Open Access
CC BY-NC-ND 4.0 · Z Gastroenterol 2025; 63(12): 1239-1245
DOI: 10.1055/a-2703-6336
Original Manuscript

Food impactions in the esophagus: incidence, causes, and treatment

Bolus-Ereignisse im Ösophagus: Inzidenz, Ursachen und Therapie

Authors

  • Florian Hentschel

    1   General Medicine, MSB Medical School Berlin GmbH, Berlin, Germany (Ringgold ID: RIN381458)
    2   Gastroenterology, HELIOS Klinikum Bad Saarow, Bad Saarow, Germany (Ringgold ID: RIN293382)
  • Isabel Elena Haller

    3   Gastroenterology, Universitätsklinikum Brandenburg an der Havel GmbH, Brandenburg, Germany (Ringgold ID: RIN72208)
  • Juliane Emily Wruck

    3   Gastroenterology, Universitätsklinikum Brandenburg an der Havel GmbH, Brandenburg, Germany (Ringgold ID: RIN72208)
  • Benedikt Seckinger

    4   Gastroenterology, Sana Kliniken Niederlausitz gGmbH, Lauchhammer, Germany (Ringgold ID: RIN169368)
  • Götz Mollenhauer

    3   Gastroenterology, Universitätsklinikum Brandenburg an der Havel GmbH, Brandenburg, Germany (Ringgold ID: RIN72208)
  • Till Bornscheuer

    3   Gastroenterology, Universitätsklinikum Brandenburg an der Havel GmbH, Brandenburg, Germany (Ringgold ID: RIN72208)
  • Stefan Lüth

    5   Center for Internal Medicine II, Städtisches Klinikum Brandenburg GmbH, Brandenburg an der Havel, Germany (Ringgold ID: RIN72208)
 

Abstract

Background

Food impactions in the esophagus are gastroenterologic emergencies. Their incidence is suspected to increase. Standard therapy is endoscopic removal. According to European guidelines from 2016, gently pushing the food forward with the endoscope is safe and efficient. This has recently been doubted, and a retrograde extraction of the food bolus was supposed to be safer.

Methods

We retrospectively analyzed all food impactions in our database over 16 years regarding incidence per year, age, sex, underlying disease, method of removal, success rate, and safety.

Results

From 2008 to 2024, 99 patients were included. Incidence of food impactions increased logarithmically from one per year to 22 per year (p < 0.001). Seventy-nine patients were male (80%; p < 0.001). Underlying diseases were reflux-associated lesions, eosinophilic esophagitis, pseudo­diverticulosis, hernia, achalasia, candidiasis, and others. Fifty-five were pushed forward prior to 2016, afterwards it were 74 (56% and 75% respectively; p < 0.001). Success rate was 100 percent for all techniques. We observed two minor adverse events in the forward-pushing group vs. one in the extraction group (not significant).

Conclusion

Our data confirms that food obstructions are on the rise and that the vast majority of patients are male. There is no typical underlying disease but a wide variety of rare diseases which together form the basis for the increase. Pushing forward is as safe as is extracting the food, so we see no necessity to change the existing guidelines.


Zusammenfassung

Hintergrund

Bolus-Impaktationen sind gastroenterologische Notfälle mit vermutlich steigender Inzidenz. Standardtherapie ist die endoskopische Entfernung. Hierbei ist gemäß Leitlinie des Jahres 2016 das sanfte Vorschieben des Bolus sicher und effizient. Dies wurde kürzlich angezweifelt und die retrograde Extraktion als sicherer empfohlen.

Methoden

Wir analysierten retrospektiv alle Bolusereignisse in unserer Datenbank über 16 Jahre hinsichtlich Inzidenz, Alter, Geschlecht, Grunderkrankung, Entfernungsmethode, Erfolgsrate und Sicherheit.

