Bile is a digestive fluid synthesized in the liver and excreted into the duodenum after food is ingested to help with the digestion of lipids and the inhibition of bacterial overgrowth in the small intestine.1 Its main component is water, and the other ingredients include bile acid, bilirubin, cholesterol, fatty acids, and lecithin.2,3 Normally, the pyloric sphincter prevents bile from entering the stomach. When the pyloric sphincter is damaged or dysfunctional, bile can reflux into the stomach. The incidence of duodenogastric reflux and its relationship with gastric and duodenal diseases are controversial. One study showed that approximately 10% of patients who underwent endoscopic examination had duodenogastric reflux.4 Duodenal fluid contains bile and pancreatic juice. Bile is strongly alkaline, and hence, bile reflux to the stomach can alter the pH balance in the stomach and cause chemical irritation of the gastric mucosa. This, in turn, results in an increase in inflammatory cells in the gastric mucosa, decrease in parietal cells, hyperplasia of mucous cells, and changes in glandular morphology. Thus, intestinal metaplasia, various forms of gastritis, gastric ulcer, gastric cancer, and esophagitis can occur.4–6 This reflux of bile into the stomach may cause bile reflux gastritis, which commonly occurs after surgeries, including partial gastrectomy, truncal vagotomy, pyloroplasty, cholecystectomy, or sphincteroplasty.7–9 Bile reflux gastritis is a common inflammatory disease caused by the chemical irritation of the stomach by alkaline ingredients in the refluxed bile. It is generally diagnosed based on the observation of swelling, redness, erosions, and bile staining of the gastric mucosa on upper endoscopy.10
Gadoxetic acid (gadolinium ethoxybenzyl diethylenetriaminepentaacetic acid; Primovist; Bayer Pharma, Berlin, Germany) is a liver-specific magnetic resonance (MR) contrast agent that is taken up by functional hepatocytes up to a maximum of 50% of the injected dose and is subsequently excreted into the bile duct. Since it is excreted into the bile duct, T1-weighted MR cholangiography (MRC) of the hepatobiliary phase can be performed. Furthermore, T1-weighted MRC can give us additional information on various pathological conditions that cannot be adequately demonstrated by T2-weighted MRC. In clinical practice, contrast media excreted from the bile duct is often seen in the gastric lumen on gadoxetic acid–enhanced MRC (gadoxetic MRC). We hypothesized that it could represent duodenogastric bile reflux and that bile reflux gastritis may be closely related to the presence of contrast material in the stomach on the delayed phase of MRC.
Therefore, this study was conducted to evaluate the relationship between biliary excreted contrast media in the stomach on gadoxetic MRC and the presence of bile reflux gastritis.
MATERIALS AND METHODS
This retrospective study was approved by our institutional review board, and informed consent was waived. Using a computerized search of our hospital's medical records, we identified patients who had undergone gadoxetic MRC at our hospital.
Between May 2009 and April 2012, 293 patients underwent gadoxetic MRC. Among them, 182 were excluded because of the absence of upper endoscopy study findings (106 patients), absence of 60-minute delayed MRC images (43 patients), absence of the passage of biliary excreted contrast media into the duodenum (30 patients), and an interval of more than 1 year between the MRC and endoscopy study (3 patients). Finally, 111 patients who underwent both gadoxetic MRC and upper endoscopy were included (Fig. 1). Data were collected by retrospectively reviewing their medical records.
MR Imaging (MRI) Protocol
All MRIs were performed on a 3-T MRI machine (Achieva; Philips Medical Systems, Best, The Netherlands) with a standard phased-array torso coil as the receiver. All 111 patients underwent gadoxetic MRC. We acquired preenhanced and enhanced fat-saturated 3-dimensional (3D) gradient-echo T1-weighted images (Table 1). The enhanced images were acquired after the administration of gadoxetic acid (Primovist; Bayer Pharma) at a dose of 0.025 mmol/kg of body weight at a flow rate of 1 mL/s, followed by a 10-mL saline flush at the same flow rate by using an intravenous power injector (Spectris Solaris; MedRad, Indianola, PA, USA). We used the enhanced-T1 high-resolution isotropic volume examination technique to acquire the 3D gradient-echo T1-weighted images. The 3D axial- and coronal-reconstruction images 60 minutes after gadoxetic acid administration were acquired using the maximum intensity projection technique.
