- Email: [email protected]
Long-Term Outcome of Endoscopic Pancreatic Necrosectomy: Final Results of the First German Multicenter Trial
Abstracts
W1414 Long-Term Outcome of Endoscopic Pancreatic Necrosectomy: Final Results of the First German Multicenter Trial Hans Seifert, Michael Biermer, Wolfgang Schmitt, Uwe Will, Michael Hocke, Christian Ju ¨Rgensen, Christian Kreitmayr, Christian Prinz, Thomas Roesch Background: Pancreatic endotherapy of necroses following acute pancreatitis consists of drainage into the upper GI tract (stomach and duodenum), followed by widening the access and removing necrotic debris by direct vision with an endoscope introduced into the cavity. Following presentation of the initial results of the German Multicenter Study on Endoscopic Pancreatic Retroperitoneal Debridement(GEPARD) we now report the final outcome including long-term follow-up results. Methods: In a multicenter setting 60 patients (70% male, mean age 60 years, range 35-83) with pancreatic abscesses and necroses treated between 1999 and 2003 were included from 1999-2003 into this study. Their clinical conditions was determined by acute/postacute pancreatitis with subacute or acute infectious signs. Drainage was mostly transgastrally by endoscopic or endosonographic access, or percutaneously (n Z 2), with subsequent balloon dilatation of the tract to introduce a gastroscope to remove necroses and debris by snares, baskets and forceps as well as by lavage and suction, usually in several sessions. Immediate clinical and morphologic outcome on hospital discharge and long-term results after a mean follow-up of 58 months (range 10-89) were recorded. Results:A median of 15 sessions (range 15-35) including initial access to the cavity were performed to remove all debris as far as possible. Acute complications were encountered in 8 cases (2 perforations, 5 bleeding, 1 pneumoperitoneum). Initially, complete and partial (minimal symptoms, remaining collection !3 cm) clinical and morphologic success could achieved in 73% of cases. 11 patients were operated due to various reasons (pancreatic surgery in 4 cases each during initial treatment and later). Follow-up showed that 8 patients had died, 1 of those initially due to an acute complication after the intervention, and 3 within 6 months after necrosectomy. 14 patients had to be readmitted for repeated therapy due to recurrence. Clinical results on long-term follow-up showed that 77% of the surviving patients were in excellent/good/moderate clinical condition, and that 16, 18 and 24 patients reported on pain, exocrine insufficiency and diabetes. Conclusions: Endoscopic pancreatic necrosectomy has successful clinical long term outcome in 3/4 of patients; recurrence rate and associated problems necessitating surgery and endoscopic reinterventions was 42%. There is a related mortality rate which is partially due to interventions (more data are presently accumulated from more centers) but mostly due to underlying disease. Further prospective studies will clarify the precise role of this transgastric retroperitoneal endotherapy.
W1415 Incidence and Mortality Following Percutaneous Endoscopic Gastrostomy (PEG) Placement in Medicare Beneficiaries from 1992 to 2003 Richard C. Wong, Doug Kuo, Gregory D. Olds, Gregory S. Cooper Background: PEG placement is commonly performed to deliver enteral nutrition and medications in patients who are unable to swallow because of conditions such as dementia and stroke. Although the rate of feeding tube placement is increasing, a detailed longitudinal analysis of mortality has not been conducted. Objective: Analysis of patient mortality following PEG placement in noncancer patients. Methods: All inpatient claims from 1992 to 2003 from a 5% random sample of cancer-free Medicare beneficiaries age 66 years and older were examined. Patients who were enrolled in Medicare Part A and were in fee-for-service arrangements were included. Study cohorts were divided into 4 groups according to when the initial PEG was placed. Survival was measured from the date of index PEG placement and for up to 3 years of follow-up. Kaplan-Meier analyses and Cox regression including adjustment for comorbidity were used. Results: 7,437 noncancer patients received PEG placement during the study period. The incidence rate of PEG placement has nearly tripled from 7.5 to 19.4 per 1000 over the study period. Adjusted overall survival at 30 days, 1 year and 3 years were 81.2%, 49% and 36.3%, respectively, with the greatest hazard of death within the first year. Women had better 1-year adjusted survival compared to men (75.6% vs. 69.5%, p ! 0.0001). Patients between the ages of 65-74 years were also associated with better survival compared to patients R 85 years of age (median survival of 8.9 months vs. 3.8 months). Those patients who had PEG placement between 2001 and 2003 had significantly lower 1-year survival (p !0.0001) [Table 1]. Conclusions: In a noncancer population, the rate of feeding tube placement has nearly tripled from 1992-2003, which is consistent with the aging population and the increased incidences of dementia and stroke. Increased survival was observed in women and younger patients. In patients who survived the first year following PEG placement, the risk of dying over the subsequent 2 years was significantly lower. With significant increases in PEG tube utilization over time, survival rates at 1 and 3 years
