AG

AA

bst

ract

s≥1 organ systems. The duration of SIRS and MMODS was defined as transient (≤48H) orpersistent (>48H). All variables were evaluated for the first 72 hours of hospitalization. Theassociation of these severity markers and mortality was evaluated using χ2 and t-testing forcategorical and continuous variables, respectively. Results: Of 303 enrolled patients withacute pancreatitis; there were 100 in their first episode that had an abdominal CT scanconfirming acute interstitial pancreatitis. There were a total of 4 (4%) deaths, none in the14 transferred patients. Persistent SIRS was found in 46 (46%) patients with 4 (8%) deaths.Persistent SIRS, obesity, and age were not significantly associated with mortality. Highermean APACHE II scores during the first 24 hours of hospitalization were seen in those whodied (16±12 vs. 8±4, p<0.01). 15 (15%) had transient and 7 (7%) had persistent organfailure. 3 (42%) of the 7 patients with persistent organ failure died which is statisticallysignificant when compared to patients without organ failure (p=0.0007). Compared to thosewithout organ failure, patients who survived persistent organ failure were more likely to bedischarged to hospice, rehabilitation, and visiting nursing care (p=0.04). The need for ICUand intubation was not different between the transient and persistent organ failure patients.Conclusion: Persistent organ failure as defined by MMODS is associated with increasedmortality and morbidity in acute interstitial pancreatitis. Higher mean APACHE II scoreswere seen in the first 24 hours of hospitalization in patients who die of acute interstitialpancreatitis. However, persistent SIRS, obesity, and age were not associated with increasedmortality in acute interstitial pancreatitis.

M1283

What Are the Criteria for Resuming Oral Diet in Acute Pancreatitis and IsThere Any Differences in the Course of Disease Due to That Criteria?JungHo Eum, Byungmoo Yoo, KyungRok Lee

Background/Aims : The most serious complication of the acute pancreatitis is infection. Todecrease this problem, it has been recommended to start the oral diet at the earliest stageas possible. Improvements of organ dysfunction, improvements of ileus and abdominal painhave been presented as the criteria for resuming oral diet in moderate to several acutepancreatitis patients. However, very few clinical studies have been done on this issue.Therefore, we aimed to determine the timing and criteria for resuming oral diets in acutepancreatitis. Patients/Method : We enrolled 25 acute pancreatitis patients admitted to Ajouuniversity hospital over the period between Apr. 2007 and Sep. 2007. Firstly, we brokethem down into two groups; the ones who resumed oral diet due to the improvement ofthe ileus and abdominal pain (Group A) and the others who did according to their wishesof intake diet(Group B). Then, we prospectively investigated the time of resuming oral diet,hospitalization time and the reoccurrence of the abdominal pain after resuming oral dietfor each group. The severity of the pancreatitis was based on the abdominal CT results (CTscores) or Ranson's criteria. The mean age for the patients was 52.2 (age 33~84) with 21male and 4 female. The causes of the pancreatitis were alcohol, gall stone and others in 10,7, 8 cases respectively. The overall mean time to resuming diet was 4.4 days (2~11 days)with 7.6 days (4~17 days) of mean hospitalization time. The overall mean CT score andRanson's Criteria were 3.04 and 1.16 respectively. Results : The number of patients were13 and 12 for Group A and Group B respectively. There were no significant differencesbetween the groups in the mean ages (52.15 for Group A and 52.25 for Group B, p>0.05)and gender composition of male vs. female (10:3 for Group A and 11:1 for Group B, p>0.05).No significant differences existed between the two groups in either the time of resumingoral diet within 4.85 days for Group A and 3.92 days for Group B(p>0.05) or the hospitaliza-tion time with 8.2 days and 7.0 days for Group A and Group B respectively(p=0.083). Painreoccurrence after oral diet showed no significant differences between the two groups in 2cases(or 15.4%) in Group A and 1 case(or 8.3%) in Group B. Conclusion : There were nosignificant differences between the groups in the time till resuming oral diet, hospitalizationtime and the frequency of recurrent abdominal pain. However, the cases who resumed theoral diet by the patients' wishes showed the smaller numbers in the hospitalization time.We believe that studies with bigger samples are highly required in the future.

M1284

Use of Angiotensin II Receptor Blockers May Decrease Risk of AcutePancreatitis - a Nested Case-Control StudyTomas Sjöberg, Luis A. Garcia Rodriguez, Mats Lindblad

There is no specific treatment against acute pancreatitis. A protective effect by AngiotensinII Receptor Blockers (ARB) on acute pancreatitis has been suggested experimentally, but todate no study has been carried out on humans. Material and Methods We conducted apopulation-based case-control study using The Health Improvement Network in the UK,comprising about 167,000 hypertensive patients in the study period 1996-2005. In multivari-able regression analysis odds ratios (ORs) were calculated with 95% confidence intervals(CIs). Adjustments included sex, age, calendar year, body mass index, tobacco smoking,alcohol, general practitioner visits, and various antihypertensive medications with regard toexposure to ARB and risk of acute pancreatitis. Results Among 633,281 person-years at risk265 new cases of acute pancreatitis were identified. Current users of ARB had a 37% reducedrisk of developing acute pancreatitis as compared to non users (OR 0.63, 95% CI (0.38-1.02). A high daily dose of ARB nearly halved the risk (OR 0.53, 95% CI 0.22-1.26). Noclear association was found between use of other antihypertensive drugs and risk of acutepancreatitis. Conclusion Our study adds support to previous experimental findings. Use ofARB may decrease risk of acute pancreatitis. More research is needed to elucidate the potentialtherapeutic role of ARB for acute pancreatitis in the clinical setting.Acute pancreatitis and use of angiotensin receptor blockers

