Omar A. Bamalan1*, Abdulaziz A. Bazuhair1, Omar Y. AlKhlaiwy1, Ahmad S. Bubshait1, Eman A. Almohawes2, Khaldoon A. Alkhums1
1General Surgery Department, Security Forces Hospital, PO Box 9003, Dammam 31413, Saudi Arabia
2Medical Imaging Department, Security Forces Hospital, PO Box 9003, Dammam 31413, Saudi Arabia
*Corresponding author: Dr Omar A. Bamalan, General Surgery Department, Security Forces Hospital, PO Box 9003, Dammam 31413, Saudi Arabia.
Received: 20 August 2026; Accepted: 27 August 2026; Published: 02 September 2026
Bowel perforation secondary to Fishbone (FB) ingestion is a common clinical entity, however factors related to FB-induced bowel perforation were rarely studied. Understanding the clinical presentation, management and biomechanics of FB-induced bowel perforation. A case report of a 63-yearold female presented to our centre with a FB-induced bowel perforation and a systematic review using the Cochrane Central Register of Controlled Trials, PubMed, MEDLINE and Web of Science databases from inception to March 2026. The inclusion criteria approved all cases with a diagnosis of FB-induced bowel perforation or related complication (e.g., intra-abdominal collection). The case was examined, investigated and operated on laparoscopically, extracting the FB and primarily repairing the ileal wall, then was discharged in a stable condition. Furthermore, the systematic review yielded a total of 174 studies, of which 30 studies were included for the qualitative and quantitative analysis. An individual’s mere ingestion of a FB poses an increased risk of perforation, if the tip of the FB directly contacts the wall tissues, regardless of the tissue thickness, site or curvature, despite anatomical and health factors being pivotal in determining outcomes in post-perforation sequalae (e.g., Sepsis).
Biomechanics; Bowel; Fishbone; Perforation
biomechanics articles, bowel articles, fishbone articles, perforation articles
The accidental ingestion of Fishbones is a clinical entity with a spectrum of clinical presentations and sequelae (i.e., asymptomatic to septic with peritonitis) [1,2]. In addition, small bowel perforation secondary to ingestion has been reported in several sites of relative narrowing (e.g., ileocecal junction, duodenojejunal junction), different tissue thickness (e.g., stomach, duodenum) and differing patient factors (e.g., multimorbid elders with polypharmacy) [3]. Therefore, in this case report and systematic review we focus on the biomechanics of small bowel perforation secondary to fishbone ingestion.
The patient was consented for photographs use and using the clinical details for research purposes under the Helsinki doctrine, in Security Forces Hospital in Dammam, Saudi Arabia. In addition, the formulated research question is: “In patients with fishbone ingestion, what are the factors, mechanics of bowel perforation, migration and impalement in different sites?” The guidelines of Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) were implemented in this review, with a PROSPERO ID: [CRD420251026514].
A search of the literature was done using the Cochrane Central Register of Controlled Trials (OvidSP), PubMed, MEDLINE (ProQuest) and Web of Science (Clarivate) databases. The available publications from inception until March 2026 were included in the review, with an inclusion criterion of studies that discuss patients with small bowel perforation secondary to fishbones, who underwent conservative or surgical management, worldwide. The included study designs are randomized studies and others (e.g., observational studies, case-control studies, case reports). This review will consider only the articles that have been conducted on human subjects and published in any language. On the other hand, the exclusion criteria encompass studies that discuss small bowel perforation secondary to foreign bodies other than fishbones (e.g., nails, glass), involvement of non-human subjects, and poor-quality studies (i.e. unclear diagnostics, and/or outcomes). After the search was done, All relevant studies were imported into and examined for duplicates using the Microsoft Excel software, then two reviewers (OB and OA) examined the titles and abstracts for possible included studies, and if any variances between the two reviewers’ discussion in regards to a study, a discussion with a third reviewer (AB) is warranted.