Ergebnisse

99 Bolusereignisse wurden ausgewertet. Von 2008 bis 2024 stieg die jährliche Inzidenz logarithmisch von 1 auf 22 (p<0,001). 79 Patienten waren männlich (80%; p<0,001). Grunderkrankungen waren Reflux, eosinophile Ösophagitis, Pseudodivertikulose, Hernie, Achalasie, Candidiasis und andere. Vor 2016 wurden 55 Boli vorgeschoben (56%), danach 74 (75%). Die Erfolgsrate lag für beide Techniken bei 100 Prozent. Wir beobachteten zwei kleinere Komplikationen in der Vorschub-Gruppe und eine in der Extraktions-Gruppe (nicht signifikant).

Diskussion

Unsere Daten bestätigen die Vermutung, dass Bolusereignisse deutlich zugenommen haben, und dass die überwiegende Mehrheit der Patienten männlich ist. Grundlage ist weniger eine einzelne typische Erkrankung, sondern eine Vielzahl eher seltener Erkrankungen zusammengenommen. Bezüglich Erfolg und Sicherheit sind Vorschub und Extraktion gleichwertig. Wir sehen daher keine Notwendigkeit, die bestehenden Leitlinien zu ändern.


Introduction

Food impactions in the esophagus are a known phenomenon in gastroenterology. Although few studies on their epidemiology exist, many authors suspect a rising incidence, with current numbers of about 10 to 20 per 100000 per year [1] [2] [3]. The risk is considerably higher in individuals with underlying esophageal diseases such as eosinophilic esophagitis (EoE), pseudodiverticulosis (DEIPD), strictures, or tumors [4] [5] [6]. In these groups, up to 50 percent of all patients may experience single or repeated food impactions at least once in their lifetime [7]. The type of impacted food will usually be meat [1] [8].

Clinically, food impactions present as emergencies with the patient unable to swallow anything including their own saliva. Chest pain can occur but it is not mandatory and, if present, usually mild [9].

Before the widespread availability of flexible endoscopy, several noninvasive methods for bolus removal had been tried, but neither nifedipine, nitroglycerin, diazepam, buscopan, glucagon, nor Coca-Cola proved successful [10] [11] [12] [13] [14]. Standard therapy since the 1980s is endoscopy, which basically gives us two ways of removing the bolus: forward into the stomach, or backwards by retrograde extraction [15] [16].

In the early days of therapeutic endoscopy, it was believed that pushing the food forward would pose an unacceptable risk of laceration, penetration, or bleeding. Consequently, retrograde removal of the impacted bolus was generally recommended, and routinely performed [17]. By the turn of the century, this dogma had been questioned by a growing number of endoscopists, and several studies came out showing that the forward-pushing technique was faster than the retrograde removal, and equally safe [18] [19] [20]. Consequently, the American Society for Gastrointestinal Endoscopy (ASGE) allowed a gentle forward-pushing as early as 2002, and explicitly recommended it in their updated guidelines in 2011 [21] [22]. By 2016, the European Society of Gastrointestinal Endoscopy (ESGE) followed suit and recommended pushing in their clinical guidelines as well [23].

Now in 2023, Heise et al. published a new series of 77 food obstructions in adults and children, of which only 5 were resolved by pushing the bolus forward. In the remaining cases, impacted food was removed by suctioning it into an overtube, a technique that was originally introduced in 1990 [24]. Based on those findings, the authors conclude that pushing forward is unsafe and recommend a change of the ESGE guidelines [25].

Since this is in contrast with otherwise published data [18] [19] [20] as well as with national and international guidelines [22] [23], we undertook this retrospective, single-center, observational study. Its objectives were to examine whether the nature of food impactions and way they are treated had changed in the last two decades, wether pushing can still be considered a safe procedure, and wether the existing guidelines need to be changed.


Patients and Methods

Brandenburg University Hospital is a tertiary center, serving a population of roughly 100000. In January 2025, we retrospectively searched the hospital’s endoscopic database (ViewPoint Versions 5 and 6, General Electric Company, USA) for every food or bolus obstruction from January 1st 2008 to December 31st 2024, thus covering the time span from eight years before the publication of the European Guidelines in 2016, until eight years after [23].

