- Email: [email protected]
Gastroesophageal Mucosal Injury after Cholecystectomy: An Indication for Surveillance?
Accepted Manuscript Gastroesophageal Mucosal Injury after Cholecystectomy: An Indication for Surveillance? Syeda Nadia Shah Gilani, MD, MRCS, Gary Alan Bass, MD, Natallia Kharytaniuk, MB BCh BAO, DOHNS, Michelle R. Downes, MD, MRCPath, Emer Frances Caffrey, MD, MRCPath, Iqbal Tobbia, MD, FRCPath, Thomas Noel Walsh, MD MCh FRCSI PII:
S1072-7515(16)31692-1
DOI:
10.1016/j.jamcollsurg.2016.12.003
Reference:
ACS 8554
To appear in:
Journal of the American College of Surgeons
Received Date: 3 November 2016 Revised Date:
3 December 2016
Accepted Date: 5 December 2016
Please cite this article as: Shah Gilani SN, Bass GA, Kharytaniuk N, Downes MR, Caffrey EF, Tobbia I, Walsh TN, Gastroesophageal Mucosal Injury after Cholecystectomy: An Indication for Surveillance?, Journal of the American College of Surgeons (2017), doi: 10.1016/j.jamcollsurg.2016.12.003. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Gastroesophageal Mucosal Injury after Cholecystectomy: An Indication for Surveillance?
Syeda Nadia Shah Gilani1,3, MD, MRCS, Gary Alan Bass1,3, MD, Natallia Kharytaniuk1, MB
RI PT
BCh BAO, DOHNS, Michelle R Downes2, MD, MRCPath, Emer Frances Caffrey2, MD, MRCPath, Iqbal Tobbia2, MD, FRCPath, Thomas Noel Walsh1,3, MD MCh FRCSI
M AN U
and Royal College of Surgeons in Ireland3, Dublin, Ireland.
SC
Departments of Surgery1 and Pathology2, Connolly Hospital, Blanchardstown, Dublin, Ireland
Disclosure Information: Nothing to disclose. Support: The Lollipop Day Fund
Abstract presented at the American College of Surgeons 101st Annual Clinical Congress,
TE D
Chicago, IL, October 2015.
EP
Correspondence address: Professor TN Walsh, Dept of Surgery, Connolly Hospital, Blanchardstown, Dublin 15, IRELAND E: [email protected] T: +353 1 646 5632353; +353 1 8202284
AC C
Running head: Cholecystetomy-Mediated Gastric Injury
1
ACCEPTED MANUSCRIPT
ABSTRACT Background: Cholecystectomy alters bile release dynamics from pulsatile meal-stimulated to continuous, and results in retrograde duodeno-gastric bile reflux(DGR). Bile is implicated in
RI PT
mucosal injury following gastric surgery, but whether cholecystectomy causes esophago-gastric mucosal inflammation, thus increasing the risk of metaplasia, is unclear.
Study Design: This study examined whether cholecystectomy-induced DGR promotes chronic
SC
inflammatory mucosal changes of the stomach and/or at the esophago-gastric junction(EGJ). Four groups of patients were studied and compared with controls. A group of patients was
M AN U
studied before and one year after cholecystectomy; two further groups were studied 5-10 years and 10-20 years post-cholecystectomy(LTPC). All underwent abdominal ultrasound and upper gastrointestinal endoscopy(EGD) with gastric antral and EGJ biopsies, noting the presence of gastric bile-pooling. Biopsy specimens were stained for Ki67 and p53 over-expression and the
TE D
bile reflux index(BRI) was calculated.
Results: At endoscopy, bile-pooling was observed in 9/26(34.6%) controls, in 8/25(32%) patients pre-cholecystectomy, in 15/25(60%) one year post-cholecystectomy(p=0.047) and
EP
23/29(79.3%) LTPC patients(p=0.001). BRI positivity at the EGJ increased from 19% of controls through 41% of LTPC patients(p=0032). Ki67 was over-expressed at EGJ in 19% of controls but
AC C
in 62% of LTPC patients(p=0.044); p53 was over-expressed at the EGJ in 19% controls compared with 66% of LTPC patients(p=0.001). Conclusions: DGR was more common in patients with gallstones than controls and its incidence doubled after cholecystectomy. This was associated with inflammatory changes in the gastric antrum and the EGJ, evident in most LTPC patients. Ki67 and p53 over-expression at the EGJ suggest cellular damage attributable to chronic bile exposure post-cholecystectomy, increasing
2
ACCEPTED MANUSCRIPT
the likelihood of dysplasia. Further studies are required to determine whether DGR-mediated esophageal mucosal injury is reversible or avoidable and if surveillance endoscopy is indicated
AC C
EP
TE D
M AN U
SC
RI PT
after cholecystectomy.
3
ACCEPTED MANUSCRIPT
INTRODUCTION Cholecystectomy – the standard of care for symptomatic cholelithiasis – alters the dynamics of bile storage and release.1 Bile is normally stored in the gallbladder in the fasting inter-digestive
RI PT
period and is propelled into the duodenum in response to meals under the influence of
cholecystokinin (CCK) mediated gallbladder contraction.1-3 Following cholecystectomy, the facility for bile storage is lost, and bile is continuously released into the duodenum, even during
SC
fasting1.
There appears to be a clear causal link between cholecystectomy and duodeno-gastric bile
M AN U
reflux.2-4 The continuous presence of bile in the duodenum permits its overflow across the pylorus and into the stomach. A significantly greater concentration of bile acids has been confirmed in naso-gastric aspirates from patients with gallstones compared to those without, and this increases further following cholecystectomy.1
TE D
Cholecystectomy also results in elevated serum levels of cholecystokinin (CCK).3 CCK is an enteric hormone that is normally inhibited by the negative feedback of the CCK-mediated bile bolus in the duodenum5, but following cholecystectomy this switch off mechanism is lost,
EP
resulting in persistent elevation of CCK levels3. Exposure of the gastric mucosa to bile has been shown to cause erythema, inflammation or ulceration of the gastric mucosa6,7. Gastric ulceration
AC C
and gastritis have been shown to attenuate CCK-mediated increases in pyloric muscle tone8, which may explain the observation of a bilious refluxate through an open pylorus on gastroscopy of patients with gastritis9. The increase in circulating CCK results in reduction of lower oesophageal sphincter pressure10 and increases the frequency of transient lower esophageal sphincter relaxation episodes (TLESREs), potentially further exposing the lower oesophageal
4
ACCEPTED MANUSCRIPT
mucosa to refluxed bile11. Increased duodeno-gastric reflux in turn predisposes to increased gastro-oesophageal reflux injury. While we know that these injuries occur, their precise molecular mechanism at the cellular level
RI PT
has yet to be elucidated. We hypothesised that cholecystectomy should be considered a model of iatrogenically-induced mucosal injury of the upper gastro-intestinal tract, resulting in bileinduced chronic cellular injury to the mucosa of the intact stomach and esophago-gastric
SC
junction. This should be quantifiable as an increase in bile reflux index (BRI) positivity and as an
esophageal junction.
PATIENTS AND METHODS Patient Cohorts/Study Groups
M AN U
increase in cellular Ki67 and p53 expression of the mucosa of the gastric antrum and gastro-
TE D
In order to determine whether mucosal injury is truly attributable to cholecystectomy, a paired cohort of patients was studied before and after cholecystectomy, and compared with longstanding post-cholecystectomy patients and further with controls. Eighty patients, divided into
EP
five study groups, were prospectively-enrolled into this observational case-control study, in accordance with institutional ethics committee approval.
AC C
Group 1 (n=26) were individuals investigated for dyspepsia, but without evidence of gallbladder disease and who were not on non-steroidal anti-inflammatory agents (NSAIDs) or proton pump inhibitors (PPIs); these acted as the non-gallstone control group of patients. Group 2 (n=25) were patients with an ultrasound-confirmed diagnosis of symptomatic cholelithiasis, and were scheduled for cholecystectomy. Because of their symptoms of dyspepsia they also underwent upper GI endoscopy.
5
ACCEPTED MANUSCRIPT
Group 3 (n=25) were the Group 2 patient-cohort described above, who underwent repeat upper GI endoscopy one year post-cholecystectomy. Groups 4 and 5 were 29 patients who presented with dyspepsia but who had previously
RI PT
undergone cholecystectomy: namely, Group 4, (n=10) had undergone cholecystectomy 5 to 10 years previously, and Group 5, (n=19) had had their cholecystectomy between 10 to 20 years previously. While recruitment of an additional cohort of patients 5-10 years post-
SC
cholecystectomy who were asymptomatic (and thus NOT complaining of dyspepsia) could address a potential confounding variable, strengthening our hypothesis-testing and the
M AN U
conclusions which may be drawn from the observed results, prospectively enrolling an adequately-powered cohort of well asymptomatic patients for an invasive procedure such as gastroscopy would have posed both ethical and logistic challenges.
