- Email: [email protected]
406 Famotidine for the Prevention of Peptic Ulcers in Users of Low-Dose Aspirin. Placebo-Controlled Prospective Trial [Famous Trial]
AGA Abstracts
which are associated with pain sensation ratings (p<0.01 and <0.1 respectively). Several significant associations were observed between colonic transit or rectal sensation thresholds or ratings and NPSR1 SNPs; specifically, with NPSR1 SNPs rs1379928 and rs2609234 (see table, which reports all p<0.1). Conclusions: The specific NPSR1 associations with colonic transit and rectal sensation in FGID suggest that genetic variation at this locus may be a risk factor for colonic dysfunction. Summary of associations between NPSR1 SNPs and at least 1 parameter of colonic transit and rectal sensation
SNPs in the TLR9, IL6, and CDH1 regions were significantly associated with sporadic IBS. Exploratory analyses identified independent genetic risk factors in both the serotoninergic (i.e. HTR2A and SLC6A4) and innate immunity categories (i.e. IL10) to be significantly associated with sporadic IBS (p<0.05). These were not previously associated with PI-IBS. Conclusion: This study provides a first roadmap comparing genetic risk factors for PI-IBS and sporadic IBS. While defects in bacterial recognition (TLR9), intestinal permeability (CDH1), and regulation of acute inflammatory mediators secretion (IL6) appear to be involved in PI-IBS, sporadic IBS genetic risk factors seem to include defects in the regulation of the serotoninergic pathway (HTR2A and SLC6A4) and secretion of anti-inflammatory cytokine (IL10). These results suggest that different pathways and distinct genetic risk factors contribute to PI-IBS and sporadic IBS susceptibility. Replication in other non-PI-IBS cohorts is needed. References: 1-Gastroenterology 2008; 134(4): Suppl 1:A122. 405 What Is the Risk of Recurrent Lower Gastrointestinal Bleeding with Continued Low-Dose Aspirin Therapy? An 8-Year Prospective Cohort Study Francis K. L. Chan, Jessica Ching, Kim WL Au, Kee Ka Man Carmen, Grace Wong, Dorothy K. Chow, Joseph J. Sung, Justin Wu Background: Among patients who develop lower gastrointestinal (GI) bleeding while on low-dose aspirin, the long-term risk of recurrent lower GI bleeding with continued aspirin therapy is unknown. Aim: To determine the long-term incidence of recurrent lower GI bleeding with resumption of low-dose aspirin after an episode of low-dose aspirin-induced lower GI bleeding. Methods: Patients were diagnosed to have low-dose aspirin-induced lower GI bleeding if they fulfilled all the criteria: 1. use of low-dose aspirin within one week of onset of bleeding, 2. documentation of melena or rectal bleeding by the admitting officer, 3. no ulcer, blood or coffee ground detected by upper endoscopy, 4. no bleeding from hemorrhoids, and 5. no concomitant use of NSAIDs, non-aspirin anti-platelet drugs, anticoagulants or corticosteroids. After the bleeding episode had settled, low-dose aspirin was resumed or permanently discontinued at the discretion of the attending physician. Patients who resumed aspirin were enrolled in the aspirin cohort, whereas those who discontinued aspirin were in the control cohort. The primary endpoint was recurrent lower GI bleeding, which was defined as recurrent melena, rectal bleeding, or anemia attributed to occult GI blood loss with no bleeding source identified by upper endoscopy. All cases of suspected recurrent lower GI bleeding were evaluated by a blinded adjudication committee according to pre-specified criteria. Only cases adjudicated to be recurrent lower GI bleeding were included in the primary analysis. Results: A total of 405 patients fulfilled the criteria of lowdose aspirin-induced lower GI bleeding between 2000 and 2007; 295 patients were enrolled in the aspirin cohort (median follow-up, 14.07 months; range, 0.07 - 96) and 110 patients were in the control cohort (median follow-up, 31.75 months; range, 0.27 - 96). One hundred and twenty-four cases of suspected recurrent lower GI bleeding were evaluated by the adjudication committee. The committee identified 35 cases of recurrent lower GI bleeding, 31 in the aspirin cohort and 4 in the control cohort. The cumulative incidence of recurrent lower GI bleeding in 8 years was 18.0% in the aspirin cohort and 4.3% in the control cohort (hazard ratio, 3.95; 95% CI, 1.39 - 11.20; p=0.005). Conclusion: Among patients with a history of low-dose aspirin-induced lower GI bleeding, continued aspirin therapy increases the risk of recurrent lower GI bleeding by almost 4 times compared to patients who discontinue aspirin. Acknowledgment: We thank Ms. Jessica Chong, Mr. K.F. Lee, Ms. Yawen Chan, and Ms. Envy Lee for their generous support in this project.