Microsoft Excel software was used, and articles were retrieved from the databases if they met the inclusion criteria. Furthermore, excluded articles were reviewed to ensure that no relevant articles were missed in the during the first screening process. Detailed items required for data extraction were gathered and organized in tables for each included study (e.g., clinical presentation, patient demographics, FB ingestion to presentations, etc.).
The Joanna Briggs Institute appraisal tools for different studies were used to assess the quality of included study and address the possible biases within (e.g., study methods and conclusions), and set an objective assessment of results’ reliability.
This is a 63 year old female, with a past medical history of Diabetes mellitus type 2 and Hypertension and a past surgical history of an ovarian cystectomy a few years ago, presented to the Emergency department (ED) with a 2 days history of generalized abdominal pain that started suddenly, post eating her regular meal, stabbing in nature, 9/10 in severity, partially improved on analgesics, aggravating by eating or drinking with no other bleed, obstructive or septic symptomology. The patient reported that on the day prior to her pain she had a seafood meal (fried fish) and probably swallowed a few, small fish bones.
Upon examination, the patient was notably in pain, vitally stable and afebrile, while her abdomen had a diffuse tenderness at the right upper and lower quadrants, no rigidity, no guarding with intact hernial orifices. Furthermore, the patient’s labs upon presentation showed elevated inflammatory marker with normal renal, pancreatic and hepatic markers. The surgical team opted for a computed tomography (CT) with intravenous contrast imaging (Figure 1a and 1b) which showed a linear bone density most likely fish bone penetrating the terminal ileum lumen and extending into right transversus abdominus and internal oblique muscles with minimal surrounding fat stranding, no free air or free fluid seen. Therefore, the patient was shifted for an emergent Laparoscopic exploration and intra-operatively, a fishbone was noted demonstrating what was described on the CT, in which it was retrieved and the defect primarily repaired (Figure 2a and 2b). Post-operatively, the patient was well, tolerating her diet and was discharged. In the outpatient clinic, a few weeks post-operatively, the patient was doing well with no complaints, ED visits or re-admission.
Study Selection
A total of 174 studies were found, 57 duplicates were removed and after a three-step filtration process, 30 studies were included after the application of the inclusion and exclusion criteria quantitative and qualitative appraisal [Table 1 and Figure 3].
Demographics and Background
The resultant included international studies including the majority from Spain [13, 17], China [19, 28-30, 32], United Kingdom [15, 23, 26], Turkey [8, 22, 31] and Japan [5, 23, 25], with a sample size of 87 participants. The number of Male patients was 57 (65.5 %) and Female patients 30 (34.5 %) with a Male-to-female ratio of 2:1. The mean age of included cases was 52.7 ± 3.8 years of age [Range 5 – 83] with 2 patients being below 18 years of age (a 5 years old Male and a 10 years old Female) [12, 32] [Table 1].
The patients past medical and surgical histories included 3 patients with ischemic heart disease requiring angioplasty [17, 24, 29], 3 hypertensive [14, 17, 28] and 2 hypothyroid patients [14, 15]. On the other hand, a patient underwent partial gastrectomy for gastric cancer at the age of 60 [13] and a Roux-en Y gastric bypass (RYGBP) in another patient [15]. However, most of the included studies did not record the patients’ background.
Diagnostics
Clinically, the chief complaint was abdominal pain of varying sites while identification of fish type and duration since ingestion was limited, in 12 studies the duration from ingestion of fish to diagnosis was 7.7 ± 2.5 days, with 5 studies being able to identify the fish type (Codfish, Rockfish, Sea bass, Tilapia and Salmon) [4-6,9,12,14,17,22,24,27,30,31]. Radiologically, the most commonly utilized imaging modality was abdominal computed tomography (A-CT) then abdominal ultrasound (A-US) and rarely a magnetic resonances imaging (MRI) was indicated or done. Furthermore, the site of impaction was mainly the ileum in 12 studies [16,4,8,10-12, 17, 23,24,26,27,31], Duodenum in 6 studies [18,20-22, 29, 30] and Colon in 5 studies (3 in the rectosigmoid, 1 in the right-colon and 1 in the cecum) [4,5,13,16,32]. In addition, there were unusual sites of impaction noted, such as the appendix [7, 28], Omental mass [25], urachal cyst [33] and Meckel’s diverticulum [10, 27]. Nonetheless, there was a statistically insignificant (P > 0.05) correlation between fishbone length to site, days till presentation and age.