TE D
EGD And Collection Of Biopsy Samples
Upper gastrointestinal endoscopy was performed on all patients by a single endoscopist (TNW), who recorded common macroscopic findings using standardized consensus criteria for
EP
inflammation. The presence or absence of macroscopic bile-pooling was noted upon the endoscopic intubation of the stomach in each patient and photographed for subsequent
AC C
verification. A minimum of four biopsies per site were taken from the gastric antrum and the EGJ. The biopsy specimens were oriented on filter paper and immediately fixed in formalin. Bile Reflux Index (BRI)
Histological features of bile reflux in the gastric antrum and EGJ were recorded using the bile reflux index (BRI) devised by Sobala et al.12 This was determined by histological examination of 4 x 1mm specimen sections, following haematoxylin and eosin (H&E) staining, by an expert
6
ACCEPTED MANUSCRIPT
histopathologist, blinded to the clinical details of the patients and to the endoscopic findings. In this system, the index is derived based on the presence and/or severity of certain histological parameters, such as: edema of the lamina propria (E), intestinal metaplasia (IM), chronic
RI PT
inflammation (CI), and Helicobacter pylori colonization in the stomach. A BRI greater than 14 indicated duodeno-gastric reflux in the Sobala study, with 70% sensitivity and 85% specificity.12 Immuno-staining For Ki67 and p53
SC
Immuno-histochemical (IHC) staining of 4 x 1mm tissue slices was undertaken to demonstrate the presence and cellular location of MIB-1/Ki67 and p53 proteins in endoscopic biopsies taken
M AN U
at the gastric antrum and EGJ (IHC Leica BOND-MAX™, BOND™ reagents & antibody kits; Leica Biosystems, Melbourne, Australia). IHC reactivity of Ki67 and p53 was determined as the nuclear staining in the expected proliferative compartment. All slides were evaluated in a blinded fashion for nuclear Ki67 and p53 staining.
TE D
For p53, <15% positive nuclei was regarded as normal expression (grade 0), 15-40% as moderate over-expression (grade 1) and >40% as strong over-expression (grade 2). For Ki67, the percentage of positive nuclei was determined and a percentage of positive cells
EP
<20% was regarded as normal expression (grade 0), 20-50% as moderate over-expression (grade 1), and >50% as strong over-expression (grade 2).
AC C
Statistical Analysis
Data were analyzed using SPSS version 21. P-values were all two-tailed and the alpha-level of significance was set at 0.05. The prevalence was shown as a percentage. The chi-squared paired test was used for testing comparison in paired group, with Bonferroni correction utilised for multiple comparisons. ANOVA was used to test multivariate analysis between and within groups. Post-hoc correlations were calculated with Dunett’s T3 test.
7
ACCEPTED MANUSCRIPT
RESULTS Patient Demographics The mean age (±SEM) for all patients was 57.3(±1.5) years; the mean age of control subjects
RI PT
was 58(±1.7) years compared with 54.2(±2.6) years in those undergoing cholecystectomy and 59.2 (± 3.1) years in patients with an antedecent history of cholecystectomy with no significant difference between the cohorts (p=0.626). The mean (±SEM) body mass index (BMI) across all
SC
groups was 27.8 (±0.54). The mean BMI of patients was 26.8(±4.3)in the control group versus 29.1(±4.5) in those undergoing cholecystectomy or with a history of cholecystectomy at
M AN U
28.3(±3.1). There was no significant difference in BMI between groups (p = 0.706). A history of smoking was present in 11/25 (44%) of control patients versus 13/26 (50%) of patients undergoing cholecystectomy (p=0.668). While our study was not powered to explore Barrett’s (with an estimated population prevalence of up to 2%) as a co-variate with DGER in progressive
TE D
gastric and esophageal mucosal injury, the incidence of Barrett’s on endoscopy across our studied cohorts was 3/80(3.75%) patients, including 1 patient in each of the control, paired (preand post- cholecystectomy) and historic cholecystectomy groups. No malignancies were detected
EP
in any of these patients.
Bile-pooling visible on endoscopy
AC C
On endoscopy note was made of the presence or absence of bile-pooling under gravity on the mucosa of the greater curve of the stomach. Bile-pooling was recorded in 9/26 (34.6%) individuals in the control group (Group 1); a value similar to the cohort of patients awaiting cholecystectomy 8/25 (32%) (p=0.843) (Group 2) (Table 1). When endoscopy was repeated at one year post-cholecystectomy in the same patient cohort, however, 15/25 (60%) now exhibited
8
ACCEPTED MANUSCRIPT
gastric bile-pooling, a significant increase directly consequent on cholecystectomy(Group 3) (p=0.047). Endoscopy performed on a cohort of patients 5-10 years post-cholecystectomy found bile-
RI PT
pooling in 7/10(70%). Endoscopy performed on a final cohort of patients that were 10-20 years post-cholecystectomy found bile-pooling in 16/19 (84.2%) patients. When these the latter two
cholecystectomy (p=0.001 versus controls). Bile Reflux Index (BRI)
SC
groups were pooled there was bile-pooling in 23/29 (79.3%) patients that were long-term post-
M AN U
Biopsies of the gastric antrum were found to have a positive BRI in 3/36 (11%) of the controls, 5/25 (20%) pre-operative patients, 9/25 (36%) of the same patients at one year postcholecystectomy and in 20/29 (69%) of the patients with a long history of cholecystectomy (p=0.001) (Figure 1,2; Table 1). There was no significant difference between the controls and
TE D
pre-cholecystectomy patients (p=0.406); nor between the same patients before and one-year postcholecystectomy (p = 0.208), but there was a marked difference between these groups and patients that had undergone cholecystectomy at least five years previously (p<0.001).
EP
When endoscopic biopsies at the EGJ were compared, BRI was positive in 5/26 (19%) controls, 3/25 (12%) pre-cholecystectomy patients, and in 5/25 (20%) patients at one year post-
AC C
cholecystectomy. It rose to 12/29 (41%) patients with a long history of cholecystectomy which was significantly higher than controls (p=0.032). Ki67
At the gastric antrum, the cellular proliferation marker Ki67 was over-expressed in 6/26 (23%) controls, compared with 17/29 patients (59%) with long-standing history of cholecystectomy (p=0.001) (Figure 3).
9
ACCEPTED MANUSCRIPT
Ki67 was over expressed at the EGJ in 5/26 (19%) of controls and in 18/29 patients (62%) patients with a long-standing history of cholecystectomy (p=0.001) (Figure 3). p53
RI PT
In the gastric antrum the tumor-suppressor gene p53 was over-expressed in 1/26 (4%) control subjects, compared with 11/25(44%) pre-operative cholecystectomy patients, 9/25(40%) one year post-cholecystectomy patients and 15/29(52%) patients with a long-standing history of
SC
cholecystectomy (Table 1). Expression of p53 was significantly higher in patients who had undergone cholecystectomy (either recent or remote), compared with controls (p=0.001).
M AN U
p53 was over-expressed at the EGJ in 5/26 (19%) controls, in 12/25 (48%) recently-operated patients and 19/29 (66%) long-standing post-cholecystectomy patients, and this was significantly higher than controls (p=0.001) (Figure 2).
TE D
DISCUSSION
In this study, cholecystectomy resulted in an increase in duodeno-gastric reflux injury that was progressive with time. The same 25 symptomatic patients underwent endoscopic evaluation
EP
before and after their cholecystectomy and were found to have excess macroscopic (visible bile pooling), microscopic (histochemical evidence of bile-mediated cellular injury) and epigenetic
AC C
(p53 and Ki-67 upregulation) changes. Whether it is prolonged mucosal contact with the bile per se or with some hitherto-uncovered substrate carried in the bile will be the subject of future work. Nevertheless, following cholecystectomy, these otherwise-identical patients were found in our paired comparison to have an increased prevalence of bile-exposed mucosa. Further elevation of inflammatory markers in patients with increased duration of bile exposure suggests that this injury is progressive. The increased incidence of bile reflux injury in pre-operative
10
ACCEPTED MANUSCRIPT
patients with symptomatic gallstone disease with respect to controls is likely to represent a proportion of patients that may have lost part or all of their bile reservoir capacity to gallstones, thus having become “functionally-cholecystectomised” at various time periods prior to this study
RI PT
and in whom little further duodeno-gastric reflux would be expected to occur following
gallbladder removal. Studies with 99Tc-HIDA radionucleotide imaging lend further weight to this hypothesis.13
SC
Duodeno-gastric reflux is identifiable as a dependent bile-pool during EGD14 and its incidence increases significantly following cholecystectomy.13,14 It is associated with: (1) an alkaline shift
M AN U
in gastric content15, (2) the development of chronic superficial gastritis6,7, and (3) the change in parietal cell activity13. Cell proliferation is thought to be one of the earliest steps in the development of pre-neoplastic change in response to inflammation. Chronic inflammation, through repetitive cycles of epithelial regeneration and aberrant growth, leads to over-expression
TE D
of the intracellular protein Ki67.16 We observed an increase in the expression of the molecular marker Ki67 in the gastric antral mucosa and at the EGJ on repeat endoscopic biopsy in our patients within one year of cholecystectomy which was associated with the increase in bile-
EP
pooling and increased BRI. This was observed to be significantly higher in patients studied at five and subsequent years after cholecystectomy. This up-regulated expression of Ki67 may
AC C
reflect chronic mucosal inflammation and repetitive cycles of epithelial regeneration17. Immunohistochemical staining for MIB-1, the Ki67 proliferation antigen, has been shown to increase expression in the Barrett’s oesophagus-dysplasia-adenocarcinoma sequence.16-18 These changes were mirrored by up-regulation of p53 at the antrum and the EGJ in our patient cohorts. Changes in p53 expression are thought to result from post-translational modifications in response to cellular stress or DNA damage.19 A number of studies have demonstrated positive
