Gastric bezoars: reassessment of clinical and radiographic findings in 19 patients
Abstract
The aim of this study was to reassess the clinical and radiographic findings in a series of patients with gastric bezoars. Radiology files revealed 19 patients with bezoars; 10 patients had CT and 10 had endoscopy before or after the barium studies. 11 patients (58%) had risk factors for gastroparesis and 6 (32%) had had previous gastric surgery, including 3 having had a gastric bypass or vertical banded gastroplasty. 18 patients (95%) had symptoms; in 10 of those patients, symptoms were present for 1 week or less (53%). On barium studies, the bezoars were round or ovoid in 17 patients (89%) and irregular in 2 (11%); mottled in 10 (53%) and homogeneous in 9 (47%); and mobile in 15 (79%) and immobile in 4 (21%). Gastroparesis was observed at fluoroscopy in 8 (62%) out of 13 patients without gastric surgery. Symptoms improved/resolved in 12 (67%) out of 18 patients. Follow-up CT or endoscopy showed resolution of the bezoars in 8 (80%) out of 10 patients; the mean interval to resolution was 12 days. Our experience suggests that gastroparesis is the single most common cause of bezoars, accounting for the majority of cases. Partial gastric resection or bariatric surgery should also be recognized as a cause of bezoar formation. These lesions may be manifested on barium studies by a spectrum of findings, appearing as mottled or homogeneous, mobile or immobile masses, sometimes filling the gastric pouch after bariatric surgery. Affected individuals often have an acute clinical presentation with symptoms for 1 week or less, and some bezoars resolve rapidly on conservative medical treatment.
Gastric bezoars, defined as conglomerate masses of food or foreign matter in the stomach, have a reported incidence of less than 1% in the general population [1]. Affected individuals may be asymptomatic or may present with a variety of symptoms, including epigastric pain, bloating, nausea, vomiting, early satiety, weight loss and upper gastrointestinal bleeding [2–4]. Symptomatic patients are sometimes treated by endoscopic dissolution or suction of the bezoar [5].
Gastric bezoar formation has frequently been attributed to trichophagia (hair swallowing) or ingestion of persimmons [4, 6–8]. Matted balls of ingested hair in the stomach, also known as trichobezoars, have been described in young women with trichophagia [4, 6–8], whereas conglomerate masses of fruit or vegetable matter in the stomach, also known as phytobezoars, have most often been described in patients who ingest unripe persimmons containing a soluble tannin that polymerises on contact with gastric acid, forms a coagulum and affixes to other gastric contents [4, 6–8].
In our experience, however, most gastric bezoars are unrelated to ingestion of hair or unripe persimmons but instead are caused by gastroparesis with decreased or absent gastric peristalsis or by surgical resection/bypass of the gastric antrum and body, resulting in poor mechanical breakdown of ingested solids that form conglomerate masses in the stomach. The rise in bariatric surgery over the past two decades, including gastric bypass and laparoscopic adjustable gastric banding, has resulted in the surgical construction of gastric pouches that are particularly susceptible to the development of bezoars secondary to impaired breakdown of ingested solids in the pouch [9–12]. Yet, surprisingly, the role of gastroparesis and, more recently, bariatric surgery in the formation of gastric bezoars has received little attention in the radiological literature. The purpose of our investigation therefore was to reassess the clinical and radiographic findings in a series of patients with gastric bezoars, particularly as these findings relate to the pathophysiology of bezoars.
Methods and materials
Patient population
A computerised search of the radiology database at our hospital revealed 33 patients with a radiographic diagnosis of gastric bezoars on upper gastrointestinal tract barium studies during a 9.5-year period from January 1998 to July 2007. Seven patients were excluded from analysis because no radiographic images were available for review (n = 2), no gastric bezoars were identified on retrospective review of the images (n = 3) or because no medical records were available for review (n = 2). 19 (73%) of the remaining 26 patients were found to have gastric bezoars based on one of the three following criteria: (1) confirmation of the bezoar on CT (n = 5) or endoscopy (n = 5); (2) treatment of the bezoar with symptomatic improvement (n = 4); and (3) predisposing factors for the development of a bezoar with associated symptoms but no treatment or follow-up studies (n = 5). These 19 patients constituted our study group.
