Keywords
Foreign bodies - GI surgery
Introduction
Foreign body ingestion is a common cause for Emergency Department (ED) presentation.
It is more common in the pediatric population compared with adults, in whom the cause
is usually accidental [1]. In adults, patients with underlying psychiatric comorbidity, the elderly, those
with alcohol intoxication, and prisoners account for the vast majority of foreign
body ingestions [2]
[3]. Foreign bodies vary widely with respect to material, shape, length, width, and
number, all of which affect the diagnostic and management approach. Imaging can assist
with risk stratification of patients based on the location, size, and number of ingested
objects as well as exclusion of complications such as perforation [4]. The majority of ingested foreign bodies pass through the alimentary tract without
complication. Endoscopic retrieval is recommended following ingestions with high risk
of complication, or ingestion of foreign bodies that are unlikely to traverse the
gastrointestinal tract, and approximately 1% of patients require surgery due to complications
or failed endoscopy [4]
[5]
[6]. Repeated ingestion of foreign bodies by people with psychiatric comorbidity and/or
prisoners may be associated with secondary gain, which makes behavioral management
challenging. Conservative management may break a cycle of repeated swallowing behavior,
and may be safe even after foreign body ingestion for which guidelines recommend endoscopic
retrieval.
The aim of this study was to review the management of patients presenting to a single
tertiary hospital with foreign body ingestion in comparison with best practice guidelines.
We were specifically interested in outcomes among patients with repeated episodes
of ingestion of foreign bodies who were managed conservatively.
Patients and methods
A retrospective review of all patients presenting with foreign body ingestion to a
tertiary hospital in Melbourne, Victoria, was conducted over the 5-year period between
January 2017 and December 2021. The hospital has a prison ward and is the referral
center for all prisoners requiring hospital admission in the state of Victoria. The
hospital’s on-call endoscopy roster was staffed by 31 endoscopists.
Inclusion and exclusion criteria
Adults aged 18 years and older presenting to the ED with foreign body ingestion were
included. Patients presenting with clinical or radiological evidence of a perforated
viscus or a food bolus were excluded. Patients were identified through the hospital
information system using the hospital coding for foreign body in the esophagus (T18.1),
stomach (T18.2), small intestine (T18.3), other parts of the alimentary tract (T18.8),
and alimentary tract unspecified (T18.9).
Variables and outcomes
Demographic data were collected from electronic medical records and included age,
gender, and financial class (private, public, prisoner). Descriptive data including
foreign body type, length and diameter, location on imaging, and symptomatology were
documented. Imaging modalities, management strategy, complications, length of stay,
and re-presentation rates were all recorded. High-risk foreign body ingestion was
defined as any of the following: sharp objects, hard object length >5 cm, diameter
>2.5 cm, button battery and/or magnet ingestion, and esophageal location [4]
[6]. Complications were defined as perforation, luminal obstruction, or fistula formation.
Statistical analysis
All statistical analysis was performed using SPSS V28. Continuous variables were assessed
using mean values and compared using the student t-test for parametric data and Mann-Whitney
U test for non-parametric data. Categorical variables were assessed using median values
and compared using the Chi-square test.
Ethics statement
St. Vincent’s Human Research Ethics Committee granted approval to project number 2022/PID06406
in accordance with the research conforming to the National Health and Medical Research
Council Act 1992 and the National Statement on Ethical Conduct in Human Research 2007
(updated July 2018).
Results
A total of 157 presentations by 63 patients with foreign body ingestion occurred between
2017 and 2021 ([Fig. 1]). Fifty percent of patients were male; the median age 30 years (interquartile range
[IQR] 25–30 years). The majority of presentations occurred in prisoners (n=104 presentations
[65%], 36 patients) ([Table 1]). Recurrent presentations were common (median=3, range 1–30); recurrent presentation
was more common in prisoners (36% vs. 20%, P=0.05). Two prison patients had a very high number of presentations (30 and 29, respectively),
accounting for almost 40% of all presentations. Of the patients, 56% had previous
documentation of a mental health disorder, and all patients with recurrent presentations
had an underlying mental health disorder. A previous history of intentional self-harm
was documented in 44% of patients. Prisoners were younger and were more likely to
be male ([Table 1]).
Fig. 1 Flowchart of patients.
