- Email: [email protected]
Mo1018 Probe-Based Confocal Laser Endomicroscopy for In Vivo Detection of Gastric Intestinal Metaplasia, Intraepithelial Neoplasia, and Carcinoma: A Multicenter, Randomized, Controlled Trial
Abstracts
pathogen (50.2% of all stents), followed by Candida albicans (48.4%), Escherichia coli (41.7%), Enterococcus faecium (29.6%), Streptococcus anginosus (16.9%), Enterobacter cloacae (14.6%), and Klebsiella oxytoca (11.3%). Polymicrobial colonization (95.8%) was significantly more common than single microbial colonization (4.2%, P<0.001); in 55.9% of all cases Candida species were involved. The clinical cholangitis rate was 18.8%. Eighty-four of 120 patients (70%) received antibiotic treatment during stent insertion, mainly aminopenicillins or cephalosporins. Conclusion: We conclude that colonization with enterococci and Candida species is an important mechanism in the formation of biliary biofilms and consecutively stent occlusion, possibly selected by broad spectrum antibiotic therapy. Therefore, empirical antimicrobial treatment in patients with stent-associated cholangitis should be guided towards enterococci and Candida species.
Mo1018 Probe-Based Confocal Laser Endomicroscopy for In Vivo Detection of Gastric Intestinal Metaplasia, Intraepithelial Neoplasia, and Carcinoma: A Multicenter, Randomized, Controlled Trial Xiu-Li Zuo*, Zhen Li, Yan-Qing Li Gastroenterology, Qilu Hospital of Shandong University, Jinan, China Background and Study Aims: Because of indistinctive endoscopic appearance of gastric intestinal metaplasia (GIM), gastric intraepithelial neoplasia (GIN) and early gastric cancer (EGC), significant such lesions may be missed during surveillance endoscopy. In this study we assessed the value of combined computed virtual chromoendoscopy (Fujinon Intelligent Color Enhancement, FICE) and probe-based confocal laser endomicroscopy (pCLE) for the detection of GIM, GIN and EGC in a multicenter, randomized, controlled trial. Patients and Methods: This is a multicenter, double-blind, randomized study. Patients were randomly assigned to receive either FICE-guided pCLE with targeted biopsies (group A) or FICE with standard biopsies (group B). Results: A total of 238 patients were finally analyzed in this study (120 in group A and 118 in group B). On a per-patient analysis, the diagnostic yield of GIM/GIN/EGC for group A and B were 73.33% and 63.56% respectively (PZ0.012). On a per-biopsy analysis, FICE-guided pCLE with targeted biopsies significantly increased the diagnostic yield of GIM/GIN/EGC as compared with FICE with standard biopsies from 31.50% (252/800) to 75.06% (313/417) (P<0.001). In addition, CLE-guided targeted biopsies led to a significant decrease in the biopsy number of 63% per patient as compared to WLE with standard biopsies (P <0.001). Conclusions: FICE-guided pCLE with targeted biopsies is superior to FICE with standard biopsies for the detection and surveillance of GIM, GIN and EGC, and the number of biopsies needed to confirm these lesions is about a half of those needed by using FICE with standard biopsies.
Mo1019 Proton Pump Inhibitor Use and Healthcare Utilization in the Treatment of Marginal Ulceration Allison Schulman*, Aoife Devery, Nitin Kumar, Michele B. Ryan, Walter W. Chan, Christopher C. Thompson Division of Gastroenterology, Brigham & Women’s Hospital, Boston, MA Introduction: Ulceration at the gastrojejunal anastomosis (GJA), also known as marginal ulceration, is a common complication in patients who have undergone Rouxen-Y gastric bypass (RYGB). Management of this condition increasingly incorporates either soluble proton pump inhibitors (PPIs) or the opening of capsules to enhance mucosal absorption. This approach of administration may lead to improved ulcer healing and decrease cost burden. Aims: To analyze resource utilization during management of marginal ulceration in RYGB patients who are taking PPIs via an open capsule (OC) vs intact capsule (IC). Methods: This was a multicenter retrospective review of a prospectively collected database. Primary outcome was total charge of ulceration management. Patients who underwent RYGB were prospectively enrolled in the Research Patient Data Registry, and those found to have marginal ulceration on endoscopy were reviewed for inclusion in this study. Maximum medical management including therapy with high dose PPI was initiated in all cases, and repeat endoscopy was performed as per surveillance recommendations until ulcer healing was confirmed. Patients were excluded if they were lost to follow-up after the index endoscopy or if information regarding method of PPI administration was unavailable. Relative utilization was determined by comparison of categorized charges incurred from time of diagnosis to time of resolution. Charge count was stopped on the day of ulcer resolution or at time of alternative therapy (ie. surgery). All charges related to ulcer management including procedural, medication, and ED visit were obtained. Professional fees used 2014 Medicare rates for metropolitan Boston. Means and medians were compared with Student’s t-test or Wilcoxon based on normality of the data, respectively. Proportional comparisons between groups were performed with Fisher’s Exact test. All statistics reported as meanSEM or median[IQR]. Results: 162 patients (age 51.80.6yr, 82.3%F) had pre-RYGB BMI of 47.00.6 kg/m2. Marginal ulceration was found 3.40.4 years after RYGB. 121 patients opened capsules and 41 did not. Baseline differences between groups are in Table 1. Mean number of endoscopic procedures required were