13 patients (68%) were women and 6 (32%) were men. The mean age was 48.1 years (range, 19–81 years). Five patients (26%) underwent abdominal CT scans before the barium study; the mean interval between CT and the barium study was 3.4 days (range, 1–6 days). Five patients (26%) underwent endoscopy before the barium study; the mean interval between endoscopy and the barium study was 18.4 days (range, 1–52 days).
Five patients (26%) underwent post-treatment CT scans after the barium study; the mean interval between the barium study and the follow-up CT was 15 days (range 1–30 days). Finally, five patients (26%) underwent follow-up endoscopy after the barium study (three were treated for their gastric bezoars and two were not); the mean interval between the barium study and follow-up endoscopy was 5.6 days (range, 1–21 days).
Examination technique
Barium studies
17 patients (89%) had single-contrast upper gastrointestinal tract examinations that included upright and recumbent single-contrast views of the oesophagus, stomach and duodenum using a 50% w/v barium suspension (Entrobar; Lafayette Pharmaceuticals, Lafayette, IN). The remaining two patients (11%) had double-contrast barium studies that included upright and recumbent double- and single-contrast views using an effervescent agent (Baros; Lafayette Pharmaceuticals) and a 250% w/v barium (E-Z-HD; E-Z-EM, Lake Success, NY) followed by a 50% w/v barium (Entrobar). All of the studies were performed using digital fluoroscopic equipment (Diagnost 76; Phillips, Eindhoven, the Netherlands; Sireskope SD; Siemens Pharmaceuticals, Munich, Germany). The studies were performed by residents, fellows or one of three attending gastrointestinal radiologists, and all were interpreted by the attending radiologists.
CT
8 patients (42%) had 10 helical CT scans of the abdomen, according to an established protocol. One CT scan was performed only with iv contrast material (100 ml of 60% iodinated contrast material (diatrizoate meglumine (Hypaque) or iohexol (Omnipaque 300); Nycomed, Zurich, Switzerland)); two were performed only with oral contrast material (500 ml of 2–3% oral contrast material (Gastrografin; Bristol-Myers Squibb, New York City, NY)) before the study; five were performed with both intravenous and oral contrast material; and two were performed without any contrast material. CT images were obtained with the patient in a supine position during inspiration. Axial images were obtained at 5-mm or 7-mm slice collimations (pitch 1.3:1; 200–220 mAs) and reconstructed with a soft-tissue algorithm.
Review of images and medical records
The images from the 19 barium studies were reviewed at a workstation by a consensus of two authors (both experienced gastrointestinal radiologists) who were blinded to the clinical and endoscopic findings. The images were analysed to determine the morphological features of these gastric bezoars, including their shape (round, ovoid or irregular), internal characteristics (homogeneous or mottled secondary to trapping of barium in the interstices of the bezoar), mobility (mobile or immobile) and density (sinking or floating in the barium pool). The presence or absence of gastric narrowing, gastric outlet obstruction or gastric surgery (resection or bariatric surgery) was also noted. The original radiological reports for the 19 studies were also reviewed by one author to determine whether there was any mention of decreased or absent gastric peristalsis (gastroparesis) and/or delayed gastric emptying of barium at fluoroscopy in the reports. Decreased gastric peristalsis was defined as diminished amplitude and/or velocity of peristalsis in the antrum and body of the stomach at fluoroscopy.
The images from the five pre-treatment CT scans were also reviewed to assess for the presence or absence of gastric bezoars. When present, the bezoars were evaluated for homogeneity (homogeneous or inhomogeneous) and attenuation characteristics (gas, soft tissue and/or contrast material).
Medical, surgical and radiographic records were also reviewed to determine the medical and surgical history (including predisposing factors for bezoars) as well as the clinical presentation, treatment and course.
IRB approval
Our institutional review board approved all aspects of this retrospective study and did not require informed consent from patients whose records were included. The study was also compliant with the Health Insurance Portability and Accountability Act.