Table 1 Demographics of prisoners vs. non-prisoners.
|
|
Prisoners
|
Non-prisoners
|
P value
|
|
*t-test, †Chi-square test, IQR, interquartile range.
|
|
Number of patients
|
36
|
27
|
|
|
Number of presentations (%)
|
104 (66)
|
53 (34)
|
|
|
Number of presentations per patient (median IQR)
|
6.5 (1–22.25)
|
2 (1–5)
|
<0.001*
|
|
Age (median, IQR)
|
28 (26–30)
|
33 (25–33)
|
<0.001*
|
|
Gender n, (% patients)
|
|
Male
|
29 (81)
|
14 (52)
|
0.005†
|
|
Female
|
7 (19)
|
12 (44)
|
|
|
Non-binary
|
0 (0)
|
1 (4)
|
|
|
Psychiatric comorbidity (% patients)
|
23 (64)
|
13 (48)
|
0.165†
|
|
History of self-harm (% patients)
|
18 (50)
|
12 (44)
|
0.492†
|
The most commonly ingested objects were batteries (23%), alleged drug-containing balloons
(17%), razor blades (16%), magnets (4%), and miscellaneous (e.g. TV/radio parts, pens,
cutlery) (40%) ([Fig. 2] and [Fig. 3]). Multiple different objects were ingested in 25 presentations (16%). High-risk
foreign body ingestion occurred in 103 of 157 presentations (66%). Sharp objects were
observed in 43 presentations, 52 presentations had an object length >5 cm, nine presentations
had an object diameter >2.5 cm, 11 presentations with foreign body were located in
the esophagus, and 49 presentations were with magnets and/or battery ingestion. High-risk
foreign body ingestion was less common in the prison population (58% vs 81%; P=0.003).
Fig. 2 A 27-year-old male prisoner ingested a 35-cm television cable. He was managed conservatively
and the object passed 3 weeks after ingestion.
Fig. 3 A 20-year-old non-binary person with recurrent presentations ingested a metal spoon
and, as part of a behavioral management approach, was initially managed conservatively.
After the spoon failed to pass on serial imaging, endoscopy was attempted. At endoscopy,
the spoon was lodged in the duodenal cap with erosion into the duodenal wall at both
ends, such that it could not be removed endoscopically. The patient proceeded to surgery.
Urgent endoscopy was performed in 45% of presentations. All presentations with a foreign
body lodged in the esophagus (n=11) were treated endoscopically. Patients managed
endoscopically were more likely to present with high-risk ingestions (78% vs. 22%,
P=0.002). Of presentations in patients undergoing endoscopy, 86% (59/69) had successful
retrieval of the foreign body. The foreign bodies passed into the small bowel by the
time of endoscopy in 12% of presentations (8/69) and were not able to be retrieved.
Two patients failed endoscopic management and proceeded to surgery. The first patient
ingested 47 magnets, while the second ingested 500 coins. In both cases, endoscopic
extraction was not possible and the patients required surgical gastrostomy. A third
case of a patient with recurrent presentations (n=5) involved ingestion of a metal
spoon ([Fig. 3]) and was initially managed conservatively. After the spoon failed to pass on serial
imaging, endoscopic retrieval was attempted. At endoscopy, the spoon was lodged in
the duodenal cap with erosion into the duodenal wall at both the proximal (D1) and
distal (D2) ends, such that it could not be removed endoscopically. The patient proceeded
to surgery.
Of the cases, 55% were managed conservatively, defined as clinical observation without
urgent endoscopy. Presentations with low-risk ingestions were more likely to be managed
conservatively (70% vs 46%, P = 0.002), while presentations with high-risk ingestions were more likely to be managed
endoscopically (54% vs 30%, P=0.002) ([Table 2]).
