Endoscopic Treatment of Gastric Bezoars: A Report of Three Cases
Article information
Abstract
Bezoars, including phytobezoars, trichobezoars, and pharmacobezoars, are accumulations of undigested substances in the gastrointestinal tract. We report three cases of gastric bezoars. Case 1: An 86-year-old woman presented with a one-month history of abdominal pain and vomiting. Esophagogastroduodenoscopy revealed gastric bezoars; consumption of 2 L of cola daily for 2 weeks resulted in complete disappearance of the bezoars. Case 2: An asymptomatic 63-year-old woman underwent esophagogastroduodenoscopy, which revealed a gastric bezoar. Cola spraying and endoscopic lithotomy were ineffective; therefore, the patient underwent laparoscopic removal of the bezoar for management of small bowel obstruction secondary to the bezoar fragments. Case 3: A 6-year-old girl with a history of pica underwent two laparoscopic surgeries 10 months apart for recurrent trichobezoars. We report our treatment approach in three patients who presented with gastric bezoars.
INTRODUCTION
Bezoars refers to undigested substances accumulating in the gastrointestinal tract to form a mass. The three most common types are phytobezoar, trichobezoar, and pharmacobezoar [1]. Phytobezoar is the most common form of bezoar, which is a mass of unabsorbed dietary fiber accumulated in the gastrointestinal tract. It is also associated with reduced motility of the gastrointestinal tract. Pharmacobezoars can occur after taking unabsorbed antacids, some cardiovascular disease drugs, vitamins, cholecstyramine.
Trichobezoars are clumps of hair in the stomach or small bowel and can occur anywhere in the gastrointestinal tract. Because it occurs when long hairs become tangled, it has the characteristic of being longer and forming a lump than a phytobezoar or a pharmacobezoar. A special type of trichobezoar that extends from the stomach to the small bowel is called Rapunzel syndrome. Rapunzel syndrome usually occurs in girls who have psychiatric disorders such as hair-pulling, trichotillomania, pica, mental retardation, or depression [2].
We have experienced cases of phytobezoars and trichobezoars, and would like to report these cases to discuss the treatment process and prevention of recurrence.
CASE REPORT
Case 1
An 86-year-old female patient visited our hospital because gastric bezoars were observed on computed tomography (CT) and esophagogastroduodenoscopy (EGD). She said that vomiting and pain in the solar plexus persisted for about two months, so she underwent an abdominal CT at a local clinic, and gastric bezoar was observed, so EGD was performed. A bezoarlike filling defect was observed in her gastric fundus on an abdominal CT performed at a local clinic (Fig. 1A). A gastric bezoar was confirmed on EGD performed two days later. She was said to have cut it into two pieces through snaring during the EGD (Fig. 1B). The patient said that she drank more than 2 L of cola every day for 10 days until she came to our hospital. At the time of her visit, she said that her defecation was going well without change.
At the time of her visit to the hospital, there were no remarkable findings in her vital signs or physical examination. Her underlying diseases included complete atrioventricular block with a pacemaker inserted, hypertension, dyslipidemia, and mild cognitive impairment. She had no history of previous abdominal surgery. She said she had not recently changed any of her medications, she had never taken any dietary supplements or persimmons, and she did not have pica. She said there were no notable findings in her EGD two years ago. She had never experienced a bezoar before, and she did not have diabetes or thyroid disease.
The patient continued to drink more than 2 L of cola every day, and a follow-up EGD performed at our hospital four days later confirmed that the bezoars had completely disappeared (Fig. 2A). She had no abdominal pain and no defecation. Because of the possibility that the bezoar had passed into the small bowel, a Hypaque swallowing test was performed, and no abnormal findings were observed (Fig. 2B).
Case 2
A 63-year-old female patient was referred because gastric bezoar was continuously observed on EGD. There were no symptoms, such as abdominal pain, and the vital signs were normal at the time of visiting the hospital. There were no unusual findings in the physical examination. Her underlying diseases included hypertension, hyperlipidemia, and depression, and she had no previous history of abdominal surgery. Hypertension was not being treated, and no medications had been recently changed. She also had no particular drug or food history. She said she was receiving periodic EGD every two years, had never had a bezoar before, and had no indigestion or other gastrointestinal symptoms. Her baseline blood tests showed no abnormal findings, and her thyroid function tests and HbA1c were also normal.
In an elective EGD performed in an outpatient setting, bezoars were observed filling the stomach body, and 1 L of cola was sprayed during the EGD. Additionally, lithotomy was performed using a snare, and the bezoar was broken into four pieces (Fig. 3). The bezoar was too hard to cut into tiny pieces, and there was a technical problem that kept slipping when trying to catch it with a snare. Also, the procedure was stopped because it was thought that if it could not cut very small, it could cause more small intestine obstruction. The patient was instructed to drink cola every day, and a follow-up EGD was performed 10 days later. As four bezoars were observed, an additional 600 mL of cola was sprayed, and the patient was advised to drink cola daily. An EGD performed two weeks later showed only three bezoars, which did not dissolve or decrease in size with cola, so surgery was requested.
