K P Koo1, L W Traverso. 1. Department of General Surgery, Virginia Mason Medical Center, Seattle, Washington, USA.
Abstract
BACKGROUND: Criteria have been suggested to help decide if an intraoperative cholangiogram (IOC) should be performed during laparoscopic cholecystectomy (LC). They are a clinical history of passing a common bile duct (CBD) stone, elevated serum amylase, elevated liver function tests, or ultrasound findings suggesting a CBD stone. What is the sensitivity and specificity of the above criteria when the presence or absence of CBD stones is already known by IOC? What is the probability that these criteria will predict a CBD stone or be normal if a stone is absent, ie, the positive predictive value (PPV) and negative predictive value (NPV)? METHODS: We reviewed 420 cases of elective LC done between May 1990 and December 1992. In our teaching hospital, IOC is routine and acted as the reference standard for the presence of CBD stones. All 420 films were reviewed as well as the results of any preoperative endoscopic retrograde cholangiopancreatography (ERCP) (30 were done). The following preoperative indicators of CBD stones were recorded: any clinical history of CBD stones; an elevated amylase, SGOT, alkaline phosphatase, or bilirubin level; and ultrasound findings. The sensitivity, specificity, PPV, and NPV were calculated. RESULTS: CBD stones were found in 12% of these elective LC cases. The sensitivity, specificity, PPV, and NPV, respectively, for each preoperative indicator were: a history suggestive of CBD stones (36%, 94%, 45%, and 91%), serum biochemistries as a group (43%, 86%, 30% and 91%), and ultrasound findings of CBD stones (22%, 92%, 28% and 89%). We compared any elevation versus > 2x from the normal range of the serum indicators and did not improve their accuracy. Combination of the indicators increased sensitivity and NPV but lowered specificity and PPV. The best predictor of CBD stones was the history (45%) and this was in a hospital with a CBD stone prevalence rate of 12%. CONCLUSION: There are no predictive tests that can sufficiently increase an observer's probability estimate of the presence or absence of CBD stones to allow for "selective" IOC decisions.
BACKGROUND: Criteria have been suggested to help decide if an intraoperative cholangiogram (IOC) should be performed during laparoscopic cholecystectomy (LC). They are a clinical history of passing a common bile duct (CBD) stone, elevated serum amylase, elevated liver function tests, or ultrasound findings suggesting a CBD stone. What is the sensitivity and specificity of the above criteria when the presence or absence of CBD stones is already known by IOC? What is the probability that these criteria will predict a CBD stone or be normal if a stone is absent, ie, the positive predictive value (PPV) and negative predictive value (NPV)? METHODS: We reviewed 420 cases of elective LC done between May 1990 and December 1992. In our teaching hospital, IOC is routine and acted as the reference standard for the presence of CBD stones. All 420 films were reviewed as well as the results of any preoperative endoscopic retrograde cholangiopancreatography (ERCP) (30 were done). The following preoperative indicators of CBD stones were recorded: any clinical history of CBD stones; an elevated amylase, SGOT, alkaline phosphatase, or bilirubin level; and ultrasound findings. The sensitivity, specificity, PPV, and NPV were calculated. RESULTS: CBD stones were found in 12% of these elective LC cases. The sensitivity, specificity, PPV, and NPV, respectively, for each preoperative indicator were: a history suggestive of CBD stones (36%, 94%, 45%, and 91%), serum biochemistries as a group (43%, 86%, 30% and 91%), and ultrasound findings of CBD stones (22%, 92%, 28% and 89%). We compared any elevation versus > 2x from the normal range of the serum indicators and did not improve their accuracy. Combination of the indicators increased sensitivity and NPV but lowered specificity and PPV. The best predictor of CBD stones was the history (45%) and this was in a hospital with a CBD stone prevalence rate of 12%. CONCLUSION: There are no predictive tests that can sufficiently increase an observer's probability estimate of the presence or absence of CBD stones to allow for "selective" IOC decisions.
Authors: V Bresadola; S Intini; G Terrosu; U Baccarani; M G Marcellino; M Sistu; F Scanavacca; F Bresadola Journal: Surg Endosc Date: 2001-05-11 Impact factor: 4.584