11
ACCEPTED MANUSCRIPT
immune-histochemical staining for p53 in dysplastic Barrett’s metaplasia and in gastric and esophageal adenocarcinoma20-23 while further work has shown over-expression of Ki67 in association with Barrett’s mucosa and esophageal adenocarcinoma24. Several studies have
RI PT
suggested that bile can directly damage DNA, with the bile constituent lithocholic acid
specifically identified as a promoter in experimental carcinogenesis through cell transformation and DNA-strand breaks.25,26 Exposure of the gastric and esophageal epithelial cells to abnormal
SC
levels of bile acids may lead to the generation of reactive oxygen species (ROS) and reactive nitrogen species (RNS) through multiple pathways involving disruptions of the cell membrane
M AN U
and mitochondria.27 This, in turn, may result in increased DNA damage and thus a tendency towards increased mutagenesis.27 Thus the molecular/epigenetic changes and early carcinogenesis, as shown by the increased expression of p53 and Ki67 that occur as a result of cholecystectomy, are similar to those identified as precursors of Barrett’s oesophagus.17,18
TE D
Patients with Barrett’s esophagus and esophageal adenocarcinoma are known to have increased duodeno-gastric bile reflux28,29, the mechanism of which is unclear. The role of duodeno-gastric reflux in the pathogenesis of Barrett’s oesophagus is, however, well established.30,31 And
adenocarcinoma.32
EP
Barrett’s oesophagus is a known pre-malignant precursor lesion for esophageal
AC C
It has been reported that gallstone disease is more prevalent in Barrett’s oesophagus.33 We have previously shown that gallbladder function is impaired in patients with Barrett’s esophagus and esophageal adenocarcinoma.34 This impaired motility and dysfunction of the gallbladder may predispose to increased duodeno-gastric reflux consequent on impaired bile storage. Impaired motility may in turn, be a contributing factor for the development of gallstones.35 Decreased motility and formation of stones all predispose to duodeno-gastric bile reflux as the function of
12
ACCEPTED MANUSCRIPT
the gallbladder becomes progressively impaired or completely non-functional.36 Cholecystectomy will worsen duodeno-gastric bile reflux in patients with a functioning gallbladder, due to the immediate loss of a bile reservoir.37
RI PT
Cholecystectomy is one of the most commonly performed major surgical procedures in the
western world with approximately 917,000 performed each year in the USA and over 50,000 performed in the England.38 It is concerning that cholecystectomy mediated mucosal injury
SC
appears to mimic the situation that pertains following partial gastrectomy for peptic ulcer
disease, when reconstruction with Bilroth I or Billroth II exposes the stomach to increased bile
M AN U
and pancreatic reflux39 resulting in an increased incidence in gastritis and ultimately a significantly greater incidence of cancer in the gastric remnant.40,41 This is avoided by the Rouxen-Y technique where bile is diverted away from the stomach39 and probably explains the lower incidence in gastric cancer following Roux-en-Y reconstruction for peptic ulcer surgery39,42. Data
TE D
linking cholecystectomy with chronic gastric and oesophageal inflammation are persuasive, which are also suggestive of an increased risk of inflammation-induced metaplasia and neoplasia.43,44 The impact of bile on gastrointestinal mucosa, described by Dixon and colleagues
EP
as the BRI, appears to be a more robust measure of injury from direct quantitative measurement of bilirubin in gastric juice. The latter has been shown by us45 and by others46-48 to be inaccurate,
AC C
often under-estimating bilirubin concentration. In conclusion, we believe that our study provide novel insights into post-cholecystectomy chronic inflammatory change in the stomach and lower oesophagus. While correlation may not necessarily imply causation, the trends towards excess bile pooling and the associated microscopic and epigenetic changes are suggestive of progressive injury as the natural history of prolonged gastric mucosal exposure to bile. Cytological changes attributable to DGR occur early
13
ACCEPTED MANUSCRIPT
in the gastric and EGJ mucosa of patients following cholecystectomy and these appear to be progressive. Ki67 and p53 cellular over-expression in response to DGR suggests these changes may be pre-cursors of chronic inflammation raising concern its long-term consequences. If
RI PT
cholecystectomy places patients at greater risk for inflammation/dysplasia/neoplasia, there may be a benefit from endoscopic surveillance to begin after 20 years and repeated at perhaps 5yearly intervals, as with Barrett’s esophagus.49 Consideration should also be given to the role of
SC
systemic anti-inflammatory agents in protecting the gastro-intestinal mucosa. Further work is required to ascertain if this mucosal injury is reversible or avoidable through medication or
AC C
EP
TE D
M AN U
lifestyle changes.
14
ACCEPTED MANUSCRIPT
REFERENCES [1] Svensson J, Gelin J, Svanvik J. Gallstones, cholecystectomy, and duodenogastric reflux of bile acid. Scand J Gastroenterol 1986;21:181-187.
RI PT
[2] Cabrol J, Navarro X, Simo-Deu J, Segura R. Evaluation of duodenogastric reflux in gallstone disease before and after simple cholecystectomy. Am J Surg 1990;160:283-286.
cholecystokinin. Am J Gastroenterol 2002;97:2189-2192.
SC
[3] McDonnell C, Bailey I, Stumpf T, et al. The effect of cholecystectomy on plasma
[4] Lujan-Mompean J, Robles-Campos R, Parrilla-Paricio P, et al. Duodenogastric reflux in
1993;176:116-118.
M AN U
patients with biliary lithiasis before and after cholecystectomy. Surg Gynecol Obstet
[5] Gomez G, Lluis F, Guo YS, et al. Bile inhibits release of cholecystokinin and neurotensin. Surgery 1986;100:363–368.
TE D
[6] Rhodes J, Barnard DE, Philips SF, et al. Increased reflux of bile into the stomach in patients with gastric ulcer. Gastroenterology 1969;57:241-252. [7] Harmon J, Doong T, Gadacz TR. Bile acids are not equally damaging to the gastric mucosa.
EP
Surgery 1978;84:79-86.
[8] Si XM, Huang L, Lv P, et al. Effects of cholecystokinin-8 induced gastric dysmotility on bile
AC C
regurgitation during stress and molecular mechanisms. Regul Pept 2006;136:64-71. [9] Rovelstad RA. The incompetent pyloric sphincter. Bile and mucosal ulceration. Am J Dig Dis 1976;21:165-173.
[10] Fisher RS, Dimarino AJ, Cohen S. Mechanism of cholecystokinin inhibition of lower esophageal sphincter pressure. Am J Physiol 1975;228:1469–1473.
15
ACCEPTED MANUSCRIPT
[11] Boulant J, Fioramonti J, Dapoigny M, et al. Cholecystokinnin and nitric oxide in transient lower oesophageal sphincter relaxation to gastric distention in dogs. Gastroenterology 1994;107:1059-1066.
RI PT
[12] Sobala G, O'Connor HJ, Dewar EP, et al. Bile reflux and intestinal metaplasia in gastric mucosa. J Clin Pathol 1993;46:235-240.
[13] Lorusso D MG, Mangini V, Messa C, et al. Duodenogastric reflux of bile acids, gastrin and
SC
parietal cells, and gastric acid secretion before and 6 months after cholecystectomy. Am J Surg 1990;159:575-578.
M AN U
[14] Kunsch S, Neesse A, Huth J, et al. Increased Duodeno-Gastro-Esophageal Reflux (DGER) in symptomatic GERD patients with a history of cholecystectomy. Z Gastroenterol 2009;47:744748.
[15] Brown T, Walton G, Cheadle WG, Larson G M. The alkaline shift in gastric pH after
TE D
cholecystectomy. Am J Surg 1989;157:58-65.
[16] Going JJ, Keith WN, Neilson L, et al. Aberrant expression of minichromosome maintenance proteins 2 and 5, and Ki-67 in dysplastic squamous oesophageal epithelium and Barrett's
EP
mucosa. Gut 2002;50:373-377.
[17] Whittles CE, Biddlestone LR, Burton A, et al. Apoptotic and proliferative activity in the
AC C
neoplastic progression of Barrett's oesophagus: a comparative study. J Pathol 1999;187:535-540. [18] Hong M, Laskin WB, Herman BE, et al. Expansion of the Ki-67 proliferative compartment correlates with degree of dysplasia in Barrett's esophagus. Cancer 1995;75:423-429. [19] Younes M, Lechago J, Chakraborty S, et al. Relationship between dysplasia, p53 protein accumulation, DNA ploidy, and Glut-l overexpression in Barretts metaplasia. Scand J Gastroenterol 2000;35:131-137.
16
ACCEPTED MANUSCRIPT
[20] Binato M, Gurski RR, Fagundes RB, et al. P53 and Ki-67 overexpression in gastroesophageal reflux disease - Barrett's esophagus and adenocarcinoma sequence. Dis Esophagus 2009;22:588-595.
RI PT
[21] Chen M, Huang J, Zhu Z, et al. Systematic review and meta-analysis of tumor biomarkers in predicting prognosis in esophageal cancer. BMC Cancer 2014;14:892.
[22] Gregson EM, Bornschein J, Fitzgerald RC. Genetic progression of Barrett’s oesophagus to
SC
oesophageal adenocarcinoma. Br J Cancer 2016;115:403-410.
[23] Clemons NJ, Phillips WA, Lord RV. Signaling pathways in the molecular pathogenesis of
M AN U
adenocarcinomas of the esophagus and gastroesophageal junction. Cancer Biol The 2013;14:782-795.
[24] Kerkhof M, Steyerberg EW, Kusters JG, et al. Aneuploidy and high expression of p53 and Ki67 is associated with neoplastic progression in Barrett esophagus. Cancer Biomark 2008;4:1-
TE D
10.
[25] Hiasa Y, Konishi Y, Kamamoto Y, et al. Effect of lithocolic acid on DL-ethionine carcinogenesis in rat liver. Gann 1971;62:239-245.
EP
[26] Narisawa T, Magadia NE, Weisburger JH, Wynder EL. Promoting effect of bile acids on colon carcinogenesis after intrarectal instillation of N-methyl-N'-nitro-N-nitrosoguanidine in rats.
AC C
J Natl Cancer Inst 1974;53:1093-1097. [27] Dvorak K, Payne CM, Chavarria M, et al. Bile acids in combination with low pH induce oxidative stress and oxidative DNA damage: relevance to the pathogenesis of Barrett's oesophagus. Gut 2007;56:763-771. [28] Bernstein H, Bernstein C, Payne C M, et al. Bile acids as carcinogens in human gastrointestinal cancers. Mutat Res 2005;589:47-65.