Management
There was a spectrum of presentations and imaging results, accordingly the management was individually tailored with three main categories. Firstly, conservative care (i.e., intravenous antibiotics, percutaneous aspiration of abscess, intensive care with follow up imaging) in which 8 patients underwent it. Secondly, endoscopic exploration and retrieval where 9 patients had symptoms alleviation with no noted complications. Lastly, surgical (open or laparoscopic) exploration and extraction was opted in 70 patients with various employed techniques (e.g., retrieval and primary repair, resection and anastomosis, etc.). There were no noted major post-operative complications or functional affection.
Bias assessment
The quality of the included studies was assessed using Joanna Briggs Institute critical appraisal checklists were utilized, to evaluate for bias categories (information bias, selection bias, confounding), study design and statistical analysis. In addition, depending upon the checklists output, the category is labelled as high, intermediate or low risk of bias. The details of the assessment of bias are provided in (Supplementary Tables 1-3).
Foreign body ingestion represents a significant clinical issue, with fishbone ingestion being one of the most frequently encountered scenarios, posing a substantial risk of bowel perforation [34]. We encountered a case of a fishbone penetrating the lumen of the terminal ileum and impaling into the abdominal wall. This prompted us to conduct a systematic review of literature on fishbone-related intestinal perforations and derive factors related to its biomechanics.
Perforation sites and patient characteristics
The systematic review identified multiple studies discussing perforations found in different anatomical sites, we revealed that the most frequent perforation sites are the ileum, followed by the duodenum and stomach. In addition, uncommon sites were found (e.g., Roux-en-Y gastric bypass entero-enterostomy site, Meckel’s diverticulum) affected by fishbones ranging from 1 to 4.5 cm in length. Notably, our findings did not reveal any correlation between the size of the fishbone and the site of Perforation.
The mean age and standard deviation for each anatomical site was calculated as follows: Stomach (59.7 ± 12.5 years), Duodenum (53.6 ± 15.7 years), Jejunum (50.5 ± 8.5 years), Ileum (45.7 ± 22.5 years) and Sigmoid (80 ± 3 years), while abdominal pain was the most frequent symptom. The majority of cases were males noted across all sites, especially in the ileum, representing 60.6%, initially presenting with symptoms lasting for a mean of 2 days, with noted outliers (e.g., extending up to 10 months [13]). The analysis of the outliers yielded no unique characteristics reported to account for the chronic nature of the condition.
Management
Immediate intervention was required in cases of duodenal perforation presenting with septic shock, however the rest were hemodynamically stable. In managing these cases, our data showed that endoscopic retrieval was feasible in hemodynamically stable patient with perforations in the stomach, duodenum, and one reported case involving the sigmoid colon with no noted complications and fishbone sizes ranging from 1.8 to 4 cm. Furthermore, most reported cases were managed via a laparoscopic approach, either involving retrieval with primary repair or bowel resection and anastomosis. A few cases necessitated a laparotomy approach, often owing to either fishbone perforation complications (e.g., splenic abscess), or challenging sites (e.g., entero-enterostomy site of a Roux-en-Y gastric bypass). The data suggest that in hemodynamically stable patients an endoscopic approach is safe for accessible sites, while the laparoscopic approach is an alternative, depending on the surgeon's expertise in laparoscopy.