403 Decreased Anti-Inflammatory and Cytoprotective Molecules in Colonic Mucosa of Patients with Diarrhea-Predominant Irritable Bowel Syndrome Yehuda . Ringel, Ian M. Carroll, R. Balfour Sartor The pathogenesis of Irritable bowel syndrome (IBS) has not yet been clearly defined. Recent studies have demonstrated chronic low-grade mucosal inflammation (e.g., increased cellularity of mast cells or T lymphocytes) (Spiller 2007) and alteration in the immune system function (e.g., IL-10/IL12 ratio) (O'Mahony et al., 2005) in subgroups of patients with IBS. However, to date no specific marker/s of intestinal inflammation nor status of antiinflammatory molecules have been clearly defined in patients with IBS. Aim: to determine the levels of pro- and anti-inflammatory markers of inflammation in patients with diarrheapredominant (D-IBS). Methods: We studied ten patients who met the Rome III criteria for D-IBS and ten healthy controls. Distal colonic mucosal biopsies were collected during an unprepped unsedated flexible sigmoidoscopy. Total cellular RNA was extracted from each biopsy and used to make cDNA. The transcription levels of the cytokines TNF, TGF-β, IL-1β, IL-8, IL-10, IL-12 and the inflammation-induced cytoprotective, anti-inflammatory molecule cyclooxygenase-2 (Cox2) were assessed by quantitative real time PCR (qPCR). The level of transcription for each gene was normalized by β-actin to yield a relative transcription level. Results: A significant 2.3 fold decrease in the relative transcription level of the antiinflammatory cytokine IL-10 (P < 0.05) and a trend for an increase (10.2 fold) in the proinflammatory cytokine IL-12 (P = 0.3) were detected in mucosal samples from D-IBS patients when compared to controls. The transcription levels of the cytokines TNF, TGF- β, IL-1β and IL-8 did not differ significant between the D-IBS patients and the controls. Additionally, a significant 7.5 fold decrease in Cox2 was observed in the D-IBS patients (P < 0.05) when compared to the control samples. Conclusion: Decreased mRNA levels of IL-10 and COX2 with increased levels of IL-12 in D-IBS patients suggest a pro-inflammatory state in the colonic mucosa of patients with this disorder. Inhibition or deletion of Cox1/2 initiate or potentiate inflammation in mouse models of colitis (Morteau et al, 2000, Berg et al., 2002) demonstrating anti-inflammatory and cytoprotective roles for Cox2 in inflammatory conditions, and both prostaglandin E, a COX 2 product, and IL-10 inhibit proinflammatory cytokine expression. The observed decrease in IL-10 and Cox2 mRNA levels in D-IBS patients suggests decreased mucosal protection and enhanced susceptibility to mild colonic inflammation in this subset of patients with IBS. (Supported by K23 DK075621, MO1 RR00046)
406 Famotidine for the Prevention of Peptic Ulcers in Users of Low-Dose Aspirin. Placebo-Controlled Prospective Trial [Famous Trial] Ali S. Taha, Caroline McCloskey, Rakesh Prasad, Vladimir Bezlyak INTRODUCTION: The use of low-dose aspirin in a wide range of conditions has been increasingly associated with peptic ulcers and their complications, with limited preventative options. AIMS & METHODS: We aimed at testing the efficacy of famotidine at standard dose of 20-mg twice daily, in the prevention of esophagitis, gastric, and duodenal ulcers in patients taking low-dose aspirin (75-325 mg daily). Adult patients were recruited from the cardio-vascular, neuro-vascular or diabetes clinics if they took aspirin with or without clopidogrel or dipyridamole, and regardless of the level of their comorbidity. Patients were endoscoped at baseline and 12 weeks after taking famotidine 20-mg twice daily or its placebo. At base-line, patients were still included if they had erosions or ulcer scars, and excluded if they had active ulcers, malignancy, or if they took warfarin, NSAIDs, or antiulcer agents. An ulcer was defined as an excavated and deep mucosal lesion measuring at least 3 mm in diameter. Smaller lesions were considered as erosions. Esophagitis was graded by the LA system. RESULTS: Of 14, 515 patients screened for suitability to consider the study, 404 were randomized. The rest were either unsuitable (not taking aspirin, or having ulcers at initial endoscopy, etc), or refused to consider the trial. At baseline assessment of the randomized group (N = 404), 277 (68.6%) were males, the median (range) age was 63 [36 - 86] years, 128 (31.8%) had H pylori gastritis, 96 (23.8%) took clopidogrel and/ or dipyridamole besides their aspirin, and 354 (87.6%) had erosions or scars at baseline endoscopy. These characteristics were evenly distributed between the famotidine and placebo groups. After 12 weeks of treatment, the primary end-points are as shown in Table 1 [intention-to-treat and logistic regression analyses; odds ratios, OR; 95% confidence intervals, CI]. Also, hematemesis , melena, cardiac events, or death were observed in 4 patients (1.96%) taking famotidine and 9 (4.50%) taking placebo. CONCLUSION: Famotidine is effective in the prevention of erosive esophagitis, gastric, and duodenal ulcers in patients taking lowdose aspirin. Also, famotidine is well tolerated in patients with cardio-vascular or neurovascular diseases, or those with diabetes, who require low-dose aspirin Table 1. Primary end-points at the completion of the Trial