17
ACCEPTED MANUSCRIPT
[29] Menges M, Muller M, Zeitz M. Increased acid and bile reflux in Barrett's esophagus compared to reflux esophagitis, and effect of proton pump inhibitor therapy. Am J Gastroenterol 2001;96:331-337.
RI PT
[30] Richter J. Importance of bile reflux in Barrett's esophagus. Dig Dis 2000;18:208-216. [31] Koek G, Sifrim D, Lerut T, et al. Multivariate analysis of the association of acid and
duodeno-gastro-oesophageal reflux exposure with the presence of oesophagitis, the severity of
SC
oesophagitis and Barrett's oesophagus. Gut 2008;57:1056-1064.
[32] Voutilainen M, Farkkila M, Juhola M, et al. Specialized columnar epithelium of the
M AN U
esophagogastric junction: prevalence and associations. The Central Finland Endoscopy Study Group. Am J Gastroenterol 1999;94:913-918.
[33] Izbeki F, Rosztoczy AI, Yobuta JS, et al. Increased prevalence of gallstone disease and impaired gallbladder motility in patients with Barrett's esophagus. Dig Dis Sci 2008;53:2268-
TE D
2275.
[34] Nassr AO, Gilani SN, Atie M, et al. Does impaired gallbladder function contribute to the development of Barrett's esophagus and esophageal adenocarcinoma? J Gastrointest Surg
EP
2011;15:908-914.
[35] Johnston DE, Kaplan MM. Pathogenesis and treatment of gallstones. N Engl J Med
AC C
1993;328:412-421.
[36] Matsuzaki J, Suzuki H, Asakura K, et al. Gallstones increase the prevalence of Barrett's esophagus. J Gastroenterol 2010;45:171-178. [37] Fein M, Bueter M, Sailer M, Fuchs KH. Effect of cholecystectomy on gastric and esophageal bile reflux in patients with upper gastrointestinal symptoms. Dig Dis Sci 2008;53:1186-1191.
18
ACCEPTED MANUSCRIPT
[38] Wu XD, Tian X, Liu MM, et al. Meta-analysis comparing early versus delayed laparoscopic cholecystectomy for acute cholecystitis. Br J Surg 2015;102:1302-1313. [39] Fukuhara K, Osugi H, Takada N, et al. Reconstructive procedure after distal gastrectomy for
RI PT
gastric cancer that best prevents duodenogastroesophageal reflux. World J Surg 2002;26:14521457.
investigation. Ann Surg 1989;210:159–164.
SC
[40] Toftgaard C. Gastric cancer after peptic ulcer surgery. A historic prospective cohort
[41] Hansson L-E, Nyrén O, Hsing AW, et al. The risk of stomach cancer in patients with gastric
M AN U
or duodenal ulcer disease. N Engl J Med 1996;335:242-249.
[42] Kojima K, Yamada H, Inokuchi M, et al. A comparison of Roux-en-Y and Billroth-I reconstruction after laparoscopy-assisted distal gastrectomy. Ann Surg 2008;247:962-967. [43] Lagergren J, Mattsson F. Cholecystectomy as a risk factor for oesophageal adenocarcinoma.
TE D
Br J Surg 2011;98:1133-1137.
[44] Dixon M, Mapstone NP, Neville PM, et al. Bile reflux gastritis and intestinal metaplasia at the cardia. Gut 2002;51:351-355.
EP
[45] Caldwell MT, Byrne PJ, Brazil N, et al. An ambulatory bile reflux monitoring system: an in vitro appraisal. Physiol Meas 1994;15:57-65.
AC C
[46] Barrett M, Myers JC, Watson DI, Jamieson GG. Detection of bile reflux: in vivo validation of the Bilitec Fibreoptic system. Dis Esophagus 2000;13:44-50. [47] Bechi P, Baldini F, Cianci F. What is the reliability of the portable spectro-photometer with a fibreoptic probe to detect bilirubin in bile refluxate? Is sensitivity of Bilitec likely to be improved? Florence: OESOKnowledge; 1998.
19
ACCEPTED MANUSCRIPT
[48] Tibbling Grahn L, Blackadder L, Franzen T, Kullman E. Gastric bile monitoring: an in vivo and in vitro study of Bilitec reliability. Scand J Gastroenterol 2002;37:1334-1337.
AC C
EP
TE D
M AN U
SC
RI PT
[49] Shaheen NG, Falk GW, Iyer PG, Gerson LB. Am J Gastroenterol 2016;111:30-50.
20
ACCEPTED MANUSCRIPT
Table 1. Endoscopic and Histopathologic Findings Study groups
n
%
n
%
n
Yes
9
34.6
8
32
15
No
17
65.4
17
68
10
Absent
23
88.5
Present
3
11.5
Absent
21
80.8
Present
5
19.2
BRI
p53 expression, antrum 25
Positive (>30%; <50%)
1
Extensive (>50%)
AC C
p53 expression, EGJ
0
%
60
23
79.3
40
6
20.7
76
12
48
9
31
6
24
13
52
20
69
21
84
17
68
17
58.6
4
16
8
32
12
41.4
96.2
14
56
17
68
14
48.3
3.8
5
20
6
24
11
37.9
0
6
24
2
8
4
13.8
EP
Normal (<30%)
n
19
TE D
EGJ
M AN U
Antrum
%
SC
Endoscopic bilepooling
>5 y Postcholecystectomy, n = 29
RI PT
Control, n = 26
1 y PostPre-cholecystectomy cholecystectomy (paired), n = 25 (paired), n = 25
Normal (<30%)
21
80.8
13
52.0
17
68.0
10
34.5
Positive (>30%; <50%)
5
19.2
5
20.0
5
20
15
51.7
Extensive (>50%)
0
0
7
28
3
12
4
13.8
Normal expression
20
76.9
24
96
14
56
12
41.4
Over-expression
6
23.1
1
4
11
44
17
58.6
Ki67 expression, antrum
21
ACCEPTED MANUSCRIPT
Ki67 expression, EGJ Normal expression
21
80.8
22
88
13
52
11
37.9
Over-expression
5
19.2
3
12
12
48
18
62.1
AC C
EP
TE D
M AN U
SC
RI PT
BRI, bile reflux index; EGJ, esophagogastric junction.
22
ACCEPTED MANUSCRIPT
FIGURE LEGENDS Figure 1. Incidence of macroscopic bile-pooling visible when the stomach was intubated with an endoscope. The proportion of patients with bile-pooling did not differ between patients awaiting
RI PT
cholecystectomy and those control patients with an intact gallbladder. However, the incidence of bile-pooling significantly increased on re-endoscopy of those same patients following
cholecystectomy. Green bar, endoscopic bile pooling; blue bar, no endoscopic bile pooling. Pre-
SC
op, preoperative; post-op, postoperative
M AN U
Figure 2. Photomicrograph of haematoxylin and eosin-stained specimen of gastric antral mucosa a patient in whom macroscopic bile reflux was seen on endoscopy; histologic bile reflux index was calculated using such slides.
TE D
Figure 3. (A) High-power light microscopy demonstrating p53 over-expression in an endoscopic biopsy taken at the gastric antrum in a patient who had undergone cholecystectomy. (B) Lowpower light microscopy in the same specimen. Immuno-histochemical brown staining
AC C
EP
demonstrates nuclear proliferation over a background haematoxylin & eosin stain.
23
ACCEPTED MANUSCRIPT
Precis Surgical endoscopists have long-noticed bile pooling upon endoscopic intubation of the stomach, more commonly in patients who have undergone cholecystectomy. There is concern that this bile
RI PT
may have a chronic irritative effect on the underlying gastric mucosa, predisposing it to
AC C
EP
TE D
M AN U
SC
metaplasia and thus examined mucosal specimens for p53 and Ki67 overexpression.
24
AC C
EP
Figure 1
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
Figure 2
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
Figure 3
ACCEPTED MANUSCRIPT
AUTHOR CONTRIBUTIONS FORM Individuals claiming authorship should meet all 3 of the following conditions in accordance with the “Consensus Statement on Surgery Journals Authorship—2005”:
RI PT
1) Authors make substantial contributions to conception and design, and/or acquisition of data, and/or analysis and interpretation of data; 2) Authors participate in drafting the article or revising it critically for important intellectual content; and 3) Authors give final approval of the version to be submitted and any revised version.
SC
Each author should have participated sufficiently in the work to take public responsibility for appropriate portions of the content. Allowing one’s name to appear as an author without having contributed significantly to the study or adding the name of an individual who has not contributed or who has not agreed to the work in its current form is considered a breach of appropriate authorship.
M AN U
Ghost-writing is NOT acceptable. No one, other than the authors listed below, should have contributed substantially to the writing and revising of the manuscript. Contributors who do not meet the criteria for authorship should be listed in the acknowledgment. Examples include: individuals who allowed their clinical experience to be included, a person who provided purely technical help, copyediting, proofreading or translation assistance (NO ghostwriters allowed), or a department Chair who provided only general support.
TE D
Groups of persons who have contributed materially to the paper, but whose contributions do not justify authorship may be listed under a heading such as “clinical investigators” or “participating investigators,” and their function or contribution should be described; for example, “served as scientific advisors,” “critically reviewed the study proposal.”] If you have any question about this, contact that editorial office before submitting your manuscript at [email protected] or 312-202-5316. *
*
*
EP
Please type each author’s LAST NAME ONLY next to the appropriate category.