AB430 GASTROINTESTINAL ENDOSCOPY Volume 83, No. 5S : 2016
1.240.05 and 1.820.25 in the OC vs IC groups, respectively (pZ0.02). Median time to ulcer healing was significantly lower in the OC vs IC groups (91.0 vs 321.0 days, pZ<0.001). Total healthcare utilization was significantly lower for those who opened capsules compared to those who ingested intact form ($7206 vs $11009, pZ0.05). Categorized charges are compared in Table 2. Conclusions: Patients with marginal ulceration following RYGB demonstrate significantly lower healthcare utilization when opening PPI capsules as compared to those who ingest the intact form. Opening of PPI medications should be considered the new standard of care in RYGB patients. Table 1. Baseline characteristics
Sex (%F) Age (mean +/- Std Dev) Pre-RYGB BMI (K/m2) Risk factors for GJA ulcers: Gastric pouch length (mean +/- Std Dev) Smoking (%) NSAID use (%) H. Pylori infection (%) Fistula (%) Foreign body (suture/staple) Diabetes (%)
Intact capsules (n[41)
Open capsules (n[121)
p-value
80.3 53.3 1.3 48.0 1.3
83.6 51.2 0.9 46.2 0.7
0.57 0.18 0.22
5.50.4 20.0 19.7 8.0 7.0 43.7 16.4
5.00.2 19.6 10.7 13.9 8.0 38.0 19.6
0.25 1.0 0.06 0.44 1.0 0.47 0.71
C.I. Z 95% confidence interval. *denotes p-value <0.05, NS denotes nonsignificance.
Table 2. Categorized charge comparison Intact capsules (n[41) Endoscopy professional Theatre, instruments, recovery Proton Pump Inhibitor Related abdominal computed tomography (CT) Related abdominal or chest x-rays Related hospital Emergency Room visits Total charge until ulcer resolution
Open capsules (n[121)
p-value
369 55 5575 833 1356 305 975 363
250 10 3795 144 141 21 507 125
0.04* 0.04* <0.0001* 0.23
55 22 3620 1728
27.7 12 2559 8115
0.23 0.56
11009 2547
7206 672
0.05*
Mo1020 Endoscopic Submucosal Dissection for Early Gastric Cancer Applying the Expanded Resection Criteria in a Western Tertiary Center Amir Klein*, Nicholas Tutticci, David J. Tate, Farzan F. Bahin, Eric Y. Lee, Michael J. Bourke Gastroenterology and Hepatology, Westmead hospital, Sydney, New South Wales, Australia Introduction and aim: Gastric cancer is a common cause of mortality worldwide. High prevalence in Asian countries has resulted in screening programs, early detection and highly effective treatment by endoscopic submucosal dissection (ESD). Advances in endoscopic technique with improved outcomes have led to the expansion of the original criteria enabling safe and curative endoscopic resection for larger lesions with more advanced features. Western experience with ESD is limited and reports on successful application of the expanded criteria for resection are few. Herein we describe the experience of a single western tertiary center with ESD for EGC and specifically the outcomes of resections according to the expanded criteria. Methods: Analysis of prospectively collected data (NCT02306707). A comparison between small lesions (original criteria) and large more advanced lesions (expanded criteria) was performed. Results: Over 60 months to October 2015 69 patients with EGCs (mean age 73, 73% male) were referred for ESD. Lesions underwent preresection evaluation by high definition white light endoscopy, narrow band imaging and endoscopic ultrasound. One lesion was referred directly for surgery and one procedure was abandoned midway due to significant fibrosis and vasculature raising concern for invasive disease. ESD was performed on 67 lesions (median lesion size 20 mm (IQR 15-30)). Lesions satisfying the expanded criteria were larger (mean 38 mm versus 18 mm, pZ0.001) and contained more invasive cancer (26% versus 0%, pZ0.01). Complete endoscopic resection was achieved in 97% at the index procedure and was similar between the two groups (median procedure time 123.5 minutes). En-bloc resection rate was 91% for the entire cohort but was significantly
www.giejournal.org
pathogen (50.2% of all stents), followed by Candida albicans (48.4%), Escherichia coli (41.7%), Enterococcus faecium (29.6%), Streptococcus anginosus (16.9%), Enterobacter cloacae (14.6%), and Klebsiella oxytoca (11.3%). Polymicrobial colonization (95.8%) was significantly more common than single microbial colonization (4.2%, P<0.001); in 55.9% of all cases Candida species were involved. The clinical cholangitis rate was 18.8%. Eighty-four of 120 patients (70%) received antibiotic treatment during stent insertion, mainly aminopenicillins or cephalosporins. Conclusion: We conclude that colonization with enterococci and Candida species is an important mechanism in the formation of biliary biofilms and consecutively stent occlusion, possibly selected by broad spectrum antibiotic therapy. Therefore, empirical antimicrobial treatment in patients with stent-associated cholangitis should be guided towards enterococci and Candida species.