Two radiologists (with 8 and 12 years of experience) performed a blinded independent review of the gadoxetic-MRC image sets comprising precontrast axial T1-weighted images and contrast-enhanced axial/coronal delayed images acquired 60 minutes after the intravenous injection of contrast media. We then evaluated the images to determine the presence of contrast media in the stomach and graded the extent of contrast media presence on a 3-point scale: grade 1, antrum; grade 2, body; grade 3, fundus. Discordant findings between the 2 radiologists were resolved by a consensus review by a third board-certified abdominal radiologist (with 30 years of experience). To determine the presence of bile reflux gastritis, endoscopic images were reviewed by an experienced gastroenterologist blinded to the results of gadoxetic MRC. Bile reflux gastritis (or duodenogastric reflux gastritis) was defined according to the Sydney classification as erythematous and/or exudative gastric mucosa with bile staining on upper endoscopy.10
Statistical analyses were performed using IBM SPSS Statistics for Windows, Version 20.0 (IBM Corp, Armonk, NY, USA). The sensitivity, specificity, and accuracy of duodenogastric bile reflux diagnosis were evaluated for the gadoxetic-MRC image sets. The Fisher exact test was used to compare the frequency of bile reflux gastritis. The linear-by-linear association test was used to evaluate the association between bile reflux gastritis and the extension grade of contrast media. A 2-sided probability value of P < 0.05 was considered statistically significant.
Among the 111 patients, 49 were men and 62 were women. The mean age of the patients was 60 years (range, 17–85 years), and the mean interval between the gadoxetic MRC and endoscopy studies was 25 days (range, 0–337 days). Among the 102 patients who had benign diseases, 50 had biliary stones, 5 had hepatic abscesses, 3 had choledochal cysts, 1 had primary sclerosing cholangitis, and 43 had benign pancreatic lesions such as cystic tumors or intraductal papillary mucinous neoplasms. Among the remaining 9 patients who had underlying malignancies, 4 had peripheral cholangiocarcinomas, 1 had gallbladder cancer, 1 had pancreatic cancer, and 3 had hepatic metastasis. Eleven of the 111 patients had previously undergone gastrojejunostomy. Most of our patients underwent endoscopy for abdominal pain, dyspepsia, or postoperative follow-up.
Among the 111 patients, 39 showed contrast media in the stomach on the 60-minute delayed images. Of these 39 patients, 13 had bile reflux gastritis (Fig. 2) and 5 showed bile in the stomach without evidence of erythematous/edematous mucosal changes. Four of the 5 patients showed chronic atrophic gastritis, and 1 showed erosive gastritis. Of the 13 patients who had bile reflux gastritis, one patient had early gastric cancer.
Of the 72 patients who did not show contrast media in the stomach, none had bile reflux gastritis and 2 showed bile staining in the stomach without evidence of erythematous/exudative mucosal changes, except for chronic atrophic gastritis (Fig. 3). Of the 70 patients who did not show contrast media or bile staining in the stomach, 39 had chronic atrophic gastritis, 12 had erosive gastritis, 11 had erythematous gastritis, and 1 had hemorrhagic gastritis.
The sensitivity of MRI for the diagnosis of bile reflux was 90% (18/20), and the specificity was 76.9% (70/91). The diagnostic accuracy was 79.3% (88/111; Table 1).
The sensitivity of MRI for the diagnosis of bile reflux gastritis was 33.3% (13/39), and the specificity was 100% (72/72). The diagnostic accuracy was 76.6% (85/111).
