Article Text
PDF
Abstract
Objective We report on the increasing incidence and outcomes from intentional foreign body ingestion (iFoBI) presenting to our hospital over a 5-year period. The aim was to assess the impact on services and to identify ways to safely mitigate against this clinical challenge.
Design/method We performed a retrospective observational study of all patients presenting to a university hospital between January 2015 and April 2020 with iFoBI with a focus on objects swallowed, timing of endoscopy and clinical outcomes.
Results 239 episodes of iFoBI in 51 individuals were recorded with a significant increase in incidence throughout the study period (Welch (5, 17.3)=15.1, p<0.001), imposing a high burden on staff and resources. Items lodged in the oesophagus were more likely to lead to mucosal injury (p=0.009) compared with elsewhere. Ingested item type and timing of endoscopy were not related to complications (p=0.78) or length of stay (p=0.8). In 12% of cases, no objects were seen at endoscopy.
Conclusion In all except those patients with oesophageal impaction of the object on radiograph, there is no need to perform endoscopic extraction out of hours. A subset of cases can avoid endoscopy with an X-ray immediately prior to the procedure as a significant proportion have passed already. We discuss more holistic approaches to deal with recurrent attendances.
- endoscopic procedures
- economic evaluation
- endoscopy
Data availability statement
All data relevant to the study are included in the article or uploaded as supplementary information.
Statistics from Altmetric.com
Significance of this study
What is already known on this topic
Intentional foreign body ingestion (iFoBI) accounts for approximately 4% of urgent endoscopy procedures and is mostly seen in patients with psychiatric illness and prisoners.
While most accidentally ingested foreign bodies pass through the digestive tract spontaneously intentional ingestion often involves more challenging objects.
Complications such as perforation are reported in 1%–5% of cases with oesophageal impaction carrying the highest risk of perforation (2%).
What this study adds
We report a sustained rise in incidence of iFoBI and its impact on hospital services.
Complications of iFoBI were in line with the published data but generally included only superficial mucosal injury.
Injury was associated with oesophageal location, but there was no relationship between mucosal injury and timing of endoscopy for objects elsewhere in the upper gastrointestinal tract.
How might it impact on clinical practice in the foreseeable future
Middle-of-the-night endoscopy can be avoided in the majority of cases (all except impacted foreign body in oesophagus).
A significant proportion of oesophagogastroduodenoscopies could be avoided by repeat radiological imaging just prior to endoscopic intervention as some of these objects will have passed.
Addressing the cause of recurrent attenders with investment in psychiatric services for patients, psychological support for staff and the use of a clinical ethics committee, including legal expertise, should be incorporated in future guidance.
Background
Intentional foreign body ingestion (iFoBI) is a serious clinical scenario. Ingested items are often metallic (batteries, nails and wire), sharp (razor blades, glass and knives) and/or large (toothbrushes and pens) and can form part of a repeated behaviour.
In the USA, incidence rates of FoBI almost doubled between the period of 2000 and 2017,1 from 3/100 000 persons in 2000 to 5.3/100 000 persons in 2017 with 14%, of the approximate 150 000 cases, intentional in nature. A single-centre UK study identified that among patients with recurrent iFoBI, the mean time between presentations was 22 days.2 Management of iFoBI in the recurrent setting has a significant impact on healthcare resources.3
iFoBI is associated with the psychiatric and prison services4 5 and a reported 80% of patients having a concomitant psychiatric diagnosis at the time of ingestion.6 Compared with single ingestion events, recurrent patients with FoBI were 2.5 times more likely to have a psychiatric diagnosis.7 The relationship between iFoBI and coexistent psychiatric diagnoses is complex. Unfortunately, long-term multidisciplinary management of this patient cohort has proven difficult, with little success.8 9
International guidelines10 11 state that endoscopic management should be guided by presenting symptoms and ingested item features (type, location and size), with emergent endoscopic management for high-risk items/item locations (table 1). If required, endoscopic management is usually successful and serious adverse events occur in <5% of cases.12 13 Intentionally ingested items are often higher risk and so necessitate general anaesthesia and endoscopic management with increased rates of serious adverse events and endoscopic failure requiring surgical intervention.14 15 These risks are amplified and accumulate within patients exhibiting recurrent iFoBI.
The clinical reality of management for 239 admissions with intentional foreign body ingestion at University Hospital Southampton (UHS) compared with European Society of Gastrointestinal Endoscopy (ESGE) guidelines
This study presents a retrospective analysis of iFoBI events over a 5-year period in a UK university hospital. The study characterises the management and outcomes in the real-world setting and compares them to guidelines. Impact on clinical services is discussed, and experienced-based refinements in practice are suggested.
