- Email: [email protected]
Predictors for Positive Outcomes in Capsule Endoscopy for the Indication of Obscure Gastrointestinal Bleeding
Abstracts
M1409 Impact of Capsule Endoscopy in Obscure Gastrointestinal Bleeding: Defining Strict Diagnostic Criteria for a Favorable Outcome Nikos Viazis, Kostis Papaxoinis, Ioannis Theodoropoulos, Spiros Sgouros, John Vlachogiannakos, Prokopis Pipis, Costas Markoglou, Alec Avgerinos Background: The most frequent indication for capsule endoscopy is to diagnose the cause of obscure gastrointestinal bleeding. Objective: To determine the impact of capsule endoscopy on the outcome of patients subjected to the test for the evaluation of gastrointestinal bleeding of obscure origin. Methods: From September 2002 till June 2004, all patients referred to our Unit for capsule endoscopy for the investigation of obscure gastrointestinal bleeding were considered for entry into the study. We defined bleeding of obscure origin as bleeding that persists or recurs after a negative initial work up including gastroscopy, colonoscopy, small bowel barium follow through or enteroclysis and push enteroscopy. Capsule endoscopy was performed with the Given M2A video capsule system. Using strict diagnostic criteria, studies were classified as having positive findings, findings of uncertain significance and no findings. According to our final classification, findings were considered positive only if we thought they were the actual source of bleeding, while lesions suspected to be the source were classified as findings of uncertain significance. Outcome was defined as continued or complete resolution of bleeding, during the follow up period. Results: Ninety six patients (M/F: 53/43, mean age: 60.84 G 16.55 years old) were enrolled in the study. Positive findings, findings of uncertain significance and no findings were identified in 41.7%, 20.8% and 37.5% of our study population respectively. The most common lesions seen were angiodysplasias of the small intestine. Therapeutic intervention was possible in 82.5% of patients with positive findings and 35% of patients with findings of uncertain significance. Complete resolution of bleeding, after a mean follow up period of one year, occurred significantly more often in patients with positive findings (68.4%), as compared to patients with findings of uncertain significance and no findings (40.8%; p Z 0.009). Complete resolution of bleeding did not differ significantly in patients with any kind of findings (positive findings and findings of uncertain significance) as compared to patients with no findings (p Z 0.198). Conclusion: Capsule endoscopy increases the diagnostic yield in the work up of patients with gastrointestinal bleeding of obscure origin. Patients with strictly defined positive findings are those that will benefit from medical therapy and have a favorable outcome.
M1410 H2-Blocker for Rebleeding From Bleeding Peptic Ulcer, Which Was Prevented By Eradication of Helicobacter pylori, Was As Effective As Proton Pump Inhibitor Kenichiro Watanabe, Atsushi Kikkawa, Ryuichi Iwakiri, Akiko Danjyo, Hibiki Ootani, Seiji Tsunada, Hiroyuki Sakata, Kazuma Fujimoto Aim: This study was to examine: 1) an effect of H2-blocker on rebleeding from bleeding peptic ulcer compared to proton pump inhibitor (PPI), and 2) if eradiation of Helicobacter pylori (H. pylori) improved the rebleeding. Patients and Methods: Patients, who were diagnosed as bleeding peptic ulcer in our emergency endoscopy unit from 1994 to 2001, was enrolled into this study. Patients received endoscopic hemostasis and were followed up for 12 months. The cases, who died within a year caused by unrelated diseases, were excluded from this study. The patients were divided into two groups according to 4 year-interval: patients in group A, who received endoscopic hemostasis from 1994 to 1997; patients in group B, who received endoscopic hemostasis from 1998 to 2001. The patients in group B were eradicated if H. pylori was positive, but all patients in group A were not eradicated. Results: Among 366 patients enrolled in this study, 320 patients were analyzed: 162 patients in group A (113 of gastric ulcer and 49 of duodenal ulcer), and 158 patients in group B (116 of gastric ulcer and 42 of duodenal ulcer). Regarding background characteristics, no significant differences were found between the two groups. In group B, H. pylori positive patients were 62 (61.4%), who were eradicated after hemostasis. Twenty four patients in group A rebled (14.8%), and 5 patients in group B rebled (3.2%) during one year follow-up period. The rate of rebleeding in group B was significantly lower than that in group A (p ! 0.01), and none of patients who were eradicated H.pylori rebled in this study. Patients without eradication was evaluated in rebleeding. Rebleeding rate in patients treated with H2-blocker (famotidine) (15/128: 11.7%) was not differed from that in patients treated with PPI (14/142: 9.9%). Conclusion: This historical control examination indicated that eradication of H. pylori decreased bleeding rate of bleeding peptic ulcer, and that rebleeing rate was not differed between H2-blocker and PPI medication in patients without eradication.