404 Validation of Genetic Risk Factors for Post-Infectious Irritable Bowel Syndrome (IBS) in Patients with Sporadic IBS Alexandra-Chloe Villani, Yuri A. Saito, Mathieu Lemire, Marroon Thabane, Joseph J. Larson, Elizabeth J. Atkinson, Janice Zimmerman, Ann E. Almazar Elder, Stephen M. Collins, Denis Franchimont, John K. Marshall Background & Aim: While familial aggregation and twin studies support the involvement of genetic risk factors in the pathogenesis of sporadic irritable bowel syndrome (IBS), previous association studies have failed to identify reproducible candidate genes. Since low-grade inflammation may also play a role in the pathogenesis of sporadic IBS, we aim to validate the three loci involved in post-infectious (PI) IBS that we had previously identified (i.e. TLR9, IL6, CDH1)1 and to screen 78 functional variants, previously evaluated in PI-IBS, in a cohort of non PI- (sporadic) IBS patients. Methods: The validation cohort included 473 sporadic IBS cases attending the Mayo Clinic and 464 healthy controls frequently-matched for age, gender, ethnicity and region. Clinical diagnoses of IBS were obtained from chart reviews. Sequenom hME assays were used to genotype 63 tagging SNPs (TLR9, IL6, CDH1) and the 78 functional variants related to: (1) serotoninergic pathways; (2) intestinal epithelial barrier function; and (3) innate immunity. Association analyses compared case-control allele frequencies using Chi-square tests. Results: None of the functional variants originally identified in the Walkerton PI-IBS cohort (two located in TLR9: rs352139 (coding, P545P) (p= 0.970) and rs5743836 (promoter,-T1237C) (p=0.452); one in CDH1: rs16260 (promoter,C160A) (p=0.528); and one in IL6: rs1800795 (promoter,-G174C) (p=0.673))1 and tagging
AGA Abstracts
A-68
407 Does High-Dose Famotidine Reduce Gastric and Duodenal Ulcers in NSAID Users? Two Double-Blind Six-Month Trials of Single-Tablet Combination Ibuprofen-Famotidine vs. Ibuprofen Alone (Reduce-1 and 2) Loren A. Laine, Michael Schiff, Mark Genovese, Alan Kivitz, George Tidmarsh, Michael E. Weinblatt
409 Prescription Rates of PPI Therapy Are High in Patients On Dual Antiplatelet Therapy in Spain, Regardless of the Presence of Risk Factors Ruben Casado, Monica Polo-Tomas, Alfonso Del Rio, James M. Scheiman, Angel Lanas
Proton pump inhibitors are recommended to decrease NSAID-associated GI injury, while standard-dose histamine2-receptor antagonists (H2RAs) are not. However, a Cochrane review suggested that high-dose H2RAs provide significant benefit in NSAID users. Importantly, the benefit of antisecretory therapy in patients taking NSAIDs plus low-dose aspirin is uncertain. In addition, most NSAID users at increased risk for GI events do not receive or adhere to protective co-therapy, and decreased adherence is associated with an increased risk of ulcers and bleeding. We performed two 24-wk double-blind, randomized trials to determine if a single-tablet combination of ibuprofen (800mg) and famotidine (26.6mg) (HZT-501) given thrice daily (providing 80mg famotidine daily) will significantly decrease ulcers as compared to ibuprofen alone. METHODS: Patients 40-80 yrs expected to require daily NSAID therapy ≥6 mos with no history of ulcer complications, negative H. pylori stool test, and baseline endoscopy showing no ulcers and <5 erosions in the upper GI tract were randomly assigned in a 2:1 ratio to HZT-501 or identical-appearing ibuprofen 800 mg tablets thrice daily. Concomitant aspirin ≤325 mg daily and anticoagulant therapy were permitted. Randomization was stratified based on aspirin/anticoagulant therapy and prior ulcer history. Study endoscopies were done at 8, 16, and 24 wks of therapy. The predefined population for primary analyses of ulcers was all patients with ≥1 follow-up study endoscopy. RESULTS: The REDUCE-1 and REDUCE-2 studies included 812 and 570 patients in their primary analysis populations. Results are shown in the Table. The differences for HZT-501 vs. ibuprofen in the life table estimates of the proportion of patients developing ulcers over 24 wks were significant in both REDUCE-1 (14.7% vs. 29.1%, p=0.0002) and REDUCE-2 (13.8% vs. 22.6%, p=0.030). CONCLUSIONS: High-dose H2RA delivered in single-tablet combination ibuprofen-famotidine significantly reduces NSAID-associated gastric and duodenal ulcers. A reduction in ulcers also was seen in the subset of NSAID users taking lowdose aspirin. Proportion of Patients with Ulcers at 24 Wks: Crude Proportions (n/N (%)); Life Table Estimates (%)
Background: Dual antiplatelet therapy (aspirin and clopidogrel) potentiates the risk of gastrointestinal (GI) bleeding beyond the risk observed with the individual agents in cardiovascular patients. Proton pump inhibitors (PPI) reduce the risk of ulcer bleeding in high-risk patients treated with aspirin or clopidogrel. No studies have determined whether GI preventive therapy with PPIs is being prescribed in cardiovascular clinical practice. Methods: We performed a retrospective cross-sectional study of medical records at the University Hospital in Zaragoza (Spain). Patients admitted to the hospital Cardiology Unit undergoing percutaneous coronary intervention (PCI) between January and December 2007 were eligible for inclusion. Data were extracted on: (1) patient demographics; (2) medications used; (3) prescription drugs; and, (4) risk factors for GI bleeding (age ≥ 70; history of peptic ulcer disease; concurrent anticoagulant, corticosteroids ≥ 10mg daily, or daily/intermittent NSAID use). Patients with ≥ 1 risk factor were defined as “high-risk” for GI bleeding. The proportion of patients discharged on PPI therapy was calculated and stratified by GI bleeding risk. Statistical differences and Odds ratios were calculated using Chi-square tests Results: 230 patients were included in the study (3 died during hospitalization and no valuable information was available for 6 patients). The mean age was 66.9±8.9, and 16.1% were women. At admission, 69 patients (30.0%) were on PPI therapy and 110/230 patients (47.8%) were already on antiplatelet therapy. 