Biomechanics
The mechanics of intestinal contraction is crucial for understanding the mechanics of fishbone-induced intestinal perforations. The contraction of smooth muscle fibres within the intestines generates a mechanical force on the mucosal wall that imparts movement to the underlying contents, facilitating flow [35]. To understand this further, we formulated an equation to estimate the force with which a fishbone might perforate the wall of the intestines [Figure 4], we utilized our formula within a theoretical scenario of a fishbone perforating ileal wall, relying on numerical estimates derived from our systematic search results [4-33]. The typical thickness of the ileal wall is 2.5 mm, with a curvature radius of 20 mm, and an estimated tissue failure stress at 1.5 MPa and given a fishbone with a tip area of 0.5 mm², travelling at a momentum 1.4 x 10-3 Kg•m/s (a typical sized fishbone propagation through 5 cm of bowel) conveys a force of 255 MPa if the tip of the fishbone comes in contact with the bowel wall. However, a force of approximately 0.19 N could theoretically perforate the ileum (this force is less than the gravitational force acting on a small coin). Accordingly, perforation occurs if force applied by the fishbone exceeds the critical stress threshold of a tissue. Furthermore, in an attempt to expand the scope of our hypothesis we altered the numerical parameters in the equation, yet it did not yield substantial changes in results, even when varying the radius or wall thickness.
Consequently, we hypothesize that the potential for fishbone penetration is not confined to a specific site and can occur at any location within the gastrointestinal tract, if the tip of the fishbone comes in contact with the wall tissues [Figure 5]. This hypothesis is further supported by our findings of fishbone-induced perforations in different anatomical sites with varying diameters and wall thicknesses [4-33]. Furthermore, our case revealed that a fishbone penetrated the ileal wall with sufficient force to be impaled into the abdominal wall as well. This suggests that the actual force exerted by a fishbone might exceed our predicted estimations
Lastly, we acknowledge that this is a systematic review with included studies having varying quality [Supplementary tables 1-3] and the possibility of a range of biases (e.g., selection and conclusion bias). However, we highlight an important area of unprecedented research showcasing the biomechanics of fishbone-induced perforation and formulating supportive evidence through physics. Therefore, more experimental trials should be initiated to further understand the reasoning of different sites perforation (i.e., rather than the well-studied distal ileum), different times till presentation and possible sequelae.
The qualitative and quantitative analysis of our case and systematic review indicates no clear correlation between fishbone size and perforation site, suggesting that an individual’s mere ingestion of a fishbone poses an increased risk of perforation at any site of the gastrointestinal tract, despite anatomical and health factors being pivotal in determining outcomes in post-perforation sequalae (e.g., Sepsis). The risk has been analysed and proven to be present regardless of bowel diameter, wall thickness and content. This highlights the necessity for heightened awareness among healthcare professionals regarding the implications of fishbone ingestion (e.g., intra-abdominal abscess). Future research should focus on elucidating the biomechanics involved and developing standardized management protocols to enhance patient care and outcomes.
This study was performed in accordance with the Declaration of Helsinki. Ethics approval for this human study was waived by Security Forces Hospital. All adult participants provided written informed consent to participate in this study. Written informed consent was obtained from the individual(s) for publication of the details of their medical case and any accompanying images.
No funding was granted
none
The authors would like to thank Naser F. Ashour for his support in editing the figures and illustrations used in this paper.
Supplementary Table 1.
|
Case report |
Question (Q) |
Low risk of bias |
Intermediate risk of bias |
High risk of bias |
|
Information bias |
3,4,7 |
Answer Yes 3 times |
Answer Yes 1/2 times |
Answer Yes 0 times |
|
Selection bias |
1,2 |
Answer Yes 2 times |
Answer Yes 1 times |
Answer Yes 0 times |
|
Confounding |
5,6 |
Answer Yes 2 times |
Answer Yes 1 times |
Answer Yes 0 times |
|
Statistical quality |
8 |
Answer Yes 1 times |
Answer Yes 0 times |
|
Cohort |
Question |
Low risk of bias |
Intermediate risk of bias |
High risk of bias |
|
Information bias |
2,3,7,8,9,10 |
Answer Yes 5/6 times |
Answer Yes 3/4 times |
Answer Yes 0/1/2 times |
|
Selection bias |
1,6 |
Answer Yes 2 times |
Answer Yes 1 times |
Answer Yes 0 times |
|
Confounding |
4,5 |
Answer Yes 2 times |
Answer Yes 1 times |
Answer Yes 0 times |
|
Statistical quality |
11 |
Answer Yes 1 times |
Answer Yes 0 times |
Supplementary Table 2.
