⁎4702 National survey of management in peptic ulcer hemorrhage.

*4702 NATIONAL SURVEY OF MANAGEMENT IN PEPTIC ULCER HEMORRHAGE. Monique E. Leerdam, Erik A. Rauws, Alfons A. Geraedts, Guido N. Tytgat, Dept of Gastroenterology, Acad Med Ctr, Amsterdam, Netherlands; Dept of Gastroenterology, Onze Lieve Vrouwe Gasthuis, Amsterdam, Netherlands. Aim:Evaluation of management in patients with peptic ulcer hemorrhage (PUH) in the Netherlands. Methods: A questionnaire was sent to gastroenterologists and internists, performing endoscopy, in every hospital in the Netherlands (N=123). Endoscopic therapy, acid suppressant therapy, endoscopic re-intervention and management of Helicobacter pylori were evaluated. Results: 90/123 (73%) questionnaires were returned. Endoscopic hemostatic therapy is given in ulcers classified as Forrest Ia/ Ib/ IIa/ IIb by respectively 89%/ 93%/ 83% and 47% of responders. Gastroenterologists perform more often endoscopic therapy in Forrest Ib (p=0.03), IIa (p=0.002) and IIb (p=0.001) ulcers compared to internists. As first modality, endoscopic injection therapy is used by 93%. Adrenaline combined with polidocanol is used in 60%, adrenaline alone in 14%, polidocanol alone in 6%, adrenaline and histoacryl/ thrombine/ fibrine/ alchohol in 7%, 4%, 1% and 1% respectively. 68% of responders add a sclerosant to adrenaline injections. Acid suppressant therapy is given by 97% of responders; 71% preferred proton pump inhibitors, 26% H2-receptorantagonists, both mainly given intravenously (80% and 92% respectively). In case of suspected rebleeding 71% perform endoscopic re-intervention (89% of gastroenterologists vs. 60% of internists, p=0.005), whereas the others refer the patient directly for surgery. Detection of H. pylori is performed by almost all responders. Eradication is confirmed by only 64% (80% of gastroenterologists vs. 50% of internists, p=0.004). Conclusions: There are important differences in management of PUH between gastroenterologists and internists in the Netherlands. Despite consensus conferences and results of randomised trials, a significant percentages of patients with PUH are still not adequately treated endoscopically, nor is H. pylori eradication confirmed in all patients. *4703 EFFECTIVENESS OF ENDOSCOPY IN CRITICALLY ILL PATIENTS WITH UPPER GASTROINTESTINAL HEMORRHAGE. Amitabh Chak, Gregory S. Cooper, Lynne E. Lloyd, Charlene Kolz, Barbara Barnhart, Richard C. Wong, Univ Hospitals of Cleveland, Cleveland, OH; Univ Hosp, Cleveland, OH; Quality Information Management Corp, Cleveland, OH, Uruguay; Quality Information Management Corp, Cleveland, OH. Background: EGD is generally indicated for the management of patients with UGIH admitted to intensive care units (ICU) but its impact on practice in the community has not been measured. Thus we examined the effectiveness of EGD factors that may vary between providers, viz. accurate initial diagnosis, performance within 24 hours of admission (early EGD), and appropriate intervention. Methods: Records of 214 consecutive patients admitted to the ICU of 10 metropolitan hospitals with UGIH were reviewed. Inaccurate initial diagnosis was defined as change in diagnosis at subsequent EGD and appropriate interventions were defined as per consensus guidelines on endoscopic therapy. Severity of illness was measured using a disease specific multivariable model developed and validated in prior studies. Unadjusted and severity-adjusted associations of EGD factors with rebleeding, surgery, death, length of hospital stay, length of ICU stay, and readmission to ICU were evaluated. Results: The primary source of hemorrhage was gastric ulcer in 81 (38%), duodenal ulcer in 51 (24%), MW tear in 18 (8%), gastritis in 16 (8%), varices in 7 (3%), and miscellaneous in 41 (19%) patients. Inaccurate diagnosis at initial EGD occurred in 8 % of patients and was associated with significant increases in risk of rebleeding (70% vs. 11%, p < 0.001), rate of surgery (20% vs. 4%, p < 0.05), length of hospital stay (median 7.5 vs. 5 days, p < 0.005), length of ICU stay (median, 4 vs. 2 days, p < 0.005), and rate of readmission to ICU (20% vs. 0.6%, p < 0.001). These associations persisted after adjusting for severity of illness. Early EGD performed in 82% of patients was associated with significant severity adjusted reductions in hospital (-33%: 95% CI [-45%, 18%]) and ICU (-20%: 95% CI [-24%, -3%]) stay. Appropriate interventions at initial EGD, performed in 84% of patients, tended to be associated with reductions in severity adjusted length of ICU stay (-18%: 95% CI [-32%, 0%]) and rate of rebleeding (OR = 0.37: 95% CI [0.13, 1.06]). Conclusions: Early, accurate EGD with appropriate therapeutic intervention is effective as practiced in the community and is associated with improved outcomes for patients with upper gastrointestinal hemorrhage admitted to the ICU. Inaccurate diagnosis at initial EGD is uncommon but has a significant adverse association with all outcome measures.