Mo1018 Probe-Based Confocal Laser Endomicroscopy for In Vivo Detection of Gastric Intestinal Metaplasia, Intraepithelial Neoplasia, and Carcinoma: A Multicenter, Randomized, Controlled Trial Xiu-Li Zuo*, Zhen Li, Yan-Qing Li Gastroenterology, Qilu Hospital of Shandong University, Jinan, China Background and Study Aims: Because of indistinctive endoscopic appearance of gastric intestinal metaplasia (GIM), gastric intraepithelial neoplasia (GIN) and early gastric cancer (EGC), significant such lesions may be missed during surveillance endoscopy. In this study we assessed the value of combined computed virtual chromoendoscopy (Fujinon Intelligent Color Enhancement, FICE) and probe-based confocal laser endomicroscopy (pCLE) for the detection of GIM, GIN and EGC in a multicenter, randomized, controlled trial. Patients and Methods: This is a multicenter, double-blind, randomized study. Patients were randomly assigned to receive either FICE-guided pCLE with targeted biopsies (group A) or FICE with standard biopsies (group B). Results: A total of 238 patients were finally analyzed in this study (120 in group A and 118 in group B). On a per-patient analysis, the diagnostic yield of GIM/GIN/EGC for group A and B were 73.33% and 63.56% respectively (PZ0.012). On a per-biopsy analysis, FICE-guided pCLE with targeted biopsies significantly increased the diagnostic yield of GIM/GIN/EGC as compared with FICE with standard biopsies from 31.50% (252/800) to 75.06% (313/417) (P<0.001). In addition, CLE-guided targeted biopsies led to a significant decrease in the biopsy number of 63% per patient as compared to WLE with standard biopsies (P <0.001). Conclusions: FICE-guided pCLE with targeted biopsies is superior to FICE with standard biopsies for the detection and surveillance of GIM, GIN and EGC, and the number of biopsies needed to confirm these lesions is about a half of those needed by using FICE with standard biopsies.
Mo1019 Proton Pump Inhibitor Use and Healthcare Utilization in the Treatment of Marginal Ulceration Allison Schulman*, Aoife Devery, Nitin Kumar, Michele B. Ryan, Walter W. Chan, Christopher C. Thompson Division of Gastroenterology, Brigham & Women’s Hospital, Boston, MA Introduction: Ulceration at the gastrojejunal anastomosis (GJA), also known as marginal ulceration, is a common complication in patients who have undergone Rouxen-Y gastric bypass (RYGB). Management of this condition increasingly incorporates either soluble proton pump inhibitors (PPIs) or the opening of capsules to enhance mucosal absorption. This approach of administration may lead to improved ulcer healing and decrease cost burden. Aims: To analyze resource utilization during management of marginal ulceration in RYGB patients who are taking PPIs via an open capsule (OC) vs intact capsule (IC). Methods: This was a multicenter retrospective review of a prospectively collected database. Primary outcome was total charge of ulceration management. Patients who underwent RYGB were prospectively enrolled in the Research Patient Data Registry, and those found to have marginal ulceration on endoscopy were reviewed for inclusion in this study. Maximum medical management including therapy with high dose PPI was initiated in all cases, and repeat endoscopy was performed as per surveillance recommendations until ulcer healing was confirmed. Patients were excluded if they were lost to follow-up after the index endoscopy or if information regarding method of PPI administration was unavailable. Relative utilization was determined by comparison of categorized charges incurred from time of diagnosis to time of resolution. Charge count was stopped on the day of ulcer resolution or at time of alternative therapy (ie. surgery). All charges related to ulcer management including procedural, medication, and ED visit were obtained. Professional fees used 2014 Medicare rates for metropolitan Boston. Means and medians were compared with Student’s t-test or Wilcoxon based on normality of the data, respectively. Proportional comparisons between groups were performed with Fisher’s Exact test. All statistics reported as meanSEM or median[IQR]. Results: 162 patients (age 51.80.6yr, 82.3%F) had pre-RYGB BMI of 47.00.6 kg/m2. Marginal ulceration was found 3.40.4 years after RYGB. 121 patients opened capsules and 41 did not. Baseline differences between groups are in Table 1. Mean number of endoscopic procedures required were