Bile reflux gastritis was significantly more common in patients with contrast media in the stomach on gadoxetic MRC (13/39 [33.3%]) than in those without contrast media (0/72 [0%]; P < 0.001; Table 2).
The result of the linear-by-linear association revealed that patients with high-grade extension of contrast media in the stomach had significantly frequent bile reflux gastritis than did those with low-grade extension (grade 0, 0%; grade 1, 16.7%; grade 2, 36.4%; grade 3, 43.8%; P < 0.001; Table 3).
Among the 11 patients who had undergone prior gastrojejunostomy, 7 showed contrast media in the stomach on the 60-min delayed images. Of these 7 patients, 4 had bile reflux gastritis (Fig. 4). Of the 4 patients who did not show contrast media in the stomach, none had bile reflux gastritis and 2 showed erosive gastritis without bile staining.
Of the 21 patients who showed contrast media in the stomach on MRI but had no bile staining in the stomach on endoscopic examination, 14 had chronic atrophic gastritis, 1 had early gastric cancer, 2 had gastric ulcer, 2 had hemorrhagic gastritis, and 11 had erosive gastritis. Many of these patients had multiple diseases at the same time.
Bile reflux to the gastric lumen is common in patients who have undergone gastric surgeries including Billroth I and II surgeries or pyloroplasty. A previous study showed that the incidence of duodenogastric reflux on endoscopic examination was 10% (428/4256) and that it was more frequent in patients undergoing surgeries such as vagotomy and resection (98/583; 16.8%) than in patients not undergoing any surgery (330/3673 [8.98%]).4 In our study, 63.6% (7/11) of patients who had undergone prior gastrojejunostomy showed contrast media in the stomach on MRI images; and 4 of these 7 patients showed bile reflux gastritis on endoscopic examination.
Bile reflux gastritis is a type of reactive gastritis induced by the irritation of the gastric mucosa by the refluxed strongly alkaline bile. Bile reflux gastritis causes various clinical manifestations, including epigastric pain, nausea, vomiting, and chest discomfort.11 In addition, cancer occurs more frequently in the remnant stomach after Billroth II reconstruction than after Billroth I reconstruction, and the influence of reflux of duodenal juice, including bile acids, has been pointed out as a cause.12 Therefore, the revision of Billroth surgery to Roux-en-Y surgery has been suggested for the correction of bile reflux.9,13–15
We can diagnose bile reflux to the stomach by various techniques, including endoscopy, intubation methods such as nasogastric aspiration of bile marker or the measurement of bile acids in fasting gastric aspirates, 24-hour pH monitoring through the nasogastric tube, and hepatobiliary scintigraphy.16 The most widely accepted methods of diagnosing duodenogastric bile reflux are the measurement of intragastric bile acid in the gastric juice aspirated through the nasogastric tube and hepatobiliary scintigraphy. However, these methods have limited diagnostic sensitivity and specificity and cause patient discomfort.17,18 Recently, a 24-hour bilirubin concentration monitoring technique has been introduced and is being widely used.16 Magnetic resonance imaging has many advantages over other tests. Magnetic resonance imaging does not release radiation and gives us more detail and correct anatomical information and is less painful.
In our study, 35.1% (39/111) of the patients who underwent 60-minute delayed gadoxetic MRC showed biliary excreted contrast media in the stomach. Eighteen of these 39 patients had bile staining in the stomach on gastroscopic examination. Moreover, 13 of 18 patients showed combined erythematous/exudative gastritis representing bile reflux gastritis. Therefore, 33% of the patients who showed contrast media in the stomach on the delayed gadoxetic MRC really had bile reflux gastritis.
Our study shows the applicability of MRI using hepatocyte-specific contrast media in detecting bile reflux. Its theoretical principle is almost identical to that of biliary scintigraphy. Gadoxetic acid is taken up by functional hepatocytes up to a maximum of 50% of the injected dose and subsequently excreted into the bile duct. Since it is excreted into the bile duct, it shows up as high signal intensity on T1-weighted MRC.