*4704 RISK FACTORS FOR EARLY REBLEEDING AFTER INITIAL ENDOSCOPIC HEMOSTASIS IN PATIENTS WITH BLEEDING PEPTIC ULCERS. Ho Gak Kim, Jeong Ki Park, Ye Dal Jung, Jong Suk Bae, Chang Hyeong Lee, Jung Dong Bae, Catholic Univ of Taegu-Hyosung, Taegu, South Korea. Background & Aims: Rebleeding after initial endoscopic hemostasis in patients with ulcer hemorrhage was been reported in 20-30%. Identification of patients who are at high risk for rebleeding would be expected to improve the outcome of endoscopic hemostasis. The purpose of this study was to evaluate the risk factors for early rebleeding after initial hemostasis in the view of clinical and endoscopic characteristics. Materials & Methods: We reviewed 99 patients who presented with bleeding peptic ulcers and were treated with endoscopic hemostasis including hypertonic saline injection, electrocautery and clipping. We compared the clinical variables (age, blood transfusioin, comorbid illness, initial systolic BP and pulse rate, hemoglobin), endoscopic characteristics of ulcer (size, number, and location of ulcer, clots on the base, bleeding stigmata, size and color of exposed vessle) and Baylor Bleeding Score between the patients who had early rebleeding (n=22) and who had no early rebleeding (n=77) within 5 days. All data were compared with Pearson’s chisquare test in both groups, and multivariate analysis was tested with logistic regression and expressed as odds ratio in 95% confidence interval. Results: The stasistically significant correlates with early rebleeding after hemostasis were number of comorbid illness (≥2) (p=0.031), volume of transfusion (≥5 units) (p=0.001), size of ulcer (>1 cm) (p=0.024), multiple ulcers (p=0.017), presence of blood clots on ulcer base (p=0.008), stigmata (active bleeding and visible vessles) (p=0.005), size of vessle (>1 mm) (p=0.001) and pearl-colored vessle rather than black-colored (p=0.001). In multivariate analysis, volume of transfusion (5.4;1.94-14.97), ulcer size (4.53;0.98-20.94), multiple ulcers (3.17;1.16-8.17) and size of exposed vessle (16.73;5.23-53.49) were significant risk factors. There was no correlation between Baylor Bleeding Score and rebleeding in our study. Old age(>60 years), smoking, NSAIDs use, initial systolic BP, initial hemoglobin, location of ulcer and H. pylori infection were not stastistically significant factors adversely affecting rebleeding. Conclusions: The risk factors for early rebleeding after hemostasis in bleeding peptic ulcer can be predicted by clinical variables and endoscopic findings. Early evaluation of risk factors such as transfusion over 5 units, large-sized ulcer, multiple ulcers and size of exposed vessle over 1 mm in initial hemostasis can improve the outcome of hemostasis. *4705 ENDOSCOPIC TREATMENT OF GASTRIC ANGIODYSPLASIA WITH ELASTIC BAND LIGATION. SHORT AND LONG-TERM RESULTS IN 22 PATIENTS. Enric Brullet, Manel Alcantara, Agusti Panades, Corporacio Parc Tauli, Sabadell, Spain; H Mar, Barcelona, Spain. The aim of the present study was to evaluate the short and long-term efficacy of elastic band ligation (EBL) to treat gastric angiodysplasia (GA). Methods. 22 patients (p.) (17 male, mean age 73 ± 6 yrs., range 63-89 yrs.) with upper gastrointestinal bleeding (UGB) (17 p.) or anemia (5 p.) diagnosed of GA were included. EBL was performed with single or multipleband ligating devices. UGB episodes and the need for transfusion were assessed during follow-up. Results. 9 p. (41%) had multiple GA lesions and 4 p. (18%) showed an actively bleeding GA on endoscopy. 16 p. underwent one session of EBL (ligation of 1 GA lesion in 14 p., 2 lesions in 1 p., and 3 lesions in 1 p.). During a mean follow-up of 20 ± 10 months, none of these p. experienced UGB or received transfusions for UGB, although one p. with a multiple myeloma received blood transfusion. Six p. needed more than 1 session of EBL due to further UGB episodes or anemia, with a mean interval between sessions of 6 months (3 days-19 months). These p. were followed for a mean of 23 ± 11 months (see Table). In four p. with multiple lesions only the largest lesion or the actively bleeding lesion were treated. No complications related to EBL were identified. Conclusion. Elastic band ligation is an effective and safe procedure to treat gastric angiodysplasia.

Results of EBL in patients needing more than one session. Patient Sessions No of EBL per session EGB episodes‡ Blood units transfused‡

1#

2#

3

4

5*

6**

2 2/1

2 1/1

2 3/4

2 1/1

3 1/2/1

4 3/2/1/1

3/1/0 6/4/2

2/0/0 4/5/2

1/0/0 2/0/0

2/0/0 0/2/0

1/0/0 2/4/0

2/3/0 2/7/0

#No lesions of GA were identified after the last session. * Cecal and ** duodenal angiodysplasia treated with argon plasma coagulation; ‡ Before 1st session / between 1st and last session / after last session

AB208

GASTROINTESTINAL ENDOSCOPY

VOLUME 51, NO. 4, PART 2, 2000