- Email: [email protected]
933 Small Intestinal Bacterial Overgrowth Recurrence After Rifaximin Therapy
Modulation By Muscarinic Acetylcholine Receptor Subtype 4 of CarbacholInduced Pepsin Secretion in Mice: Relation to D Cells/Somatostatin Nobuaki Takahashi, Takuya Endo, Kikuko Amagase, Eiji Nakamura, Susumu Okabe, Minoru Matsui, Koji Takeuchi Background/Aim: Muscarinic acetylcholine receptors (mAChRs) consist of five subtypes (M1~M5) and are widely expressed to mediate diverse autonomic functions in peripheral organs, including the gastrointestinal tract. Although both M1- and M3- receptors reportedly play a role in the regulation of pepsin secretion in response to acetylcholine, we recently found that carbachol (CCh)-induced pepsin secretion was markedly decreased in M4 knockout (KO) mice. In the present study, we demonstrated, using M1~M5 KO mice, the importance of M4 receptors in the cholinergic regulation of pepsin secretion and investigated how this secretion is modulated by the activation of M4 receptors. Methods: C57BL/6J mice of wildtype (WT) and M1-, M2-, M3-, M4- or M5-KO were used. Under urethane anesthesia, the abdomen was incised, the cardiac portion was ligated, and an acute fistula prepared with a polyethylene tube was provided in the stomach through a pylorus. Then, the stomach was instilled with saline (0.4 ml) through the fistula, and the solution was changed every 20 min. CCh (30 µg/kg) was given SC as a single injection. Atropine (0.3 mg/kg) or CYN154806 [somatostatin-2 receptor (SST2R) antagonist 0.1-3mg/kg] was given SC 20 min before CCh. Expressions of D cells and M4 receptors were examined immunohistochemically by double staining with anti-somatostatin and anti-M4 receptor antibodies. Results: CCh caused an increase of pepsin secretion in WT mice, and the effect was completely inhibited by prior administration of atropine. The stimulatory effect of CCh was similarly observed in the animals lacking M1-, M2- or M5-receptors but significantly decreased in M3- or M4KO mice, as compared to WT; especially, the response was all but completely abolished in M4-KO mice. CYN154806, the SST2R antagonist, significantly reversed the decreased pepsin response to CCh in M4- but not M3-KO mice. The M4-KO mice showed basal pepsin secretion much less than WT mice, and these changes were also reversed by CYN154806. By contrast, somatostatin decreased pepsin secretion under basal and CCh-stimulated conditions. Furthermore, the immuno- histochemical study showed the localization of M4 receptors on D cells in the mouse stomach. Conclusion: These results suggest that under cholinergic stimulation the secretion of pepsin is mediated mainly by the activation of M4-receptors and partly through M3-receptors but does not involve other muscarinic receptor subtypes. Somatostatin has an inhibitory effect on pepsin secretion through SST2 receptors. It is assumed that the activation of M4-receptors inhibits the release of somatostatin from D cells and results in enhancement of pepsin response.
931 Rifaximin Versus Metronidazole for the Treatment of Small Bowel Bacterial Overgrowth Maurizio Gabrielli, Ernesto Cristiano Lauritano, Emidio Scarpellini, Andrea Lupascu, Veronica Ojetti, Alessia Cazzato, Davide Roccarina, Maria Elena Ainora, Giovanni Cammarota, Michele Serricchio, Giovanni Gasbarrini, Giovanni Ghirlanda, Antonio Gasbarrini Background: Few controlled trials concerning choice and duration of antibiotic therapy for small intestinal bacterial overgrowth are available. Aims: to assess efficacy, safety and tolerability of rifaximin with respect to metronidazole for the treatment of small intestinal bacterial overgrowth. Patients: We enrolled 124 consecutive small intestinal bacterial overgrowth patients. Methods: Diagnosis of small intestinal bacterial overgrowth based on the clinical history and positive glucose breath test. Patients were randomised in two 7-day treatment groups: rifaximin 1200 mg and metronidazole 750 mg. Glucose breath test was reassessed 1 month after. Compliance and side-effect incidence were also evaluated. Results: No dropout were observed in rifaximin group. Five drops-out occurred in metronidazole group. The glucose breath test normalization rate was significantly higher in rifaximin with respect to metronidazole group (64.5%, 40/62 versus 45.2%, 28/62; OR 1.5, 95% CI 1.07-4.54). The overall prevalence of adverse events was significantly lower in rifaximin with respect to metronidazole group. Conclusion: Rifaximin showed an higher SIBO decontamination rate than metronidazole at the tested doses, both with a significant gain in terms of tolerability. Either the present study or recent evidencies suggest that rifaximin represents a good choice for the management of patients affected by SIBO.
929 Luminal L-Glutamate Enhances Gastroduodenal Mucosal Defenses in Rats Yasutada Akiba, Misa Mizumori, Jonathan D. Kaunitz
932
Taste receptor families have been localized in the gastrointestinal (GI) tract, suggesting that the GI mucosa detects the luminal nutrients and chemicals in order to maintain physiological processes such as digestion, absorption, secretion, and motility. L-glutamate, which is the primary nutrient conferring the umami or proteinaceous taste, stimulates gastric vagal afferents with release of NO and 5-HT, suggesting a nutrient-sensing pathway for L-glutamate. We hypothesized that luminal L-glutamate affects mucosal defense mechanisms in the rat gastroduodenum. Methods: Under isoflurane anesthesia, the gastric or duodenal mucosa was exposed and topically perfused with pH 7.0 buffer with or without L-glutamate (0.1-10 mM). Intracellular pH (pHi) of the epithelial cells, blood flow and mucus gel thickness (MGT) were simultaneously and continuously measured in the stomach or duodenum using In Vivo microscopic system. Some rats were pretreated with indomethacin (5 mg/kg, sc) or capsaicin (125 mg/kg, sc). Results: Luminal L-glutamate dose-dependently increased pHi and MGT, but had no effect on blood flow in the duodenum. L-glutamate (10 mM)-induced cellular alkalinization and mucus secretion were inhibited by pretreatment with indomethacin or capsaicin, suggesting the involvement of prostaglandin (PG) synthesis and capsaicin-sensitive afferent nerves in the L-glutamate effects. Similar effects were reproduced by PGE2 iv (0.3 mg/kg). L-glutamate effects were also observed in the stomach. The metabotropic glutamate receptors mGluR1 and 4, the taste receptors T1R1 and 3, but not T1R2, and the calciumsensing receptor (CaSR) were expressed in the gastroduodenal mucosa, as measured with real time RT-PCR, supporting the hypothesis that glutamate receptors are present in the gastroduodenal epithelium. Conclusions: Intracellular alkalinization and increased mucus secretion induced by luminal L-glutamate suggest that L-glutamate may protect the gastroduodenal mucosa from acid-induced injury. Activation of glutamate receptors may enhance mucosal defenses via activation of PG production and afferent nerves.
