AG

AA

bst

ract

srecruited into the European Prospective Investigation into Cancer-Norfolk (EPIC-Norfolk)with 23 658 completing 7-DFDs at enrolment. The cohort was monitored for 14 years fornew symptomatic gallstones, with diagnoses confirmed by review of the clinical notes. The7-DFDs recorded one week's diet including drink type, brands and volumes consumed. Arepresentative sample of 3 970 and also those with gallstone disease had their 7-DFDs codedby nutritionists using a computer program containing nutrient information on 11 000 foodand drink items. A cohort analysis used Cox regression modelling to estimatehazard ratios(HR) for men and women, comparing 1, 2 or 3+ cups (1 cup=250mls) of caffeinated coffeeand tea intake per day versus zero intake as well as a binary variable of any caffeinatedcoffee vs none and decaffeinated coffee vs none. The analyses were adjusted for the riskfactors: age, body mass index, energy intake, alcohol intake and physical activity in men,and also parity and hormone replacement therapy use in women. Results: 177 women(mean age of diagnosis 66.5 yrs SD=9.5 yrs) and 90 men (64.2 yrs SD=9.2 yrs) developedsymptomatic gallstones. In men, caffeinated coffee intake was associated with a decreasedrisk of symptomatic gallstones (3+ cups day vs zero intake HR=0.48, 95% CI=0.26-0.89)with a significant trend across categories (per 1 cup increase HR=0.79, p=0.022). Drinkingany caffeinated coffee compared to zero consumption in men, the HR was 0.58 (95% CI=0.38-0.90, p=0.015). If all men consumed three or more cups of caffeinated coffee a day,then 26% of symptomatic gallstones could be prevented. No effects were found for tea anddecaffeinated coffee consumption in men and any drink intake in women (any caffeinatedcoffee vs zero consumption HR=0.84 95% CI=0.60-1.17, p=0.30). Conclusion: The datasuggests that caffeinated coffee may reduce the risk of symptomatic gallstones in men butnot in women. Coffee should be measured in future studies investigating the aetiology ofgallstones in men.

Su1278

Functional CT and MR Biliary Imaging in Patients With UnexplainedAbdominal Pain and ControlsMark Topazian, Jeff L. Fidler, John M. Knudsen, Douglas A. Collins, Kiaran P. McGee,Brian Lahr, Johnson Thistle

Background: The pathophysiology of suspected biliary pain in patients without gallstonesis unclear. Gallbladder ejection fraction (GBEF) measured by biliary scintigraphy (HIDA)fails to predict response to cholecystectomy in some patients. We hypothesized that CT andMR imaging during sincalide infusion would identify dynamic anatomical changes of thegallbladder (GB) or bile ducts potentially responsible for otherwise unexplained pain. Aims:To identify potentially abnormal GB or bile duct morphology before and during GB contrac-tion. Methods: 33 patients with unexplained upper abdominal pain scheduled for HIDAwere enrolled; 30 completed the protocol. All subjects had normal GB ultrasound and serumliver tests. After an overnight fast, CT and MR images of the biliary tree were obtained atbaseline and 20, 40, and 60 minutes after initiation of a 45 minute sincalide infusion identicalto that used during HIDA. Intravenous contrast agents excreted preferentially into the biliarytree were administered 1 hour prior to imaging (iodipamide meglumine for CT, gadobenatedimeglumine for MR). CT, MR and HIDA were performed on separate days. Subsequently,30 normal volunteer controls matched to cases by sex and age underwent MR and sincalideinfusion with the same imaging protocol. Results: Gallbladder volumes and GBEF were notstatistically different when measured by CT or MR. Gallbladder contraction (at 20 minutes)and re-filling (at 60 minutes) occurred more rapidly by CT and MR than by HIDA (p<0.001for each); GBEF measured by MR and HIDA was not significantly different at 40 minutes(p=0.23). Two patients (7%) had gallbladder stones detected by CT and/or MR that hadbeen missed by transabdominal ultrasound. Normal GBEF (measured by MR in controlsubjects) was 49% to 95% at 40 minutes, and there was no significant difference in GBEFbetween patients and controls (p=0.74). GB shape was categorized as straight, curved, orfolded; there was a significant difference in GB shape at baseline and during contractionbetween patients and controls (p<.001 for both). Asymmetrical patterns of GB contraction(neck > body, straightening, kinking, focal luminal collapse) were observed in 10 (33%)patients and 4 (14%) controls (p=0.13). During GB contraction, pain similar to the patient'susual pain occurred in 6 of 10 (60%) patients with a folded GB and 8 of 21 (38%) patientswith a straight or curved GB (p=0.44). Common hepatic duct diameter measured by MRdid not change significantly during gallbladder contraction in either patients or controls.Conclusions: Functional CT and MR biliary imaging demonstrate differences in GB morpho-logy between patients with unexplained abdominal pain and controls. CT and MR alsodemonstrate GB stones missed by other modalities, and provide more accurate assessmentof GB volume, contraction, and re-filling than HIDA.

