The value of serum CA19‐9 in predicting primary sclerosing cholangitis in patients with ulcerative colitis

Abstract

AbstractIn the present study, we postulated that using CA19‐9 in inflammatory bowel disease (IBD) patients may be a useful method for the screening of primary sclerosing cholangitis (PSC). Considering the presence of other extraintestinal manifestations (EIMs) in IBD such as arthritis that may associate with the elevated level of CA19‐9, in the present study we aimed to compare the CA19‐9 level in ulcerative colitis (UC) patients with PSC or other EIMs and also evaluate the value of CA19‐9 in the detection of PSC in UC patients. In the present cross‐sectional study, 192 patients with UC were selected. EIMs were diagnosed based on patients recall and after a thorough physical examination. PSC was diagnosed by magnetic resonance cholangiopancreatography by expert gastroenterologist and radiologist. Serum level of CA19‐9 was measured using chemiluminescent enzyme immunometric assay. About 6.77% of participants had PSC and 24.47% of them had other EIMs. The results showed that the CA19‐9 level in patients with PSC (median [interquartile range (IQR)]: 93 [10‐125]) was significantly higher than that of patients diagnosed with other EIMs (median [IQR]: 11 [6‐17.30]) or with no EIMs (median [IQR]: 7.15 [4.82‐11.00]). CA19‐9 value of >75 U/mL could predict PSC by the sensitivity of 69% and specificity of 99% with an area under the ROC curve (AUC) of 0.83 in UC patients and alkaline phosphatase (ALP) > 299.5 could predict PSC with a sensitivity of 100% and a specificity of 99% and AUC of 0.99. In addition, 23.07% of PSC patients had CA19‐9 level of >129 U/mL. According to the results, ALP had higher sensitivity and specificity for the screening of PSC compared to CA19‐9 in UC patients. Moreover, 23.07% of PSC patients had CA19‐9 level of >129 U/mL. So it could be concluded that cut‐off of >129 U/mL may not be an accurate cut‐off for the diagnosis of CCA in IBD patients.