Five days before surgery, the patient visited the emergency room with severe abdominal pain and vomiting. At admission to the emergency room, the vital signs were blood pressure 165/101 mmHg, heart rate 85 times per minute, respiratory rate 20 times, and body temperature 36.6°C. General blood tests showed hemoglobin 14.7 g/dL, white blood cells 14780/uL, and platelets 236000/uL, and serum biochemical tests showed aspartate aminotransferase 72 IU/L, alanine aminotransferase 63 IU/L, and C-reactive protein 7.93 mg/L. All serum electrolytes were in the normal range, and there were no specific findings on electrocardiogram and chest X-ray. However, abdominal X-rays showed mechanical intestinal obstruction (Fig. 4A).
In the abdominal CT performed, small bowel dilatation with tapered transition in the pelvic ileal loop was observed, and there was bezoar-like mottled density near the transition point (Fig. 4B). She was immediately hospitalized for surgery. After hospitalization, abdominal pain disappeared as the patient was gassed out, and the ileus improved on the abdominal X-ray (Fig. 4C). It is presumed that one of the pieces that had previously been broken during the EGD escaped to the small bowel, causing mechanical ileus and that the intestinal obstruction was resolved by naturally escaping to the colon after hospitalization.
She electively underwent laparoscopic gastrostomy and removal of bezoar on the scheduled surgery day. Almost the entire small bowel was observed during surgery, but bezoars in the small bowel were not identified. Three bezoars measuring 15×10 cm, 10×10 cm, and 10×7 cm were observed in the stomach and removed. Hypaque swallowing test was performed on the second day after surgery, and no gastric filling defect was observed. She was discharged 10 days after surgery without any complications.
Case 3
A 6-year-old girl was admitted to the emergency room of our hospital with intermittent vomiting that had started one month ago. Symptoms such as abdominal pain and loss of appetite were accompanied. When she arrived at the hospital, her patient’s vital signs were normal, and physical examination showed no unusual findings. Approximately 10 months prior to her admission, the patient had undergone laparoscopic surgery for trichobezoar. She had visited the hospital due to abdominal pain, and intussusception of the jejunal loop was confirmed in the abdominal CT performed during her last hospitalization, so laparoscopic exploration and reduction were performed (Fig. 5). The patient was diagnosed with pica at the Department of Psychiatry during her previous admission. Still, she has not been followed up since. She had no other past history or psychiatric family history.
In general blood tests, hemoglobin was 14.0 g/dL, white blood cells were 13520/uL, and platelets were 777000/uL. Serum biochemistry tests showed that albumin 3.8 g/dL, total bilirubin 0.2 mg/dL, aspartate aminotransferase 29 IU/L, alanine aminotransferase 15 IU/L, blood urea nitrogen 11.6 mg/dL, blood creatinine was 0.4 mg/dL, and C-reactive protein 4.43 mg/L. Serum electrolytes were all within the normal range, and the electrocardiogram and chest X-ray showed no unusual findings. A simple abdominal X-ray examination showed non-specific gas distension and an enlarged stomach (Fig. 6A). Mechanical ileus was not observed.
A large trichobezoar was observed during EGD to identify the cause of vomiting. Bezoar was attempted to be removed using a snare, but due to its large diameter and hardness, it could not pass through the lower esophageal sphincter. The trichobezoar mass was observed to extend past the pylorus to the duodenum when it was pulled (Fig. 6B). It was explained to the patient’s parents that it was necessary to surgically remove the mass, which was beyond the scope of endoscopic removal. However, they had concerns and strong fears about reoperation because the patient had undergone abdominal surgery 10 months ago and wanted endoscopic removal if possible. After explaining that there was a high possibility of failure, and that emergency surgery may be necessary, endoscopic removal of the bezoar was attempted again. The procedure was performed in the endoscopy room in the presence of a pediatrician, and the drugs used to induce sedation were 90 mg of ketamine and 2.5 mg of midazolam.
A plan was made to retrieve the bezoar by segmenting it with a snare, but since it was made of hair, it became tangled with the snare. Because it may be difficult to retrieve the snare, the upper part of the wire was cut, and only the endoscope was retrieved with the wire hanging over the esophagus. Afterward, an attempt was made to segment the bezoar by electrocoagulation, but it was not easily cut because it was hair. However, while applying heat, the wire of the snare tangled in the bezoar was able to be separated, and it was carefully collected to minimize damage to the mucous membrane. The total time required was 1 hour and 20 minutes.
It was judged that additional endoscopic removal attempts were meaningless, so after explaining to her parents that endoscopic removal of the bezoar had failed, the patient was referred to a surgeon for surgery. The patient underwent laparoscopic exploration for trichobezoar the next day. A 4-cm incision was made in the anterior wall of the gastric body, and gastrostomy and bezoar removal were performed. The total time for surgery was 1 hour and 15 minutes. Hypaque swallowing test performed on the second day after surgery showed no contrast material leakage from the stomach, and a 30-minute delayed image confirmed that the contrast media was injected into the proximal jejunum. Abdominal pain resolved after the surgery, and she was discharged six days after the surgery without any complications.
