3418 Biliary obstruction following billroth ii gastrectomy: comparison of ercp and mrcp.

Abstract

Objective: Endoscopic retrograde cholangiopancreatography (ERCP) is more technically difficult post-Billroth II gastrectomy (BII). Magnetic resonance cholangiopancreatography (MRCP) is a new noninvasive radiological technique available to evaluate the pancreaticobiliary tree. Our objective was to compare clinical outcomes and direct medical costs associated with each of these modalities in post-BII patients presenting with suspected biliary obstruction. Methods: Using decision analysis software (Data 3.5, TreeAge Software Inc.) we considered a hypothetical cohort of 100 patients. Two strategies were considered as the first step in the imaging of the biliary system: i) ERCP and ii) MRCP. The results of these imaging modalities were divided into four categories: common bile duct (CBD) stones, malignant obstruction, benign obstruction and no abnormality. The relative frequency of these results, test sensitivities, test failure rates, complications and procedural costs were estimated from a retrospective review of hospital charts, expert health care professional estimation, local costs and a MEDLINE searh. Direct medical costs ($Cdn) from a third party payer perspective, diagnostic accuracy, total number of procedures and untoward events were then calculated. Results: This model estimated the total cost per patient associated with the ERCP and MRCP strategies at $2,059 and $1,751 respectively. The ERCP strategy resulted in 4.3 more correct diagnoses, 29 fewer procedures and 0.6 fewer untoward events per 100 patients. Compared with MRCP, the ERCP strategy produced cost effectiveness ratios of $7,163/correct diagnosis, $1,047/procedure avoided and $51,333/untoward event prevented. Sensitivity analysis demonstrated that the ERCP strategy was less expensive than the MRCP strategy if the probability of a CBD stone was greater than 45%. However, there were fewer untoward events with the MRCP strategy with an MRCP sensitivity greater than 91% for CBD stones. Conclusion: At present in Canada, ERCP is more expensive but has better clinical outcomes than MRCP as an initial diagnostic test in post-BII patients with suspected biliary obstruction.With improved sensitivity and accessibility, MRCP may prove to be the diagnostic test of choice. Objective: Endoscopic retrograde cholangiopancreatography (ERCP) is more technically difficult post-Billroth II gastrectomy (BII). Magnetic resonance cholangiopancreatography (MRCP) is a new noninvasive radiological technique available to evaluate the pancreaticobiliary tree. Our objective was to compare clinical outcomes and direct medical costs associated with each of these modalities in post-BII patients presenting with suspected biliary obstruction. Methods: Using decision analysis software (Data 3.5, TreeAge Software Inc.) we considered a hypothetical cohort of 100 patients. Two strategies were considered as the first step in the imaging of the biliary system: i) ERCP and ii) MRCP. The results of these imaging modalities were divided into four categories: common bile duct (CBD) stones, malignant obstruction, benign obstruction and no abnormality. The relative frequency of these results, test sensitivities, test failure rates, complications and procedural costs were estimated from a retrospective review of hospital charts, expert health care professional estimation, local costs and a MEDLINE searh. Direct medical costs ($Cdn) from a third party payer perspective, diagnostic accuracy, total number of procedures and untoward events were then calculated. Results: This model estimated the total cost per patient associated with the ERCP and MRCP strategies at $2,059 and $1,751 respectively. The ERCP strategy resulted in 4.3 more correct diagnoses, 29 fewer procedures and 0.6 fewer untoward events per 100 patients. Compared with MRCP, the ERCP strategy produced cost effectiveness ratios of $7,163/correct diagnosis, $1,047/procedure avoided and $51,333/untoward event prevented. Sensitivity analysis demonstrated that the ERCP strategy was less expensive than the MRCP strategy if the probability of a CBD stone was greater than 45%. However, there were fewer untoward events with the MRCP strategy with an MRCP sensitivity greater than 91% for CBD stones. Conclusion: At present in Canada, ERCP is more expensive but has better clinical outcomes than MRCP as an initial diagnostic test in post-BII patients with suspected biliary obstruction.With improved sensitivity and accessibility, MRCP may prove to be the diagnostic test of choice.