Results
Clinical findings
Predisposing factors
11 patients (58%) had known risk factors for gastroparesis, including narcotic medications in 5, vagotomy in 2 (1 pyloroplasty and vagotomy and 1 oesophagogastrectomy, gastric pull-through and vagotomy), hypothyroidism in 1, diabetes in 1, bilateral lung transplants for cystic fibrosis in 1 and idiopathic gastroparesis in 1. Six patients (32%) had surgical resection/bypass of the gastric antrum and body, including a partial gastrectomy (Billroth II) with Roux-en-Y reconstruction in two, a Whipple's procedure with hemigastrectomy in one, a Roux-en-Y gastric bypass in two and a vertical banded gastroplasty in one. In the eight patients with gastric surgery (including the two with vagotomies), the mean interval between surgery and the barium study showing a gastric bezoar was 13.7 years (range, 14 months to 27 years). Thus, 17 (90%) of the 19 patients had risk factors (gastroparesis or surgical resection or bypass of the gastric antrum or body) for the development of gastric bezoars. The remaining two patients (10%) had no known risk factors for developing bezoars.
Presenting symptoms
18 (95%) of the 19 patients with gastric bezoars had one or more presenting symptoms, including nausea in 13 patients (68%), vomiting in 9 (47%), abdominal pain in 8 (42%), dysphagia in 5 (26%), bloating in 3 (16%), weight loss in 2 (11%), anorexia in 1 (5%) and melaena in 1 (5%). The remaining patient was asymptomatic. Excluding two patients, one with symptoms for 14 years and the other with symptoms for 4 years, the mean duration of symptoms prior to the barium study showing the gastric bezoar was 24 days (range, 1 day to 7 months). 10 patients (53%) had symptoms for 1 week or less.
All five patients with dysphagia had mobile bezoars that intermittently could have obstructed the gastric cardia, accounting for their symptoms. The patient with melaena also had erosive gastritis and a pyloric channel ulcer.
Imaging findings
Barium studies
The bezoars appeared on barium studies as round or ovoid masses in the stomach in 17 patients (89%) and as irregular masses in 2 (11%), including 1 patient with a bilobed bezoar and 1 with a bezoar that filled virtually the entire stomach, conforming to the gastric wall (Figure 1). The masses had a mottled appearance secondary to barium entering the interstices of the bezoar in 10 patients (53%) (Figure 1a) and a relatively homogeneous appearance in the remaining 9 (47%) (Figures 2–5). The bezoars were freely mobile within the stomach in 15 patients (79%) (including 3 with partial gastrectomies and 1 with an oesophagogastrectomy and gastric pull-through) (Figures 2 and 3) and immobile in 4 patients (21%) (including 1 with a giant bezoar filling virtually the entire stomach (Figure 1a) and 3 who had undergone bariatric surgery — 2 gastric bypasses and 1 vertical banded gastroplasty — in whom the bezoars filled the pouch and were fixed in position by the pouch) (Figures 4 and 5). In the 15 patients with mobile bezoars, 10 (67%) floated in the barium pool (Figures 2 and 3), 4 (27%) sank to the bottom of the pool and 1 (6%) alternately floated and sank in the barium pool.
Decreased or absent gastric peristalsis was observed at fluoroscopy in 8 (62%) of the 13 patients who had not undergone previous surgical resection/bypass of the gastric antrum and body (Figures 1–3), and all 8 of these patients had known risk factors for the development of gastroparesis. Gastric dilatation was found in 12 patients (63%), including 3 with bezoars in dilated gastric pouches after bariatric surgery.
None of the 19 patients with gastric bezoars had findings of mechanical gastric outlet obstruction to the flow of barium, and none of the 6 patients who underwent surgical resection/bypass of the gastric antrum and body had strictures at the surgical anastomoses.
Abdominal CT
In all five patients with pre-treatment CT scans, the gastric bezoars appeared as inhomogeneous, round or ovoid masses containing areas of soft tissue density intermixed with gas and oral contrast material in four patients (80%) (Figure 2b) and intermixed with gas but no oral contrast material in one (20%) (Figure 1b).
Endoscopy
All five patients who had endoscopy before the barium studies were found to have gastric bezoars. One patient also had erosive gastritis and a pyloric channel ulcer.