Methods
Inclusion criteria
We collected data concerning all admissions to adult inpatient services at University Hospital Southampton with iFoBI from 1 January 2015 to 31 March 2020.
Accidental FoBI, food bolus obstruction and admissions with the deliberate swallowing of a liquid substance to self-harm (eg, bleach) were excluded.
Data collection
Admissions meeting the inclusion criteria were identified from the electronic patient record (EPR) by searching hospital discharge summaries containing International Classification of Diseases (ICD) codes T18.0–T.18.9 (excluding T18.12, T18.4 and T18.5) and also by searching the endoscopy database for procedures coded: ‘ingestion of foreign body’. Episodes that were managed conservatively, endoscopically and surgically were identified. Data about the individual and the treatment they received on each admission were collected from the EPR.
This was a service evaluation using retrospective routinely collected clinical data. Only the clinical team responsible for managing the FoBI care pathway had access to the data. The clinical team conducted all analyses.
Data analysis
Descriptive statistics were calculated using SPSS for Mac V.26. We tested for associations between categorical variables using χ2 test and compared means between continuous variables with a normal distribution using the independent t-test. To compare medians of parameters with a skewed distribution we used a non-parametric exact test. To calculate the correlation between continuous variables we used the Pearson’s correlation and a one-way between-group analysis of variance was used to test the effect of time (in years) on the monthly rate of admissions with iFoBI. Statistical significance in all analyses was assumed at p<0.05.
Results
Between 1 January 2015 and 31 March 2020 there were 239 admissions with iFoBI. This time period included a significant increase in the monthly frequency of admissions (Welch (5, 17.3)=15.1, p<0.001). Post hoc comparisons using the Tukey test indicated that the significant increase in admissions occurred between 2017 and 2018 (figure 1).
Frequency of admission for iFOBI over the period evaluated.
One hundred and eighty-seven (78%) of admissions involved female patients. Mean age in years was 23 (SD=8 years). Two hundred and twenty-three (93%) involved an individual with a history of psychiatric illness with the most common background of psychiatric diagnosis being emotionally unstable personality disorder (101; 33%). Forty-two per cent were from a forensic psychiatric institution, 48 (25%) were from other psychiatric institutions and 85 (36%) were from the community.
In total, 51 different individuals were admitted with iFoBI. The most frequent attender was admitted on 30 occasions, 23 individuals attended on multiple occasions and 27 (52%) had just a single episode. A wide range of items was ingested including batteries (21.8%), miscellaneous sharp metal items (20.5%), pens (14.2%), lighters (10.9%), toothbrushes (7.9%), wire (7.5%) and a mixture of items (17.2%).
Figure 2 shows the management and outcomes for the 239 admissions and table 1 displays how our practice compares to the European Society of Gastrointestinal Endoscopy guidelines. All but one case had an X-ray to confirm item position soon after admission and over 90% (218) of the patients then underwent attempted item retrieval via oesophagogastroduodenoscopy (OGD) with a general anaesthetic in 150 (69%). At OGD, items were found within the body or antrum of the stomach (144; 67%), the gastric fundus (14; 7%), the oesophagus (24; 11%) and duodenum (8; 4%). On 24 occasions (12% of OGDs), the item(s) were not seen and had presumably passed beyond the second part of the duodenum. On half of these occasions, the item that could not be found was a battery. The equipment used in extraction was recorded on reports from 179 procedures. The most common equipment used to remove the item was a snare (140: 78%) but on nine (5%) occasions a snare was used in combination with a net or forceps. A net was used alone in 12 (7%) procedures and forceps were used alone in 13 (7%) procedures. The use of a protective overtube was recorded on 106 (48%) of procedures.
Outcomes for each episode during the period evaluated. iFoBI, intentional foreign body ingestion; OGD, oesophagogastroduodenoscopy.
The majority of procedures were performed in daylight hours (08:00 to 20:00) (162/218; 74%). The overall mean length of stay for each admission was 2.3 days (SD=5 days) and there was a positive correlation with the time lag between the admission and first OGD (R2=0.198; p<0.001). However, there was no association between time of day the procedure was performed and length of stay for the patient (2.4 days for night-time procedures, 2.5 for daytime procedures; p=0.8).
As displayed in figure 2, item(s) could not be removed at endoscopy on only 16 (7%) occasions, 12 were managed conservatively and 4 required surgical recovery due to inability to retrieve endoscopically. Of the surgical cases, all items were recovered from the stomach intraoperatively. The items recovered from each case were: (1) large glass shard, (2) two scissors, (3) two steak knives, and (4) ‘various metal objects’. Initial endoscopies were completed between 5 and 10 hours from admission and subsequent surgical retrievals within 24 hours from endoscopic attempt. No significant postoperative complications were reported, and length of stay ranged from 8 to 13 days in the cases.