Reflux gastritis is generally diagnosed based on the presence of hemorrhagic, erythematous, or exudative gastritis with bile staining on the gastric mucosa, as confirmed on endoscopy.10 Interestingly, in several patients, no bile staining was observed on upper endoscopy although they had reflux of contrast media in the stomach. A previous study showed that gastroscopy had lower accuracy and predictive value than did hepatobiliary scintigraphy or gastric pH monitoring in the assessment of duodenogastric bile reflux.19 We believe that 60-minute delayed MR images can also show higher accuracy and predictive values than gastroscopy, since MR images are the result of 60-minute data, whereas gastroscopy would be similar to taking a snapshot at the time of endoscopic examination. Therefore, we suggest that the sensitivity of gadoxetic acid–enhanced MRI is superior to that of gastroscopy in the diagnosis of bile reflux gastritis. This alternative diagnostic tool can be used for diagnosing bile reflux gastritis. Moreover, our findings demonstrated that gadoxetic acid–enhanced MRI could also indicate duodenogastric bile reflux.
The Sydney classification system defined bile reflux gastritis only when hemorrhagic, erythematous, or exudative gastritis was present along with mucosal bile staining. Many of our patients showed chronic atrophic gastritis or erosive gastritis. However, the correlation between duodenogastric bile reflux and atrophic or erosive gastritis is still controversial.5,20
We previously conducted a study to determine the optimal time for delineating the bile duct on gadoxetic MRC. The T1-weighted gadoxetic-MRC images were acquired at 20, 30, 50, and 60 minutes after contrast administration and at 30 minutes after a fatty meal, and the time sequence of biliary excretion of gadoxetic acid in 40 normal healthy subjects was evaluated. Although complete filling of the distal common bile duct with contrast was seen on the 30-minute delayed preprandial images, the T1-weighted gadoxetic-MRC images acquired 60 minutes after contrast injection proved to be the optimal images for evaluating the bile duct.21 Usually, the hepatobiliary phase images of the liver can be acquired after a 20- or 30-minute delay; therefore, a 60-minute delayed image acquired using gadoxetic acid is not common in clinical practice. However, we know that the 60-minute delayed images can reveal greater extension of the biliary excreted contrast media and provide varied clinical information.22–28
In practice, gadoxetic MRC costs much more than upper endoscopy does. Therefore, it might seem unreasonable to perform MRC just to diagnose bile reflux gastritis. Our study showed that the reflux of contrast media was an evidence of the presence of bile reflux. One third of the patients who showed contrast media in the stomach really had bile reflux gastritis on upper endoscopy examinations. We thought that biliary excreted contrast media in the stomach on the 60-minute delayed gadoxetic-MRC images could suggest the presence of duodenogastric bile reflux, and this was an added benefit of gadoxetic MRC. Therefore, if the reflux of contrast media in the stomach is seen on the hepatobiliary-phase gadoxetic-MRC images, we recommend performing endoscopy to confirm the diagnosis of bile reflux gastritis. Magnetic resonance cholangiography is a useful imaging modality with many advantages such as the absence of irradiation, a noninvasive nature, good resolution, wide field of view, and large development potential. Therefore, in the future, researchers should aim to develop more dedicated MRI techniques for diagnosing gastrointestinal diseases.
False-positive results can be obtained if there are anatomical variations in the biliary tree or gastrointestinal tract of a patient, or if the patient ingested fluids that could appear as high signal intensity on T1-weighted images during the study.29 However, none of our patients had any anatomical variations and continued fasting during the studies.
Our study has many limitations. First, we correlated the amount of contrast media in the stomach with the presence of bile reflux gastritis. However, in our routine gadoxetic-MRC protocol, the patient, after undergoing MR cholangiopancreatography for 20 minutes, does not lie down but moves around until it is time to acquire the 60-minute delayed images. Therefore, the amount of refluxed contrast media in the stomach may not represent the absolute amount of refluxed contrast media. This might limit the generalizability of the present results.