Table 2 Low-risk vs. high-risk foreign body ingestion.
|
|
Low-risk
|
High-risk
|
P value
|
|
*t-test., †Chi-square test, IQR, interquartile range; LOS, length of stay.
|
|
Number of patients
|
27
|
39
|
|
|
Number of presentations (%)
|
54 (34)
|
103 (66)
|
|
|
Number of presentations per patient (median IQR)
|
4 (1–21)
|
6 (1–12.5)
|
0.121*
|
|
Age (median, IQR)
|
26.5 (25–30)
|
28 (26–30)
|
0.733*
|
|
Gender n, (% patients)
|
|
Male
|
18 (67)
|
24 (62)
|
0.153†
|
|
Female
|
8 (30)
|
15 (38)
|
|
|
Non-binary
|
1 (3)
|
0 (0)
|
|
|
Psychiatric comorbidity (% patients)
|
17 (41)
|
28 (72)
|
0.037†
|
|
Management (% presentations)
|
0.002†
|
|
Conservative
|
38 (70)
|
47 (46)
|
|
|
Endoscopic
|
16 (30)
|
56 (54)
|
|
|
LOS (median, IQR)
|
2 (1–3)
|
1 (1–3)
|
0.044*
|
|
Re-presentation (% presentations)
|
18 (33)
|
35 (34)
|
0.799†
|
Management did not differ between patients presenting for the first time and recurrent
presenters (conservative management 51% vs 56%, P=0.460). Management did not differ between first-time and recurrent presenters with
high-risk ingestion (conservative management 46% vs. 50%; P=0.694). However, in patients presenting with low-risk ingestions, recurrent presenters
were more likely to be managed conservatively compared with first-time presenters
(conservative management 83% recurrent presentations vs. 54% first presentations;
P=0.042).
As previously discussed, one case required surgical intervention for a metal spoon
lodged in the duodenal cap with erosion into the duodenal wall. In the remainder of
the cohort, no cases of perforation, luminal obstruction, or fistula occurred. The
median length of stay (LOS) for all foreign body ingestion presentations was 2 days
(range 1–13 days) and did not differ between patients receiving conservative vs. endoscopic
management (median 2.1 (range 1–13) days vs. 2.4 (range 1–7) days; P=0.408). However, in high-risk ingestions, conservative management had a shorter LOS
compared with endoscopic management (median LOS 1 day vs 2 days; P=0.044).
Thirty-day re-presentation with further foreign body ingestion was common (31%), with
a median of three presentations per patient. Patients with intentional ingestion (34%
vs. 0%; P=0.021), underlying mental health disorders (38% vs. 0%; P <0.01), and documented history of self-harm (36% vs. 20%; P=0.05) were more likely to re-present with foreign body ingestion.
High-risk ingestions managed conservatively
Forty-seven presentations (24 patients) involved high-risk ingestions that were managed
without endoscopy ([Table 3]). In this cohort, the most commonly ingested foreign bodies were razor blades (41%),
batteries (14%), and drug-containing balloons (9%). The most common reason for pursuing
conservative management was that the object had passed the duodenum on imaging (n=28,
60%) and endoscopy was felt to be futile. There were four presentations (9%) in which
there was a history provided of ingestion of a high-risk object, but radiology was
negative and endoscopy was not performed. Endoscopy was refused by the patient for
three presentations. Conservative management was pursued in 12 presentations (25%)
(7 patients), all of which were recurrent presentations by patients in whom a behavioral
strategy was being pursued.
Table 3 High-risk foreign body ingestion managed conservatively.
|
Total number of high-risk presentations
|
102
|
|
IQR, interquartile range.
|
|
Number of presentations with high-risk ingestion managed conservatively (n)
|
47
|
|
Age (median, IQR)
|
29 (27–32.5)
|
|
Gender n, (%)
|
|
Male
|
23 (49%)
|
|
Female
|
24 (51%)
|
|
First presentation (n)
|
15
|
|
Sharp object (n)
|
27
|
|
Length >5 cm (n)
|
17
|
|
Diameter >2.5 cm (n)
|
2
|
|
Button battery and/or magnet (n)
|
8
|
|
Esophageal location (n)
|
0
|
Re-presentation with further foreign body ingestion was common (34%) in high-risk
ingestions managed conservatively. In more detail, 12 presentations (7 patients) had
conservative management as part of a behavioral management strategy after recurrent
presentations with foreign body ingestions. This was a multidisciplinary plan developed
to manage admitted secondary gain associated with hospital transfer, analgesia, and
sedation after multiple recurrent presentations in a small number of prisoners. The
multidisciplinary team included gastroenterologists, emergency physicians, psychiatrists,
nurses, and prison clinical staff Conservative management was then pursued for foreign
body ingestion in this subgroup, as long as the foreign body had passed the esophagus
and was not associated with clinical suspicion of peritonitis. No cases of perforation,
luminal obstruction, or fistula occurred in this behavioral management cohort. In
five of seven patients, all prisoners, there were no more ingestion episodes after
the decision was made to pursue conservative management for all foreign body ingestion
(median follow-up 12 months). The remaining two patients continued to ingest foreign
bodies, characterized by crescendo presentations with evidence of decreased frequency
of presentation after institution of conservative management (Supplementary Fig. 1 and Supplementary Fig. 2). This included the patient who required surgical removal of the metal spoon.