112/230 patients (48.7%) were at high risk for GI bleeding (96/230 = 41.7% > 70 yrs old); 52% (34/112) of these high-risk patients were on a PPI at admission. At discharge, 220/221 patients (99.5%) were on dual antiplatelet therapy, and 171 (74.3%) patients were on PPI therapy. Among the high-risk patients not admitted on PPI therapy, the opportunity to initiate therapy was missed in 18 patients (23%) (95% CI: 13.7% - 32.4%). PPI therapy rates at discharge were similar in patients with no risk factors (73.5%), 1 risk factor (71.6%) or > 1 risk factor (90.9%) (p = 0.166). Conclusions: New prescription of PPI therapy is high among patients undergoing PCI on dual antiplatelet therapy in our center, regardless of the presence of risk factors. Still, almost a quarter of high-risk patients were not started on a PPI following initiation of dual antiplatelet therapy. Future efforts need to be implemented to further improve the rates of GI prevention strategies in the high-risk population. Studies to assess the GI and CV impact of these strategies in low and high risk patients are warranted. 410 Proton Pump Inhibitor Prescription in High-Risk Cardiac Patients: A Missed Opportunity? Liz Guastello, Sameer D. Saini, Angel Lanas, James M. Scheiman Background: Patients undergoing percutaneous coronary intervention (PCI) require therapy with aspirin and clopidogrel. Though such therapy decreases the risk of cardiac events, it also increases the risk of life-threatening gastrointestinal (GI) bleeding. No study has determined whether proton-pump inhibitors (PPIs) are being used appropriately to reduce GI bleeding risk in such patients. Methods: We performed a retrospective cross-sectional study of medical records at the University of Michigan Hospital. Patients admitted to the hospital status-post PCI between August 2007 and December 2007 were eligible for inclusion. Data were extracted on: (1) patient demographics; (2) medications used; (3) indication for PPI use; and, (4) risk factors for GI bleeding (age ≥ 70; history of peptic ulcer disease; concurrent warfarin, corticosteroids ≥ 10mg daily, or daily NSAIDs). Patients with ≥ 1 risk factor were defined as “high-risk” for GI bleeding. The proportion of patients discharged on PPI therapy was calculated and stratified by GI bleeding risk. The chi-squared test was used to assess for statistical significance. Results: 199 patients were included in the study. The mean age was 63 years, and 23% were women. The median length of hospital stay was 2 days. At admission, 68 patients (34%) were on PPI therapy (56% for heartburn). 79 patients (40%) were at high risk for GI bleeding; 48% of these high-risk patients were on a PPI at admission. Highrisk patients were more likely than low-risk patients to be on PPI therapy at the time of admission (OR = 2.78, p<0.001, 95% CI 1.52-5.09). At discharge, 196 patients (98%) were on dual antiplatelet therapy, and 82 patients (41%) were on PPI therapy. Among the highrisk patients not admitted on PPI therapy, the opportunity to initiate therapy was missed in 80% (95% CI: 0.65-0.91). Overall, 42% of high-risk patients were discharged without PPI, though high-risk patients were more likely than low-risk patients to be discharged on a PPI (OR = 3.25, p<0.001, 95% CI: 1.80-5.89). Conclusions: While high-risk patients were significantly more likely to be discharged on PPI therapy than low risk patients, 80% of eligible high-risk patients were not started on a PPI following initiation of dual antiplatelet therapy. Such patients represent a “missed opportunity” for intervention. Future efforts to educate clinicians on the importance of utilizing PPIs for bleeding prophylaxis in high-risk patients are warranted.
*p<0.05, HZT-501 vs. ibuprofen (separate comparisons for crude proportions and life table estimates) 408 Increased Aspirin Use and Upper Gastrointestinal Bleeding Rates in Socially Deprived Patients Yana Vinogradova, Matthew Leighton, Anthony Avery, Richard F. Logan, Denise Kendrick, John C. Atherton, Anthony Shonde, Julia Hippisley-Cox, Chris J. Hawkey INTRODUCTION: Cardiovascular morbidity is associated with social deprivation but it is not clear whether this results in greater use of aspirin nor whether more aspirin associated gastro toxicity results. QRESEARCH is a new larger database that includes information on deprivation. AIMS & METHODS: We used QRESEARCH to investigate aspirin use in relation to the Townsend Index of deprivation and to co-morbidity and whether this was associated with a higher incidence of gastrointestinal bleeding. All patients aged 45-100 years with continuous aspirin use (definition: new prescription ≤8 weeks after previous) at any time during the period 1 April 2003 - 31 March 2007 were included. RESULTS: Four hundred and fifty nine practices met inclusion criteria to generate 4.87 million person years of observation. Over 99% of aspirin use was for ≤300mg per day. Aspirin use was higher at younger ages in men but similar between sexes by age 80. It increased from 9.9% in 2003 to 14% in 2007. Use was higher in deprived compared to affluent areas (11.8% in lowest quintile versus 8.6% in highest quintile in 2003, 15.9% and 12.5% in 2007). Of patients with coronary heart disease 55.2% were prescribed aspirin in 2003 and 59.1% in 2007,
A-69
AGA Abstracts
AGA Abstracts
50.8% and 54.2% for stroke, 33.7% and 44.1% for diabetes, 24.3% and 29.2% for hypertension. Coronary heart disease was more common in deprived than affluent areas (12.0% versus 8.8%). The incidence of upper gastrointestinal bleeding rose with age (from 44.3 per 100,000 at aged 60 to 325.2 per 100,000 at age 85 for women and 62.7 per 100,000 and 378.4 per 100,000 for men) but not over the four years of the study (80.1 per 100,000 in 2003/4, 72.3 per 100,000 in 2006/7) more men than women were affected and rates were higher in deprived than affluent areas (100.9 per 100,000 versus 58.3 per 100,000). Upper gastrointestinal bleeding was more common in patients with cardiovascular disease, diabetes or hypertension. Aspirin increased the risk of upper gastrointestinal bleeding, even after adjustment for age, sex, deprivation and co-morbidities (1.59, 1.47-1.73). CONCLUSION: Patients in deprived areas are prescribed more aspirin than in affluent areas and have higher rates of upper gastrointestinal bleeding. Aspirin may contribute to this. However, upper gastrointestinal bleeding appears not to be increasing despite increased use of aspirin.