Clinical follow-up
11 patients (58%) underwent one or more forms of treatment, including dietary restrictions in 8 (42%) (small, frequent meals in 1, a low-residue diet in 1, a liquid diet in 5 and withholding of tube feeds in 1); drug therapy in 3 (16%) (administration of metoclopramide (Reglan; Robins Pharmaceuticals, Richmond, VA) in 2 and withholding of narcotics in 1); endoscopic intervention in 2 (mechanical dissolution of the bezoar in 2 and mechanical suction of the bezoar in the other); and surgical intervention in 1 (5%) (a subtotal gastrectomy in 1 of the 2 patients who previously underwent a partial gastrectomy). Seven patients (37%) had no specific treatment for their gastric bezoars. The treatment was unknown for the remaining patient (5%).
Symptoms improved (n = 4) or resolved (n = 5) in 9 (82%) of the 11 patients who underwent medical, endoscopic or surgical treatment for their gastric bezoars (including all three who underwent endoscopic or surgical treatment). The remaining two patients (18%) who were treated had persistent symptoms without clinical improvement. Conversely, symptoms persisted (n = 3) or worsened (n = 1) in four (57%) of the seven patients who were not treated. The remaining three patients (43%) who were not treated had clinical improvement in their symptoms. Thus, symptoms improved or resolved in 12 (67%) of the 18 patients with gastric bezoars who had clinical follow-up.
Imaging and endoscopic follow-up
10 patients (53%) with gastric bezoars had follow-up imaging studies after the original barium study, including abdominal CT in 5 and endoscopy in 5. Follow-up CT in three patients with improvement (n = 1) or resolution (n = 2) of symptoms after treatment revealed no evidence of a residual gastric bezoar. Similarly, follow-up endoscopy in three patients with improvement (n = 1) or resolution (n = 2) of symptoms after treatment revealed no evidence of a gastric bezoar. Conversely, follow-up CT in two patients with recurrent symptoms after endoscopic treatment revealed recurrent gastric bezoars. Finally, follow-up endoscopy in two patients with improved symptoms but no treatment revealed spontaneous resolution of the bezoars.
Thus, follow-up CT or endoscopy showed resolution of the gastric bezoars in 8 (80%) of the 10 patients in whom such studies were performed. The mean interval between the initial barium study showing a bezoar and the follow-up study showing resolution of the bezoar in these 8 patients was 12 days (range, 1–35 days). When the eight patients were stratified on the basis of treatment, two (25%) had mechanical dissolution or suction of the bezoar, four (50%) had medical treatment and two (25%) had no form of treatment.
Discussion
Our experience suggests that gastroparesis is the single most common cause of gastric bezoars, so it is important for radiologists to be aware of this association. In our study, 11 (58%) of 19 patients with bezoars had known risk factors for gastroparesis (58%), and markedly decreased or absent gastric peristalsis was observed at fluoroscopy in 8 (62%) of 13 patients who had not undergone previous gastric surgery. The medical and surgical literature also contains multiple reports of gastric bezoars in patients with gastroparesis secondary to vagotomy, diabetes, hypothyroidism, cystic fibrosis, lung transplants and other predisposing conditions [1, 3, 13–16]. Except for one review, however, in which 6% of patients with gastroparesis had gastric bezoars on barium studies [17], the association between gastroparesis and bezoars has not been addressed in the radiological literature. Whatever the aetiology of the gastroparesis, decreased or absent gastric peristalsis results in poor mechanical breakdown of ingested solids that serve as the nidus for the formation of conglomerate masses in the stomach.
Three of our patients (16%) with gastric bezoars underwent partial gastrectomy, including a Billroth II in two and a Whipple's procedure in one. It is well recognised that gastric bezoars may develop as a complication of partial gastrectomy for gastric ulcers or other abnormalities [2, 18–20]. These bezoars are thought to result from gastric resection and vagotomy with loss of gastric motility and the normal mixing function of the distal part of the stomach [19, 20]. In our study, none of the patients with gastric bezoars after partial gastrectomy had anastomotic strictures, so their bezoars presumably developed because of surgical absence of the gastric antrum and body, the part of the stomach normally responsible for breaking down ingested solids by active peristalsis. Thus, even in the absence of anastomotic strictures, these patients had the functional equivalent of severe gastroparesis as the cause of their bezoars.