The table below shows the results of the different JBI questionnaires (cohort studies and Case report studies).
Available on: https://jbi.global/critical-appraisal-tools
|
Cohort |
Q1 |
Q2 |
Q3 |
Q4 |
Q5 |
Q6 |
Q7 |
Q8 |
Q9 |
Q10 |
Q11 |
|
Campos et al, 2020 [16] |
NA |
NA |
Yes |
UC |
UC |
Yes |
Yes |
UC |
NA |
NA |
Yes |
|
Case Reports |
Q1 |
Q2 |
Q3 |
Q4 |
Q5 |
Q6 |
Q7 |
Q8 |
|||
|
Dupont et al, 2024 [4] |
Yes |
Yes |
Yes |
Yes |
Yes |
Yes |
Yes |
Yes |
|||
|
Imoto et al, 2024 [5] |
No |
Yes |
Yes |
Yes |
Yes |
Yes |
No |
Yes |
|||
|
Saleem et al, 2023 [6] |
Yes |
UC |
Yes |
Yes |
Yes |
Yes |
Yes |
Yes |
|||
|
Das et al, 2023 [7] |
UC |
Yes |
Yes |
Yes |
Yes |
Yes |
Yes |
Yes |
|||
|
Demir et al, 2022 [8] |
UC |
UC |
Yes |
Yes |
UC |
UC |
UC |
No |
|||
|
Xia et al, 2022 [9] |
UC |
Yes |
Yes |
Yes |
Yes |
Yes |
Yes |
Yes |
|||
|
Baca et al, 2022 [10] |
Yes |
UC |
Yes |
Yes |
Yes |
Yes |
No |
Yes |
|||
|
Crowley et al, 2021 [11] |
No |
UC |
UC |
Yes |
UC |
Yes |
UC |
Yes |
|||
|
Nguyen et al, 2021 [12] |
Yes |
Yes |
Yes |
Yes |
Yes |
Yes |
Yes |
Yes |
|||
|
Kanaoka et al, 2021 [13] |
Yes |
Yes |
Yes |
Yes |
Yes |
Yes |
Yes |
Yes |
|||
|
Silva et al, 2021 [14] |
UC |
Yes |
Yes |
Yes |
Yes |
Yes |
Yes |
Yes |
|||
|
Traynor et al, 2020 [15] |
UC |
UC |
Yes |
Yes |
Yes |
Yes |
Yes |
UC |
|||
|
Saleh et al, 2019 [17] |
UC |
Yes |
Yes |
Yes |
Yes |
Yes |
Yes |
Yes |
|||
|
Lemaître et al, 2018 [18] |
UC |
UC |
Yes |
Yes |
No |
No |
No |
No |
|||
|
Sierra-Ruiz et al, 2016 [19] |
UC |
Yes |
Yes |
Yes |
Yes |
Yes |
Yes |
Yes |
|||
|
Jiménez-Fuertes et al, 2016 [20] |
UC |
UC |
UC |
Yes |
Yes |
Yes |
UC |
UC |
|||
|
Maree et al, 2015 [21] |
Yes |
UC |
Yes |
Yes |
Yes |
Yes |
Yes |
Yes |
|||
|
Tumay et al, 2015 [22] |
UC |
UC |
UC |
Yes |
Yes |
Yes |
UC |
Yes |
|||
|
Hakeem et al, 2015 [23] |
UC |
Yes |
UC |
Yes |
Yes |
Yes |
Yes |
Yes |
|||
|
Beecher et al, 2015 [24] |
Yes |
Yes |
Yes |
Yes |
Yes |
Yes |
UC |
Yes |
|||
|
Yamamoto et al, 2007 [25] |
Yes |
Yes |
Yes |
Yes |
Yes |
Yes |
Yes |
Yes |
|||
|
Law et al, 2003 [26] |
UC |
Yes |
Yes |
Yes |
Yes |
Yes |
UC |
Yes |
|||
|
Henrik Christensen, 1999 [27] |
Yes |
UC |
Yes |
No |
Yes |
Yes |
UC |
Yes |
|||
|
Beh et al, 2016 [28] |
Yes |
Yes |
Yes |
Yes |
Yes |
Yes |
UC |
Yes |
|||
|
Gao et al, 2024 [29] |
UC |
UC |
UC |
UC |
Yes |
Yes |
UC |
Yes |
|||
|
Yi et al, 2020 [30] |
Yes |
UC |
UC |
UC |
Yes |
Yes |
UC |
Yes |
|||
|
Mutlu et al, 2012 [31] |
Yes |
UC |
Yes |
Yes |
Yes |
No |
No |
Yes |
|||
|
Lau et al, 2023 [32] |
Yes |
UC |
UC |
Yes |
Yes |
Yes |
UC |
Yes |
|||
|
Zin et al, 2021 [33] |
Yes |
Yes |
Yes |
Yes |
Yes |
Yes |
Yes |
Yes |
Abbreviations: NA=not applicable, Q=question, UC=unclear
Supplementary Table 3.