which are associated with pain sensation ratings (p<0.01 and <0.1 respectively). Several significant associations were observed between colonic transit or rectal sensation thresholds or ratings and NPSR1 SNPs; specifically, with NPSR1 SNPs rs1379928 and rs2609234 (see table, which reports all p<0.1). Conclusions: The specific NPSR1 associations with colonic transit and rectal sensation in FGID suggest that genetic variation at this locus may be a risk factor for colonic dysfunction. Summary of associations between NPSR1 SNPs and at least 1 parameter of colonic transit and rectal sensation
SNPs in the TLR9, IL6, and CDH1 regions were significantly associated with sporadic IBS. Exploratory analyses identified independent genetic risk factors in both the serotoninergic (i.e. HTR2A and SLC6A4) and innate immunity categories (i.e. IL10) to be significantly associated with sporadic IBS (p<0.05). These were not previously associated with PI-IBS. Conclusion: This study provides a first roadmap comparing genetic risk factors for PI-IBS and sporadic IBS. While defects in bacterial recognition (TLR9), intestinal permeability (CDH1), and regulation of acute inflammatory mediators secretion (IL6) appear to be involved in PI-IBS, sporadic IBS genetic risk factors seem to include defects in the regulation of the serotoninergic pathway (HTR2A and SLC6A4) and secretion of anti-inflammatory cytokine (IL10). These results suggest that different pathways and distinct genetic risk factors contribute to PI-IBS and sporadic IBS susceptibility. Replication in other non-PI-IBS cohorts is needed. References: 1-Gastroenterology 2008; 134(4): Suppl 1:A122. 405 What Is the Risk of Recurrent Lower Gastrointestinal Bleeding with Continued Low-Dose Aspirin Therapy? An 8-Year Prospective Cohort Study Francis K. L. Chan, Jessica Ching, Kim WL Au, Kee Ka Man Carmen, Grace Wong, Dorothy K. Chow, Joseph J. Sung, Justin Wu Background: Among patients who develop lower gastrointestinal (GI) bleeding while on low-dose aspirin, the long-term risk of recurrent lower GI bleeding with continued aspirin therapy is unknown. Aim: To determine the long-term incidence of recurrent lower GI bleeding with resumption of low-dose aspirin after an episode of low-dose aspirin-induced lower GI bleeding. Methods: Patients were diagnosed to have low-dose aspirin-induced lower GI bleeding if they fulfilled all the criteria: 1. use of low-dose aspirin within one week of onset of bleeding, 2. documentation of melena or rectal bleeding by the admitting officer, 3. no ulcer, blood or coffee ground detected by upper endoscopy, 4. no bleeding from hemorrhoids, and 5. no concomitant use of NSAIDs, non-aspirin anti-platelet drugs, anticoagulants or corticosteroids. After the bleeding episode had settled, low-dose aspirin was resumed or permanently discontinued at the discretion of the attending physician. Patients who resumed aspirin were enrolled in the aspirin cohort, whereas those who discontinued aspirin were in the control cohort. The primary endpoint was recurrent lower GI bleeding, which was defined as recurrent melena, rectal bleeding, or anemia attributed to occult GI blood loss with no bleeding source identified by upper endoscopy. All cases of suspected recurrent lower GI bleeding were evaluated by a blinded adjudication committee according to pre-specified criteria. Only cases adjudicated to be recurrent lower GI bleeding were included in the primary analysis. Results: A total of 405 patients fulfilled the criteria of lowdose aspirin-induced lower GI bleeding between 2000 and 2007; 295 patients were enrolled in the aspirin cohort (median follow-up, 14.07 months; range, 0.07 - 96) and 110 patients were in the control cohort (median follow-up, 31.75 months; range, 0.27 - 96). One hundred and twenty-four cases of suspected recurrent lower GI bleeding were evaluated by the adjudication committee. The committee identified 35 cases of recurrent lower GI bleeding, 31 in the aspirin cohort and 4 in the control cohort. The cumulative incidence of recurrent lower GI bleeding in 8 years was 18.0% in the aspirin cohort and 4.3% in the control cohort (hazard ratio, 3.95; 95% CI, 1.39 - 11.20; p=0.005). Conclusion: Among patients with a history of low-dose aspirin-induced lower GI bleeding, continued aspirin therapy increases the risk of recurrent lower GI bleeding by almost 4 times compared to patients who discontinue aspirin. Acknowledgment: We thank Ms. Jessica Chong, Mr. K.F. Lee, Ms. Yawen Chan, and Ms. Envy Lee for their generous support in this project.