Inclusion criteria were:

  • Patient 18 years or older

  • Admitted to emergency department for suspected esophageal food impaction

  • Underwent endoscopy for food removal within 24 hours after admission.

Exclusion criteria were:

  • All cases in which the obstruction spontaneously resolved between admission and endoscopy

  • All cases in which the stuck object was other than food (4 times tablets, 3 dislocated stents, 1 battery, 1 tracheal cannula, 1 bone splinter, 1 razor blade, 1 hemostatic clip, 1 suture, 1 dislocated Lynx band)

  • All cases in which the documentation regarding type of bolus and / or way of removal was incomplete.

For patients who had more than one episode of bolus obstructions, each episode was counted as one case. This resulted in 99 cases that were included in the study.

Over the time span of 16 years, 12 different endoscopists were involved in treating food impactions. Endoscopy was generally carried out under nurse administered propofol sedation (NAPS) and circulatory monitoring. No patient was intubated. Gastroscopes used were Fujinon EG Series (EG 530CT, EG 530WR, EG 580NW2, EG 600WR; Fujifilm Group, Japan), and Olympus GIF Series (GIF 1T -140, GIF-160, GIF-G145, GIF-XTQ160, GIF-H190, GIF-HQ190; Olympus Corp, Japan). If not pushed forward, food was removed by plain suction, or by the use of forceps, 3- or 5-prong graspers, nets (all: MTW Endoskopie W. Haag KG, Germany), or loops (Fujinon, Fujifilm Group, Japan). However, we did not use an overtube as a suction device.

Data were stored in a spreadsheet program (Excel 2016, Microsoft Corp., USA). Statistical computation was performed using Prism 9.0.0 (GraphPad Software Inc., La Jolla, USA). Binary and numeric data is presented as count and/or percentage. Differences between groups were assessed by chi-square (Χ2) test. A p value < 0.05 was considered significant.

This study has been approved by the ethics committee of the Brandenburg Chamber of Physicians, Ref No. 2024–142-BO-ff. All procedures were in accordance with the standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.


Results

From January 2018 to December 2024, our clinic performed endoscopy in 99 fully documented cases of esophageal food impactions. Median age of patients was 65 years, minimum was one year, maximum was 94 years. Eighty percent of patients were male, 20 % were female (p < 0.001) ([Fig. 1]). Counting food impactions per year, there was a nearly logarithmic increase, with zero to four annual impactions prior to 2016, and up to 22 annual impactions afterwards. Coefficient of correlation (R2) was 0.82 ([Fig. 2]).

Zoom
Fig. 1 Esophageal food obstructions 2008 to 2024 by sex. p < 0.001.
Zoom
Fig. 2 Esophageal food obstructions in Brandenburg from 2008 to 2024. Given are absolute numbers per year and 2nd degree polynomic trend line. R2 = coefficient of correlation. If one would omit the “corona year” 2020, R2 would be greater than 0.9.

The majority of causes for food impactions were reflux-associated, like peptic strictures, mucosal scarring, impaired peristalsis, or ulcers within Barret’s mucosa in 26 cases (25 %, hereinafter referred to as “GERD”), EoE in 16 cases (15 %), and pseudodiverticulosis, hernia, and unspecific esophagitis in 7 cases respectively (7%); these five making up almost two thirds of all patients. The rest was due to a much broader spectrum of diagnoses; in descending order: unspecific esophagitis, achalasia, rings, post-OP stenosis, candidiasis, esophageal carcinoma, radiation esophagitis, scarring after caustic injury, metastasis, and ulcer. In 12 cases (11%), no cause for the bolus obstruction could be identified ([Fig. 3]).

Zoom
Fig. 3 Causes of esophageal food obstructions in Brandenburg. Given are percentages of 99 cases from 2008 to 2024. GERD = complications of gastro-esophageal reflux disease, like scarring, strictures, stenoses, or Barrett’s ulcer; EoE = eosinophilc esophagitis; post OP = anastomotic stricture; Metastasis = compression by lymph note metastases from bronchial carcinoma or gastric carcinoma.

