Abstract: Objective:To investigate clinical efficacy and prognostic factors of hilar cholangiocarcinoma.
Methods:The retrospective casecontrol study was conducted. The clinicopathological data of 322 patients with hilar cholangiocarcinoma who were admitted to the Xiangya Hospital of Central South University between December 2005 and November 2015 were collected. Preoperative staging and classification of tumor and treatment planning were carried out according to the results of laboratory and imaging examinations. Observation indexes: (1) clinical features and results of assisted examinations; (2) treatments and results of pathological examination; (3) followup and survival; (4) prognostic factors analysis: gender, age, preoperative highest total bilirubin (TBil), preoperative carcinoembryonic antigen (CEA),preoperative CA199, preoperative CA242, preoperative CA125, treatment methods and TNM staging. The followup of outpatient examination and telephone interview was performed to detect patients′ survival up to November 2016. Survival curve was drawn using the KaplanMeier method. Survival and univariate analyses were done using the Logrank test, and multivariate analysis was done using the Cox proportional hazard model.
Results:(1) Clinical features and results of assisted examinations: among the 322 patients, there were 301 patients with a chief complaint of jaundice. Of the 322 patients, the preoperative highest levels of TBil, DBil, ALT and AST in 322 patients were 3.9-785.2 μmol/L, 1.6-410.2 μmol/L, 14.8-484.5 U/L and 21.4-539.8 U/L, respectively. Levels of ALP and GGT in 272 patients were 93.8-1 890.0 U/L and 2.0-1 832.8 U/L, respectively. Seventyseven of 292 patients had an elevated CEA level, 272 of 298 patients had an elevated CA199 level, 153 of 260 patients had an elevated CA242 level and 86 of 260 patients had an elevated CA125 level. According to BismuthCorlette type, 24 patients were detected in type Ⅰ, 115 in type Ⅱ, 55 in type Ⅲa, 63 in type Ⅲb and 65 in type Ⅳ. (2) Treatments and results of pathological examination: 0f the 322 patients, 104 patients underwent radical resection, including 79 with hilar bile duct resection (9 combined with vascular resection and reconstruction) and 25 with extended hepatic lobectomy (16 combined with caudate lobectomy), and 218 patients underwent palliative treatments, including 134 with external biliary drainage and 84 with internal biliary drainage. Five patients were dead in the perioperative period, of which 2 died of acute liver failure, 1 died of systemic infection and multiple organ failure, 1 died of acute renal failure and 1 died of acute suppurative cholangitis, septic shock and disseminated intravascular coagulation. Of 263 patients receiving pathological examination, adenocarcinoma was detected in 253 patients (12 with highdifferentiated adenocarcinoma, 85 with moderatedifferentiated adenocarcinoma, 33 with lowdifferentiated adenocarcinoma and 123 with indefinite differentiation), mucinous adenocarcinoma in 5 patients, cholangiocarcinoma in 3 patients and neuroendocrine carcinoma in 2 patients. TNM staging of 322 patients: stage Ⅰ was detected in 8 patients，stage Ⅱ in 53 patients, stage Ⅲ in 132 patients, stage Ⅳ in 96 patients and indefinite stage in 33 patients. (3) Follow up and survival: among the 322 patients, 296 were followed up for 12-132 months, with a median followup time of 65 months, including 94 with radical resection and 202 with palliative treatments. Among the 296 patients, the median survival time and 1, 3, 5year survival rates were 10 months, 47.1%, 20.2% and 9.5%, respectively. Of 296 patients with followup, median survival time and 1, 3, 5year survival rates were 31 months, 84.0%, 46.2%, 25.0% in 94 patients receiving radical resection and 7 months, 29.9%, 8.1% and 2.3% in 202 patients receiving palliative treatment, respectively, with a statistically significant difference between the 2 groups (x²=78.777, P<0.05). Among the 94 patients receiving followup and radical resection, the median survival time and 1, 3, 5year survival rates were 31 months, 82.1%, 45.1%, 25.7% in 73 patients undergoing hilar bile duct resection and 35 months, 90.5%, 49.8%, 22.1% in 21 patients undergoing hepatic lobectomy, respectively, with no statistically significant difference (x²=0.186, P>0.05). Among the 73 patients undergoing hilar bile duct resection, median survival time and 1, 3, 5year survival rates were 16 months, 57.1%, 0, 0 in 7 patients combined with vascular resection and reconstruction and 34 months, 84.6%, 49.5%, 27.5% in 66 patients undergoing simplex hilar bile duct resection, respectively, showing a statistically significant difference (x²=11.977, P<0.05). (4) Prognostic factors analysis: results of univariate analysis showed that preoperative highest TBil, preoperative CEA, preoperative CA242, preoperative CA125, treatment methods and TNM staging were related factors affecting prognosis of patients with hilar cholangiocarcinoma (x²=25.009, 18.671, 9.359, 33.628, 94.729, 77.136, P<0.05). Multivariate analysis showed that preoperative highest TBil≥342.0 μmol/L, preoperative CEA≥5.00 μg/L, palliative treatments, TNM stage Ⅲ and Ⅳ were the independent risk factors affecting the poor prognosis of patients with hilar cholangiocarcinoma (HR=2.270, 2.147, 3.166, 2.351, 95% confidence interval: 1.587-3.247, 1.446-3.188, 2.117-4.734, 1.489-3.712, P<0.05).
Conclusions:Prognosis of hilar cholangiocarcinoma is still unsatisfactory. The R₀ resection is the key in radical surgery. Preoperative highest TBil≥342.0 μmol/L, preoperative CEA≥5.00 μg/L, palliative treatments, TNM stage Ⅲ-Ⅳ are independent risk factors affecting the poor prognosis of patients with hilar cholangiocarcinoma.