Three patients (16%) with gastric bezoars also underwent bariatric surgery, including a Roux-en-Y gastric bypass in two and a vertical banded gastroplasty in one. Although none of the patients had strictures or obstruction of the gastric pouch, all three developed bezoars within the pouch, presumably because of surgical bypass of the gastric antrum and body, the part of the stomach responsible for breaking down ingested solids; therefore these patients had the functional equivalent of severe gastroparesis. Other studies from the surgical literature have anecdotally reported gastric bezoars in patients with vertical banded gastroplasty, Roux-en-Y gastric bypass and laparoscopic adjustable gastric banding [9–12], but this complication of bariatric surgery has not been described previously in the radiological literature. With the rising popularity of bariatric surgery for treatment of morbid obesity, gastric bezoars may be encountered with greater frequency as a complication of this surgery in the future.
As in our study, patients with gastric bezoars typically present with nausea, vomiting, epigastric pain, early satiety, bloating, weight loss and, occasionally, upper gastrointestinal bleeding [2–4]. Dysphagia was another relatively frequent symptom in our patients with bezoars, occurring in 5 (26%) of 19 cases. None of these patients had other abnormalities on barium studies to account for their dysphagia, and we postulate that mobile bezoars could intermittently obstruct the gastric cardia, causing dysphagia in these individuals. Other investigators have also reported distal oesophageal obstruction and associated dysphagia as a complication of gastric bezoars [21].
Gastric bezoars have classically been described on barium studies as mobile masses in the stomach that float in the barium pool and have a mottled appearance due to trapping of barium in the interstices of the lesion [4, 7, 18]. In our study, however, 9 (47%) of 19 patients with gastric bezoars had relatively homogeneous masses (Figures 2–5) and 4 (21%) had immobile masses at fluoroscopy, including 1 patient with a giant bezoar filling virtually the entire stomach (Figure 1a) and 3 patients with bariatric surgery in whom the bezoars filled the gastric pouch and were fixed in position by the pouch (Figures 4 and 5). Even when the bezoars were freely mobile, four (21%) of these lesions sank to the bottom of the barium pool because of their high density. Gastric bezoars may therefore be manifested by a spectrum of findings on barium studies, depending on the cohesiveness and density of these lesions. It is also important to recognise the value of a dynamic fluoroscopic examination for assessing the mobility of a gastric bezoar and its relationship to the gastric remnant/pouch in patients who have undergone previous gastric surgery.
It seems reasonable that gastric bezoars might develop slowly over a long period of time. In our study, however, the majority of patients (10 out of 19, or 53%) had symptoms for 1 week or less. Furthermore, the mean interval between the initial barium study showing a bezoar and a follow-up study (CT or endoscopy) showing resolution of the bezoar in 8 patients was only 12 days, and 6 (75%) of these 8 were treated conservatively without endoscopic or surgical intervention. In other studies, gastric bezoars have also been reported to form in less than 1 month and to resolve in less than 7 days on conservative treatment [3, 21, 22]. The data therefore suggest that some patients with gastric bezoars have acute clinical symptoms of relatively short duration and that these lesions can form and heal rapidly on conservative treatment.
In our study, symptoms improved or resolved in six (75%) out of eight patients who were treated by dietary restrictions, drug therapy and withholding of narcotics, in all three (100%) who were treated by endoscopic or surgical intervention and in three (43%) out of seven who were not treated. Our experience suggests that endoscopic or surgical intervention is not required for most gastric bezoars and that these lesions can resolve spontaneously or on conservative treatment with dietary restrictions or drug therapy in many patients.
Our investigation has the inherent limitations of a retrospective study, including selection bias. Our review of the data also was limited by incomplete medical records, which did not specify patient diet in many cases. As a result, we cannot completely exclude trichophagia or persimmon ingestion in some patients, though there was no evidence to suggest this possibility in the medical records. Nor did the records contain other potentially useful information about the eating habits of these patients, which could be important if inadequate mastication or rapid deglutition contributed to the formation of gastric bezoars in some cases. Finally, not all patients had endoscopic confirmation of bezoars, so some of these individuals could have had false-positive barium studies.