www.mosby.com/gie
M1411 Predictors for Positive Outcomes in Capsule Endoscopy for the Indication of Obscure Gastrointestinal Bleeding Victor Wong, Joanna Law, Robert Enns, Harpinder Nagi, Hongbin Zhang, Monica Choi Introduction: Capsule endoscopy (CE) is primarily used for visualization of the small bowel mucosa in patients with obscure gastrointestinal bleeding (OGIB). The yield of positive findings will depend on the patient group. Careful patient selection will improve diagnostic yield and maximize resource utilization. The purpose of this study is to determine the risk factors or patient characteristics that can help predict which patients are likely to have positive findings after undergoing CE. Methods: Risk factors and patient characteristics were identified and exact logistic regression was employed for univariate analysis and factor selecting. Factors analyzed were indication for procedure, history of NSAIDs, alcohol, smoking, anti-platelet/anticoagulant use, PPI use, steroid use, use of SSRI, abdominal pain, change in bowel pattern, presence of red blood per rectum (RBPR), history of melena, number of previous endoscopies, small bowel imaging and transfusion requirements prior to the procedure. Results: A single centre Canadian assessment of 212 patients who underwent CE between 12/01 and 03/04 for OGIB. Patients ranged from 17-93 years (mean Z 60). For the investigation of OGIB, 3 factors were determined to correlate with a positive study. The use of oral steroids had an odds ratio of 4.911 (95% CI 1.575-17.939), iron deficiency anemia requiring supplemental iron had an odds ratio of 1.936 (95% CI 1.016-3.728), and where transfusion requirements exceeded 10 units, an odds ratio of 1.360 (95% CI 1.057-1.761). Conversely, the RBPR tended to lower the odds of a positive study (OR 0.274, 95% CI 0.092-0.727). Conclusion: There is limited literature reporting predictors of positive capsule studies. We have found that patients with OGIB who are on oral steroids or iron supplementation, and patients with large transfusion requirements are more likely to have sources of blood loss identified by CE. The presence of RBPR was negatively correlated with identifying a bleeding source. Using these predictors may help maximize resource utilization by improving patient selection for CE.
M1412 The Study of Correlation of Stereoscopic Finding and Pathologic Finding in Early Gastric Cancer by Endoscopic Incision and Submucosal Dissection Method Jae Hoon Yang, Jun Hwan Wi, Joo Young Cho, In Seop Jung, Su Jin Hong, Chang Beom Ryu, Joon Seong Lee, So Young Jin, Moon Sung Lee, Chan Sup Shim Background/Aim: The indication of EMR (Endoscopic mucosal resection) has been extended in the treatment of early gastric cancer. Recently, EISD (Endoscopic incision and submucosal dissection) can resect larger lesion endoscopically, However, complete resection cannot be confirmed without pathologic results. After EMR, additional procedure is difficult because of fibrous tissue. The aim of this study are to evaluate the correlation between stereoscopic finding and pathologic finding and to evaluate whether the immediate stereoscopic finding is useful to decide the need of additional procedure after EISD. Material/Method: Early gastric cancer or tubular adenoma with high grade dysplasia of 20 specimens in 17 patients were analyzed. Stereoscopic finding of free lateral margin and area of the lesion were compared with the pathologic result by TDI scope eyeÔ software program. Result: 20 specimens of 17 patients (mean age:59 ¡3⁄4 13 years, M:F Z 12:5) consisted of 17 case of early gastric cancer and 3 case of tubular adenoma with high grade dysplasia. The stereoscopic findings of mean of free lateral margin and area of the lesions were 2.1 ¡3⁄4 1.3 mm, 313.7 ¡3⁄4 294.0 mm2. The pathologic findings of mean of free lateral margin and area of the lesion were 2.7 ¡3⁄4 2.1 mm, 283.5 ¡3⁄4 285.0 mm2. In free lateral margin, the differences of stereoscopic findings and pathologic findings were 0.6 ¡3⁄4 1.5 mm, so it is underestimated. In overestimated 3 case, stereoscopic finding and pathologic finding were within 2 mm, 2 case were prospected positive free lateral margin in stereoscopic finding. Stereoscopic finding and pathologic finding have significant correlation in lesion area and free lateral margin. (free lateral margin: r2 Z 0.496, p ! 0.001, area:r2 Z 0.964, p ! 0.001). In area of lesion, the differences of stereoscopic findings and pathologic findings were 30.2 ¡3⁄4 55.9 mm2, so it is overestimated. Conclusion: the correlation of stereoscopic finding and pathologic finding was recognized. It is suggested that free lateral margin measured in stereoscopic finding is helpful to decide the completeness of procedure and the need of additional procedure. Further study is needed to clinical correlations of stereoscopic measuring area of the lesion.