T : 11501$$CH204-02-08 16:47:09 Page 376Layout: 11501B : e

A-376AGA Abstracts

Risk of acute pancreatitis presented as odds ratios (OR) and 95% confidence intervals (95%CI) in a fuly adjusted model (**)

M1285

Serum Concentrations of Leptin and Resistin As Acute Inflammatory PhaseFactors in Patients with Acute PancreatitisEwa Malecka-Panas, Bartosz Lesniowski, Anna Kumor

Bartosz Lesniowski 1, Anna Kumor 2, Piotr Daniel 1, Miroslawa Pietruczuk 2, Ewa Malecka-Panas1 1 Department of Digestive Tract Disorders , Medical University of Lodz - Poland 2Department of Laboratory Medicine, Medical University of Lodz - Poland Acute pancreatitis(AP) is the severe disease with high mortality and morbidity rates of not clearly elucidatedpathomechanism. The possible pathogenic role of adipocytokines in AP has been recentlysuggested. Adipocytes hyperplasia is a trigger factor that initiates chronic inflammatory statewith adipokines release, among them leptin, resistin, and TNF-alpha. The aim of this studywas to evaluate the serum concentration of leptin and resistin as regards to CRP as possibleacute phase factors in AP. Material and methods: 39 patients with AP and 40 age matchedhealthy subjects were enrolled in this study. In all patients AP was classified as gradeB according to Balthazar's score In both groups BMI values have been obtained. Serumconcentrations of leptin and resistin were measured with ELISA in the first day of hospitalis-ation. Serum concentration of CRP was measured with turbidimetric assay. . Results: Nosignificant differences in BMI between AP patients and control group have been found.Increased levels of resistin were noted in AP compared to the healthy subjects (8,38 ± 4,87vs 3,58 ± 1,51 ng/ml; p<0,05). The serum concentrations of leptin were 3 times higherthan in controls (16,7 ± 8,1 vs 6,1 ± 1,9 ng/ml; p<0,05). Similarly CRP serum concentrationsin AP patients were significantly higher than in control group (23,2 ± 8,75 vs 3,95 ± 1,96;p<0,01). The positive correlations between serum resistin and CRP concentrations (r=0,57;p<0,05) and between leptin concentration and CRP (r=0,61; p<0,05) have been observed.In conclusion, the results of this study may suggest, that the serum concentrations of leptinand resistin may play an important role in early diagnostics of AP and possibly be used asmarkers of acute inflammatory response. Correspondence to: Bartosz Lesniowski Dept ofDigestive Tract Disorders Medical University of Lodz, 90-153 Lodz, Poland Kopcinskiegostr. 22 phone:+4842 6776664 fax +4842 6786480 mobile:+48660704744 email:lesniowski1-973@tlen.pl

M1587

The Mucin Phenotypes and Claudin Expressions in EsophagealAdenocarcinoma and the Surrounding MucosaTakayo Sakano, Ryoji Kushima, Shizuki Takemura, Ken-ichi Mukaisho, Michael Vieth,Manfred Stolte, Kaiyo Takubo, Takanori Hattori

Background and aims: Most esophageal adenocarcinomas arise from Barrett's esophagus(BE) via the metaplasia-dysplasia-carcinoma sequence. Over the past 50 years, the definitionof BE has changed, and it is recommended that BE should be diagnosed only when specializedintestinal epithelium is observed. However, in our experience, the background mucosa ofdistal esophageal adenocarcinomas may not always be the intestinalized mucosa but couldalso be the gastric cardiac type mucosa. We investigated mucin phenotypes and claudinexpressions in esophageal glandular neoplastic lesions and the surroundings to elucidatethe histogenesis of esophageal adenocarcinoma. Materials and methods: Histologic andimmunohistochemical analyses were performed to examine 70 glandular neoplastic lesionsand the surroundings of the esophagus in German patients. Based on the histologicalexamination, the neoplastic lesions were classified into low-grade dysplasia, high-gradedysplasia, and adenocarcinoma. Regarding the surrounding mucosa, BE was determinedaccording to the definition of American Colleges of Gastroenterology. And for immunohisto-chemical analyses, MUC5AC, MUC6, and claudin-18 were used as gastric markers, andMUC2, CD10, CDX2, claudin-3 and -4 were used as intestinal markers. Mucin phenotypeswere determined by MUCs and CD10 expressions. Results: Of the 70 lesions examined,4.4%, 36.8%, and 58.8% were classified as low-grade dysplasia, high-grade dysplasia, andadenocarcinoma, respectively. Moreover, 18.4% of the adenocarcinomas invaded into thesubmucosal area. Both the high-grade dysplasias and the adenocarcinomas mainly expressedgastric type mucin. Moreover, most cases of adenocarcinomas with complete gastric mucinphenotypes were also positive for intestinal claudins. Regarding the surrounding mucosa,gastric cardiac type mucosa without intestinal elements was observed in half of the cases,whereas mucosa with specialized intestinal epithelium was observed in the remaining cases.High-grade dysplasias and adenocarcinomas with complete gastric phenotypes showed tend-encies to develop in gastric cardiac type mucosa. However, almost half of the gastric cardiactype mucosa showed expressions of intestinal-claudins (claudin-3 and -4). Conclusions:Most of the esophageal glandular neoplastic lesions showed gastric mucin phenotypes. Notonly BE but gastric cardiac type mucosa without intestinal epithelium were detected as thesurrounding mucosa of the esophageal adenocarcinomas. The gastric cardiac type mucosaof esophagus which has the formation of intestinal tight junction is different from genuinegastric mucosa, and may be associated with carcinogenesis.