403 Decreased Anti-Inflammatory and Cytoprotective Molecules in Colonic Mucosa of Patients with Diarrhea-Predominant Irritable Bowel Syndrome Yehuda . Ringel, Ian M. Carroll, R. Balfour Sartor The pathogenesis of Irritable bowel syndrome (IBS) has not yet been clearly defined. Recent studies have demonstrated chronic low-grade mucosal inflammation (e.g., increased cellularity of mast cells or T lymphocytes) (Spiller 2007) and alteration in the immune system function (e.g., IL-10/IL12 ratio) (O'Mahony et al., 2005) in subgroups of patients with IBS. However, to date no specific marker/s of intestinal inflammation nor status of antiinflammatory molecules have been clearly defined in patients with IBS. Aim: to determine the levels of pro- and anti-inflammatory markers of inflammation in patients with diarrheapredominant (D-IBS). Methods: We studied ten patients who met the Rome III criteria for D-IBS and ten healthy controls. Distal colonic mucosal biopsies were collected during an unprepped unsedated flexible sigmoidoscopy. Total cellular RNA was extracted from each biopsy and used to make cDNA. The transcription levels of the cytokines TNF, TGF-β, IL-1β, IL-8, IL-10, IL-12 and the inflammation-induced cytoprotective, anti-inflammatory molecule cyclooxygenase-2 (Cox2) were assessed by quantitative real time PCR (qPCR). The level of transcription for each gene was normalized by β-actin to yield a relative transcription level. Results: A significant 2.3 fold decrease in the relative transcription level of the antiinflammatory cytokine IL-10 (P < 0.05) and a trend for an increase (10.2 fold) in the proinflammatory cytokine IL-12 (P = 0.3) were detected in mucosal samples from D-IBS patients when compared to controls. The transcription levels of the cytokines TNF, TGF- β, IL-1β and IL-8 did not differ significant between the D-IBS patients and the controls. Additionally, a significant 7.5 fold decrease in Cox2 was observed in the D-IBS patients (P < 0.05) when compared to the control samples. Conclusion: Decreased mRNA levels of IL-10 and COX2 with increased levels of IL-12 in D-IBS patients suggest a pro-inflammatory state in the colonic mucosa of patients with this disorder. Inhibition or deletion of Cox1/2 initiate or potentiate inflammation in mouse models of colitis (Morteau et al, 2000, Berg et al., 2002) demonstrating anti-inflammatory and cytoprotective roles for Cox2 in inflammatory conditions, and both prostaglandin E, a COX 2 product, and IL-10 inhibit proinflammatory cytokine expression. The observed decrease in IL-10 and Cox2 mRNA levels in D-IBS patients suggests decreased mucosal protection and enhanced susceptibility to mild colonic inflammation in this subset of patients with IBS. (Supported by K23 DK075621, MO1 RR00046)
406 Famotidine for the Prevention of Peptic Ulcers in Users of Low-Dose Aspirin. Placebo-Controlled Prospective Trial [Famous Trial] Ali S. Taha, Caroline McCloskey, Rakesh Prasad, Vladimir Bezlyak INTRODUCTION: The use of low-dose aspirin in a wide range of conditions has been increasingly associated with peptic ulcers and their complications, with limited preventative options. AIMS & METHODS: We aimed at testing the efficacy of famotidine at standard dose of 20-mg twice daily, in the prevention of esophagitis, gastric, and duodenal ulcers in patients taking low-dose aspirin (75-325 mg daily). Adult patients were recruited from the cardio-vascular, neuro-vascular or diabetes clinics if they took aspirin with or without clopidogrel or dipyridamole, and regardless of the level of their comorbidity. Patients were endoscoped at baseline and 12 weeks after taking famotidine 20-mg twice daily or its placebo. At base-line, patients were still included if they had erosions or ulcer scars, and excluded if they had active ulcers, malignancy, or if they took warfarin, NSAIDs, or antiulcer agents. An ulcer was defined as an excavated and deep mucosal lesion measuring at least 3 mm in diameter. Smaller lesions were considered as erosions. Esophagitis was graded by the LA system. RESULTS: Of 14, 515 patients screened for suitability to consider the study, 404 were randomized. The rest were either unsuitable (not taking aspirin, or having ulcers at initial endoscopy, etc), or refused to consider the trial. At baseline assessment of the randomized group (N = 404), 277 (68.6%) were males, the median (range) age was 63 [36 - 86] years, 128 (31.8%) had H pylori gastritis, 96 (23.8%) took clopidogrel and/ or dipyridamole besides their aspirin, and 354 (87.6%) had erosions or scars at baseline endoscopy. These characteristics were evenly distributed between the famotidine and placebo groups. After 12 weeks of treatment, the primary end-points are as shown in Table 1 [intention-to-treat and logistic regression analyses; odds ratios, OR; 95% confidence intervals, CI]. Also, hematemesis , melena, cardiac events, or death were observed in 4 patients (1.96%) taking famotidine and 9 (4.50%) taking placebo. CONCLUSION: Famotidine is effective in the prevention of erosive esophagitis, gastric, and duodenal ulcers in patients taking lowdose aspirin. Also, famotidine is well tolerated in patients with cardio-vascular or neurovascular diseases, or those with diabetes, who require low-dose aspirin Table 1. Primary end-points at the completion of the Trial