Lymphocytic Follicles and Aggregates in Colonic Biopsies with Non-Specific Changes Are Related to Duration of Diarrhea Nilesh Shah, Emily F. Shen, Wee Hoe Gan, Tian Ming Tu, Khay Guan Yeoh, Shen Liang, Manuel Salto-Tellez Background and aims: Non-specific changes (non-specific chronic inflammation) in patients endoscoped because of diarrhoea is, arguably, the commonest diagnosis in colorectal biopsy interpretation, but it is of little clinical significance. In this study we aim to find histological diagnostic criteria associated with long-during chronic diarrhoea within this poorly understood group of patients. Method: Detailed clinical features and histological findings were analysed in a group of 49 patients with chronic diarrhoea, with near-normal histological appearance and no clear-cut known etiological agent after subsequent clinical and pathological, re-appraisal amounting to a retrospective review of 358 colonic biopsies fragments. We made a detailed assessment of more than thirty morphological markers of inflammation and related architectural changes in the colonic mucosa. Results: In the colonic mucosa of our study group the large LFA median count was 3 (range, 0-10), the total (large and small) LFA median count was 4 (range, 0-10) and the mean density of LFA was 0.49 +/- 0.22. These values were significantly higher than control group [large LFA 1 (range, 0-3), p <0.001; total LFA 2 (range, 0-4), p = 0.003 and density of LFA 0.23 +/- 0.14, p = 0.001]. The means of total LFA per biopsy were 0.59 +/- 0.32) for the study group and 0.31 +/0.21 for the control group (p = 0.005). The duration of diarrhoea was correlated among others, with the following parameters: number of large LFA (p = 0.001, Spearman's rho = 0.480), number of total LFA (p = 0.003, Spearman's rho = 0.433), density of LFA (p = 0.009, Spearman's rho = 0.385), and means of total lymphoid follicles per biopsy (p = 0.004, Spearman's rho = 0.429). Conclusion: Our analysis shows that, in biopsy samples with otherwise “non-specific changes” there is a statistically significant correlation between the increase number of lymphoid follicles and aggregates and the length of the diarrhoea before and/or after the time of biopsy. These findings challenge the view of LFA as “physiological” components of the colonic mucosa above a certain threshold and, more importantly, allow the prediction of the chronic behaviour of a subset of patients unpredictable until now. In essence, this observation may indicate the presence of a new diagnostic entity, the Lymphocytic Follicle and Aggregate Colitis.
930 Gender Differences in Duodenal Bicarbonate Secretion Cause Gender Differences in Human Duodenal Ulcer Biguang Tuo, Guorong Wen, Jinqi Wei, Yalin Zhang, Hui Dong Background: It has long been observed that prevalence in duodenal ulcer is lower in women than age-matched men; however, the causes of these gender differences are currently unknown. Since duodenal mucosal bicarbonate secretion (DMBS) protects the duodenum against acid-induced injury, we hypothesize that estrogen attributes to gender differences in human duodenal ulcer via a specific stimulation of DMBS. Methods: Male or female healthy volunteers were divided into two different age-matched groups: 20 ~29 years old and 60 ~ 69 years old. The proximal duodenum was flushed and isolated. After basal HCO3secretion was observed for 60 minutes, the duodenum was perfused with 100 mM HCl or 2 mg 17β-estradiol to study the effects of duodenal acidification or estrogen on duodenal HCO3- secretion, which was measured by back titration. Protein expression of estrogen receptors was detected in human duodenal tissues by Western blot analysis. Results: Basal DMBS was significantly higher in women than in men of 20 ~ 29 years old group (189.5 ± 23.5 in women vs. 120.7 ± 16.2 µmol/cm-h in men, n=9, p<0.05). However, the basal DMBS was comparable in both genders of 60~69 years old group (123.1 ± 17.8 in women
933 Small Intestinal Bacterial Overgrowth Recurrence After Rifaximin Therapy Ernesto Cristiano Lauritano, Maurizio Gabrielli, Lucia Sparano, Alessia Cazzato, Andrea Lupascu, Maria Luisa Novi, Sandra Sottili, Giovanna Vitale, Valentina Cesario, Michele Serricchio, Giovanni Cammarota, Giovanni Gasbarrini, Giovanni Ghirlanda, Antonio Gasbarrini Objectives: Current treatment for small intestinal bacterial overgrowth (SIBO) is based on courses of broad spectrum antibiotics. No data concerning SIBO recurrence are available. The aims of the present study were to investigate SIBO recurrence as assessed by glucose
A-137
AGA Abstracts
AGA Abstracts
vs. 109.8 ± 16.2 µmol/cm-h in men, n=9, p>0.05). Luminal perfusion of HCl resulted in a robust increase in DMBS in both 20~29 years old and 60~69 years old groups. However, acid-stimulated net peak DMBS was significantly higher in women than in men of 20~29 years old group (n=9, P<0.05), whereas net peak DMBS was not significantly different between women and men in 60~69 years old (n=9, P>0.05). Further study found that luminal perfusion of 17-β-estradiol significantly increased DMBS in men and women of both 20~29 years old and 60~69 years old groups (n=9, P<0.001), indicating estrogen has a direct effect on human DMBS. Western blot analysis from human proximal duodenal tissues showed that estrogen receptor subtypes ERα and ERβ proteins were expressed at similar levels in human duodenal tissues taken by biopsy from male and female healthy subjects. Conclusion: gender differences in human DMBS exist between premenopausal women and the age-matched men, but not between postmenopausal women and the agematched men. Estrogen may be involved in the gender differences of DMBS and protect the human duodenum by activating specific estrogen receptors and then stimulating DMBS, which may explain the previously observed lower prevalence of duodenal ulcer in women.