AC C
Study conception and design: WALSH, GILANI, BASS
Acquisition of data: GILANI, TOBBIA, DOWNES, CAFFREY, KHARYTANIUK
Analysis and interpretation of data: BASS, GILANI, WALSH
Drafting of manuscript: BASS, GILANI, WALSH
Critical revision: BASS, GILANI, TOBBIA, DOWNES, CAFFREY, KHARYTANIUK, WALSH
S1072-7515(16)31692-1
DOI:
10.1016/j.jamcollsurg.2016.12.003
Reference:
ACS 8554
To appear in:
Journal of the American College of Surgeons
Received Date: 3 November 2016 Revised Date:
3 December 2016
Accepted Date: 5 December 2016
Please cite this article as: Shah Gilani SN, Bass GA, Kharytaniuk N, Downes MR, Caffrey EF, Tobbia I, Walsh TN, Gastroesophageal Mucosal Injury after Cholecystectomy: An Indication for Surveillance?, Journal of the American College of Surgeons (2017), doi: 10.1016/j.jamcollsurg.2016.12.003. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Gastroesophageal Mucosal Injury after Cholecystectomy: An Indication for Surveillance?
Syeda Nadia Shah Gilani1,3, MD, MRCS, Gary Alan Bass1,3, MD, Natallia Kharytaniuk1, MB
RI PT
BCh BAO, DOHNS, Michelle R Downes2, MD, MRCPath, Emer Frances Caffrey2, MD, MRCPath, Iqbal Tobbia2, MD, FRCPath, Thomas Noel Walsh1,3, MD MCh FRCSI
M AN U
and Royal College of Surgeons in Ireland3, Dublin, Ireland.
SC
Departments of Surgery1 and Pathology2, Connolly Hospital, Blanchardstown, Dublin, Ireland
Disclosure Information: Nothing to disclose. Support: The Lollipop Day Fund
Abstract presented at the American College of Surgeons 101st Annual Clinical Congress,
TE D
Chicago, IL, October 2015.
EP
Correspondence address: Professor TN Walsh, Dept of Surgery, Connolly Hospital, Blanchardstown, Dublin 15, IRELAND E: [email protected] T: +353 1 646 5632353; +353 1 8202284
AC C
Running head: Cholecystetomy-Mediated Gastric Injury
1
ACCEPTED MANUSCRIPT
ABSTRACT Background: Cholecystectomy alters bile release dynamics from pulsatile meal-stimulated to continuous, and results in retrograde duodeno-gastric bile reflux(DGR). Bile is implicated in
RI PT
mucosal injury following gastric surgery, but whether cholecystectomy causes esophago-gastric mucosal inflammation, thus increasing the risk of metaplasia, is unclear.
Study Design: This study examined whether cholecystectomy-induced DGR promotes chronic
SC
inflammatory mucosal changes of the stomach and/or at the esophago-gastric junction(EGJ). Four groups of patients were studied and compared with controls. A group of patients was
M AN U
studied before and one year after cholecystectomy; two further groups were studied 5-10 years and 10-20 years post-cholecystectomy(LTPC). All underwent abdominal ultrasound and upper gastrointestinal endoscopy(EGD) with gastric antral and EGJ biopsies, noting the presence of gastric bile-pooling. Biopsy specimens were stained for Ki67 and p53 over-expression and the
TE D
bile reflux index(BRI) was calculated.
Results: At endoscopy, bile-pooling was observed in 9/26(34.6%) controls, in 8/25(32%) patients pre-cholecystectomy, in 15/25(60%) one year post-cholecystectomy(p=0.047) and
EP
23/29(79.3%) LTPC patients(p=0.001). BRI positivity at the EGJ increased from 19% of controls through 41% of LTPC patients(p=0032). Ki67 was over-expressed at EGJ in 19% of controls but
AC C
in 62% of LTPC patients(p=0.044); p53 was over-expressed at the EGJ in 19% controls compared with 66% of LTPC patients(p=0.001). Conclusions: DGR was more common in patients with gallstones than controls and its incidence doubled after cholecystectomy. This was associated with inflammatory changes in the gastric antrum and the EGJ, evident in most LTPC patients. Ki67 and p53 over-expression at the EGJ suggest cellular damage attributable to chronic bile exposure post-cholecystectomy, increasing
2
ACCEPTED MANUSCRIPT
the likelihood of dysplasia. Further studies are required to determine whether DGR-mediated esophageal mucosal injury is reversible or avoidable and if surveillance endoscopy is indicated
AC C
EP
TE D
M AN U
SC
RI PT
after cholecystectomy.
3
ACCEPTED MANUSCRIPT
INTRODUCTION Cholecystectomy – the standard of care for symptomatic cholelithiasis – alters the dynamics of bile storage and release.1 Bile is normally stored in the gallbladder in the fasting inter-digestive
RI PT
period and is propelled into the duodenum in response to meals under the influence of
cholecystokinin (CCK) mediated gallbladder contraction.1-3 Following cholecystectomy, the facility for bile storage is lost, and bile is continuously released into the duodenum, even during
SC
fasting1.
There appears to be a clear causal link between cholecystectomy and duodeno-gastric bile
M AN U
reflux.2-4 The continuous presence of bile in the duodenum permits its overflow across the pylorus and into the stomach. A significantly greater concentration of bile acids has been confirmed in naso-gastric aspirates from patients with gallstones compared to those without, and this increases further following cholecystectomy.1
TE D
Cholecystectomy also results in elevated serum levels of cholecystokinin (CCK).3 CCK is an enteric hormone that is normally inhibited by the negative feedback of the CCK-mediated bile bolus in the duodenum5, but following cholecystectomy this switch off mechanism is lost,
EP
resulting in persistent elevation of CCK levels3. Exposure of the gastric mucosa to bile has been shown to cause erythema, inflammation or ulceration of the gastric mucosa6,7. Gastric ulceration
AC C
and gastritis have been shown to attenuate CCK-mediated increases in pyloric muscle tone8, which may explain the observation of a bilious refluxate through an open pylorus on gastroscopy of patients with gastritis9. The increase in circulating CCK results in reduction of lower oesophageal sphincter pressure10 and increases the frequency of transient lower esophageal sphincter relaxation episodes (TLESREs), potentially further exposing the lower oesophageal
4
ACCEPTED MANUSCRIPT
mucosa to refluxed bile11. Increased duodeno-gastric reflux in turn predisposes to increased gastro-oesophageal reflux injury. While we know that these injuries occur, their precise molecular mechanism at the cellular level
RI PT
has yet to be elucidated. We hypothesised that cholecystectomy should be considered a model of iatrogenically-induced mucosal injury of the upper gastro-intestinal tract, resulting in bileinduced chronic cellular injury to the mucosa of the intact stomach and esophago-gastric
SC
junction. This should be quantifiable as an increase in bile reflux index (BRI) positivity and as an
esophageal junction.
PATIENTS AND METHODS Patient Cohorts/Study Groups
M AN U
increase in cellular Ki67 and p53 expression of the mucosa of the gastric antrum and gastro-
TE D
In order to determine whether mucosal injury is truly attributable to cholecystectomy, a paired cohort of patients was studied before and after cholecystectomy, and compared with longstanding post-cholecystectomy patients and further with controls. Eighty patients, divided into
EP
five study groups, were prospectively-enrolled into this observational case-control study, in accordance with institutional ethics committee approval.
AC C
Group 1 (n=26) were individuals investigated for dyspepsia, but without evidence of gallbladder disease and who were not on non-steroidal anti-inflammatory agents (NSAIDs) or proton pump inhibitors (PPIs); these acted as the non-gallstone control group of patients. Group 2 (n=25) were patients with an ultrasound-confirmed diagnosis of symptomatic cholelithiasis, and were scheduled for cholecystectomy. Because of their symptoms of dyspepsia they also underwent upper GI endoscopy.
5
ACCEPTED MANUSCRIPT
Group 3 (n=25) were the Group 2 patient-cohort described above, who underwent repeat upper GI endoscopy one year post-cholecystectomy. Groups 4 and 5 were 29 patients who presented with dyspepsia but who had previously
RI PT
undergone cholecystectomy: namely, Group 4, (n=10) had undergone cholecystectomy 5 to 10 years previously, and Group 5, (n=19) had had their cholecystectomy between 10 to 20 years previously. While recruitment of an additional cohort of patients 5-10 years post-
SC
cholecystectomy who were asymptomatic (and thus NOT complaining of dyspepsia) could address a potential confounding variable, strengthening our hypothesis-testing and the
M AN U
conclusions which may be drawn from the observed results, prospectively enrolling an adequately-powered cohort of well asymptomatic patients for an invasive procedure such as gastroscopy would have posed both ethical and logistic challenges.
TE D
EGD And Collection Of Biopsy Samples
Upper gastrointestinal endoscopy was performed on all patients by a single endoscopist (TNW), who recorded common macroscopic findings using standardized consensus criteria for
EP
inflammation. The presence or absence of macroscopic bile-pooling was noted upon the endoscopic intubation of the stomach in each patient and photographed for subsequent
AC C
verification. A minimum of four biopsies per site were taken from the gastric antrum and the EGJ. The biopsy specimens were oriented on filter paper and immediately fixed in formalin. Bile Reflux Index (BRI)
Histological features of bile reflux in the gastric antrum and EGJ were recorded using the bile reflux index (BRI) devised by Sobala et al.12 This was determined by histological examination of 4 x 1mm specimen sections, following haematoxylin and eosin (H&E) staining, by an expert
6
ACCEPTED MANUSCRIPT
histopathologist, blinded to the clinical details of the patients and to the endoscopic findings. In this system, the index is derived based on the presence and/or severity of certain histological parameters, such as: edema of the lamina propria (E), intestinal metaplasia (IM), chronic
RI PT
inflammation (CI), and Helicobacter pylori colonization in the stomach. A BRI greater than 14 indicated duodeno-gastric reflux in the Sobala study, with 70% sensitivity and 85% specificity.12 Immuno-staining For Ki67 and p53
SC
Immuno-histochemical (IHC) staining of 4 x 1mm tissue slices was undertaken to demonstrate the presence and cellular location of MIB-1/Ki67 and p53 proteins in endoscopic biopsies taken
M AN U
at the gastric antrum and EGJ (IHC Leica BOND-MAX™, BOND™ reagents & antibody kits; Leica Biosystems, Melbourne, Australia). IHC reactivity of Ki67 and p53 was determined as the nuclear staining in the expected proliferative compartment. All slides were evaluated in a blinded fashion for nuclear Ki67 and p53 staining.