Volume 61, No. 5 : 2005 GASTROINTESTINAL ENDOSCOPY AB185
M1409 Impact of Capsule Endoscopy in Obscure Gastrointestinal Bleeding: Defining Strict Diagnostic Criteria for a Favorable Outcome Nikos Viazis, Kostis Papaxoinis, Ioannis Theodoropoulos, Spiros Sgouros, John Vlachogiannakos, Prokopis Pipis, Costas Markoglou, Alec Avgerinos Background: The most frequent indication for capsule endoscopy is to diagnose the cause of obscure gastrointestinal bleeding. Objective: To determine the impact of capsule endoscopy on the outcome of patients subjected to the test for the evaluation of gastrointestinal bleeding of obscure origin. Methods: From September 2002 till June 2004, all patients referred to our Unit for capsule endoscopy for the investigation of obscure gastrointestinal bleeding were considered for entry into the study. We defined bleeding of obscure origin as bleeding that persists or recurs after a negative initial work up including gastroscopy, colonoscopy, small bowel barium follow through or enteroclysis and push enteroscopy. Capsule endoscopy was performed with the Given M2A video capsule system. Using strict diagnostic criteria, studies were classified as having positive findings, findings of uncertain significance and no findings. According to our final classification, findings were considered positive only if we thought they were the actual source of bleeding, while lesions suspected to be the source were classified as findings of uncertain significance. Outcome was defined as continued or complete resolution of bleeding, during the follow up period. Results: Ninety six patients (M/F: 53/43, mean age: 60.84 G 16.55 years old) were enrolled in the study. Positive findings, findings of uncertain significance and no findings were identified in 41.7%, 20.8% and 37.5% of our study population respectively. The most common lesions seen were angiodysplasias of the small intestine. Therapeutic intervention was possible in 82.5% of patients with positive findings and 35% of patients with findings of uncertain significance. Complete resolution of bleeding, after a mean follow up period of one year, occurred significantly more often in patients with positive findings (68.4%), as compared to patients with findings of uncertain significance and no findings (40.8%; p Z 0.009). Complete resolution of bleeding did not differ significantly in patients with any kind of findings (positive findings and findings of uncertain significance) as compared to patients with no findings (p Z 0.198). Conclusion: Capsule endoscopy increases the diagnostic yield in the work up of patients with gastrointestinal bleeding of obscure origin. Patients with strictly defined positive findings are those that will benefit from medical therapy and have a favorable outcome.
M1410 H2-Blocker for Rebleeding From Bleeding Peptic Ulcer, Which Was Prevented By Eradication of Helicobacter pylori, Was As Effective As Proton Pump Inhibitor Kenichiro Watanabe, Atsushi Kikkawa, Ryuichi Iwakiri, Akiko Danjyo, Hibiki Ootani, Seiji Tsunada, Hiroyuki Sakata, Kazuma Fujimoto Aim: This study was to examine: 1) an effect of H2-blocker on rebleeding from bleeding peptic ulcer compared to proton pump inhibitor (PPI), and 2) if eradiation of Helicobacter pylori (H. pylori) improved the rebleeding. Patients and Methods: Patients, who were diagnosed as bleeding peptic ulcer in our emergency endoscopy unit from 1994 to 2001, was enrolled into this study. Patients received endoscopic hemostasis and were followed up for 12 months. The cases, who died within a year caused by unrelated diseases, were excluded from this study. The patients were divided into two groups according to 4 year-interval: patients in group A, who received endoscopic hemostasis from 1994 to 1997; patients in group B, who received endoscopic hemostasis from 1998 to 2001. The patients in group B were eradicated if H. pylori was positive, but all patients in group A were not eradicated. Results: Among 366 patients enrolled in this study, 320 patients were analyzed: 162 patients in group A (113 of gastric ulcer and 49 of duodenal ulcer), and 158 patients in group B (116 of gastric ulcer and 42 of duodenal ulcer). Regarding background characteristics, no significant differences were found between the two groups. In group B, H. pylori positive patients were 62 (61.4%), who were eradicated after hemostasis. Twenty four patients in group A rebled (14.8%), and 5 patients in group B rebled (3.2%) during one year follow-up period. The rate of rebleeding in group B was significantly lower than that in group A (p ! 0.01), and none of patients who were eradicated H.pylori rebled in this study. Patients without eradication was evaluated in rebleeding. Rebleeding rate in patients treated with H2-blocker (famotidine) (15/128: 11.7%) was not differed from that in patients treated with PPI (14/142: 9.9%). Conclusion: This historical control examination indicated that eradication of H. pylori decreased bleeding rate of bleeding peptic ulcer, and that rebleeing rate was not differed between H2-blocker and PPI medication in patients without eradication.