928
AGA Abstracts
breath test (GBT) after antibiotic treatment and conditions associated to SIBO recurrence. Methods: Eighty (80) consecutive patients affected by SIBO and decontaminated by rifaximin (1200 mg per day for 1 week) were enrolled. Diagnosis of SIBO was based on GBT. GBT was reassessed at 3, 6, and 9 months after decontamination. GBT positivity recurrence, predisposing conditions and gastrointestinal symptoms were evaluated. Results: Ten (10/ 80, 12.6%), 22 (22/80, 27.5%) and 35 (35/80, 43.7%) patients showed positivity to GBT at 3, 6 and 9 months after successful antibiotic decontamination, respectively. At multivariate analysis, older age (OR 1.09, 95% CI 1.02-1.16), history of appendectomy (OR 5.9, 95% CI 1.45-24.19) and chronic use of proton pump inhibitors (OR 3.52, 95% CI 1.07-11.64) were significantly associated to GBT positivity recurrence. All gastrointestinal symptoms significantly increased at 3, 6 and 9 months in patients with evidence of GBT positivity recurrence. Conclusions: GBT positivity recurrence rate was high after antibiotic decontamination. Older age, history of appendectomy and chronic use of proton pump inhibitors were associated with GBT positivity recurrence. Patients with evidence of GBT positivity recurrence showed gastrointestinal symptoms relapse thus suggesting SIBO recurrence. Cyclic courses of non-absorbable antibiotics could be necessary in order to prevent SIBO recurrence.
significantly decreased by DSM (ANOVA adjusted for Rotavirus): Peru: DSM = 102.0 (65.5) g/Kg, placebo = 118.8 (92.5) g/Kg (p = 0.032), (14 % difference). Rotavirus positive children: DSM = 146.9 (90.1) g/Kg, placebo = 187.9 (122.1) g/Kg, (p = 0.039), (22 % difference). Malaysia: DSM = 87.9 (81.2) g/Kg, placebo = 90.7 (94.0) g/Kg, (p = 0.007) (3.1 % difference). Rotavirus positive children: DSM = 91.8 (103.0) g/kg, placebo = 184.5 (192.4) g/Kg, (p = 0.002), (50.3 % difference). Pooled studies: DSM = 94.5 (74.4) g/Kg, placebo = 104.1 (94.2) g/Kg, (p = 0.002) (9.2 % difference). Rotavirus positive children: DSM = 124.3 (98.3) g/ Kg, placebo = 186.8 (147.2) g/Kg, (p = 0.001), (33.5% difference. Median [95% IC] time to recovery: Peru: DSM = 68.2 [60.2-85.0] h, placebo = 118.9 [94.9-140.5] h, (p<0.001); Malaysia: DSM = 25.1 [20.5-29.0] h, placebo = 32.6 [27.5-39.3] h, (p<0.001). DSM improved anal irritation and was well tolerated. Conclusion: Two randomized double-blind placebocontrolled studies in 602 children with acute watery diarrhea showed that diosmectite, in association with oral rehydration salts, significantly decreases stool output and duration of diarrhea, especially in rotavirus-positive children and is well tolerated. 936
934
Determination of Urinary Collection Timeframes to Enhance Measurement of Gastrointestinal (GI) Site Specific Permeability in Adults and Children Mark E. McOmber, Robert J. Shulman
®
Evaluation of the Efficacy, Safety and Tolerability of Prucalopride (Resolor ) Given Subcutaneously in Patients Undergoing Elective Partial Colectomies Susan Galandiuk, Greet Beyens, Jannie Ausma, Lieve Vandeplassche
Background:GI permeability testing using urinary recovery of orally administered mono- and disaccharides is used commonly. Combinations of sugars are used to measure permeability in different areas of the GI tract (i.e., site specific permeability). However, human studies have not been published describing optimal collection timeframes for site specific permeability testing. This study was done to establish collection timeframes that would allow for simultaneous permeability comparisons at different gastrointestinal sites in both adults and children. Design/Methods:Twelve adults (median age 34 yr.; range 21-57) and 11 children (median age 11 yr.; 5-17 yr. old) were studied. Each subject drank 127.5 mL of a solution containing sucrose (10 g/dL), lactulose (5 g/dL), mannitol (1 g/dL), and sucralose (1 g/dL). Urine was collected over 24 hr. The first 12 hr. of collection were performed while fasting, followed by another 12 hr. of modified diet (i.e., avoiding artificial sweeteners). Sugar concentrations were measured using high-performance liquid chromatography. At each time point we calculated the ratios of the various sugars as well as the percent of the total amount of sugar ultimately recovered (% total recovered). Results:Gastric Permeability: In adults median (range) Sucrose % total recovered at 30 min. was 12% (4-36) and at 90 min. was 33% (6100). Corresponding values in children were 11% (4-34) and 31% (15-75). Small Intestinal Permeability: In adults Lactulose/Mannitol ratio peaked at a median of 11 hr (range 7-20) and Lactulose % total recovered was > 95% at a median of 14 hr (range 8-20). Corresponding values in children were 12 hr (range 9-20) and 15 hr (range 10-24). Colonic Permeability: In adults Sucralose/Lactulose ratio peaked at a median of 17 hr (range 2-26) and Sucralose % total recovered was > 95% at a median of 20 hr (range 10-24). Corresponding values in children were 17 hr (range 1-23) and 22 hr (range 14-24). Conclusions:1) Urinary collection timeframes are comparable between adults and children; 2) Gastric permeability can be assessed in 90 min of collection but optimally longer; 3) Small intestinal permeability optimally requires 12 hr of collection; 4) Colonic permeability optimally requires 17 hr of collection; 5) The relative timing of L and SL urinary recovery (i.e., Lactulose preceding Sucralose) supports the evidence that they measure small intestinal and colonic permeability, respectively.