DISCUSSION
This was a case study of three patients in whom bezoars were identified. In the first case, the bezoar was completely removed just by taking a large amount of cola. However, in the second and third cases, all endoscopic removal attempts failed, and the bezoars were removed through laparoscopic gastrostomy. In particular, the third case was a patient with pica who underwent surgery twice due to repeated occurrences of trichobezoars. In the case of trichobezoars due to pica, subsequent treatment for the underlying psychiatric disease is essential, and it has been experienced that endoscopic removal of bezoar has limited effectiveness.
Phytobezoars, the most common bezoar, are mainly composed of fiber from fruits and vegetables. In particular, persimmon tannin is reported to be the main cause. The cause of the first and second cases was unclear, but there was a possibility of phytobezoars or pharmacobezoars. In the first case, the gastric bezoar was utterly removed just by consuming cola. Of course, drinking 2 L of cola every day for two weeks would have been challenging.
Trichobezoars are commonly reported in girls with pica or mental retardation. Bezoars are formed when hair clumps up in the stomach of a patient who has pulled out hair and swallowed it. Over time, this lump extends to the small bowel, called Rapunzel syndrome. Table 1 shows the cases of trichobezoars reported so far in Korea [3-9]. Among the reported cases of trichobezoars, there were many cases in which repeated surgeries were performed, such as the patient we experienced. It has been reported that the same problem may occur repeatedly if a fundamental solution to the child’s stress or psychiatric illness is not supported [10].
Symptoms caused by gastric bezoars are often found as abdominal pain and vomiting caused by gastrointestinal obstruction. However, cases of patients presenting to the hospital after perforation or peritonitis, or even leading to death, have been reported [11]. Diagnosis is confirmed by abdominal CT or EGD.
There is no established treatment protocol for bezoars, but there are mainly enzymatic dissolution methods, endoscopic removal, and laparoscopic surgery [12]. Enzymatic dissolution methods include taking digestive enzymes such as cellulose, papain, and acetylcysteine or cola. Cola activates proteolytic enzymes in the stomach, and the carbon dioxide contained in cola is thought to penetrate the bezoar and digest the fiber of the bezoar. Additionally, the sodium bicarbonate in cola has a mucolytic effect and can promote the dissolution of bezoars. However, it is only effective against phytobezoar. Endoscopic treatment involves physically crushing and retrieving the pieces using endoscope accessories. Forceps, polypectomy snare, electrohydraulic lithotriptor, laser, and argon plasma are used. However, since there is no standard for judging the effectiveness of endoscopic treatment, the determination of the effectiveness of endoscopic treatment may vary depending on the endoscopist.
In the case of trichobezoar, there may be limitations to endoscopic treatment. In particular, in the case of Rapunzel syndrome, where the bezoar extends long into the small bowel, endoscopic treatment is more difficult. However, recently, a case of successful removal of small bowel bezoars through single-balloon enteroscopy has been reported [8]. There is a domestic case in which a trichobezoar was first cut using an electrosurgical knife during a single-balloon enteroscopy under general anesthesia. Then, the pieces were successfully retrieved. There has also been a successful case overseas of removing a trichobezoar connected from the stomach to the duodenum through electrocautery [13]. However, if complications such as small bowel obstruction or perforation occur, laparoscopic surgery is recommended from the beginning.
We have experienced several patients with bezoars. In one case, bezoars disappeared with only cola ingestion, and in another case, laparoscopic gastrostomy was performed because both cola ingestion as endoscopic lithotomy failed. Recently, with the development of endoscopic techniques, many cases of successful retrieval of bezoars that have extended into the small bowel have been reported. However, if these bezoars recur, it is important to find and correct the root cause.
Notes
Availability of Data and Material
Data sharing not applicable to this article as no datasets were generated or analyzed during the study.
Conflicts of Interest
Younghee Choe, a contributing editor of the Korean Journal of Helicobacter and Upper Gastrointestinal Research, was not involved in the editorial evaluation or decision to publish this article. All remaining authors have declared no conflicts of interest.
Funding Statement
None
Authors’ Contribution
Conceptualization: Younghee Choe, Byung-Wook Kim. Project administration: Younghee Choe. Resources: Joon Sung Kim, Byung-Wook Kim. Supervision: Younghee Choe. Writing—original draft: Younghee Choe. Writing—review & editing: all authors. Approval of final manuscript: all authors.
Ethics Statement
This study was a retrospective study based on patients’ electronic medical records, and the research subjects’ personal information was not exposed during the entire research process. It was approved for exemption from obtaining informed consent and exempt review from the Institutional Review Board of Incheon St. Mary’s Hospital, Catholic University of Korea (OC24ZISI0067).
Acknowledgements
None