Conclusions
In summary, our experience suggests that gastroparesis is the single most common cause of gastric bezoars, accounting for the majority of cases, so it is important for radiologists to be aware of this association. Although it has been well documented that partial gastrectomy can lead to bezoar formation, bariatric surgery should be recognised as another cause of gastric bezoars that may be encountered with increasing frequency with the greater number of bariatric procedures (especially Roux-en-Y gastric bypass and laparoscopic adjustable gastric banding) being performed. Bezoars may be manifested by a spectrum of findings on barium studies, appearing as mottled or homogeneous masses that float or sink in the barium pool or, less frequently, as immobile masses that fill virtually the entire stomach or gastric pouch after bariatric surgery. Affected individuals often have an acute clinical presentation with symptoms for 1 week or less, and some bezoars can resolve rapidly on medical treatment without need for endoscopic intervention.
A 21-year-old woman with a giant gastric bezoar. (a) A frontal spot image from a single-contrast upper gastrointestinal tract examination shows the bezoar as a mottled mass (large arrows) that conforms to the gastric wall, filling virtually the entire stomach. Note the presence of a small hiatal hernia (small arrows). Marked gastroparesis was observed at fluoroscopy. (b) An unenhanced axial CT image shows an inhomogeneous mass (arrows) in the stomach containing areas of soft tissue density intermixed with multiple tiny bubbles of gas. This patient presented with nausea and vomiting that improved on dietary restriction. Follow-up CT 30 days after the original barium study showed resolution of the bezoar. A 59-year-old woman with a gastric bezoar. (a) A steep left posterior oblique spot image from a double-contrast upper gastrointestinal tract examination shows an ovoid, relatively homogeneous mass (arrows) in the stomach, floating in the barium pool. Gastroparesis was observed at fluoroscopy. The bezoar was also detected at endoscopy 2 days before the barium study. (b) An intravenous enhanced axial CT image shows a heterogeneous mass (arrows) in the stomach containing soft-tissue of varying density, intermixed with oral contrast material and tiny bubbles of gas. This patient presented with nausea, vomiting and abdominal pain that improved after endoscopic suction of the bezoar and treatment with a liquid diet and metoclopramide. A 60-year-old woman with a gastric bezoar after an oesophagogastrectomy, gastric pull-through and vagotomy. A left posterior oblique spot image from a single-contrast upper gastrointestinal tract examination shows an ovoid, homogeneous mass (black arrows) floating in the barium pool within the intrathoracic stomach, with the proximal tip of the bezoar (white arrow) projecting above the pool. Gastroparesis was observed at fluoroscopy. This patient presented with intermittent nausea, abdominal bloating and dysphagia that improved without treatment. Follow-up endoscopy 21 days after the barium study showed resolution of the bezoar. A 56-year-old woman with a bezoar in a gastric pouch after a vertical banded gastroplasty. A frontal spot image from a single-contrast upper gastrointestinal tract examination shows a vertical banded gastroplasty with a relatively homogeneous mass (white arrows) filling almost the entire pouch (large black arrows) above the banded segment (small black arrow), which was incompletely filled on this view. Also note barium in the remainder of the stomach. This patient presented with nausea and vomiting that improved on a diet restricted to clear liquids. Follow-up endoscopy 1 day later showed resolution of the bezoar. A 60-year-old woman with a bezoar in a gastric pouch after a gastric bypass with Roux-en-Y reconstruction. (a) A frontal spot image from a single-contrast upper gastrointestinal tract examination shows the bezoar as an ovoid, homogeneous mass (white arrows) filling the gastric pouch. Also note barium in the jejunal loops (large black arrows) distal to the gastrojejunal anastomosis (not well visualised). There is also breakdown of the gastric staple line with barium entering the body and antrum of the stomach (small black arrows). (b) A right posterior oblique spot image shows the bezoar (large black arrows) in the gastric pouch, with barium entering the proximal jejunum via a patent gastrojejunal anastomosis (small black arrows). This view also delineates the site of breakdown of the staple line (large white arrow), with barium entering the gastric body and antrum (small white arrows). This patient's treatment and clinical course were unknown.