Objective: The objective of this study was to determine a safe and effective dose of prucalopride, a selective high-affinity 5-HT4 receptor agonist, in shortening the duration of post operative ileus in patients undergoing elective partial colectomies. Methods: This was a phase IIb, parallel-group, randomized, double-blind, placebo-controlled, dose-ranging study in patients scheduled to undergo elective partial colectomies with primary anastomoses by laparotomy. Patients with concomitant diseases affecting lower gastro-intestinal motility were excluded. Patients were randomized to receive subcutaneous (s.c.) injections (1 ml) of either placebo or 0.5 mg, 2 mg or 4 mg prucalopride once daily. The first dose was administered in the recovery room, within 2 hours after surgery. Subsequent doses were administered at breakfast time for 3 days post-surgery. Assessments were made pre-operatively, in the recovery room and at each mealtime from post-operative day (POD) 1 to either POD 12 or discharge. Efficacy variables are shown in the table below. Adverse events (AEs), laboratory tests, and electrocardiograms (ECGs) were evaluated for safety purposes. Results: The intentto-treat (ITT) population consisted of 317 patients (81.4% Caucasian, 53% male, mean age 55.6±0.8 years). Efficacy results are shown in the table below. Median time to first flatus or stool was shorter in all prucalopride groups than in the placebo group, reaching statistical significance in the 4 mg group. The median time to discharge was 1 day earlier in the prucalopride groups when compared to the placebo group (p=ns). Most frequently reported AEs were nausea, vomiting, fever, and insomnia. Incidence rates of AEs, serious AEs en AEs leading to discontinuation did not differ between prucalopride and placebo groups. There were no clinically relevant differences in ECG parameters between treatment groups. Conclusions: The results of this study demonstrate that prucalopride, given subcutaneously once daily, reduces the time to first flatus or stool in patients that underwent elective partial colectomies. Moreover, patients treated with prucalopride could be discharged from the hospital 1 day earlier then patients treated with placebo. Prucalopride was safe and well tolerated at all doses studied.
937 Mortality Associated with Gastrointestinal (GI) Bleeding Events: Comparing Short-Term Clinical Outcomes of Patients Hospitalized for Upper GI Bleed and Acute Myocardial Infarction (Ami) in a US Managed Care Setting Gergana Zlateva, Henry J. Henk, Victoria Zarotsky, Byron L. Cryer, Charles M. Wilcox Background: Outside of specialist settings, there is limited awareness about the severe outcomes associated with gastrointestinal (GI) bleeds in comparison with acute myocardial infarction (AMI). GI bleeds require intensive medical intervention leading to escalated health resource utilization, costs and sometimes mortality. Understanding how mortality among patient with GI bleeding events compares to the mortality associated with an AMI may be valuable in providing context for understanding the impact of the costs and consequences of GI bleeding events. Objectives: To estimate the 30-, 60-, and 90-day mortality among patients hospitalized for upper GI-bleeds compared to a propensity-matched control cohort of patients with acute myocardial infarction (AMI). Methods: US national health plan claims data (1999-2003) were used to identify patients hospitalized with a GI bleeding event. Patient with evidence of GI bleed or AMI in the 6 months prior to the index date were excluded. These patients were propensity-matched to AMI patients with no evidence of GI bleed from the same US health plan. The propensity score model included patient demographics (e.g., age, gender), medication use, factors associated with mortality (e.g., Charlson Comorbidity Index (CCI) score), and medical care resource utilization history. Results: 12,437 upper GI-bleed patients and 22,847 AMI patients who met inclusion criteria were identified. Propensity score matching yielded 6,923 matched pairs. Prior to matching, the GI bleed sample had a higher mean CCI and higher baseline medical care resource utilization than the AMI sample. The matched cohorts were found to have a similar CCI and to be similar on nearly all utilization and cost measures (excepting ER costs). A comparison of outcomes among the matched cohorts found that AMI patients had higher rates of mortality (4.35% vs. 2.54%; p<.0001) and rehospitalization (2.56% vs. 1.79%; p=0.002), while GI bleed patients were more likely to have a repeat procedure (72.38% vs. 44.95%; p<0.001) following their initial hospitalization. The majority of the difference in overall 30-day mortality between GI bleed and AMI patients was accounted for by mortality during the initial hospital stay (1.91% vs. 3.58%); 30- 60- and 90-day mortality excluding deaths within the initial hospitalization did not differ between the cohorts. Conclusions: GI bleeding events result in significant mortality similar to that of an AMI after adjusting for the initial hospitalization. Future study will need to evaluate the cost and resource utilization associated with GI bleeding events leading to mortality.