Discussion
Foreign body ingestion was a common presentation to our health service over the period
of review. Our health service currently holds the prison contract for the state of
Victoria, and this likely contributed to the relatively high rates of foreign body
ingestion, with two-thirds of presentations occurring in prisoners. People who present
recurrently with foreign body ingestion are an uncommon, but very challenging patient
population, especially among prisoners, and consume very high levels of healthcare
resources. Prisoners may present with foreign body ingestion for secondary gain. However,
intentions can vary widely and include suicidal ideation, self-mutilation, masochism,
genuine accidental ingestions, and drug trafficking [7]
[8]. Psychiatric comorbidity is common among patients with recurrent ingestions [2]
[3]
[9]
[10]. Recurrent ingestions may represent a self-harm behavior; they may also involve
secondary gain.
In this experience, two-thirds of presentations were classified as high-risk foreign
body ingestions. Conservative management was the most common management approach,
either because the foreign body had passed into the small bowel by the time of presentation
to hospital, or as part of a multidisciplinary behavioral management strategy after
very frequent re-presentations. Non-endoscopic management was safe. In the cohort
of high-risk ingestions managed conservatively, one case failed conservative management
and required surgery. The data suggest that in the appropriate clinical context, conservative
management is safe in this cohort.
Our data also highlight that a significant minority of patients with low-risk foreign
body ingestion proceeded to endoscopy. The data highlight the need for ongoing education
and defined clinical pathways to manage patients with low-risk ingestions as well
as high-risk ingestions.
The data suggest that a multidisciplinary behavioral management strategy that does
not involve endoscopy may be safely developed for a subset of patients with recurrent
foreign body ingestion and complex psychopathology ([Fig. 4]). The one case in our experience in which conservative management failed involved
a metal spoon that lodged between the duodenal cap and the wall of the second part
of the duodenum. Recent data suggest that foreign body length is a key characteristic
in predicting perforation or failure to progress, necessitating surgical intervention
[11]. Other caveats to consider include patients presenting with symptoms of luminal
obstruction, radiological evidence of foreign body in the esophagus, ingestion of
multiple magnets, and ingestion of button batteries. In such cases, endoscopy should
be considered due to the risk of complications. The decision to consider a strategy
of conservative management should involve multidisciplinary discussion including gastroenterologists,
emergency physicians, surgeons, psychiatrists, social workers, and where relevant,
prison clinical staff.
Fig. 4 Proposed management of recurrent high-risk foreign body ingestion.
The rate of re-presentation with foreign body ingestion within 30 days in this challenging
prisoner population was high. The prison population was particularly challenging with
higher rates of re-presentation compared with the general population. Re-presentation
was more common in patients with underlying psychiatric comorbidity. The data highlight
the complexity of this patient population and the need for holistic, multidisciplinary
management approaches.
There are a number of limitations to our study. First, this was a single-center study,
and the data may lack generalisability, particularly given our health service cares
for the state’s prison population. Two patients accounted for over one-third of all
presentations, which may introduce selection bias and affect generalizability. Although
complications as a result of conservative management were rare, patients may have
presented to other health services unbeknownst to our unit, although this is unlikely
in the prison population. Furthermore, patients presenting to multiple health services
can also affect re-presentation rates. Finally, patients with underlying mental illness
may re-present to hospital with other forms of self-harm, and this could not be identified
within the limits of this review.
Conclusions
Patients with recurrent foreign body ingestion represent a challenging population
and consume high levels of healthcare resources. Patients with recurrent presentations,
a history of mental health disorder, and suspected secondary gain can be managed safely
with a conservative, multidisciplinary approach in the appropriate clinical context.