Su1279

Intracorporeal Laser Lithotripsy (Ill) Using a Mini-Cholangiopancreatoscope(Spyglass®): A New Option for Difficult Pancreatic StonesSarah Leblanc, Ariane Vienne, Marianne Gaudric, Jean-Christophe Duchmann, JeanBoyer, Luigi Mangialavori, Stanislas Chaussade, Frederic Prat

Aim: Lithotripsy of pancreatic stones is generally accomplished using extracorporeal shockwave lithotripsy (ESWL) requiring a specific heavy equipment. The aim of this study wasto evaluate Intracorporeal Laser Lithotripsy (ILL) with a new single-operator mini-endoscope(Spyglass®) during ERCP, for its performance, feasibility and safety in treatment of difficultpancreatic lithiasis. Methods: Five patients (4 men, 1 woman, aged 43 to 65 years) underwentendoscopic management of difficult pancreatic stones using Spyglass® system (Boston Scient-ific, USA) and ILL in our center in 2009-2010. ILL was performed with a holmium lasergenerator (Auriga® laser system, StarMmedTec, Germany) and a 1.2mm fibre, with anenergy output setting of 80mJ and a pulse rate of 5-10Hz. Data including symptoms, previoustreatment, technique of ERCP, complications and follow up were reviewed. Patients werefollowed after ILL with outpatient visits and control ERCPs. Results: Pancreatic stones weresecondary to chronic alcoholic pancreatitis (n=4) or a congenital bilio-pancreatic junctionanomaly (n=1). All patients were symptomatic with pain and loss weight. Pancreatic ductobstruction (with visualization of pancreatic stones and upstream pancreatic duct dilatation)were confirmed in all cases by CTscan and EUS. The median diameter of pancreatic stones

S-448AGA Abstracts

was 8 mm (range 3-10). In all cases, a previous ERCP had been performed with a sphinc-terotomy and endoscopic transampullary drainage by endoprothesis. Balloon dilatation wasnecessary in 3 cases. No patient had undergone previous ESWL. The mean time betweenfirst ERCP and the ILL-Spyglass-ERCP was 5.8months (range 3-13). Direct visualization ofpancreatic stones was deemed good in 4/5 patients. Holmium laser lithotripsywas successfullyused in all patients. After stone disintegration, endoscopic extraction of stones and fragmentswas performed using a Dormia basket or balloon catheter. Main ductal clearance was achievedin 5 patients after 1 procedure. In 4 patients, a new stent was placed to calibrate a residualstenosis, without recurrence of pancreatic stone with a median follow-up of 7 months (range3-11). No recurrence of pain was recorded during study follow-up. No procedure-relatedcomplication was observed. Conclusion : Spyglass® is a new miniature endoscope withapplications in the management of bilio-pancreatic diseases. Using this sytem, Laser Litho-tripsy can be applied under direct visual control for pancreatic stones. This procedure seemsto be an effective and safe new option for difficult-to-extract pancreatic stones.