* p<0.01 when compared with placebo 935 The Clay Diosmectite in Association with Oral Rehydration Salts Reduces Stool Output and Diarrhea Duration in Children with Acute Watery Diarrhoea: Results of 2 Randomized Double Blind Placebo-Controlled Studies Jimmy Lee Kok Foo, Christophe Dupont, Philippe Garnier, Helene Mathiex-Fortunet, Eduardo Salazar-Lindo Introduction: Study aimed at assessing the effect of the clay diosmectite (DSM) on stool output and duration of diarrhea in children with acute watery diarrhea. Patients and Methods: Two parallel double-blind, placebo-controlled studies in 1-36 months old male children were conducted in Peru (n=300) and Malaysia (n=302). Children were included if they had ≥ 3 watery stools/day for less than 72h, weight/height ratio ≥ 80 %, no intravenous rehydration need, no gross blood in stools, fever <39°C, no current treatment with antidiarrheal or antibiotic agents. Rotavirus in stool was sought. DSM dosage was 3g (1-12 months) or 6g (13-36 months) tid over at least 3 days, then half the dosage until recovery. All children received oral rehydration salts (ORS), according to WHO guidelines. Primary criterion was stool output in g/Kg of body-weight (B-W) for the first 72 h. Other criteria were time to recovery (h), stool frequency and consistency, anal irritation (%). Results: Both treatment groups in ITT population of Peru and Malaysia were similar in their demographic characteristics: mean (SD) age: Peru = 12.5 (6.1) months, Malaysia = 15.9 (8.5) months; B-W: Peru = 9.35 (1.67) Kg, Malaysia = 9.02 (2.05) Kg. ORS consumption was also similar: Peru = 1426 (983) ml, Malaysia = 1022 (674) ml. Rotavirus frequency was: 22% in Peru, 12% in Malaysia. Mean (SD) stool output for the first 72 hours was
AGA Abstracts
A-138
931 Rifaximin Versus Metronidazole for the Treatment of Small Bowel Bacterial Overgrowth Maurizio Gabrielli, Ernesto Cristiano Lauritano, Emidio Scarpellini, Andrea Lupascu, Veronica Ojetti, Alessia Cazzato, Davide Roccarina, Maria Elena Ainora, Giovanni Cammarota, Michele Serricchio, Giovanni Gasbarrini, Giovanni Ghirlanda, Antonio Gasbarrini Background: Few controlled trials concerning choice and duration of antibiotic therapy for small intestinal bacterial overgrowth are available. Aims: to assess efficacy, safety and tolerability of rifaximin with respect to metronidazole for the treatment of small intestinal bacterial overgrowth. Patients: We enrolled 124 consecutive small intestinal bacterial overgrowth patients. Methods: Diagnosis of small intestinal bacterial overgrowth based on the clinical history and positive glucose breath test. Patients were randomised in two 7-day treatment groups: rifaximin 1200 mg and metronidazole 750 mg. Glucose breath test was reassessed 1 month after. Compliance and side-effect incidence were also evaluated. Results: No dropout were observed in rifaximin group. Five drops-out occurred in metronidazole group. The glucose breath test normalization rate was significantly higher in rifaximin with respect to metronidazole group (64.5%, 40/62 versus 45.2%, 28/62; OR 1.5, 95% CI 1.07-4.54). The overall prevalence of adverse events was significantly lower in rifaximin with respect to metronidazole group. Conclusion: Rifaximin showed an higher SIBO decontamination rate than metronidazole at the tested doses, both with a significant gain in terms of tolerability. Either the present study or recent evidencies suggest that rifaximin represents a good choice for the management of patients affected by SIBO.
929 Luminal L-Glutamate Enhances Gastroduodenal Mucosal Defenses in Rats Yasutada Akiba, Misa Mizumori, Jonathan D. Kaunitz
932
Taste receptor families have been localized in the gastrointestinal (GI) tract, suggesting that the GI mucosa detects the luminal nutrients and chemicals in order to maintain physiological processes such as digestion, absorption, secretion, and motility. L-glutamate, which is the primary nutrient conferring the umami or proteinaceous taste, stimulates gastric vagal afferents with release of NO and 5-HT, suggesting a nutrient-sensing pathway for L-glutamate. We hypothesized that luminal L-glutamate affects mucosal defense mechanisms in the rat gastroduodenum. Methods: Under isoflurane anesthesia, the gastric or duodenal mucosa was exposed and topically perfused with pH 7.0 buffer with or without L-glutamate (0.1-10 mM). Intracellular pH (pHi) of the epithelial cells, blood flow and mucus gel thickness (MGT) were simultaneously and continuously measured in the stomach or duodenum using In Vivo microscopic system. Some rats were pretreated with indomethacin (5 mg/kg, sc) or capsaicin (125 mg/kg, sc). Results: Luminal L-glutamate dose-dependently increased pHi and MGT, but had no effect on blood flow in the duodenum. L-glutamate (10 mM)-induced cellular alkalinization and mucus secretion were inhibited by pretreatment with indomethacin or capsaicin, suggesting the involvement of prostaglandin (PG) synthesis and capsaicin-sensitive afferent nerves in the L-glutamate effects. Similar effects were reproduced by PGE2 iv (0.3 mg/kg). L-glutamate effects were also observed in the stomach. The metabotropic glutamate receptors mGluR1 and 4, the taste receptors T1R1 and 3, but not T1R2, and the calciumsensing receptor (CaSR) were expressed in the gastroduodenal mucosa, as measured with real time RT-PCR, supporting the hypothesis that glutamate receptors are present in the gastroduodenal epithelium. Conclusions: Intracellular alkalinization and increased mucus secretion induced by luminal L-glutamate suggest that L-glutamate may protect the gastroduodenal mucosa from acid-induced injury. Activation of glutamate receptors may enhance mucosal defenses via activation of PG production and afferent nerves.