404 Validation of Genetic Risk Factors for Post-Infectious Irritable Bowel Syndrome (IBS) in Patients with Sporadic IBS Alexandra-Chloe Villani, Yuri A. Saito, Mathieu Lemire, Marroon Thabane, Joseph J. Larson, Elizabeth J. Atkinson, Janice Zimmerman, Ann E. Almazar Elder, Stephen M. Collins, Denis Franchimont, John K. Marshall Background & Aim: While familial aggregation and twin studies support the involvement of genetic risk factors in the pathogenesis of sporadic irritable bowel syndrome (IBS), previous association studies have failed to identify reproducible candidate genes. Since low-grade inflammation may also play a role in the pathogenesis of sporadic IBS, we aim to validate the three loci involved in post-infectious (PI) IBS that we had previously identified (i.e. TLR9, IL6, CDH1)1 and to screen 78 functional variants, previously evaluated in PI-IBS, in a cohort of non PI- (sporadic) IBS patients. Methods: The validation cohort included 473 sporadic IBS cases attending the Mayo Clinic and 464 healthy controls frequently-matched for age, gender, ethnicity and region. Clinical diagnoses of IBS were obtained from chart reviews. Sequenom hME assays were used to genotype 63 tagging SNPs (TLR9, IL6, CDH1) and the 78 functional variants related to: (1) serotoninergic pathways; (2) intestinal epithelial barrier function; and (3) innate immunity. Association analyses compared case-control allele frequencies using Chi-square tests. Results: None of the functional variants originally identified in the Walkerton PI-IBS cohort (two located in TLR9: rs352139 (coding, P545P) (p= 0.970) and rs5743836 (promoter,-T1237C) (p=0.452); one in CDH1: rs16260 (promoter,C160A) (p=0.528); and one in IL6: rs1800795 (promoter,-G174C) (p=0.673))1 and tagging
AGA Abstracts
A-68
407 Does High-Dose Famotidine Reduce Gastric and Duodenal Ulcers in NSAID Users? Two Double-Blind Six-Month Trials of Single-Tablet Combination Ibuprofen-Famotidine vs. Ibuprofen Alone (Reduce-1 and 2) Loren A. Laine, Michael Schiff, Mark Genovese, Alan Kivitz, George Tidmarsh, Michael E. Weinblatt
409 Prescription Rates of PPI Therapy Are High in Patients On Dual Antiplatelet Therapy in Spain, Regardless of the Presence of Risk Factors Ruben Casado, Monica Polo-Tomas, Alfonso Del Rio, James M. Scheiman, Angel Lanas
Proton pump inhibitors are recommended to decrease NSAID-associated GI injury, while standard-dose histamine2-receptor antagonists (H2RAs) are not. However, a Cochrane review suggested that high-dose H2RAs provide significant benefit in NSAID users. Importantly, the benefit of antisecretory therapy in patients taking NSAIDs plus low-dose aspirin is uncertain. In addition, most NSAID users at increased risk for GI events do not receive or adhere to protective co-therapy, and decreased adherence is associated with an increased risk of ulcers and bleeding. We performed two 24-wk double-blind, randomized trials to determine if a single-tablet combination of ibuprofen (800mg) and famotidine (26.6mg) (HZT-501) given thrice daily (providing 80mg famotidine daily) will significantly decrease ulcers as compared to ibuprofen alone. METHODS: Patients 40-80 yrs expected to require daily NSAID therapy ≥6 mos with no history of ulcer complications, negative H. pylori stool test, and baseline endoscopy showing no ulcers and <5 erosions in the upper GI tract were randomly assigned in a 2:1 ratio to HZT-501 or identical-appearing ibuprofen 800 mg tablets thrice daily. Concomitant aspirin ≤325 mg daily and anticoagulant therapy were permitted. Randomization was stratified based on aspirin/anticoagulant therapy and prior ulcer history. Study endoscopies were done at 8, 16, and 24 wks of therapy. The predefined population for primary analyses of ulcers was all patients with ≥1 follow-up study endoscopy. RESULTS: The REDUCE-1 and REDUCE-2 studies included 812 and 570 patients in their primary analysis populations. Results are shown in the Table. The differences for HZT-501 vs. ibuprofen in the life table estimates of the proportion of patients developing ulcers over 24 wks were significant in both REDUCE-1 (14.7% vs. 29.1%, p=0.0002) and REDUCE-2 (13.8% vs. 22.6%, p=0.030). CONCLUSIONS: High-dose H2RA delivered in single-tablet combination ibuprofen-famotidine significantly reduces NSAID-associated gastric and duodenal ulcers. A reduction in ulcers also was seen in the subset of NSAID users taking lowdose aspirin. Proportion of Patients with Ulcers at 24 Wks: Crude Proportions (n/N (%)); Life Table Estimates (%)
Background: Dual antiplatelet therapy (aspirin and clopidogrel) potentiates the risk of gastrointestinal (GI) bleeding beyond the risk observed with the individual agents in cardiovascular patients. Proton pump inhibitors (PPI) reduce the risk of ulcer bleeding in high-risk patients treated with aspirin or clopidogrel. No studies have determined whether GI preventive therapy with PPIs is being prescribed in cardiovascular clinical practice. Methods: We performed a retrospective cross-sectional study of medical records at the University Hospital in Zaragoza (Spain). Patients admitted to the hospital Cardiology Unit undergoing percutaneous coronary intervention (PCI) between January and December 2007 were eligible for inclusion. Data were extracted on: (1) patient demographics; (2) medications used; (3) prescription drugs; and, (4) risk factors for GI bleeding (age ≥ 70; history of peptic ulcer disease; concurrent anticoagulant, corticosteroids ≥ 10mg daily, or daily/intermittent NSAID use). Patients with ≥ 1 risk factor were defined as “high-risk” for GI bleeding. The proportion of patients discharged on PPI therapy was calculated and stratified by GI bleeding risk. Statistical differences and Odds ratios were calculated using Chi-square tests Results: 230 patients were included in the study (3 died during hospitalization and no valuable information was available for 6 patients). The mean age was 66.9±8.9, and 16.1% were women. At admission, 69 patients (30.0%) were on PPI therapy and 110/230 patients (47.8%) were already on antiplatelet therapy. 