Su1280

Persistent Gallbladder Sludge: Definitive Results of a Long Term Follow upAfter Successful Oral Bile Acid DissolutionCarlos Guma, Marcelo Thome, Luis A. Viola, Romina Moretti

Background and Aim: The gallbladder biliary sludge has been observed in a wide range ofclinical disorders (prolonged fast, pregnancy, drugs, enteral nutrition). However, it has beenidentified as no-innocent, pathological and persistent where the evolution to pancreatitis,cholecystitis or stone progression may be the natural course. The outcome of the pathologicalgallbladder sludge (GS) after oral bile acid dissolution is unknown. In this prospectiveexperience a population with complete biliary sludge clearance was examined clinically andthrough ultrasonography (US) during a protocol follow-up to know: gallbladder sludgerecurrence (GSR), and development of cholelithiasis. Materials and Methods: In a prospectivestudy from December 1991 to December 2008 in 221 patients (pts) the diagnosis of GSwithout stones was made according to the international US criterion. Three months (mo)later a second US control was performed in 179 pts (81%) the GS disappeared, while in42 pts (19%) the persistence of the GS was detected. These last were defined as carriers of“persistent gallbladder sludge” (PGS). One hundred percent (42/42) presented biliary pain,16.6 % (7/42) acalculous cholecystitis and 7% (3/42) recurrent acute pancreatitis. Immedi-ately an oral bile acid therapy was started 12 mglkgld of ursodeoxycholic acid (UDCA) fora six-mo period with monthly clinical examination and US every 2 mo. Between 2 and 6mo of treatment (mean 3.8 mo) in 100 % (42/42) of the pts the GS dissolution was achieved.These 42 were the “study group” and a follow up with clinical and US examination every6 mo was performed with the administration of only a written low fat diet. The results wereevaluated for: GSR and eventual development of cholelithiasis. A minimum follow up of 12mo was necessary for the evaluation in “results”. Results: Forty two pts presented PGS (22female, 25 -72 years, mean 49), in 3 of them the follow-up was lost, then 39 pts withoutGS were observed for a mean period of 6 years (range 3-17). Seventy four percent of these(29/39) were asymptomatic and with US normal for 6 years (range 3-17).While the remainder26% (10/39) presented complications: development of microcholelithiasis in 4 cases between3 and 5 years of evolution (mean 3.5 years), they were treated with laparoscopic cholecystec-tomy. The other 6 cases developed GSR between 2. and 7 years (mean 5 years) and weresuccessfully retreated with UDCA. Conclusions: The 74% of the pts with pathologicalgallbladder sludge presented a good outcome with a low fat diet, after a UDCA therapy.The 26 % remainder presented symptomatic biliary complications. News investigations arenecessary to confirm this data.

Su1281

A Pharmacokinetic-Pharmacodynamic (PK-PD)-Based Treatment Strategy WithCeftriaxone (CTRX) for Acute Cholangitis That Was Proven to Be Effective inActual Clinical PracticeMasao Toki, Yasuharu Yamaguchi, Kenji Nakamura, Shin'ichi Takahashi

[Aim] Themost important factors determining the prognosis of patients with acute cholangitisare prompt biliary drainage and a proper choice of antibiotics. In this study, using ceftriaxone(CTRX), which has good penetrability into bile and is recommended as the drug of firstchoice for severe cases of cholangitis in the “Guidelines for the management of acutecholangitis and cholecystitis in Japan,” we tested the differences between the theoreticalclinical effect as estimated by PK-PD analysis, and the actual measured blood and biliarylevels and the effect as judged in actual clinical practice. Because once- and twice-dailyadministration of CTRX has been approved in Japan, the above results were also stratifiedby the number of daily doses. [Methods] From Oct 08 to Dec 09 of the 24 cases that fulfilledthe diagnostic criteria for acute cholangitis and were classified as having severe or moderatedisease that could progress to severe disease in the severity classification in the above-mentioned guidelines, the 23 cases that fulfilled the following conditions were enrolled inthis study: (1) endoscopic nasobiliary drainage (ENBD) possible before antibiotic administra-tion; (2) no history of previous endoscopic sphincterotomy; (3) bile collection is possible2 and 24 hours after ENBD. Twenty-four subjects were randomly assigned to 2 groups(group A: once-daily administration of 2g CTRX; group B: twice-daily administration of 1gCTRX), and the serum and biliary levels of CTRX were measured 2 and 24 hours after thefirst dose of CTRX. In addition, the MICs of CTRX for the isolated bacterial strains fromthe 24 cases were measured. The theoretical clinical effect of CTRX as estimated by the PK-PD parameters and the effect in actual clinical practice (judged by the body temperature,WBC count and serum CRP) were investigated by the number of daily administrations,based on the serum CTRX level and the PK-PD breakpoint when CTRX was administeredat 2g/day. [Results] One case out of the 24 was excluded because of self-removal of theENBD tube, and the remaining 23 cases were included in the analysis. The minimum serumand biliary levels of the drug in groups A and B at 24 hours after the first dose were 9.1g/ml and 2.9g/ml, 9.2g/ml and 2.5g/ml, respectively. The MICs of CTRX for all the isolatedbacteria were plotted below the minimum serum and biliary levels at 24 hours after thefirst dose, except one belonging to the genus Enterococcus, for which CTRX is not applicable(see figure). The efficacy rate in actual clinical practice was 100% in both groups A and B.