References
1 Ahn Y-H, Maturu P, Steinheber FU, Goldman JM. Association of diabetes mellitus with gastric bezoar formation. Arch Intern Med 1987;147:527–8.
2 Diettrich NA, Gau FC. Postgastrectomy phytobezoars—endoscopic diagnosis and treatment. Arch Surg 1985;120:432–5.
3 Hayes PG, Rotstein OD. Gastrointestinal phytobezoars: presentation and management. Can J Surg 1986;29:419–20.
4 Eisenberg RL, Levine MS. Miscellaneous abnormalities of the stomach and duodenum. In: Gore RM, Levine MS, editors. Gastrointestinal radiology, 3rd edn. Philadelphia, USA: Saunders, 2008: 679–706.
5 Blam ME, Lichtenstein GR. A new endoscopic technique for the removal of gastric phytobezoars. Gastrointest Endosc 2000;52:404–8.
6 DeBakey M, Ochsner A. Bezoars and concretions. Surgery 1938;4:934–63.
7 Eisenberg RL. Gastrointestinal radiology: a pattern approach, 4th edn. Philadelphia, USA: Lippincott Williams and Wilkins, 2003: 262–3, 274–5.
8 Halpert RD. Gastrointestinal imaging: the requisites, 3rd edn. Philadelphia, USA: Mosby Elsevier, 2006: 52–3.
9 Deitel M, Bendago M. Endoscopy of vertical banded gastroplasty. Am Surg 1989;55:287–90.
10 White NB, Gibbs KE, Goodwin A, Teixeira J. Gastric bezoar complicating laparoscopic adjustable gastric banding, and review of literature. Obes Surg 2003;13:948–50.
11 Veronelli A, Ranieri R, Laneri M, Montorsi M, Bianchi P, Cosentino F, et al. Gastric bezoars after adjustable gastric banding. Obes Surg 2004;14:796–7.
12 Pinto D, Carrodeguas L, Soto F, Lascano C, Cho M, Szomstein S, et al. Gastric bezoar after laparoscopic Roux-en-Y gastric bypass. Obes Surg 2006;16:365–8.
13 Kaplan LR. Hypothyroidism presenting as gastric phytobezoar. Am J Gastroenterol 1980;74:168–9.
14 Dellon ES, Morgan DR, Mohanty SP, Davis K, Aris RM. High incidence of gastric bezoars in cystic fibrosis patients after lung transplantation. Transplantation 2006;81:1141–6.
15 Bodet-Milin C, Querellou S, Oudoux A, Haloun A, Horeau-Llanglard D, Carlier T, et al. Delayed gastric emptying scintigraphy in cystic fibrosis patients before and after lung transplantation. J Heart Lung Transplant 2006;25:1077–83.
16 Folch E, Shakoor H, Gomez J, Hogan K, Mason D, Murphy S, et al. Gastric bezoar after lung transplantation in non-cystic fibrosis patients and review of the literature. J Heart Lung Transplant 2007;26:739–41.
17 Levin AA, Levine MS, Rubesin SE, Laufer I. An 8-year review of barium studies in the diagnosis of gastroparesis. Clin Radiol 2008;63:407–14.
18 Sparberg M, Nielsen A, Andruczak R. Bezoar following gastrectomy. Am J Dig Dis 1968;13:579–83.
19 Szemes GC, Amberg JR. Gastric bezoars after partial gastrectomy. Radiology 1968;90:765–8.
20 Goldstein HM, Cohen LE, Hagen RO, Wells RF. Gastric bezoars: a frequent complication in the postoperative ulcer patient. Radiology 1973;107:341–4.
21 Reissman P, Fich A, Eid A, Rivkind A. Esophageal phytobezoar causing acute dysphagia: a rare complication of gastric bezoar. J Clin Gastroenterol 1994;18:159–60.
22 Ripollés T, García-Aguayo J, Martínez M-J, Gil P. Gastrointestinal bezoars: sonographic and CT characteristics. AJR Am J Roentgenol 2001;177:65–9.