Second, the MRI scans performed in our study were aimed at the diagnosis of biliary or pancreatic diseases; thus, we did not use an effervescent agent or antimotility medication for better imaging of the gastrointestinal tract. Therefore, the patients' stomachs were not fully distended, and this might have produced motion artifacts related to the gastrointestinal tracts. Because of this reason, we did not try to diagnose gastritis by using MR images. To our knowledge, only few studies have evaluated the diagnostic ability of MRI for the diagnosis of inflammatory diseases of the stomach.30,31 Mucosal enhancement, gastric wall thickening, perigastric infiltration, and increased signal intensity of the gastric wall on T2-weighted images could indicate gastric wall inflammation. Moreover, the remnant stomach after subtotal gastrectomy usually shows edematous wall thickening, and patients undergoing such surgeries frequently have bile reflux gastritis.32–34 Further research is needed to consider the use of MRI for the diagnosis of gastritis.
Third, our sample size was small. This was a single-center retrospective study. Thus, it was difficult to acquire the 60-minute delayed images for every patient because of the long durations required for completing imaging. In our hospital, 60-minute delayed gadoxetic-MRC images are used for evaluating biliary pathology. A well-designed prospective study will be necessary to generalize the results of the current study.
In conclusion, biliary excreted contrast media in the stomach suggests duodenogastric bile reflux, and one third of the patients who showed contrast media in the stomach had bile reflux gastritis. The benefit of gadoxetic MRC is that it shows very high specificity to predict the presence of bile reflux gastritis. Moreover, the amount of reflux contrast media was significantly correlated with the presence of bile reflux gastritis.
1. Ridlon JM, Kang DJ, Hylemon PB. Bile salt biotransformations by human intestinal bacteria. J Lipid Res
. 2006;47:241–259. PubMed PMID: 16299351. Epub 2005/11/22. eng.
2. Frommherz L, Bub A, Hummel E, et al. Age-related changes of plasma bile acid concentrations in healthy adults—results from the cross-sectional KarMeN study. PLoS One
. 2016;11:e0153959. PubMed PMID: 27092559. Pubmed Central PMCID: PMC4836658. Epub 2016/04/20. eng.
3. Hall JE, Guyton AC. Guyton and Hall textbook of medical physiology
. 12th. Philadelphia, PA: edSaunders/Elsevier; 2011. xix, 1091 p.
4. Radev D, Kotsev I, Panaiotov P, et al. [The incidence of duodenogastric reflux and its relation to stomach and duodenal diseases]. Khirurgiia
. 1990;43:61–64. PubMed PMID: 2097428. Epub 1990/01/01. Chestota na duodenogastralniia refluks i vruzkata mu sus zaboliavaniiata na stomakha i duodenuma [Bulgarian].
5. Sobala GM, O'Connor HJ, Dewar EP, et al. Bile reflux and intestinal metaplasia in gastric mucosa. J Clin Pathol
. 1993;46:235–240. PubMed PMID: 8463417. Pubmed Central PMCID: PMC501177. Epub 1993/03/01. eng.
6. Korzon M, Szarszewski A, Kamińska B, et al. Cholescintigraphy in the diagnosis of duodenogastric reflux in children, preliminary report. Rocz Akad Med Bialymst
. 1995;40:673–677. PubMed PMID: 8775325. Epub 1995/01/01.
7. Atak I, Ozdil K, Yücel M, et al. The effect of laparoscopic cholecystectomy on the development of alkaline reflux gastritis and intestinal metaplasia. Hepatogastroenterology
. 2012;59:59–61. PubMed PMID: 22260822. Epub 2012/01/21.
8. Woodfield CA, Levine MS. The postoperative stomach. Eur J Radiol
. 2005;53:341–352. PubMed PMID: 15741008. Epub 2005/03/03.