The MICs for the isolated bacterial strains were below 16g/ml, which is the PK-PD breakpointof CTRX at 2g/day. [Conclusion] The results suggested that CTRX can be recommended asthe drug of first choice for moderate to severe cases of acute cholangitis.

Su1282

Gastrointestinal and Psychological Symptoms Improve Persistently AfterLaparoscopic CholecystectomyKjell-Arne Ung, Magnus Simren, Marie-Louise Gustavsson, Asa Nilsson, Antal Bajor, EvaJakobsson

Introduction: Atypical abdominal symptoms may incorrectly be attributed to gall bladderstones discovered by chance at ultrasound. High rate of post-cholecystectomy symptomshas been reported and a group of patients may experience negative outcome after the surgery.Patient related psychological factors may also influence the decision for surgery. Aim: Toassess short- and long-term effects of laparoscopic cholecystectomy on the patients gastrointe-stinal symptoms and psychological well-being. Methods: The patients were investigatedbefore, 3 and 12 months after elective cholecystectomy. They completed two validated self-administered questionnaires: the Gastrointestinal Symptom Rating Scale (GSRS) with aquestion on eating dysfunction added and Psychological General Well Being (PGWB) andbody mass index (BMI) was registered. Results: One-hundred and two patients (69 female,mean age:48 (SD:14.5) and 33 male, mean age:59, SD:12) were investigated. The total GSRSscore was significantly better after 3 months (mean:1.80, SD:1.30, p<0.0001) which persistedafter one year (mean:1.81, SD:1.29, p<0.0001) compared with before the operation(mean:2.01, SD:1.30). The domains abdominal pain improved after 3 months (mean>:1.54,SD:0.79, p<0.0001) and 12 months (mean:1.52, SD:0.81, p<0.0001) vs before (2.05,SD:1.16) as well as reflux after 3 months (mean:1.41, SD:0.76, p<0.0001) and 12 months(mean:1.40 SD:0.82, p=0.0002) vs. before (mean:1.69 SD:1.06) and eating dysfunction after3 months (mean:1.25, SD0.65, p<0.0001) and 12 months (mean:1.14, SD:0.38, p<0.0001)vs before (mean:1.67, SD:1.18) . Diarrea, constipation and indigestion did not changesignificantly. The total PGWB score improved from mean:103 before surgery to mean:108both 3 and 12 months after the surgery, p<0.0001. The domains general health, depressionmode, anxiety and vitality were significantly improved after 3 months (P<0.0001, p=0.01,p<0.0001 and p<0.0001 respectively) which persisted after 12 months (p<0.0001, p<0.0001,p<0.0001 and p<0.0001). Positive well being and self control did not show significantchanges. The BMI had increased after one year (Mean:27.8, SD:4.3 vs. 28.1, SD:4.2,p<0.0001). Conclusions: The gastrointestinal and psychological scores improved significantlyboth in the short- and long-term after laparoscopic cholecystectomy. Thus, the patientsbenefit from the operation and the selection of patients seems to be accurate. The increasedBMI may be due to less abdominal pain and eating dysfunction.

Su1283

Cholecystectomy or Gallbladder in Situ After Endoscopic Bile Duct StoneRemoval in Korean; Single Center StudySo Young Choi, Chang Min Cho, Min-Kyu Jung

Background The delayed complications and long term prognosis after endoscopic removalof common bile duct stone are important issues. Cholecystectomy is generally recommendedafter endoscopic sphicterotomu and clearance of bile duct stones. However the necessityfor subsequent cholecystectomy to prevent pancreatobiliary complication remains controver-sial. Aim In this study, we compared the treatment strategies of prophylactic cholecystectomyand gallbladders left in situ for preventing the pancreatobiliary complications. Materialand Method This study was conducted prospective in single tertiary academic center. Werandomized patients after encoscopic sphicerotomy and clearance of their bile duct stonesto receive cholecystectomy or expectant management. The primary outcome was furtherbiliary complications. Results 90 patients entered into the trial. The number of cholecystec-tomy was 27 cases and control group was 63. Delayed pancreatobiliary complicationdeveloped 0/27 in cholecystectomy group and 11/63 (cholangitis =2, cholecystitis=4, GBcolic=4, GB empyema=1)in control group (p<0.001). Conclusion The cholecystectomy, afterendoscopic removal of CBD stone, is useful for preventing the pancreatobiliary complication.

