batteries, pins and magnets [4]. Preventive measures, e.g., pressure on manufacturers to package items appropriately for children below 3 years of age, effectively reduced toy ingestions in the U.S. in 2011 [8]. Furthermore, food and toys should not be marketed together to prevent children from ingesting toys. Legal measures to protect children from the growing problem of unintentional button battery ingestion are required and should include child-resistant packaging for batteries, child-resistant closures for all consumer products that use button batteries, and warnings regarding the potential danger of ingestion [52, 53].

Finally, few limitations of this study should be underlined. First, as we incorporated a retrospective study design, the conditions may not be ideal and lack some relevant information not available from the electronic patient file. In addition, we only captured information from patients who presented to our hospital, which, as in many other studies, underestimates the real frequency of accidental ingestions among children. Additionally, comparability among related studies is limited, as patient selection and categorization of objects may differ, and age groups may vary [5].

This study shows a significant increase in ingested household items and liquids associated with complications over 13 years in the studied population.

Appropriate care for children with FBI requires proper management, skilled physicians, adequate equipment and, if necessary, inpatient admission. The implementation of clear diagnostic and therapeutic algorithms in paediatric emergency departments may lower the progression of injuries and complications prospectively.

Additional preventive measures, including public health interventions and superior consumer safety standards, are necessary to stop the observed trend and should particularly focus on the preschool toddler group.