Mucosal injuries were recorded on the reports of 71 (33%) OGDs. There were no reports of serious complications (perforation, stricture or obstruction) or mortality, either by the FoB or during the extraction process. The associations between reported mucosal injury and features of each episode are displayed in table 2. Notably, item position at OGD was significantly associated with mucosal injury (p=0.009) (table 2). Two cases with deep ulceration involved glass items and a battery that had become impacted in the lower oesophagus and one involved an AAA battery in the stomach which had been bitten. These items were removed at endoscopy at 4, 5 and 13 hours from admission respectively.
The association between complications from iFoBI seen at OGD and characteristics of each patient and each admission
Discussion
We report the management and outcomes for patients admitted with iFoBI to a single UK tertiary centre. We have shown that the incidence of these events significantly increased over a 5-year period and that they tended to involve young, female individuals with psychiatric comorbidities.
An underlying explanation is likely multifactorial with psychological, environmental and societal factors all important.8 Given this study focused on a single university hospital providing medical care to local inpatient psychiatric units where recurrent iFoBI individuals were regular inpatients at similar times, a complex network of social interactions cannot be discounted as a contributory factor. Using our retrospective hospital data, it is difficult to speculate further about the aetiology. However, other authors have also reported short-lived clusters of iFoBI events6 and an overall increase in the frequency of these events.1 16 An interesting further study would be to determine whether the increase we have reported is replicated across the UK.
The management of admissions with iFoBI we report is broadly in keeping with international guidelines.10 These guidelines emphasise the importance of timely endoscopy; however, in our centre this had a huge impact on our service with collateral consequences for operating theatres, anaesthetics and bed occupancy. As well as significant financial cost (circa £1500 per episode17), work taken place outside of normal working hours further stresses the system. This has placed considerable stress on our staff, some of whom have experienced the well-described countertransference anger associated with feeling powerless and controlled by the often manipulative individuals.9 18
This workload is contrasted against the low morbidity that was associated with the admissions. Notably, in keeping with other authors, we report no abdominal perforations and only three occasions where the FoB caused ‘deep ulceration’.3 Where mucosal injury did occur, there was no association with length of time to endoscopy and no association with object type. There was no significant difference in the number of reported mucosal injuries between OGDs performed at night or in the daytime.
The key feature relating to complications was the presence of objects lodged in the oesophagus at the time of endoscopy, and as per existing guidelines.10 However, based on our experiences in other situations we believe a less urgent approach than currently stipulated in the guidelines can be contemplated. This would facilitate targeted resource allocation towards high-risk cases such as when sharp objects are within the oesophagus. Specifically, as the time delay before endoscopy was not associated with complications or increased rates of mucosal injury we suggest non-oesophageal extractions can wait until daylight hours. We hypothesise that such a controlled response may remove this instant gratification of ‘the procedure’ and deter repeat attenders. Several individuals attended our hospital with iFoBI on dozens of occasions. In our hospital, a clinical ethics committee, including legal expertise, exists to provide support and wisdom for those managing these challenging cases. The widespread adoption of such a service could support other centre when managing high volumes of admissions involving frequent attenders.
There are important limitations to be aware of on interpretation of our results. First, while all cases were identified in a systematic, reproducible manner based on ICD diagnostic codes, the retrospective nature of the study leaves it open to selection bias. Additionally, a single-centre study design allows regional environmental, social and logistical factors to impact on results and so effects the generalisability of the findings. Finally, the inclusion of cases requiring inpatient admission could underestimate the true number of iFoBI events occurring, missing those managed in the emergency department only, as well as providing a disproportionate number of high-risk ingested objects and subsequent clinical consequences.
Most case series in this field acknowledge this as a challenging area in need of further study.6 7 Available treatment options are multidisciplinary and focus on underlying psychopathology.4 Ultimately, no current management paradigm effectively prevents people from swallowing objects and with no obvious solution, mitigation is the current best outcome and future studies targeted at reducing the impact on receiving institutions, and its staff, are important.
Data availability statement
All data relevant to the study are included in the article or uploaded as supplementary information.
Ethics statements
Patient consent for publication
Acknowledgments
We acknowledge the hard work of our departmental colleagues who performed the endoscopies along with ourselves and also the anaesthetic, endoscopy and theatre teams who assisted with clinical management.
References
Footnotes
Twitter @helpatologist, @marktheliverdoc
Contributors MW conceived, planned, cowrote and submitted the study and is the guarantor for the content. SY and RB collected the data, analysed it and cowrote the paper. NT, BS, IR and PB collected the data, contributed to planning of the study and reviewed the manuscript and data.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.