www.mosby.com/gie
M1411 Predictors for Positive Outcomes in Capsule Endoscopy for the Indication of Obscure Gastrointestinal Bleeding Victor Wong, Joanna Law, Robert Enns, Harpinder Nagi, Hongbin Zhang, Monica Choi Introduction: Capsule endoscopy (CE) is primarily used for visualization of the small bowel mucosa in patients with obscure gastrointestinal bleeding (OGIB). The yield of positive findings will depend on the patient group. Careful patient selection will improve diagnostic yield and maximize resource utilization. The purpose of this study is to determine the risk factors or patient characteristics that can help predict which patients are likely to have positive findings after undergoing CE. Methods: Risk factors and patient characteristics were identified and exact logistic regression was employed for univariate analysis and factor selecting. Factors analyzed were indication for procedure, history of NSAIDs, alcohol, smoking, anti-platelet/anticoagulant use, PPI use, steroid use, use of SSRI, abdominal pain, change in bowel pattern, presence of red blood per rectum (RBPR), history of melena, number of previous endoscopies, small bowel imaging and transfusion requirements prior to the procedure. Results: A single centre Canadian assessment of 212 patients who underwent CE between 12/01 and 03/04 for OGIB. Patients ranged from 17-93 years (mean Z 60). For the investigation of OGIB, 3 factors were determined to correlate with a positive study. The use of oral steroids had an odds ratio of 4.911 (95% CI 1.575-17.939), iron deficiency anemia requiring supplemental iron had an odds ratio of 1.936 (95% CI 1.016-3.728), and where transfusion requirements exceeded 10 units, an odds ratio of 1.360 (95% CI 1.057-1.761). Conversely, the RBPR tended to lower the odds of a positive study (OR 0.274, 95% CI 0.092-0.727). Conclusion: There is limited literature reporting predictors of positive capsule studies. We have found that patients with OGIB who are on oral steroids or iron supplementation, and patients with large transfusion requirements are more likely to have sources of blood loss identified by CE. The presence of RBPR was negatively correlated with identifying a bleeding source. Using these predictors may help maximize resource utilization by improving patient selection for CE.
M1412 The Study of Correlation of Stereoscopic Finding and Pathologic Finding in Early Gastric Cancer by Endoscopic Incision and Submucosal Dissection Method Jae Hoon Yang, Jun Hwan Wi, Joo Young Cho, In Seop Jung, Su Jin Hong, Chang Beom Ryu, Joon Seong Lee, So Young Jin, Moon Sung Lee, Chan Sup Shim Background/Aim: The indication of EMR (Endoscopic mucosal resection) has been extended in the treatment of early gastric cancer. Recently, EISD (Endoscopic incision and submucosal dissection) can resect larger lesion endoscopically, However, complete resection cannot be confirmed without pathologic results. After EMR, additional procedure is difficult because of fibrous tissue. The aim of this study are to evaluate the correlation between stereoscopic finding and pathologic finding and to evaluate whether the immediate stereoscopic finding is useful to decide the need of additional procedure after EISD. Material/Method: Early gastric cancer or tubular adenoma with high grade dysplasia of 20 specimens in 17 patients were analyzed. Stereoscopic finding of free lateral margin and area of the lesion were compared with the pathologic result by TDI scope eyeÔ software program. Result: 20 specimens of 17 patients (mean age:59 ¡3⁄4 13 years, M:F Z 12:5) consisted of 17 case of early gastric cancer and 3 case of tubular adenoma with high grade dysplasia. The stereoscopic findings of mean of free lateral margin and area of the lesions were 2.1 ¡3⁄4 1.3 mm, 313.7 ¡3⁄4 294.0 mm2. The pathologic findings of mean of free lateral margin and area of the lesion were 2.7 ¡3⁄4 2.1 mm, 283.5 ¡3⁄4 285.0 mm2. In free lateral margin, the differences of stereoscopic findings and pathologic findings were 0.6 ¡3⁄4 1.5 mm, so it is underestimated. In overestimated 3 case, stereoscopic finding and pathologic finding were within 2 mm, 2 case were prospected positive free lateral margin in stereoscopic finding. Stereoscopic finding and pathologic finding have significant correlation in lesion area and free lateral margin. (free lateral margin: r2 Z 0.496, p ! 0.001, area:r2 Z 0.964, p ! 0.001). In area of lesion, the differences of stereoscopic findings and pathologic findings were 30.2 ¡3⁄4 55.9 mm2, so it is overestimated. Conclusion: the correlation of stereoscopic finding and pathologic finding was recognized. It is suggested that free lateral margin measured in stereoscopic finding is helpful to decide the completeness of procedure and the need of additional procedure. Further study is needed to clinical correlations of stereoscopic measuring area of the lesion.
Volume 61, No. 5 : 2005 GASTROINTESTINAL ENDOSCOPY AB185