April 2000 Results: STEP DOWN resulted in more symptom free months (7.01 per year) when compared to STEP UP (6.75 per year). Cost per patient and incremental cost per symptom free month for each country was calculated (TABLE). Conclusion: Initial PPI therapy results in more symptom free months than initial therapy with an H2RA. This clinical advantage is achieved at an incremental cost largely dependent on local differences in pharmaceutical prices. TREATMENT COSTS AND INCREMENTAL COST EFFECTIVENESS(ICE) COUNTRY STEP·OOWN IUS$) STEP·UP IUS$) ICEIUS$) USA 2237 2103 517 FRANCE 1851 1798 204 GERMANY 1730 1671 227 ITALY 1854 1802 199 SPAIN 1761 1709 201 UK 1721 1681 156 1321 USE OF RESOURCES IN DYSPEPSIA WITH AND WITHOUT PRE-ENDOSCOPY; A 1 YEAR FOLLOW-UP IN GENERAL PRAC- TICE. A NATURALISTIC TRIAL. Juergen F. Riemann, H. D. Allescher, Guido Adler, 1. Hartung, M. P. Manns, M. Wienbeck, M. Classen, P. Enck, Clin, Ludwigshafen, Germany; Tech Univ, Muenchen, Germany; Univ of Ulm, Ulm, Germany; Institute for statistics, Univ Dortmund, Dortmund, Germany; Med Dept, Univ Hannover, Hannover, Germany; III Med Dept, Zentralklinikum Augsburg, Augsburg, Germany; liMed Clin, TU Muenchen, Meunchen, Germany; Dept of Surg, Univ Tuebingen, Tuebingen, Germany. Earlier studies have suggested that withholding early endoscopy to dys- peptic patients will lead to increased consumption of medical services and medication. Others concluded that selecting patients for endoscopy on the basis of alarm signs and symptoms suggestive of organic disease would be a more cost-effective approach. Materials and Methods: PRESTO is a naturalistic study, following adult dyspeptic patients with a standardized questionnaire 1, 6, and 12 months after initial diagnosis, involving symp- toms, use of resources, and well-being data (VAS, PGWBI). In this trial 983 centers participated with 3000 dyspeptic patients which, based on the judgement of the treating physician, were either pre-endoscoped and then treated (ET) or directly given an empirical pharmacotherapy (PT). Another subgroup received no initial therapy (O'I'), Results: The table summarizes the results. At the end of one year follow-up the mean number of unplanned visits per patient because of dyspeptic symptoms was 0.76 (PT), 0.69 (ET) and 0.52 (O'T), After one month the number of endoscopies was higher in ET than in PT but became similar after six months. The overall number of organic diseases of the digestive tract diagnosed in the follow-up period was small. Regarding additional diagnostic work-up significantly more labs and sonographies were performed in ET than in PT or O'T, After 6 months pharmacotherapy was continued more often in ET than in PT (p<0.05), this difference becoming smaller after 12 months. Conclusions: In this study the use of medical resources and services during one year follow-up was smaller in the PT and O'T groups than in ET. These results suggest that in the majority of patients presenting with symptoms of dyspepsia, starting of a tentative treatment or simply patient s counseling before proceeding to invasive diagnostic procedures is cost-effective. n Recurrent Endoscopy Or- Lab X·ray Sono Continued Work dyspeptic ganlc pharma· days symptoms dis. cotherapy lost 1-6 6·12 1st 2-6 1·6 6·12 mths mths ET 626 17% 15% 11.7% 1.9% 1.6% 6.9%' 2.6% 3.4%' 28% 21% 2.0% PT 1945 19% 17% 2.3% 2.5% 2.0% 3.7% 2.0% 1.8% 23% 20% 1.3% OT 430 14% 12% 2.8% 2.3% 2.2% 2.8% 2.8% 2.5% 11% 9% 1.3% 'significant AGAA217 1322 TREATMENT OF ACUTE ULCER BLEEDING BASED ON DOPP- LER CLASSIFICATION - CLINICAL OUTCOMES. Anika Rosenbaum, Claus Benz, Joachim C. Arnold, Henning E. Adamek, Bernd Kohler, Juergen F. Riemann, Clin Ludwigshafen, Ludwigshafen, Germany; Clin Bruchsal, Bruchsal, Germany. Clinical parameters and initial endoscopic findings are used to assess the risk of recurrence in bleeding peptic ulcers and to determine the appropri- ate treatment. However, visual assessment of bleeding lesions using the Forrest classification (FI - FIll) shows high interobserver variability. Therefore the Doppler classification has been introduced. The aim of our study was to evaluate a treatment strategy based mainly on Doppler findings. Patients I Methods: We retrospectively analyzed 123 patients who had been treated for bleeding peptic ulcers from 1996 to 1999. All cases had been managed according to the following algorithm: Emergency endoscopy immediately after admission, assessment of Forrest and Doppler stages, and treatment as follows. Actively bleeding lesions (FI I 01) were treated endoscopically. Lesions with stigmata of recent bleeding (FII) or no stigmata (FIll) gave either a positive (Dll) or a negative (DIII) signal on Doppler examination. Every Dll lesion was treated endoscopically regard- less to the Forrest stage. After endoscopic intervention a follow-up endos- copy was performed, including re-treatment if the lesion still proved Doppler positive. Treatment was repeated until the Doppler signal disap- peared. Patients with DIII lesions received drug therapy and clinical observation for signs of rebleeding. We assessed rebleeding, surgery, and bleeding related mortality, during hospitalization and during a follow-up of 2-46 months. Results: The table summarizes the results. Conclusion: With this concept no long-term recurrence was seen and the rates of surgery and bleeding related mortality were very low compared to previous treatment studies. Endoscopic Doppler ultrasound improves safety and effectiveness in the management of acute ulcer bleeding. n Rebleedlng Surgery Bleeding related mortality acute follow·up 01 56 9(16.1%) 0 2(3.6%) 2(3.6%) 011 27 7(25.9%) 0 2(7.5%) 0 0111 24 2(8.3%) 0 0 0 NoDoppler on 16 2(12.5%) 0 0 0 flrstexam Total 123 20 (16.3%) 4(3.2%) 2(1.6%) 1323 FULL PUBLICATION OF ABSTRACTS PRESENTED AT THE BRITISH SOCIETY OF GASTROENTEROLOGY. David S. Sanders, Martyn J. Carter, Paul Hurlstone, Nigel Hoggard, Alan J. Lobo, The Gastroenterology and Liver Unit, Sheffield, United Kingdom. Background & Aims: The presentation of abstracts at scientific meetings is a means of rapidly conveying the results of novel research and provides the opportunity for peer review prior to submission for publication. The quality of such a meeting is likely to be reflected in the rate of subsequent publication in peer-reviewed, indexed journals and in the impact factor of those journals. Our aim was to determine the publication rate of abstracts presented at a single BSG. meeting. Methods: All abstracts presented at the British Society of Gastroenterology (BSG) meeting of March 1994 (n=255) were assessed for subsequent publication. A PubMed search was performed encompassing 1993 to present using cross-referencing of first author, senior author and at least one key word. The abstract and subse- quent paper were then examined in tandem to ensure they represented the same study. Data regarding the study type (as either clinical or basic science), significance of findings, sample size, journal of publication, impact factor (IF) and lag time to full publication were recorded. Data on whether the abstract was submitted/published at the AGA meeting of the same year were also collected. Statistical analyses were performed using contingency tables and chi-squared statistics Results: 165 abstracts (64.7%) were subsequently published. Median lag-time to full publication was 20 months. 58(22.7%) abstracts were published in high IF journals (IF>4). 92(36%) abstracts were published in gastroenterology journals. 31(12%) were accepted by GUT. There were no statistical differences between publication rates when sub-dividing abstracts according to type, category or sample size. Acceptance at the AGA significantly increased the likeli- hood of publication (p = 0.002; Odds Ratio 2.8 (95% CI:1.5-5.3) Conclu- sion: Acceptance of abstracts by the BSG meeting suggests almost a 2 in 3 chance of subsequent full publication, which compares favorably with previous published series.