Lymphocytic Follicles and Aggregates in Colonic Biopsies with Non-Specific Changes Are Related to Duration of Diarrhea Nilesh Shah, Emily F. Shen, Wee Hoe Gan, Tian Ming Tu, Khay Guan Yeoh, Shen Liang, Manuel Salto-Tellez Background and aims: Non-specific changes (non-specific chronic inflammation) in patients endoscoped because of diarrhoea is, arguably, the commonest diagnosis in colorectal biopsy interpretation, but it is of little clinical significance. In this study we aim to find histological diagnostic criteria associated with long-during chronic diarrhoea within this poorly understood group of patients. Method: Detailed clinical features and histological findings were analysed in a group of 49 patients with chronic diarrhoea, with near-normal histological appearance and no clear-cut known etiological agent after subsequent clinical and pathological, re-appraisal amounting to a retrospective review of 358 colonic biopsies fragments. We made a detailed assessment of more than thirty morphological markers of inflammation and related architectural changes in the colonic mucosa. Results: In the colonic mucosa of our study group the large LFA median count was 3 (range, 0-10), the total (large and small) LFA median count was 4 (range, 0-10) and the mean density of LFA was 0.49 +/- 0.22. These values were significantly higher than control group [large LFA 1 (range, 0-3), p <0.001; total LFA 2 (range, 0-4), p = 0.003 and density of LFA 0.23 +/- 0.14, p = 0.001]. The means of total LFA per biopsy were 0.59 +/- 0.32) for the study group and 0.31 +/0.21 for the control group (p = 0.005). The duration of diarrhoea was correlated among others, with the following parameters: number of large LFA (p = 0.001, Spearman's rho = 0.480), number of total LFA (p = 0.003, Spearman's rho = 0.433), density of LFA (p = 0.009, Spearman's rho = 0.385), and means of total lymphoid follicles per biopsy (p = 0.004, Spearman's rho = 0.429). Conclusion: Our analysis shows that, in biopsy samples with otherwise “non-specific changes” there is a statistically significant correlation between the increase number of lymphoid follicles and aggregates and the length of the diarrhoea before and/or after the time of biopsy. These findings challenge the view of LFA as “physiological” components of the colonic mucosa above a certain threshold and, more importantly, allow the prediction of the chronic behaviour of a subset of patients unpredictable until now. In essence, this observation may indicate the presence of a new diagnostic entity, the Lymphocytic Follicle and Aggregate Colitis.
930 Gender Differences in Duodenal Bicarbonate Secretion Cause Gender Differences in Human Duodenal Ulcer Biguang Tuo, Guorong Wen, Jinqi Wei, Yalin Zhang, Hui Dong Background: It has long been observed that prevalence in duodenal ulcer is lower in women than age-matched men; however, the causes of these gender differences are currently unknown. Since duodenal mucosal bicarbonate secretion (DMBS) protects the duodenum against acid-induced injury, we hypothesize that estrogen attributes to gender differences in human duodenal ulcer via a specific stimulation of DMBS. Methods: Male or female healthy volunteers were divided into two different age-matched groups: 20 ~29 years old and 60 ~ 69 years old. The proximal duodenum was flushed and isolated. After basal HCO3secretion was observed for 60 minutes, the duodenum was perfused with 100 mM HCl or 2 mg 17β-estradiol to study the effects of duodenal acidification or estrogen on duodenal HCO3- secretion, which was measured by back titration. Protein expression of estrogen receptors was detected in human duodenal tissues by Western blot analysis. Results: Basal DMBS was significantly higher in women than in men of 20 ~ 29 years old group (189.5 ± 23.5 in women vs. 120.7 ± 16.2 µmol/cm-h in men, n=9, p<0.05). However, the basal DMBS was comparable in both genders of 60~69 years old group (123.1 ± 17.8 in women
933 Small Intestinal Bacterial Overgrowth Recurrence After Rifaximin Therapy Ernesto Cristiano Lauritano, Maurizio Gabrielli, Lucia Sparano, Alessia Cazzato, Andrea Lupascu, Maria Luisa Novi, Sandra Sottili, Giovanna Vitale, Valentina Cesario, Michele Serricchio, Giovanni Cammarota, Giovanni Gasbarrini, Giovanni Ghirlanda, Antonio Gasbarrini Objectives: Current treatment for small intestinal bacterial overgrowth (SIBO) is based on courses of broad spectrum antibiotics. No data concerning SIBO recurrence are available. The aims of the present study were to investigate SIBO recurrence as assessed by glucose
A-137
AGA Abstracts
AGA Abstracts
vs. 109.8 ± 16.2 µmol/cm-h in men, n=9, p>0.05). Luminal perfusion of HCl resulted in a robust increase in DMBS in both 20~29 years old and 60~69 years old groups. However, acid-stimulated net peak DMBS was significantly higher in women than in men of 20~29 years old group (n=9, P<0.05), whereas net peak DMBS was not significantly different between women and men in 60~69 years old (n=9, P>0.05). Further study found that luminal perfusion of 17-β-estradiol significantly increased DMBS in men and women of both 20~29 years old and 60~69 years old groups (n=9, P<0.001), indicating estrogen has a direct effect on human DMBS. Western blot analysis from human proximal duodenal tissues showed that estrogen receptor subtypes ERα and ERβ proteins were expressed at similar levels in human duodenal tissues taken by biopsy from male and female healthy subjects. Conclusion: gender differences in human DMBS exist between premenopausal women and the age-matched men, but not between postmenopausal women and the agematched men. Estrogen may be involved in the gender differences of DMBS and protect the human duodenum by activating specific estrogen receptors and then stimulating DMBS, which may explain the previously observed lower prevalence of duodenal ulcer in women.