Su1284

Risk Factors and Clinical Outcome of Biliary Bacteremia Associated WithAntibiotic ResistanceYoung Kyung Sung, Limwha Song, Cheol-In Kang, Jong Kyun Lee, Kwang Hyuck Lee,Kyu Taek Lee

Background and Aim: Antibiotic resistant microorganisms have been increasingly noted inKorean patients with biliary bacteremia. The present study investigated the risk factors andclinical outcome of biliary bacteremia associated with antibiotic resistance. Methods: FromJanuary 2000 to April 2010, 556 biliary bacteremic events in 411 patients were retrospectivelyanalyzed. Also we further divided the biliary bacteremia into nosocomial infection (N=396)and community acquired infection (N=160). Results: Antibiotic resistant microorganismswere detected in 39.0 % of 556 biliary bacteremia. The antibiotics resistance rate of 3rdcephalosporin in common gram negative pathogens (Escherichia coli, Klebsiella spp., Enterob-acter spp. and Citrobacter spp.) was 36.5% and the rate of ESBL producing Escherichia coliand Klebsiella spp. was 27.6%. The ampicillin resistance rate in Enterococcus spp. was44.8%, the ceftazidime resistance rate in Pseudomonas spp.was 51.7%, and the methicillinresistance rate in Staphylococcus spp. and Streptococcus spp. was 32.7%. The rates ofantibiotic resistance in biliary bacteremia increased from 27.5% in the first 5 years to 48.2%in the last 5 years (p<0.001). Nosocomial infection (p<0.001), previous antibiotics use within90 days (p<0.001), previous hospitalization within 90 days (p=0.006), hospital days beforebacteremia (p=0.040) and indwelling biliary drainage (p=0.045) were significant independentrisk factors for antibiotic resistance acquisition in biliary bacteremia. In the 30-day mortalityassociated biliary bacteremia, Pitt bacteremia score (p<0.001), hospital days before bacteremia(p=0.005), biliary malignancy (p=0.001), liver cirrhosis (p=0.012) and use of appropriate

S-449 AGA Abstracts

antibiotics (p=0.007) were independent factors. Conclusion: The antibiotic resistant micro-organisms have increased recently in biliary bacteremia and were isolated more commonin nosocomial biliary bacteremia. The severity score and underlying diseases are moreimportant factors for the outcome of biliary bacteremia.

Su1285

Can Sphincter of Oddi Dysfunction (SOD) Be Caused by Roux-en-Y GastricBypass Surgery (RYGB)?Kara Bradford, Laith H. Jamil, Kapil Gupta, Neel K. Mann, Simon K. Lo