TE D
For p53, <15% positive nuclei was regarded as normal expression (grade 0), 15-40% as moderate over-expression (grade 1) and >40% as strong over-expression (grade 2). For Ki67, the percentage of positive nuclei was determined and a percentage of positive cells
EP
<20% was regarded as normal expression (grade 0), 20-50% as moderate over-expression (grade 1), and >50% as strong over-expression (grade 2).
AC C
Statistical Analysis
Data were analyzed using SPSS version 21. P-values were all two-tailed and the alpha-level of significance was set at 0.05. The prevalence was shown as a percentage. The chi-squared paired test was used for testing comparison in paired group, with Bonferroni correction utilised for multiple comparisons. ANOVA was used to test multivariate analysis between and within groups. Post-hoc correlations were calculated with Dunett’s T3 test.
7
ACCEPTED MANUSCRIPT
RESULTS Patient Demographics The mean age (±SEM) for all patients was 57.3(±1.5) years; the mean age of control subjects
RI PT
was 58(±1.7) years compared with 54.2(±2.6) years in those undergoing cholecystectomy and 59.2 (± 3.1) years in patients with an antedecent history of cholecystectomy with no significant difference between the cohorts (p=0.626). The mean (±SEM) body mass index (BMI) across all
SC
groups was 27.8 (±0.54). The mean BMI of patients was 26.8(±4.3)in the control group versus 29.1(±4.5) in those undergoing cholecystectomy or with a history of cholecystectomy at
M AN U
28.3(±3.1). There was no significant difference in BMI between groups (p = 0.706). A history of smoking was present in 11/25 (44%) of control patients versus 13/26 (50%) of patients undergoing cholecystectomy (p=0.668). While our study was not powered to explore Barrett’s (with an estimated population prevalence of up to 2%) as a co-variate with DGER in progressive
TE D
gastric and esophageal mucosal injury, the incidence of Barrett’s on endoscopy across our studied cohorts was 3/80(3.75%) patients, including 1 patient in each of the control, paired (preand post- cholecystectomy) and historic cholecystectomy groups. No malignancies were detected
EP
in any of these patients.
Bile-pooling visible on endoscopy
AC C
On endoscopy note was made of the presence or absence of bile-pooling under gravity on the mucosa of the greater curve of the stomach. Bile-pooling was recorded in 9/26 (34.6%) individuals in the control group (Group 1); a value similar to the cohort of patients awaiting cholecystectomy 8/25 (32%) (p=0.843) (Group 2) (Table 1). When endoscopy was repeated at one year post-cholecystectomy in the same patient cohort, however, 15/25 (60%) now exhibited
8
ACCEPTED MANUSCRIPT
gastric bile-pooling, a significant increase directly consequent on cholecystectomy(Group 3) (p=0.047). Endoscopy performed on a cohort of patients 5-10 years post-cholecystectomy found bile-
RI PT
pooling in 7/10(70%). Endoscopy performed on a final cohort of patients that were 10-20 years post-cholecystectomy found bile-pooling in 16/19 (84.2%) patients. When these the latter two
cholecystectomy (p=0.001 versus controls). Bile Reflux Index (BRI)
SC
groups were pooled there was bile-pooling in 23/29 (79.3%) patients that were long-term post-
M AN U
Biopsies of the gastric antrum were found to have a positive BRI in 3/36 (11%) of the controls, 5/25 (20%) pre-operative patients, 9/25 (36%) of the same patients at one year postcholecystectomy and in 20/29 (69%) of the patients with a long history of cholecystectomy (p=0.001) (Figure 1,2; Table 1). There was no significant difference between the controls and
TE D
pre-cholecystectomy patients (p=0.406); nor between the same patients before and one-year postcholecystectomy (p = 0.208), but there was a marked difference between these groups and patients that had undergone cholecystectomy at least five years previously (p<0.001).
EP
When endoscopic biopsies at the EGJ were compared, BRI was positive in 5/26 (19%) controls, 3/25 (12%) pre-cholecystectomy patients, and in 5/25 (20%) patients at one year post-
AC C
cholecystectomy. It rose to 12/29 (41%) patients with a long history of cholecystectomy which was significantly higher than controls (p=0.032). Ki67
At the gastric antrum, the cellular proliferation marker Ki67 was over-expressed in 6/26 (23%) controls, compared with 17/29 patients (59%) with long-standing history of cholecystectomy (p=0.001) (Figure 3).
9
ACCEPTED MANUSCRIPT
Ki67 was over expressed at the EGJ in 5/26 (19%) of controls and in 18/29 patients (62%) patients with a long-standing history of cholecystectomy (p=0.001) (Figure 3). p53
RI PT
In the gastric antrum the tumor-suppressor gene p53 was over-expressed in 1/26 (4%) control subjects, compared with 11/25(44%) pre-operative cholecystectomy patients, 9/25(40%) one year post-cholecystectomy patients and 15/29(52%) patients with a long-standing history of
SC
cholecystectomy (Table 1). Expression of p53 was significantly higher in patients who had undergone cholecystectomy (either recent or remote), compared with controls (p=0.001).
M AN U
p53 was over-expressed at the EGJ in 5/26 (19%) controls, in 12/25 (48%) recently-operated patients and 19/29 (66%) long-standing post-cholecystectomy patients, and this was significantly higher than controls (p=0.001) (Figure 2).
TE D
DISCUSSION
In this study, cholecystectomy resulted in an increase in duodeno-gastric reflux injury that was progressive with time. The same 25 symptomatic patients underwent endoscopic evaluation
EP
before and after their cholecystectomy and were found to have excess macroscopic (visible bile pooling), microscopic (histochemical evidence of bile-mediated cellular injury) and epigenetic
AC C
(p53 and Ki-67 upregulation) changes. Whether it is prolonged mucosal contact with the bile per se or with some hitherto-uncovered substrate carried in the bile will be the subject of future work. Nevertheless, following cholecystectomy, these otherwise-identical patients were found in our paired comparison to have an increased prevalence of bile-exposed mucosa. Further elevation of inflammatory markers in patients with increased duration of bile exposure suggests that this injury is progressive. The increased incidence of bile reflux injury in pre-operative
10
ACCEPTED MANUSCRIPT
patients with symptomatic gallstone disease with respect to controls is likely to represent a proportion of patients that may have lost part or all of their bile reservoir capacity to gallstones, thus having become “functionally-cholecystectomised” at various time periods prior to this study
RI PT
and in whom little further duodeno-gastric reflux would be expected to occur following
gallbladder removal. Studies with 99Tc-HIDA radionucleotide imaging lend further weight to this hypothesis.13
SC
Duodeno-gastric reflux is identifiable as a dependent bile-pool during EGD14 and its incidence increases significantly following cholecystectomy.13,14 It is associated with: (1) an alkaline shift
M AN U
in gastric content15, (2) the development of chronic superficial gastritis6,7, and (3) the change in parietal cell activity13. Cell proliferation is thought to be one of the earliest steps in the development of pre-neoplastic change in response to inflammation. Chronic inflammation, through repetitive cycles of epithelial regeneration and aberrant growth, leads to over-expression
TE D
of the intracellular protein Ki67.16 We observed an increase in the expression of the molecular marker Ki67 in the gastric antral mucosa and at the EGJ on repeat endoscopic biopsy in our patients within one year of cholecystectomy which was associated with the increase in bile-
EP
pooling and increased BRI. This was observed to be significantly higher in patients studied at five and subsequent years after cholecystectomy. This up-regulated expression of Ki67 may
AC C
reflect chronic mucosal inflammation and repetitive cycles of epithelial regeneration17. Immunohistochemical staining for MIB-1, the Ki67 proliferation antigen, has been shown to increase expression in the Barrett’s oesophagus-dysplasia-adenocarcinoma sequence.16-18 These changes were mirrored by up-regulation of p53 at the antrum and the EGJ in our patient cohorts. Changes in p53 expression are thought to result from post-translational modifications in response to cellular stress or DNA damage.19 A number of studies have demonstrated positive
11
ACCEPTED MANUSCRIPT
immune-histochemical staining for p53 in dysplastic Barrett’s metaplasia and in gastric and esophageal adenocarcinoma20-23 while further work has shown over-expression of Ki67 in association with Barrett’s mucosa and esophageal adenocarcinoma24. Several studies have
RI PT
suggested that bile can directly damage DNA, with the bile constituent lithocholic acid
specifically identified as a promoter in experimental carcinogenesis through cell transformation and DNA-strand breaks.25,26 Exposure of the gastric and esophageal epithelial cells to abnormal
SC
levels of bile acids may lead to the generation of reactive oxygen species (ROS) and reactive nitrogen species (RNS) through multiple pathways involving disruptions of the cell membrane