AB360 GASTROINTESTINAL ENDOSCOPY Volume 65, No. 5 : 2007
have decreased. Further research is needed to explore the underlying causes of this observed disparity. Table 1 Cohorts
19921994
Incidence rate 7.5 Rate ratio [95% CI] 1.0 1 month survival (%) 82.2 1 year survival (%) 51.2 3 year survival (%) 30.5
19951997
19982000
20012003
10.6 16.6 19.4 Overall 1.4 [1.29-1.52] 2.2 [2.04-2.37] 2.6 [2.38-2.76] 82.0 81.9 78.5 81.2 53.5 50.8 41.0 49.0 36.8 28.8 N/A 36.3
W1416 Endoscopic Ultrasound: A Meta-Analysis of Test Performance in Suspected Choledocholithiasis Frances Tse, Louis W. Liu, David Armstrong, Alan N. Barkun, Paul Moayyedi Objectives: To precisely estimate the diagnostic accuracy of endoscopic ultrasound (EUS) in suspected choledocholithiasis and to explore reasons for variation in diagnostic accuracy. Methods: MEDLINE and EMBASE were used to identify prospective cohort studies in which the results of EUS were compared with the results of an acceptable reference standard including endoscopic retrograde cholangiopancreatography, intraoperative cholangiography, or surgical exploration. Two independent reviewers extracted standardized data and assessed trial quality using validated criteria. Random effects model was used to estimate the sensitivity, specificity, likelihood and diagnostic odds ratio, and a summary receiver operating characteristic curve was constructed. All predefined potential sources of heterogeneity were explored by subgroup analysis and meta-regression. Results: Of 165 studies identified, 27 were included (2673 patients). Study quality was generally low. EUS had a high overall pooled sensitivity of 0.94 (95% CI 0.93-0.96), specificity of 0.95 (95% CI 0.94 - 0.96), positive likelihood ratio of 22.41 (95% CI 12.53-40.08), negative likelihood ratio of 0.09 (0.06-0.12), diagnostic odds ratio of 312.15 (163.42596.24), and an area under the curve of 0.98 that was not affected by patient presentation. Three variables appeared to yield higher diagnostic odds ratio: higher disease prevalence, adequate time interval between index test and reference standards, and presence of verification bias. Conclusion: EUS is a non-invasive imaging test with excellent overall sensitivity and specificity for diagnosing choledocholithiasis. EUS should, therefore, be used to select patients for therapeutic ERCP in order to minimize the risk of complications and death associated with unnecessary diagnostic ERCP.
SROC Curve for All Studies
www.giejournal.org
W1414 Long-Term Outcome of Endoscopic Pancreatic Necrosectomy: Final Results of the First German Multicenter Trial Hans Seifert, Michael Biermer, Wolfgang Schmitt, Uwe Will, Michael Hocke, Christian Ju ¨Rgensen, Christian Kreitmayr, Christian Prinz, Thomas Roesch Background: Pancreatic endotherapy of necroses following acute pancreatitis consists of drainage into the upper GI tract (stomach and duodenum), followed by widening the access and removing necrotic debris by direct vision with an endoscope introduced into the cavity. Following presentation of the initial results of the German Multicenter Study on Endoscopic Pancreatic Retroperitoneal Debridement(GEPARD) we now report the final outcome including long-term follow-up results. Methods: In a multicenter setting 60 patients (70% male, mean age 60 years, range 35-83) with pancreatic abscesses and necroses treated between 1999 and 2003 were included from 1999-2003 into this study. Their clinical conditions was determined by acute/postacute pancreatitis with subacute or acute infectious signs. Drainage was mostly transgastrally by endoscopic or endosonographic access, or percutaneously (n Z 2), with subsequent balloon dilatation of the tract to introduce a gastroscope to remove necroses and debris by snares, baskets and forceps as well as by lavage and suction, usually in several sessions. Immediate clinical and morphologic outcome on hospital discharge and long-term results after a mean follow-up of 58 months (range 10-89) were recorded. Results:A median of 15 sessions (range 15-35) including initial access to the cavity were performed to remove all debris as far as possible. Acute complications were encountered in 8 cases (2 perforations, 5 bleeding, 1 pneumoperitoneum). Initially, complete and partial (minimal symptoms, remaining collection !3 cm) clinical and morphologic success could achieved in 73% of cases. 11 patients were operated due to various reasons (pancreatic surgery in 4 cases each during initial treatment and later). Follow-up showed that 8 patients had died, 1 of those initially due to an acute complication after the intervention, and 3 within 6 months after necrosectomy. 14 patients had to be readmitted for repeated therapy due to recurrence. Clinical results on long-term follow-up showed that 77% of the surviving patients were in excellent/good/moderate clinical condition, and that 16, 18 and 24 patients reported on pain, exocrine insufficiency and diabetes. Conclusions: Endoscopic pancreatic necrosectomy has successful clinical long term outcome in 3/4 of patients; recurrence rate and associated problems necessitating surgery and endoscopic reinterventions was 42%. There is a related mortality rate which is partially due to interventions (more data are presently accumulated from more centers) but mostly due to underlying disease. Further prospective studies will clarify the precise role of this transgastric retroperitoneal endotherapy.