9. Lindecken KD, Salm B. [The effectiveness of Braun's anastomosis in Billroth II surgery. The role of hepatobiliary sequence scintigraphy (HBSS) in the diagnosis of bile flow following stomach resection]. Rofo
. 1993;159:158–160. PubMed PMID: 8353262. Epub 1993/08/01. Zur Effektivitat der Braunschen Fusspunktanastomose bei der Operation nach Billroth II. Stellenwert der hepatobiliaren Sequenzszintigraphie (HBSS) bei der Diagnostik des Galleflusses nach Magenresektion [German].
10. Tytgat GN. The Sydney System: endoscopic division. Endoscopic appearances in gastritis/duodenitis. J Gastroenterol Hepatol
. 1991;6:223–234. PubMed PMID: 1912432. Epub 1991/05/01.
11. Kleba T. [Gastroscopic criteria and most frequent pain in bile reflux gastritis]. Polski merkuriusz lekarski : organ Polskiego Towarzystwa Lekarskiego
. 1998;4:190–192. PubMed PMID: 9770993. Epub 1998/10/15. Kryteria gastroskopowe oraz najczestsze dolegliwosci w zolciowym zapaleniu blony sluzowej zoladka [Polish].
12. Orlando R 3rd, Welch JP. Carcinoma of the stomach after gastric operation. Am J Surg
. 1981;141:487–491. PubMed PMID: 6164300. Epub 1981/04/01.
13. Strignano P, Collard JM, Michel JM, et al. Duodenal switch operation for pathologic transpyloric duodenogastric reflux. Ann Surg
. 2007;245:247–253. PubMed PMID: 17245178. Pubmed Central PMCID: Pmc1876986. Epub 2007/01/25.
14. Madura JA. Primary bile reflux gastritis: diagnosis and surgical treatment. Am J Surg
. 2003;186:269–273. PubMed PMID: 12946831. Epub 2003/08/30.
15. Klingler PJ, Perdikis G, Wilson P, et al. Indications, technical modalities and results of the duodenal switch operation for pathologic duodenogastric reflux. Hepatogastroenterology
. 1999;46:97–102. PubMed PMID: 10228771. Epub 1999/05/06. eng.
16. Chen TF, Yadav PK, Wu RJ, et al. Comparative evaluation of intragastric bile acids and hepatobiliary scintigraphy in the diagnosis of duodenogastric reflux. World J Gastroenterol
. 2013;19:2187–2196. PubMed PMID: 23599645. Pubmed Central PMCID: PMC3627883. Epub 2013/04/20.
17. Müller-Lissner SA, Fimmel CJ, Sonnenberg A, et al. Novel approach to quantify duodenogastric reflux in healthy volunteers and in patients with type I gastric ulcer. Gut
. 1983;24:510–518. PubMed PMID: 6852631. Pubmed Central PMCID: PMC1420006. Epub 1983/06/01.
18. Thomas WE, Jackson PC, Cooper MJ, et al. The problems associated with scintigraphic assessment of duodenogastric reflux. Scand J Gastroenterol Suppl
. 1984;92:36–40. PubMed PMID: 6588532. Epub 1984/01/01.
19. Stein HJ, Smyrk TC, DeMeester TR, et al. Clinical value of endoscopy
and histology in the diagnosis of duodenogastric reflux disease. Surgery
. 1992;112:796–803. discussion −4. PubMed PMID: 1411953. Epub 1992/10/01.
20. Matsuhisa T, Arakawa T, Watanabe T, et al. Relation between bile acid reflux into the stomach and the risk of atrophic gastritis and intestinal metaplasia: a multicenter study of 2283 cases. Dig Endosc
. 2013;25:519–525. PubMed PMID: 23363381. Epub 2013/02/01.
21. Lee SW, Cha SH, Chung HH, et al. Functional magnetic resonance cholangiography with Gd-EOB-DTPA: a study in healthy volunteers. Magn Reson Imaging
. 2014;32:385–391. PubMed PMID: 24529920. Epub 2014/02/18.