M AN U
and mitochondria.27 This, in turn, may result in increased DNA damage and thus a tendency towards increased mutagenesis.27 Thus the molecular/epigenetic changes and early carcinogenesis, as shown by the increased expression of p53 and Ki67 that occur as a result of cholecystectomy, are similar to those identified as precursors of Barrett’s oesophagus.17,18
TE D
Patients with Barrett’s esophagus and esophageal adenocarcinoma are known to have increased duodeno-gastric bile reflux28,29, the mechanism of which is unclear. The role of duodeno-gastric reflux in the pathogenesis of Barrett’s oesophagus is, however, well established.30,31 And
adenocarcinoma.32
EP
Barrett’s oesophagus is a known pre-malignant precursor lesion for esophageal
AC C
It has been reported that gallstone disease is more prevalent in Barrett’s oesophagus.33 We have previously shown that gallbladder function is impaired in patients with Barrett’s esophagus and esophageal adenocarcinoma.34 This impaired motility and dysfunction of the gallbladder may predispose to increased duodeno-gastric reflux consequent on impaired bile storage. Impaired motility may in turn, be a contributing factor for the development of gallstones.35 Decreased motility and formation of stones all predispose to duodeno-gastric bile reflux as the function of
12
ACCEPTED MANUSCRIPT
the gallbladder becomes progressively impaired or completely non-functional.36 Cholecystectomy will worsen duodeno-gastric bile reflux in patients with a functioning gallbladder, due to the immediate loss of a bile reservoir.37
RI PT
Cholecystectomy is one of the most commonly performed major surgical procedures in the
western world with approximately 917,000 performed each year in the USA and over 50,000 performed in the England.38 It is concerning that cholecystectomy mediated mucosal injury
SC
appears to mimic the situation that pertains following partial gastrectomy for peptic ulcer
disease, when reconstruction with Bilroth I or Billroth II exposes the stomach to increased bile
M AN U
and pancreatic reflux39 resulting in an increased incidence in gastritis and ultimately a significantly greater incidence of cancer in the gastric remnant.40,41 This is avoided by the Rouxen-Y technique where bile is diverted away from the stomach39 and probably explains the lower incidence in gastric cancer following Roux-en-Y reconstruction for peptic ulcer surgery39,42. Data
TE D
linking cholecystectomy with chronic gastric and oesophageal inflammation are persuasive, which are also suggestive of an increased risk of inflammation-induced metaplasia and neoplasia.43,44 The impact of bile on gastrointestinal mucosa, described by Dixon and colleagues
EP
as the BRI, appears to be a more robust measure of injury from direct quantitative measurement of bilirubin in gastric juice. The latter has been shown by us45 and by others46-48 to be inaccurate,
AC C
often under-estimating bilirubin concentration. In conclusion, we believe that our study provide novel insights into post-cholecystectomy chronic inflammatory change in the stomach and lower oesophagus. While correlation may not necessarily imply causation, the trends towards excess bile pooling and the associated microscopic and epigenetic changes are suggestive of progressive injury as the natural history of prolonged gastric mucosal exposure to bile. Cytological changes attributable to DGR occur early
13
ACCEPTED MANUSCRIPT
in the gastric and EGJ mucosa of patients following cholecystectomy and these appear to be progressive. Ki67 and p53 cellular over-expression in response to DGR suggests these changes may be pre-cursors of chronic inflammation raising concern its long-term consequences. If
RI PT
cholecystectomy places patients at greater risk for inflammation/dysplasia/neoplasia, there may be a benefit from endoscopic surveillance to begin after 20 years and repeated at perhaps 5yearly intervals, as with Barrett’s esophagus.49 Consideration should also be given to the role of
SC
systemic anti-inflammatory agents in protecting the gastro-intestinal mucosa. Further work is required to ascertain if this mucosal injury is reversible or avoidable through medication or
AC C
EP
TE D
M AN U
lifestyle changes.
14
ACCEPTED MANUSCRIPT
REFERENCES [1] Svensson J, Gelin J, Svanvik J. Gallstones, cholecystectomy, and duodenogastric reflux of bile acid. Scand J Gastroenterol 1986;21:181-187.
RI PT
[2] Cabrol J, Navarro X, Simo-Deu J, Segura R. Evaluation of duodenogastric reflux in gallstone disease before and after simple cholecystectomy. Am J Surg 1990;160:283-286.
cholecystokinin. Am J Gastroenterol 2002;97:2189-2192.
SC
[3] McDonnell C, Bailey I, Stumpf T, et al. The effect of cholecystectomy on plasma
[4] Lujan-Mompean J, Robles-Campos R, Parrilla-Paricio P, et al. Duodenogastric reflux in
1993;176:116-118.
M AN U
patients with biliary lithiasis before and after cholecystectomy. Surg Gynecol Obstet
[5] Gomez G, Lluis F, Guo YS, et al. Bile inhibits release of cholecystokinin and neurotensin. Surgery 1986;100:363–368.
TE D
[6] Rhodes J, Barnard DE, Philips SF, et al. Increased reflux of bile into the stomach in patients with gastric ulcer. Gastroenterology 1969;57:241-252. [7] Harmon J, Doong T, Gadacz TR. Bile acids are not equally damaging to the gastric mucosa.
EP
Surgery 1978;84:79-86.
[8] Si XM, Huang L, Lv P, et al. Effects of cholecystokinin-8 induced gastric dysmotility on bile
AC C
regurgitation during stress and molecular mechanisms. Regul Pept 2006;136:64-71. [9] Rovelstad RA. The incompetent pyloric sphincter. Bile and mucosal ulceration. Am J Dig Dis 1976;21:165-173.
[10] Fisher RS, Dimarino AJ, Cohen S. Mechanism of cholecystokinin inhibition of lower esophageal sphincter pressure. Am J Physiol 1975;228:1469–1473.
15
ACCEPTED MANUSCRIPT
[11] Boulant J, Fioramonti J, Dapoigny M, et al. Cholecystokinnin and nitric oxide in transient lower oesophageal sphincter relaxation to gastric distention in dogs. Gastroenterology 1994;107:1059-1066.
RI PT
[12] Sobala G, O'Connor HJ, Dewar EP, et al. Bile reflux and intestinal metaplasia in gastric mucosa. J Clin Pathol 1993;46:235-240.
[13] Lorusso D MG, Mangini V, Messa C, et al. Duodenogastric reflux of bile acids, gastrin and
SC
parietal cells, and gastric acid secretion before and 6 months after cholecystectomy. Am J Surg 1990;159:575-578.
M AN U
[14] Kunsch S, Neesse A, Huth J, et al. Increased Duodeno-Gastro-Esophageal Reflux (DGER) in symptomatic GERD patients with a history of cholecystectomy. Z Gastroenterol 2009;47:744748.
[15] Brown T, Walton G, Cheadle WG, Larson G M. The alkaline shift in gastric pH after
TE D
cholecystectomy. Am J Surg 1989;157:58-65.
[16] Going JJ, Keith WN, Neilson L, et al. Aberrant expression of minichromosome maintenance proteins 2 and 5, and Ki-67 in dysplastic squamous oesophageal epithelium and Barrett's
EP
mucosa. Gut 2002;50:373-377.
[17] Whittles CE, Biddlestone LR, Burton A, et al. Apoptotic and proliferative activity in the
AC C
neoplastic progression of Barrett's oesophagus: a comparative study. J Pathol 1999;187:535-540. [18] Hong M, Laskin WB, Herman BE, et al. Expansion of the Ki-67 proliferative compartment correlates with degree of dysplasia in Barrett's esophagus. Cancer 1995;75:423-429. [19] Younes M, Lechago J, Chakraborty S, et al. Relationship between dysplasia, p53 protein accumulation, DNA ploidy, and Glut-l overexpression in Barretts metaplasia. Scand J Gastroenterol 2000;35:131-137.
16
ACCEPTED MANUSCRIPT
[20] Binato M, Gurski RR, Fagundes RB, et al. P53 and Ki-67 overexpression in gastroesophageal reflux disease - Barrett's esophagus and adenocarcinoma sequence. Dis Esophagus 2009;22:588-595.
RI PT
[21] Chen M, Huang J, Zhu Z, et al. Systematic review and meta-analysis of tumor biomarkers in predicting prognosis in esophageal cancer. BMC Cancer 2014;14:892.
[22] Gregson EM, Bornschein J, Fitzgerald RC. Genetic progression of Barrett’s oesophagus to
SC
oesophageal adenocarcinoma. Br J Cancer 2016;115:403-410.
[23] Clemons NJ, Phillips WA, Lord RV. Signaling pathways in the molecular pathogenesis of
M AN U
adenocarcinomas of the esophagus and gastroesophageal junction. Cancer Biol The 2013;14:782-795.
[24] Kerkhof M, Steyerberg EW, Kusters JG, et al. Aneuploidy and high expression of p53 and Ki67 is associated with neoplastic progression in Barrett esophagus. Cancer Biomark 2008;4:1-
TE D
10.
[25] Hiasa Y, Konishi Y, Kamamoto Y, et al. Effect of lithocolic acid on DL-ethionine carcinogenesis in rat liver. Gann 1971;62:239-245.
EP
[26] Narisawa T, Magadia NE, Weisburger JH, Wynder EL. Promoting effect of bile acids on colon carcinogenesis after intrarectal instillation of N-methyl-N'-nitro-N-nitrosoguanidine in rats.
AC C
J Natl Cancer Inst 1974;53:1093-1097. [27] Dvorak K, Payne CM, Chavarria M, et al. Bile acids in combination with low pH induce oxidative stress and oxidative DNA damage: relevance to the pathogenesis of Barrett's oesophagus. Gut 2007;56:763-771. [28] Bernstein H, Bernstein C, Payne C M, et al. Bile acids as carcinogens in human gastrointestinal cancers. Mutat Res 2005;589:47-65.