Seven patients were admitted more than once. One patient had three episodes because of peptic stricture (PS), three patients had two episodes each, also because of PS. Two were admitted twice for DEIPD, and one was admitted twice for EoE.

Stuck food was removed by pushing forward in 58 cases (59 %), by pulling it out in 28 cases (28 %), or by a combination of both in 13 cases (13 %). When comparing the times before and after the introduction of the European guidelines in 2016, there was an increase of pushing, alone or in combination, from 56 percent to 75 percent (p < 0.001) ([Fig. 4]).

Zoom
Fig. 4 After the introduction of the European guidelines in Feb 2016, the pushing method (solely or in combination with pulling) increased from 56 percent to 75 percent (p < 0.001). Numbers are percentages; absolute numbers for the time span from 2008 to 2016 are much smaller than from 2017 on (compare [Fig. 2]).

Regardless of the method used, in all cases the food bolus was removed during the initial emergent endoscopy. This denotes a technical success rate of 100 percent.

Overall, we observed three adverse events: One minor bleeding after pushing forward which stopped spontaneously and did not require intervention, one equally minor bleeding after pushing forward and performing bougienage in the same session, and one esophageo-tracheal fistula after removal with net and grasper which was most likely pre-existing.


Discussion

We initially undertook this study to compare different endoscopic methods of removing esophageal food impactions. However, before we could come to this comparison, one incidental finding caught our attention: The steep and significant increase of esophageal bolus impactions over the last sixteen years. Although this increase has been assumed by many, our study is the first to actually prove it for a larger European collective [26] [27]. Looking at the curve, there is only one true outlier in 2020, and this is likely due to the German corona lockdown in that year, with closed restaurants and restrictions on private barbecue parties [28]. If we omitted 2020 from the analysis, the formal correlation would be even stronger with a coefficient greater than 0.9 ([Fig. 2]). Another finding is the vast majority of male patients. Also, this is not unexpected, but we did not think it would be so pronounced [29] ([Fig. 1]).

Of note, less than one third of these obstructions were caused by common diagnoses like GERD or hernia.

The rest were due to a variety of “rare” diseases like EoE, pseudodioverticulosis (DEIPD), achalasia, esophageal carcinoma, chemical or radiation esophagitis, scarring, and others ([Fig. 3]). So even if these diseases are rare, they are diagnoses on the rise, and their rising incidence combined would explain the exponential increase in bolus obstructions we observed [30] [31]. This is especially true for EoE with its worldwide increasing incidence [32] [33] [34] [35] [36] [37] [38], and for DEIPD, a disease that has long been underestimated and has only recently come to the attention of the gastroenterological community [7] [39] [40].

Of note, seven patients had more than one episode of food impaction. This may be astounding at first, but given the fact that most of the underlying diseases are chronic by nature, it is easily explained. (We had one patient suffering from DEIPD who had five episodes in four years but was excluded from the study because of incomplete documentation [41]).

Regarding the method of removal, we took a pragmatic approach: With the patient sedated, we first gave the bolus a gentle push with the endoscope. If this gentle pushing did not work we did not force it but changed our strategy and now tried to remove it with various tools in one piece. If that also failed, we broke it up into fragments until the remains could either be pushed forward, or pulled out completely ([Fig. 4]). However, this sequence was not due to a written standard operation procedure (SOP), but due to the individual decisions of 12 endoscopists over a time span of 16 years. Against this background, we find two aspects remarkable: First, a success rate of 100 percent, and secondly, the negligible number of three minor adverse events, two of them probably unrelated to the procedure. This suggests that the individual decisions were right, and that all methods applied were vastly, and equally, safe.

These results are not extraordinary. Even the first series on forward-pushing from the beginning of the century showed success rates between 40 and 97 percent, with minor adverse events between zero and 3 percent [18] [19] [20]. This trend continued, so that in 2023, a meta-analysis of 18 studies with 3528 participants could calculate a technical success rate of 97.5% for pushing vs. 88.4% pulling, with adverse event rates of 4.03% vs. 2.22%. None of these differences where significant [42].

