Clinical discrimination between choledocholithiasis and biliopancreatic malignancy based on a new biochemical model.
Abstract
Purpose: This study aims to establish predictive laboratory
tests, which could confidently assist for an initial clinical discrimination
of choledocholithiasis from biliopancreatic malignancy,
before an invasive endoscopic or surgical diagnosis.
Results: A total of 174 patients, who underwent ERCP
were analyzed. Patients with final diagnosis of choledocholithiasis
(137 patients) and biliopancreatic adenocarcinoma
(37 patients) had their biochemistry parameters compared
using Mann-Whittney test. Cut-off values for each parameter
were defined. The cut-off values that provided the best
trade-off between sensitivity and specificity were: 8.65 mg/
dL for total bilirubin, 276 U/L for serum alkaline phosphatase
and 306 IU/ml for CA19-9. A patient most probably
(96%) suffers from cancer, if he has high incriminating values
in all these three parameters. Conclusions: A simple, reproducible,
easy-to-obtain predictive model with laboratory
tests, successfully differentiates choledocholithiasis from
malignant biliopancreatic diseases and could be useful for a
more cost-effective investigation and treatment of patients
with such pathology.
tests, which could confidently assist for an initial clinical discrimination
of choledocholithiasis from biliopancreatic malignancy,
before an invasive endoscopic or surgical diagnosis.
Results: A total of 174 patients, who underwent ERCP
were analyzed. Patients with final diagnosis of choledocholithiasis
(137 patients) and biliopancreatic adenocarcinoma
(37 patients) had their biochemistry parameters compared
using Mann-Whittney test. Cut-off values for each parameter
were defined. The cut-off values that provided the best
trade-off between sensitivity and specificity were: 8.65 mg/
dL for total bilirubin, 276 U/L for serum alkaline phosphatase
and 306 IU/ml for CA19-9. A patient most probably
(96%) suffers from cancer, if he has high incriminating values
in all these three parameters. Conclusions: A simple, reproducible,
easy-to-obtain predictive model with laboratory
tests, successfully differentiates choledocholithiasis from
malignant biliopancreatic diseases and could be useful for a
more cost-effective investigation and treatment of patients
with such pathology.