22. Choi IY, Yeom SK, Cha SH, et al. Diagnosis of biliary stone disease: T1-weighted magnetic resonance cholangiography with Gd-EOB-DTPA versus T2-weighted magnetic resonance cholangiography. Clin Imaging
. 2014;38:164–169. PubMed PMID: 24359645. Epub 2013/12/24.
23. Choi IY, Cha SH, Yeom SK, et al. Diagnosis of acute cholecystitis: value of contrast agent in the gallbladder and cystic duct on Gd-EOB-DTPA enhanced MR cholangiography. Clin Imaging
. 2014;38:174–178. PubMed PMID: 24359644. Epub 2013/12/24.
24. Yeom SK, Lee SW, Cha SH, et al. Biliary reflux detection in anomalous union of the pancreatico-biliary duct patients. World J Gastroenterol
. 2012;18:952–959. PubMed PMID: 22408355. Pubmed Central PMCID: PMC3297055. Epub 2012/03/13.
25. Yacoub JH, Yousuf A, Agrawal G, et al. Evaluation of the gallbladder and cystic duct patency with gadoxetate disodium enhanced MR cholangiography: prospective comparison of patients with normal gallbladder function and acute cholecystitis. Acta Radiol
. 2015;56:782–789. PubMed PMID: 25009279. Epub 2014/07/11.
26. Corwin MT, Lamba R, McGahan JP. Functional MR cholangiography of the cystic duct and sphincter of Oddi using gadoxetate disodium: is a 30-minute delay long enough? J Magn Reson Imaging
. 2013;37:993–998. PubMed PMID: 23001618. Epub 2012/09/25.
27. Cieszanowski A, Stadnik A, Lezak A, et al. Detection of active bile leak with Gd-EOB-DTPA enhanced MR cholangiography: comparison of 20-25 min delayed and 60-180 min delayed images. Eur J Radiol
. 2013;82:2176–2182. PubMed PMID: 24012454. Epub 2013/09/10.
28. Lee NK, Kim S, Lee JW, et al. Biliary MR imaging with Gd-EOB-DTPA and its clinical applications. Radiographics
. 2009;29:1707–1724. PubMed PMID: 19959517. Epub 2009/12/05.
29. Guan J, Zhang L, Chu JP, et al. Congenital left intrahepatic bile duct draining into gastric wall mimicking biliary reflux gastritis. World J Gastroenterol
. 2015;21:3425–3428. PubMed PMID: 25805955. Pubmed Central PMCID: PMC4363778. Epub 2015/03/26.
30. Han SG, Chen Y, Qian ZH, et al. Eosinophilic gastroenteritis associated with eosinophilic cystitis: computed tomography and magnetic resonance imaging findings. World J Gastroenterol
. 2015;21:3139–3145. PubMed PMID: 25780317. Pubmed Central PMCID: PMC4356939. Epub 2015/03/18.
31. Cengic I, Tureli D, Aydin H, et al. Magnetic resonance enterography in refractory iron deficiency anemia: a pictorial overview. World J Gastroenterol
. 2014;20:14004–14009. PubMed PMID: 25320540. Pubmed Central PMCID: PMC4194586. Epub 2014/10/17.
32. Swartz DE, Mobley E, Felix EL. Bile reflux after Roux-en-Y gastric bypass: an unrecognized cause of postoperative pain. Surg Obes Relat Dis
. 2009;5:27–30. PubMed PMID: 19095503. Epub 2008/12/20.
33. Park DJ, Lee HJ, Jung HC, et al. Clinical outcome of pylorus-preserving gastrectomy in gastric cancer in comparison with conventional distal gastrectomy with Billroth I anastomosis. World J Surg
. 2008;32:1029–1036. PubMed PMID: 18256877. Epub 2008/02/08.
34. Zobolas B, Sakorafas GH, Kouroukli I, et al. Alkaline reflux gastritis: early and late results of surgery. World J Surg
. 2006;30:1043–1049. PubMed PMID: 16736335. Epub 2006/06/01.