Background: Performing ERCP in RYGB patients is exceedingly difficult. These proceduresare typically reserved for very symptomatic patients who are felt for certain to have biliaryor pancreatic pathologies. Our impression is that a significant proportion of these patientshave SOD. How common SOD is in RYGB patients is unknown. Aims: To report ourexperience on the likelihood of SOD in symptomatic patients who had altered gastrointestinalanatomy due to RYGB. Methods: Retrospective review was performed on all RYGB patientswho underwent ERCP at our institution. All cases were done with a double balloon entero-scope (DBE). The procedure technique had been presented in 2010 DDW. Results: 83 DBE-ERCPs were performed on 58 patients (82.7% women). Average BMI was 29.6 (range 19.5-43.5). Mean age was 50 (range 29-84). All had undergone extensive imaging studies beforetheir ERCP. Indications included symptomatic classical biliary or pancreatic abdominal pain(94.8%), abnormal LFT(58.6%), pancreatitis(24%), CBD stone (17.2%), bile leak(8.6%),cholangitis(6.7%), jaundice(5%), biliary stricture (3.4%) and liver abscess (1.7%). 29 patients(50%) had no obvious biliary pathology on ERCP, had no gastroduodenal ulcer on DBEand were felt to have probable SOD. Manometry could not be performed in these patients.28/29 (96.5% women) patients had undergone cholecystectomy (CCY) before ERCP. Meanage of SOD patients was 47.6 (range 29-65) and BMI was 28 (range 19-43). 55.2% hadSOD I, 34.5% had SOD II, and 10.3% were SOD III (See Table 1). 11/29 (37.9%) patientshad pain pre-CCY, 62% developed pain post-CCY. Interestingly, 24/29 (82.7%) patientsdeveloped biliary or pancreatic pain after their RYGB. 28 patients (96.6%) underwentsuccessful biliary sphincterotomies; 1 patient had a pancreatic sphincterotomy. Complicationspost ERCP were limited to pancreatitis (10 %) and severe self-limited unexplained abdominalpain (3.4%). Conclusion: 50% of our RYGB patients referred for classical pancreaticobiliarycomplaints had symptom complex that fits into the category of SOD. 90% were type I andII, indicating that there was physicality to their complaints. >80% of our patients developedSOD symptoms following their RYGB surgery. To our knowledge, this is the first suchobservation in RYGB patients. The causal relationship between SOD and RYGB should befurther investigated.Table 1.

Su1286

Rates of Cholecystectomy (CCY) and Symptomatic Gallstone Disease (Sgsd)Seem to Be Plateauing: A Population Based Study in Allegheny County, PADhiraj Yadav, Thiruvengadam Muniraj, Michael R. O'Connell

Background: After a consistent increase in obesity rates in the United States over the pastfew decades, recent data indicates that this trend may be slowing. Obesity is an importantrisk factor for gallstone disease. We therefore determined the rates of sGSD and CCY inAllegheny County, Pennsylvania (PA) during years 2000-2005. Methods: The PennsylvaniaHealth Care Cost Containment Council (PHC4), an independent state agency, collects dataon all hospitalizations and selected outpatient and ambulatory procedures in PA. We usedthe PHC4 dataset to identify all unique White and Black Allegheny County residents whounderwent a CCY and/or were discharged with a first time primary diagnosis for sGSDduring years 2000-2005. For each patient we noted the demographics, type of CCY (openCCY or laparoscopic CCY, setting (outpatient or inpatient), length of stay and vital status.Age-, sex- and race-adjusted hospitalization rates (overall and for each year) per 100,000population (adjusted to 2000 US population) were calculated using direct standardization.Results: During 2000-2005, 10624 unique White or Black residents had an incident hospitaladmission for sGSD and 24817 underwent a CCY. Patients undergoing CCY were younger(54 +/- 18 vs. 59 +/- 20 yrs, p<0.05), more likely to be female (71 vs. 66%, p<0.05) andequally likely to be White (91 vs. 89%) when compared with patients hospitalized for sGSD.The hospitalization rate for sGSD remained stable during the study period (130.1, 95% CI- 127.6-132.6), while the CCY rates showed a small (7%) but significant increase from301.6 (95% CI - 292.2-311.0) to 322.1 (95% CI - 312.1-332.2). Rates for both conditionsincreased with age and were significantly higher in females (vs. males) [CCY - 2.3 times,sGSD - 1.8 times], and in Whites (vs. Blacks) [CCY - 1.6 times, sGSD - 1.2 times]. Overall,85% of all CCY were performed laparoscopically. The rates for laparoscopic CCY increasedsignificantly (16%, p<0.05) while those for open CCY decreased significantly (36%, p<0.05)during the study period. The proportion of patients who underwent outpatient CCY increasedfrom 47 to 52%, and virtually all outpatient CCY were noted to have laparoscopic CCY.While 90% patients <65 years of age underwent a laparoscopic CCY, the prevalence of openCY was much higher in the elderly (25% in 65 years and older, 32% in 85 years older).The overall mortality for sGSD and CCY was 0.8% and it increased with age from virtually0% in patients <55 years to over 3% in patients 85 years and older. The mortality washigher after open CCY compared with laparoscopic CCY (3.4 vs. 0.3, p<0.05). Conclusions:The rates of sGSD and CCY seem to be plateauing. An increasing number of CCY are being

AG

AA

bst

ract

s