W1415 Incidence and Mortality Following Percutaneous Endoscopic Gastrostomy (PEG) Placement in Medicare Beneficiaries from 1992 to 2003 Richard C. Wong, Doug Kuo, Gregory D. Olds, Gregory S. Cooper Background: PEG placement is commonly performed to deliver enteral nutrition and medications in patients who are unable to swallow because of conditions such as dementia and stroke. Although the rate of feeding tube placement is increasing, a detailed longitudinal analysis of mortality has not been conducted. Objective: Analysis of patient mortality following PEG placement in noncancer patients. Methods: All inpatient claims from 1992 to 2003 from a 5% random sample of cancer-free Medicare beneficiaries age 66 years and older were examined. Patients who were enrolled in Medicare Part A and were in fee-for-service arrangements were included. Study cohorts were divided into 4 groups according to when the initial PEG was placed. Survival was measured from the date of index PEG placement and for up to 3 years of follow-up. Kaplan-Meier analyses and Cox regression including adjustment for comorbidity were used. Results: 7,437 noncancer patients received PEG placement during the study period. The incidence rate of PEG placement has nearly tripled from 7.5 to 19.4 per 1000 over the study period. Adjusted overall survival at 30 days, 1 year and 3 years were 81.2%, 49% and 36.3%, respectively, with the greatest hazard of death within the first year. Women had better 1-year adjusted survival compared to men (75.6% vs. 69.5%, p ! 0.0001). Patients between the ages of 65-74 years were also associated with better survival compared to patients R 85 years of age (median survival of 8.9 months vs. 3.8 months). Those patients who had PEG placement between 2001 and 2003 had significantly lower 1-year survival (p !0.0001) [Table 1]. Conclusions: In a noncancer population, the rate of feeding tube placement has nearly tripled from 1992-2003, which is consistent with the aging population and the increased incidences of dementia and stroke. Increased survival was observed in women and younger patients. In patients who survived the first year following PEG placement, the risk of dying over the subsequent 2 years was significantly lower. With significant increases in PEG tube utilization over time, survival rates at 1 and 3 years
AB360 GASTROINTESTINAL ENDOSCOPY Volume 65, No. 5 : 2007
have decreased. Further research is needed to explore the underlying causes of this observed disparity. Table 1 Cohorts
19921994
Incidence rate 7.5 Rate ratio [95% CI] 1.0 1 month survival (%) 82.2 1 year survival (%) 51.2 3 year survival (%) 30.5
19951997
19982000
20012003
10.6 16.6 19.4 Overall 1.4 [1.29-1.52] 2.2 [2.04-2.37] 2.6 [2.38-2.76] 82.0 81.9 78.5 81.2 53.5 50.8 41.0 49.0 36.8 28.8 N/A 36.3
W1416 Endoscopic Ultrasound: A Meta-Analysis of Test Performance in Suspected Choledocholithiasis Frances Tse, Louis W. Liu, David Armstrong, Alan N. Barkun, Paul Moayyedi Objectives: To precisely estimate the diagnostic accuracy of endoscopic ultrasound (EUS) in suspected choledocholithiasis and to explore reasons for variation in diagnostic accuracy. Methods: MEDLINE and EMBASE were used to identify prospective cohort studies in which the results of EUS were compared with the results of an acceptable reference standard including endoscopic retrograde cholangiopancreatography, intraoperative cholangiography, or surgical exploration. Two independent reviewers extracted standardized data and assessed trial quality using validated criteria. Random effects model was used to estimate the sensitivity, specificity, likelihood and diagnostic odds ratio, and a summary receiver operating characteristic curve was constructed. All predefined potential sources of heterogeneity were explored by subgroup analysis and meta-regression. Results: Of 165 studies identified, 27 were included (2673 patients). Study quality was generally low. EUS had a high overall pooled sensitivity of 0.94 (95% CI 0.93-0.96), specificity of 0.95 (95% CI 0.94 - 0.96), positive likelihood ratio of 22.41 (95% CI 12.53-40.08), negative likelihood ratio of 0.09 (0.06-0.12), diagnostic odds ratio of 312.15 (163.42596.24), and an area under the curve of 0.98 that was not affected by patient presentation. Three variables appeared to yield higher diagnostic odds ratio: higher disease prevalence, adequate time interval between index test and reference standards, and presence of verification bias. Conclusion: EUS is a non-invasive imaging test with excellent overall sensitivity and specificity for diagnosing choledocholithiasis. EUS should, therefore, be used to select patients for therapeutic ERCP in order to minimize the risk of complications and death associated with unnecessary diagnostic ERCP.
SROC Curve for All Studies
www.giejournal.org