928
AGA Abstracts
breath test (GBT) after antibiotic treatment and conditions associated to SIBO recurrence. Methods: Eighty (80) consecutive patients affected by SIBO and decontaminated by rifaximin (1200 mg per day for 1 week) were enrolled. Diagnosis of SIBO was based on GBT. GBT was reassessed at 3, 6, and 9 months after decontamination. GBT positivity recurrence, predisposing conditions and gastrointestinal symptoms were evaluated. Results: Ten (10/ 80, 12.6%), 22 (22/80, 27.5%) and 35 (35/80, 43.7%) patients showed positivity to GBT at 3, 6 and 9 months after successful antibiotic decontamination, respectively. At multivariate analysis, older age (OR 1.09, 95% CI 1.02-1.16), history of appendectomy (OR 5.9, 95% CI 1.45-24.19) and chronic use of proton pump inhibitors (OR 3.52, 95% CI 1.07-11.64) were significantly associated to GBT positivity recurrence. All gastrointestinal symptoms significantly increased at 3, 6 and 9 months in patients with evidence of GBT positivity recurrence. Conclusions: GBT positivity recurrence rate was high after antibiotic decontamination. Older age, history of appendectomy and chronic use of proton pump inhibitors were associated with GBT positivity recurrence. Patients with evidence of GBT positivity recurrence showed gastrointestinal symptoms relapse thus suggesting SIBO recurrence. Cyclic courses of non-absorbable antibiotics could be necessary in order to prevent SIBO recurrence.
significantly decreased by DSM (ANOVA adjusted for Rotavirus): Peru: DSM = 102.0 (65.5) g/Kg, placebo = 118.8 (92.5) g/Kg (p = 0.032), (14 % difference). Rotavirus positive children: DSM = 146.9 (90.1) g/Kg, placebo = 187.9 (122.1) g/Kg, (p = 0.039), (22 % difference). Malaysia: DSM = 87.9 (81.2) g/Kg, placebo = 90.7 (94.0) g/Kg, (p = 0.007) (3.1 % difference). Rotavirus positive children: DSM = 91.8 (103.0) g/kg, placebo = 184.5 (192.4) g/Kg, (p = 0.002), (50.3 % difference). Pooled studies: DSM = 94.5 (74.4) g/Kg, placebo = 104.1 (94.2) g/Kg, (p = 0.002) (9.2 % difference). Rotavirus positive children: DSM = 124.3 (98.3) g/ Kg, placebo = 186.8 (147.2) g/Kg, (p = 0.001), (33.5% difference. Median [95% IC] time to recovery: Peru: DSM = 68.2 [60.2-85.0] h, placebo = 118.9 [94.9-140.5] h, (p<0.001); Malaysia: DSM = 25.1 [20.5-29.0] h, placebo = 32.6 [27.5-39.3] h, (p<0.001). DSM improved anal irritation and was well tolerated. Conclusion: Two randomized double-blind placebocontrolled studies in 602 children with acute watery diarrhea showed that diosmectite, in association with oral rehydration salts, significantly decreases stool output and duration of diarrhea, especially in rotavirus-positive children and is well tolerated. 936
934
Determination of Urinary Collection Timeframes to Enhance Measurement of Gastrointestinal (GI) Site Specific Permeability in Adults and Children Mark E. McOmber, Robert J. Shulman
®
Evaluation of the Efficacy, Safety and Tolerability of Prucalopride (Resolor ) Given Subcutaneously in Patients Undergoing Elective Partial Colectomies Susan Galandiuk, Greet Beyens, Jannie Ausma, Lieve Vandeplassche
Background:GI permeability testing using urinary recovery of orally administered mono- and disaccharides is used commonly. Combinations of sugars are used to measure permeability in different areas of the GI tract (i.e., site specific permeability). However, human studies have not been published describing optimal collection timeframes for site specific permeability testing. This study was done to establish collection timeframes that would allow for simultaneous permeability comparisons at different gastrointestinal sites in both adults and children. Design/Methods:Twelve adults (median age 34 yr.; range 21-57) and 11 children (median age 11 yr.; 5-17 yr. old) were studied. Each subject drank 127.5 mL of a solution containing sucrose (10 g/dL), lactulose (5 g/dL), mannitol (1 g/dL), and sucralose (1 g/dL). Urine was collected over 24 hr. The first 12 hr. of collection were performed while fasting, followed by another 12 hr. of modified diet (i.e., avoiding artificial sweeteners). Sugar concentrations were measured using high-performance liquid chromatography. At each time point we calculated the ratios of the various sugars as well as the percent of the total amount of sugar ultimately recovered (% total recovered). Results:Gastric Permeability: In adults median (range) Sucrose % total recovered at 30 min. was 12% (4-36) and at 90 min. was 33% (6100). Corresponding values in children were 11% (4-34) and 31% (15-75). Small Intestinal Permeability: In adults Lactulose/Mannitol ratio peaked at a median of 11 hr (range 7-20) and Lactulose % total recovered was > 95% at a median of 14 hr (range 8-20). Corresponding values in children were 12 hr (range 9-20) and 15 hr (range 10-24). Colonic Permeability: In adults Sucralose/Lactulose ratio peaked at a median of 17 hr (range 2-26) and Sucralose % total recovered was > 95% at a median of 20 hr (range 10-24). Corresponding values in children were 17 hr (range 1-23) and 22 hr (range 14-24). Conclusions:1) Urinary collection timeframes are comparable between adults and children; 2) Gastric permeability can be assessed in 90 min of collection but optimally longer; 3) Small intestinal permeability optimally requires 12 hr of collection; 4) Colonic permeability optimally requires 17 hr of collection; 5) The relative timing of L and SL urinary recovery (i.e., Lactulose preceding Sucralose) supports the evidence that they measure small intestinal and colonic permeability, respectively.