Insofar, our data are in concordance with the international literature for more than two decades, but in stark contrast to the aforementioned results of Heise et al. [25].

Regarding retrograde extraction, some endoscopists suspected this technique to be the more dangerous one because pulling food through the pharynx of a sedated but not intubated patient would carry the risk of aspiration. Although this is theoretically possible it has rarely been observed in clinical studies, and all authors came to the conclusion that an intubation is not necessary. [43] [44] [45]. Our own data, with 41 uncomplicated extractions in non-intubated patients, support these findings and in this case we are in concordance with Heise at al. Whether special devices such as caps and hoods or the new OTSG Xcavator (Ovesco Endoscopy) would further improve these already excellent results remains to be seen [46].

As a conclusion, we see no reason to alter the existing guidelines or to restrict the use of gentle forward-pushing in the management of food obstructions [22] [23]. On the other hand, we see no reason to restrict the use of retrograde extraction devices either. Rather, we suggest sticking with the proven pragmatic approach based on the experience and preference of the performing endoscopist.


Abbreviations

ASGE: American Society for Gastrointestinal Endoscopy
DEIPD: Diffuse Esophageal Intramural Pseudo-diverticulosis
EoE: Eosinophilic Esophagitis
ESGE: European Society for Gastrointestinal Endoscopy
NAPS: Nurse administered Propofol Sedation


Compliance with Ethical Standards

This study has been approved by the ethics committee of the Brandenburg Chamber of Physicians, Ref No. 2024–142-BO-ff. All procedures were in accordance with the standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.


Contributions

FH planned the study, performed endoscopy and wrote the manuscript. IEH and JEW cared for patients, researched data and literature, performed statistic computation and co-wrote the manuscript. GM, TB, and SL cared for patients, performed endoscopy, researched literature, and edited the manuscript.


Data Availability

All relevant data are within the paper.



Conflict of Interest

The authors declare that they have no conflict of interest.

Acknowledgement

The Authors like to thank Many Große and all nurses and assistants in the endoscopy department.


Correspondence

Dr. med. Florian Hentschel
General Medicine, MSB Medical School Berlin GmbH
Berlin
Germany   

Publication History

Received: 14 February 2025

Accepted after revision: 21 May 2025

Article published online:
08 October 2025

© 2025. The Author(s). This is an open access article published by Thieme under the terms of the Creative Commons Attribution-NonDerivative-NonCommercial-License, permitting copying and reproduction so long as the original work is given appropriate credit. Contents may not be used for commercial purposes, or adapted, remixed, transformed or built upon. (https://creativecommons.org/licenses/by-nc-nd/4.0/).

Georg Thieme Verlag KG
Oswald-Hesse-Straße 50, 70469 Stuttgart, Germany


Zoom
Fig. 1 Esophageal food obstructions 2008 to 2024 by sex. p < 0.001.
Zoom
Fig. 2 Esophageal food obstructions in Brandenburg from 2008 to 2024. Given are absolute numbers per year and 2nd degree polynomic trend line. R2 = coefficient of correlation. If one would omit the “corona year” 2020, R2 would be greater than 0.9.
Zoom
Fig. 3 Causes of esophageal food obstructions in Brandenburg. Given are percentages of 99 cases from 2008 to 2024. GERD = complications of gastro-esophageal reflux disease, like scarring, strictures, stenoses, or Barrett’s ulcer; EoE = eosinophilc esophagitis; post OP = anastomotic stricture; Metastasis = compression by lymph note metastases from bronchial carcinoma or gastric carcinoma.
Zoom
Fig. 4 After the introduction of the European guidelines in Feb 2016, the pushing method (solely or in combination with pulling) increased from 56 percent to 75 percent (p < 0.001). Numbers are percentages; absolute numbers for the time span from 2008 to 2016 are much smaller than from 2017 on (compare [Fig. 2]).