|
Table 3. Assessment of bias of all included studies |
||||
|
Author |
Information Bias |
Selection Bias |
Confounding |
Statistical Quality |
|
Campos et al, 2020 [16] |
- |
+/- |
- |
+ |
|
Case Reports |
||||
|
Dupont et al, 2024 [4] |
+ |
+ |
+ |
+ |
|
Imoto et al, 2024 [5] |
+/- |
+/- |
+ |
+ |
|
Saleem et al, 2023 [6] |
+ |
+/- |
+ |
+ |
|
Das et al, 2023 [7] |
+ |
+/- |
+ |
+ |
|
Demir et al, 2022 [8] |
+/- |
- |
- |
- |
|
Xia et al, 2022 [9] |
+ |
+/- |
+ |
+ |
|
Baca et al, 2022 [10] |
+/- |
+/- |
+ |
+ |
|
Crowley et al, 2021 [11] |
+/- |
- |
+/- |
+ |
|
Nguyen et al, 2021 [12] |
+ |
+ |
+ |
+ |
|
Kanaoka et al, 2021 [13] |
+ |
+ |
+ |
+ |
|
Silva et al, 2021 [14] |
+ |
+/- |
+ |
+ |
|
Traynor et al, 2020 [15] |
+ |
- |
+ |
- |
|
Saleh et al, 2019 [17] |
+ |
+/- |
+ |
+ |
|
Lemaître et al, 2018 [18] |
+/- |
- |
- |
- |
|
Sierra-Ruiz et al, 2016 [19] |
+ |
+/- |
+ |
+ |
|
Jiménez-Fuertes et al, 2016 [20] |
+/- |
- |
+ |
- |
|
Maree et al, 2015 [21] |
+ |
+/- |
+ |
+ |
|
Tumay et al, 2015 [22] |
+/- |
- |
+ |
+ |
|
Hakeem et al, 2015 [23] |
+/- |
+/- |
+ |
+ |
|
Beecher et al, 2015 [24] |
+/- |
+ |
+ |
+ |
|
Yamamoto et al, 2007 [25] |
+ |
+ |
+ |
+ |
|
Law et al, 2003 [26] |
+/- |
+/- |
+ |
+ |
|
Henrik Christensen, 1999 [27] |
+/- |
+/- |
+ |
+ |
|
Beh et al, 2016 [28] |
+/- |
+ |
+ |
+ |
|
Gao et al, 2024 [29] |
- |
- |
+ |
+ |
|
Yi et al, 2020 [30] |
- |
+/- |
+ |
+ |
|
Mutlu et al, 2012 [31] |
+/- |
+/- |
+/- |
+ |
|
Lau et al, 2023 [32] |
+/- |
+/- |
+ |
+ |
|
Zin et al, 2021 [33] |
+ |
+ |
+ |
+ |
Abbreviations: (+) Low risk of bias, (+/-) intermediate risk of bias, (-) high risk of bias, (NA) not applicable.