Objective: The objective of this study was to determine a safe and effective dose of prucalopride, a selective high-affinity 5-HT4 receptor agonist, in shortening the duration of post operative ileus in patients undergoing elective partial colectomies. Methods: This was a phase IIb, parallel-group, randomized, double-blind, placebo-controlled, dose-ranging study in patients scheduled to undergo elective partial colectomies with primary anastomoses by laparotomy. Patients with concomitant diseases affecting lower gastro-intestinal motility were excluded. Patients were randomized to receive subcutaneous (s.c.) injections (1 ml) of either placebo or 0.5 mg, 2 mg or 4 mg prucalopride once daily. The first dose was administered in the recovery room, within 2 hours after surgery. Subsequent doses were administered at breakfast time for 3 days post-surgery. Assessments were made pre-operatively, in the recovery room and at each mealtime from post-operative day (POD) 1 to either POD 12 or discharge. Efficacy variables are shown in the table below. Adverse events (AEs), laboratory tests, and electrocardiograms (ECGs) were evaluated for safety purposes. Results: The intentto-treat (ITT) population consisted of 317 patients (81.4% Caucasian, 53% male, mean age 55.6±0.8 years). Efficacy results are shown in the table below. Median time to first flatus or stool was shorter in all prucalopride groups than in the placebo group, reaching statistical significance in the 4 mg group. The median time to discharge was 1 day earlier in the prucalopride groups when compared to the placebo group (p=ns). Most frequently reported AEs were nausea, vomiting, fever, and insomnia. Incidence rates of AEs, serious AEs en AEs leading to discontinuation did not differ between prucalopride and placebo groups. There were no clinically relevant differences in ECG parameters between treatment groups. Conclusions: The results of this study demonstrate that prucalopride, given subcutaneously once daily, reduces the time to first flatus or stool in patients that underwent elective partial colectomies. Moreover, patients treated with prucalopride could be discharged from the hospital 1 day earlier then patients treated with placebo. Prucalopride was safe and well tolerated at all doses studied.
937 Mortality Associated with Gastrointestinal (GI) Bleeding Events: Comparing Short-Term Clinical Outcomes of Patients Hospitalized for Upper GI Bleed and Acute Myocardial Infarction (Ami) in a US Managed Care Setting Gergana Zlateva, Henry J. Henk, Victoria Zarotsky, Byron L. Cryer, Charles M. Wilcox Background: Outside of specialist settings, there is limited awareness about the severe outcomes associated with gastrointestinal (GI) bleeds in comparison with acute myocardial infarction (AMI). GI bleeds require intensive medical intervention leading to escalated health resource utilization, costs and sometimes mortality. Understanding how mortality among patient with GI bleeding events compares to the mortality associated with an AMI may be valuable in providing context for understanding the impact of the costs and consequences of GI bleeding events. Objectives: To estimate the 30-, 60-, and 90-day mortality among patients hospitalized for upper GI-bleeds compared to a propensity-matched control cohort of patients with acute myocardial infarction (AMI). Methods: US national health plan claims data (1999-2003) were used to identify patients hospitalized with a GI bleeding event. Patient with evidence of GI bleed or AMI in the 6 months prior to the index date were excluded. These patients were propensity-matched to AMI patients with no evidence of GI bleed from the same US health plan. The propensity score model included patient demographics (e.g., age, gender), medication use, factors associated with mortality (e.g., Charlson Comorbidity Index (CCI) score), and medical care resource utilization history. Results: 12,437 upper GI-bleed patients and 22,847 AMI patients who met inclusion criteria were identified. Propensity score matching yielded 6,923 matched pairs. Prior to matching, the GI bleed sample had a higher mean CCI and higher baseline medical care resource utilization than the AMI sample. The matched cohorts were found to have a similar CCI and to be similar on nearly all utilization and cost measures (excepting ER costs). A comparison of outcomes among the matched cohorts found that AMI patients had higher rates of mortality (4.35% vs. 2.54%; p<.0001) and rehospitalization (2.56% vs. 1.79%; p=0.002), while GI bleed patients were more likely to have a repeat procedure (72.38% vs. 44.95%; p<0.001) following their initial hospitalization. The majority of the difference in overall 30-day mortality between GI bleed and AMI patients was accounted for by mortality during the initial hospital stay (1.91% vs. 3.58%); 30- 60- and 90-day mortality excluding deaths within the initial hospitalization did not differ between the cohorts. Conclusions: GI bleeding events result in significant mortality similar to that of an AMI after adjusting for the initial hospitalization. Future study will need to evaluate the cost and resource utilization associated with GI bleeding events leading to mortality.
* p<0.01 when compared with placebo 935 The Clay Diosmectite in Association with Oral Rehydration Salts Reduces Stool Output and Diarrhea Duration in Children with Acute Watery Diarrhoea: Results of 2 Randomized Double Blind Placebo-Controlled Studies Jimmy Lee Kok Foo, Christophe Dupont, Philippe Garnier, Helene Mathiex-Fortunet, Eduardo Salazar-Lindo Introduction: Study aimed at assessing the effect of the clay diosmectite (DSM) on stool output and duration of diarrhea in children with acute watery diarrhea. Patients and Methods: Two parallel double-blind, placebo-controlled studies in 1-36 months old male children were conducted in Peru (n=300) and Malaysia (n=302). Children were included if they had ≥ 3 watery stools/day for less than 72h, weight/height ratio ≥ 80 %, no intravenous rehydration need, no gross blood in stools, fever <39°C, no current treatment with antidiarrheal or antibiotic agents. Rotavirus in stool was sought. DSM dosage was 3g (1-12 months) or 6g (13-36 months) tid over at least 3 days, then half the dosage until recovery. All children received oral rehydration salts (ORS), according to WHO guidelines. Primary criterion was stool output in g/Kg of body-weight (B-W) for the first 72 h. Other criteria were time to recovery (h), stool frequency and consistency, anal irritation (%). Results: Both treatment groups in ITT population of Peru and Malaysia were similar in their demographic characteristics: mean (SD) age: Peru = 12.5 (6.1) months, Malaysia = 15.9 (8.5) months; B-W: Peru = 9.35 (1.67) Kg, Malaysia = 9.02 (2.05) Kg. ORS consumption was also similar: Peru = 1426 (983) ml, Malaysia = 1022 (674) ml. Rotavirus frequency was: 22% in Peru, 12% in Malaysia. Mean (SD) stool output for the first 72 hours was
AGA Abstracts
A-138