Cholangiocarcinoma or IgG4-associated cholangitis: how feasible it is to avoid unnecessary surgical interventions?

OBJECTIVE: To evaluate the experience of a tertiary hepatopancreaticobiliary (HPB) center in the diagnostic approach and management of patients with suspicion of cholangiocarcinoma (CCa), focusing on excluding patients with IgG4-associated cholangitis (IAC) from unnecessary major surgical interventions.

METHODS: Between January 2008 and September 2010, a total number of 152 patients with suspicion of CCa underwent evaluation through a HPB multidisciplinary team meeting. Patients without tissue diagnosis were managed surgically or medically on the basis of probable presence of IAC as underlying pathology. Serology, immunostaining, and imaging were reviewed and analyzed according to the HISORt (Histology, Imaging, Serology, Other organ involvement, Response to therapy) criteria for IAC.

RESULTS: Tissue diagnosis during the diagnostic workup was achieved in 104 patients (68%), whereas the remaining 48 were classified as "highly suspicious for CCa" (n = 35) or as "probable IAC" (n = 13). Among 16 "highly suspicious for CCa" patients who underwent surgery, pathology revealed 2 patients harboring IAC (n = 1) and a benign chronic inflammatory biliary stricture (n = 1), respectively. Among the 13 patients with primarily medical management as "probable IAC," final diagnosis was CCa (n = 3) and IAC (n = 9), while 1 patient had no proven diagnosis. The accuracy of serum IgG4 for diagnosis of IAC reached 60%. Sensitivity and specificity of immunostaining for IAC in biopsy specimens were 56% and 89%, respectively. Imaging features suggesting IAC yielded sensitivity, specificity, and accuracy of 75%, 89%, and 83%, respectively. Initial imaging was revised at the referral institute in 75% of IAC patients (P = 0.009), while an isolated stricture (P = 0.038), a biliary mass (P = 0.006), and normal pancreas on computed tomography (P = 0.01) were statistically significant parameters for distinguishing between CCa and IAC. The mean time for establishing a diagnosis of IAC was 12.4 months (range: 2.5-32 months).

CONCLUSIONS: Differential diagnosis between CCa and IAC mandates high index of suspicion and low threshold for referral in high volume institutes. The delayed establishment of diagnosis particularly for CCa needs to be balanced versus avoiding unnecessary surgery for IAC. Imaging features may be most helpful for optimal management.