17
ACCEPTED MANUSCRIPT
[29] Menges M, Muller M, Zeitz M. Increased acid and bile reflux in Barrett's esophagus compared to reflux esophagitis, and effect of proton pump inhibitor therapy. Am J Gastroenterol 2001;96:331-337.
RI PT
[30] Richter J. Importance of bile reflux in Barrett's esophagus. Dig Dis 2000;18:208-216. [31] Koek G, Sifrim D, Lerut T, et al. Multivariate analysis of the association of acid and
duodeno-gastro-oesophageal reflux exposure with the presence of oesophagitis, the severity of
SC
oesophagitis and Barrett's oesophagus. Gut 2008;57:1056-1064.
[32] Voutilainen M, Farkkila M, Juhola M, et al. Specialized columnar epithelium of the
M AN U
esophagogastric junction: prevalence and associations. The Central Finland Endoscopy Study Group. Am J Gastroenterol 1999;94:913-918.
[33] Izbeki F, Rosztoczy AI, Yobuta JS, et al. Increased prevalence of gallstone disease and impaired gallbladder motility in patients with Barrett's esophagus. Dig Dis Sci 2008;53:2268-
TE D
2275.
[34] Nassr AO, Gilani SN, Atie M, et al. Does impaired gallbladder function contribute to the development of Barrett's esophagus and esophageal adenocarcinoma? J Gastrointest Surg
EP
2011;15:908-914.
[35] Johnston DE, Kaplan MM. Pathogenesis and treatment of gallstones. N Engl J Med
AC C
1993;328:412-421.
[36] Matsuzaki J, Suzuki H, Asakura K, et al. Gallstones increase the prevalence of Barrett's esophagus. J Gastroenterol 2010;45:171-178. [37] Fein M, Bueter M, Sailer M, Fuchs KH. Effect of cholecystectomy on gastric and esophageal bile reflux in patients with upper gastrointestinal symptoms. Dig Dis Sci 2008;53:1186-1191.
18
ACCEPTED MANUSCRIPT
[38] Wu XD, Tian X, Liu MM, et al. Meta-analysis comparing early versus delayed laparoscopic cholecystectomy for acute cholecystitis. Br J Surg 2015;102:1302-1313. [39] Fukuhara K, Osugi H, Takada N, et al. Reconstructive procedure after distal gastrectomy for
RI PT
gastric cancer that best prevents duodenogastroesophageal reflux. World J Surg 2002;26:14521457.
investigation. Ann Surg 1989;210:159–164.
SC
[40] Toftgaard C. Gastric cancer after peptic ulcer surgery. A historic prospective cohort
[41] Hansson L-E, Nyrén O, Hsing AW, et al. The risk of stomach cancer in patients with gastric
M AN U
or duodenal ulcer disease. N Engl J Med 1996;335:242-249.
[42] Kojima K, Yamada H, Inokuchi M, et al. A comparison of Roux-en-Y and Billroth-I reconstruction after laparoscopy-assisted distal gastrectomy. Ann Surg 2008;247:962-967. [43] Lagergren J, Mattsson F. Cholecystectomy as a risk factor for oesophageal adenocarcinoma.
TE D
Br J Surg 2011;98:1133-1137.
[44] Dixon M, Mapstone NP, Neville PM, et al. Bile reflux gastritis and intestinal metaplasia at the cardia. Gut 2002;51:351-355.
EP
[45] Caldwell MT, Byrne PJ, Brazil N, et al. An ambulatory bile reflux monitoring system: an in vitro appraisal. Physiol Meas 1994;15:57-65.
AC C
[46] Barrett M, Myers JC, Watson DI, Jamieson GG. Detection of bile reflux: in vivo validation of the Bilitec Fibreoptic system. Dis Esophagus 2000;13:44-50. [47] Bechi P, Baldini F, Cianci F. What is the reliability of the portable spectro-photometer with a fibreoptic probe to detect bilirubin in bile refluxate? Is sensitivity of Bilitec likely to be improved? Florence: OESOKnowledge; 1998.
19
ACCEPTED MANUSCRIPT
[48] Tibbling Grahn L, Blackadder L, Franzen T, Kullman E. Gastric bile monitoring: an in vivo and in vitro study of Bilitec reliability. Scand J Gastroenterol 2002;37:1334-1337.
AC C
EP
TE D
M AN U
SC
RI PT
[49] Shaheen NG, Falk GW, Iyer PG, Gerson LB. Am J Gastroenterol 2016;111:30-50.
20
ACCEPTED MANUSCRIPT
Table 1. Endoscopic and Histopathologic Findings Study groups
n
%
n
%
n
Yes
9
34.6
8
32
15
No
17
65.4
17
68
10
Absent
23
88.5
Present
3
11.5
Absent
21
80.8
Present
5
19.2
BRI
p53 expression, antrum 25
Positive (>30%; <50%)
1
Extensive (>50%)
AC C
p53 expression, EGJ
0
%
60
23
79.3
40
6
20.7
76
12
48
9
31
6
24
13
52
20
69
21
84
17
68
17
58.6
4
16
8
32
12
41.4
96.2
14
56
17
68
14
48.3
3.8
5
20
6
24
11
37.9
0
6
24
2
8
4
13.8
EP
Normal (<30%)
n
19
TE D
EGJ
M AN U
Antrum
%
SC
Endoscopic bilepooling
>5 y Postcholecystectomy, n = 29
RI PT
Control, n = 26
1 y PostPre-cholecystectomy cholecystectomy (paired), n = 25 (paired), n = 25
Normal (<30%)
21
80.8
13
52.0
17
68.0
10
34.5
Positive (>30%; <50%)
5
19.2
5
20.0
5
20
15
51.7
Extensive (>50%)
0
0
7
28
3
12
4
13.8
Normal expression
20
76.9
24
96
14
56
12
41.4
Over-expression
6
23.1
1
4
11
44
17
58.6
Ki67 expression, antrum
21
ACCEPTED MANUSCRIPT
Ki67 expression, EGJ Normal expression
21
80.8
22
88
13
52
11
37.9
Over-expression
5
19.2
3
12
12
48
18
62.1
AC C
EP
TE D
M AN U
SC
RI PT
BRI, bile reflux index; EGJ, esophagogastric junction.
22
ACCEPTED MANUSCRIPT
FIGURE LEGENDS Figure 1. Incidence of macroscopic bile-pooling visible when the stomach was intubated with an endoscope. The proportion of patients with bile-pooling did not differ between patients awaiting
RI PT
cholecystectomy and those control patients with an intact gallbladder. However, the incidence of bile-pooling significantly increased on re-endoscopy of those same patients following
cholecystectomy. Green bar, endoscopic bile pooling; blue bar, no endoscopic bile pooling. Pre-
SC
op, preoperative; post-op, postoperative
M AN U
Figure 2. Photomicrograph of haematoxylin and eosin-stained specimen of gastric antral mucosa a patient in whom macroscopic bile reflux was seen on endoscopy; histologic bile reflux index was calculated using such slides.
TE D
Figure 3. (A) High-power light microscopy demonstrating p53 over-expression in an endoscopic biopsy taken at the gastric antrum in a patient who had undergone cholecystectomy. (B) Lowpower light microscopy in the same specimen. Immuno-histochemical brown staining
AC C
EP
demonstrates nuclear proliferation over a background haematoxylin & eosin stain.
23
ACCEPTED MANUSCRIPT
Precis Surgical endoscopists have long-noticed bile pooling upon endoscopic intubation of the stomach, more commonly in patients who have undergone cholecystectomy. There is concern that this bile
RI PT
may have a chronic irritative effect on the underlying gastric mucosa, predisposing it to
AC C
EP
TE D
M AN U
SC
metaplasia and thus examined mucosal specimens for p53 and Ki67 overexpression.
24
AC C
EP
Figure 1
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
Figure 2
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
Figure 3
ACCEPTED MANUSCRIPT
AUTHOR CONTRIBUTIONS FORM Individuals claiming authorship should meet all 3 of the following conditions in accordance with the “Consensus Statement on Surgery Journals Authorship—2005”:
RI PT
1) Authors make substantial contributions to conception and design, and/or acquisition of data, and/or analysis and interpretation of data; 2) Authors participate in drafting the article or revising it critically for important intellectual content; and 3) Authors give final approval of the version to be submitted and any revised version.
SC
Each author should have participated sufficiently in the work to take public responsibility for appropriate portions of the content. Allowing one’s name to appear as an author without having contributed significantly to the study or adding the name of an individual who has not contributed or who has not agreed to the work in its current form is considered a breach of appropriate authorship.
M AN U
Ghost-writing is NOT acceptable. No one, other than the authors listed below, should have contributed substantially to the writing and revising of the manuscript. Contributors who do not meet the criteria for authorship should be listed in the acknowledgment. Examples include: individuals who allowed their clinical experience to be included, a person who provided purely technical help, copyediting, proofreading or translation assistance (NO ghostwriters allowed), or a department Chair who provided only general support.
TE D
Groups of persons who have contributed materially to the paper, but whose contributions do not justify authorship may be listed under a heading such as “clinical investigators” or “participating investigators,” and their function or contribution should be described; for example, “served as scientific advisors,” “critically reviewed the study proposal.”] If you have any question about this, contact that editorial office before submitting your manuscript at [email protected] or 312-202-5316. *
*
*
EP
Please type each author’s LAST NAME ONLY next to the appropriate category.
AC C
Study conception and design: WALSH, GILANI, BASS
Acquisition of data: GILANI, TOBBIA, DOWNES, CAFFREY, KHARYTANIUK
Analysis and interpretation of data: BASS, GILANI, WALSH
Drafting of manuscript: BASS, GILANI, WALSH
Critical revision: BASS, GILANI, TOBBIA, DOWNES, CAFFREY, KHARYTANIUK, WALSH