AB430 GASTROINTESTINAL ENDOSCOPY Volume 83, No. 5S : 2016
1.240.05 and 1.820.25 in the OC vs IC groups, respectively (pZ0.02). Median time to ulcer healing was significantly lower in the OC vs IC groups (91.0 vs 321.0 days, pZ<0.001). Total healthcare utilization was significantly lower for those who opened capsules compared to those who ingested intact form ($7206 vs $11009, pZ0.05). Categorized charges are compared in Table 2. Conclusions: Patients with marginal ulceration following RYGB demonstrate significantly lower healthcare utilization when opening PPI capsules as compared to those who ingest the intact form. Opening of PPI medications should be considered the new standard of care in RYGB patients. Table 1. Baseline characteristics
Sex (%F) Age (mean +/- Std Dev) Pre-RYGB BMI (K/m2) Risk factors for GJA ulcers: Gastric pouch length (mean +/- Std Dev) Smoking (%) NSAID use (%) H. Pylori infection (%) Fistula (%) Foreign body (suture/staple) Diabetes (%)
Intact capsules (n[41)
Open capsules (n[121)
p-value
80.3 53.3 1.3 48.0 1.3
83.6 51.2 0.9 46.2 0.7
0.57 0.18 0.22
5.50.4 20.0 19.7 8.0 7.0 43.7 16.4
5.00.2 19.6 10.7 13.9 8.0 38.0 19.6
0.25 1.0 0.06 0.44 1.0 0.47 0.71
C.I. Z 95% confidence interval. *denotes p-value <0.05, NS denotes nonsignificance.
Table 2. Categorized charge comparison Intact capsules (n[41) Endoscopy professional Theatre, instruments, recovery Proton Pump Inhibitor Related abdominal computed tomography (CT) Related abdominal or chest x-rays Related hospital Emergency Room visits Total charge until ulcer resolution
Open capsules (n[121)
p-value
369 55 5575 833 1356 305 975 363
250 10 3795 144 141 21 507 125
0.04* 0.04* <0.0001* 0.23
55 22 3620 1728
27.7 12 2559 8115
0.23 0.56
11009 2547
7206 672
0.05*
Mo1020 Endoscopic Submucosal Dissection for Early Gastric Cancer Applying the Expanded Resection Criteria in a Western Tertiary Center Amir Klein*, Nicholas Tutticci, David J. Tate, Farzan F. Bahin, Eric Y. Lee, Michael J. Bourke Gastroenterology and Hepatology, Westmead hospital, Sydney, New South Wales, Australia Introduction and aim: Gastric cancer is a common cause of mortality worldwide. High prevalence in Asian countries has resulted in screening programs, early detection and highly effective treatment by endoscopic submucosal dissection (ESD). Advances in endoscopic technique with improved outcomes have led to the expansion of the original criteria enabling safe and curative endoscopic resection for larger lesions with more advanced features. Western experience with ESD is limited and reports on successful application of the expanded criteria for resection are few. Herein we describe the experience of a single western tertiary center with ESD for EGC and specifically the outcomes of resections according to the expanded criteria. Methods: Analysis of prospectively collected data (NCT02306707). A comparison between small lesions (original criteria) and large more advanced lesions (expanded criteria) was performed. Results: Over 60 months to October 2015 69 patients with EGCs (mean age 73, 73% male) were referred for ESD. Lesions underwent preresection evaluation by high definition white light endoscopy, narrow band imaging and endoscopic ultrasound. One lesion was referred directly for surgery and one procedure was abandoned midway due to significant fibrosis and vasculature raising concern for invasive disease. ESD was performed on 67 lesions (median lesion size 20 mm (IQR 15-30)). Lesions satisfying the expanded criteria were larger (mean 38 mm versus 18 mm, pZ0.001) and contained more invasive cancer (26% versus 0%, pZ0.01). Complete endoscopic resection was achieved in 97% at the index procedure and was similar between the two groups (median procedure time 123.5 minutes). En-bloc resection rate was 91% for the entire cohort but was significantly
www.giejournal.org