112/230 patients (48.7%) were at high risk for GI bleeding (96/230 = 41.7% > 70 yrs old); 52% (34/112) of these high-risk patients were on a PPI at admission. At discharge, 220/221 patients (99.5%) were on dual antiplatelet therapy, and 171 (74.3%) patients were on PPI therapy. Among the high-risk patients not admitted on PPI therapy, the opportunity to initiate therapy was missed in 18 patients (23%) (95% CI: 13.7% - 32.4%). PPI therapy rates at discharge were similar in patients with no risk factors (73.5%), 1 risk factor (71.6%) or > 1 risk factor (90.9%) (p = 0.166). Conclusions: New prescription of PPI therapy is high among patients undergoing PCI on dual antiplatelet therapy in our center, regardless of the presence of risk factors. Still, almost a quarter of high-risk patients were not started on a PPI following initiation of dual antiplatelet therapy. Future efforts need to be implemented to further improve the rates of GI prevention strategies in the high-risk population. Studies to assess the GI and CV impact of these strategies in low and high risk patients are warranted. 410 Proton Pump Inhibitor Prescription in High-Risk Cardiac Patients: A Missed Opportunity? Liz Guastello, Sameer D. Saini, Angel Lanas, James M. Scheiman Background: Patients undergoing percutaneous coronary intervention (PCI) require therapy with aspirin and clopidogrel. Though such therapy decreases the risk of cardiac events, it also increases the risk of life-threatening gastrointestinal (GI) bleeding. No study has determined whether proton-pump inhibitors (PPIs) are being used appropriately to reduce GI bleeding risk in such patients. Methods: We performed a retrospective cross-sectional study of medical records at the University of Michigan Hospital. Patients admitted to the hospital status-post PCI between August 2007 and December 2007 were eligible for inclusion. Data were extracted on: (1) patient demographics; (2) medications used; (3) indication for PPI use; and, (4) risk factors for GI bleeding (age ≥ 70; history of peptic ulcer disease; concurrent warfarin, corticosteroids ≥ 10mg daily, or daily NSAIDs). Patients with ≥ 1 risk factor were defined as “high-risk” for GI bleeding. The proportion of patients discharged on PPI therapy was calculated and stratified by GI bleeding risk. The chi-squared test was used to assess for statistical significance. Results: 199 patients were included in the study. The mean age was 63 years, and 23% were women. The median length of hospital stay was 2 days. At admission, 68 patients (34%) were on PPI therapy (56% for heartburn). 79 patients (40%) were at high risk for GI bleeding; 48% of these high-risk patients were on a PPI at admission. Highrisk patients were more likely than low-risk patients to be on PPI therapy at the time of admission (OR = 2.78, p<0.001, 95% CI 1.52-5.09). At discharge, 196 patients (98%) were on dual antiplatelet therapy, and 82 patients (41%) were on PPI therapy. Among the highrisk patients not admitted on PPI therapy, the opportunity to initiate therapy was missed in 80% (95% CI: 0.65-0.91). Overall, 42% of high-risk patients were discharged without PPI, though high-risk patients were more likely than low-risk patients to be discharged on a PPI (OR = 3.25, p<0.001, 95% CI: 1.80-5.89). Conclusions: While high-risk patients were significantly more likely to be discharged on PPI therapy than low risk patients, 80% of eligible high-risk patients were not started on a PPI following initiation of dual antiplatelet therapy. Such patients represent a “missed opportunity” for intervention. Future efforts to educate clinicians on the importance of utilizing PPIs for bleeding prophylaxis in high-risk patients are warranted.
*p<0.05, HZT-501 vs. ibuprofen (separate comparisons for crude proportions and life table estimates) 408 Increased Aspirin Use and Upper Gastrointestinal Bleeding Rates in Socially Deprived Patients Yana Vinogradova, Matthew Leighton, Anthony Avery, Richard F. Logan, Denise Kendrick, John C. Atherton, Anthony Shonde, Julia Hippisley-Cox, Chris J. Hawkey INTRODUCTION: Cardiovascular morbidity is associated with social deprivation but it is not clear whether this results in greater use of aspirin nor whether more aspirin associated gastro toxicity results. QRESEARCH is a new larger database that includes information on deprivation. AIMS & METHODS: We used QRESEARCH to investigate aspirin use in relation to the Townsend Index of deprivation and to co-morbidity and whether this was associated with a higher incidence of gastrointestinal bleeding. All patients aged 45-100 years with continuous aspirin use (definition: new prescription ≤8 weeks after previous) at any time during the period 1 April 2003 - 31 March 2007 were included. RESULTS: Four hundred and fifty nine practices met inclusion criteria to generate 4.87 million person years of observation. Over 99% of aspirin use was for ≤300mg per day. Aspirin use was higher at younger ages in men but similar between sexes by age 80. It increased from 9.9% in 2003 to 14% in 2007. Use was higher in deprived compared to affluent areas (11.8% in lowest quintile versus 8.6% in highest quintile in 2003, 15.9% and 12.5% in 2007). Of patients with coronary heart disease 55.2% were prescribed aspirin in 2003 and 59.1% in 2007,
A-69
AGA Abstracts
AGA Abstracts
50.8% and 54.2% for stroke, 33.7% and 44.1% for diabetes, 24.3% and 29.2% for hypertension. Coronary heart disease was more common in deprived than affluent areas (12.0% versus 8.8%). The incidence of upper gastrointestinal bleeding rose with age (from 44.3 per 100,000 at aged 60 to 325.2 per 100,000 at age 85 for women and 62.7 per 100,000 and 378.4 per 100,000 for men) but not over the four years of the study (80.1 per 100,000 in 2003/4, 72.3 per 100,000 in 2006/7) more men than women were affected and rates were higher in deprived than affluent areas (100.9 per 100,000 versus 58.3 per 100,000). Upper gastrointestinal bleeding was more common in patients with cardiovascular disease, diabetes or hypertension. Aspirin increased the risk of upper gastrointestinal bleeding, even after adjustment for age, sex, deprivation and co-morbidities (1.59, 1.47-1.73). CONCLUSION: Patients in deprived areas are prescribed more aspirin than in affluent areas and have higher rates of upper gastrointestinal bleeding. Aspirin may contribute to this. However, upper gastrointestinal bleeding appears not to be increasing despite increased use of aspirin.