Aims Our aim was to predict the malignancy in patients with cholestasis before ERCP.
Methods In a retrospective single-center study, we investigated the ERCP records of patients
between 12.2016- 04.2021. All patients had abdominal imaging before the ERCP. Laboratory
parameters were derived at the time of admission. Statistical analyses were performed
using SPSS. Receiver operating characteristic (ROC) curve analysis was used to determine
the cut-off values for predicting malignancy pre-ERCP.
Results 267 patients were included in the analysis. In ERCP, 35 (13.1%) patients had normal
common bile duct (CBD), 56 (20.9%) patients had dilated CBD, 18 (6.7%) patients had
sludge in CBD, stone extracted from 128 (48%) patients in ERCP, 25 (9.4%) patient
had a periampullary tumor and 5 (1.9%) patients had CBD tumor. There were significant
differences between groups by means of ALP, total bilirubin, direct bilirubin, diameter
of CBD. According to ROC curve analysis, the best cut-off ALP value to differentiate
between patients with malignancy from control group was 285 (Sens:53,3 ; Spec:80,6
PPV:25,8 ; NPV:93,2), best cut-off total bilirubin value was 3 (Sens:73,3 ; Spec:73,4
; PPV:25,9 ; NPV:95,6), best cut-off direct bilirubin value was 2 (Sens:73,3 ; Spec:72
; PPV:25 ; NPV:95,5), and best cut-off CBD value was 10 (Sens:80,8 ; Spec:50,9 ; PPV:16,8
; NPV:95,6).
Conclusions Among all parameters, ALP, total bilirubin, direct bilirubin, the diameter of CBD
were statistically significant. The cut-off values were 285, 3, 2, 10 respectively.
Thus, both parameters may be used to predict malignancy before ERCP. Large-scale,
prospective studies are needed for further conclusions.
