Back to Annual Meeting Program

BACKGROUND: Non-alcoholic fatty liver disease (NAFLD) leads to a chronic inflammatory state hypothesized as carcinogenic. The clinical significance of NAFLD for patients diagnosed with ICC is not known. The aim of this study was to evaluate the significance of NAFLD on overall survival (OS) for patients with ICC.
METHODS: In this single-institution, retrospective cohort study, all patients who underwent curative resection for ICC from 1997 to 2011 were identified. NAFLD was defined as pathologic evidence of steatosis or steatohepatitis in uninvolved liver parenchyma at time of resection or preoperative liver biopsy. Patients with clinical history or pathologic evidence of underlying liver disease other than NAFLD (n = 17) and patients who died of perioperative complications (n=2) were excluded from survival analyses. Kaplan-Meier estimates and Cox regression proportional hazards were used to evaluate predictors of OS.
RESULTS: One-hundred and thirty-eight patients (median age 60.5; 39.9% male) underwent curative resection for ICC with pathologic evidence of NAFLD in 18 (13.0%) patients. Median follow-up was 29 months (IQR 43) and median OS was 55 months. Age, sex, ASA, MELD score, CA 19-9, and BMI were similar between NAFLD and non-NAFLD patients. AJCC 7th Edition T-Stage was significantly lower in NAFLD patients compared with non-NAFLD patients: T1 50% vs 44%; T2a 22% vs 2.5%; T2b 6% vs 23%; T3 11% vs 13%, and T4 11% vs 18% (p = 0.02). However, AJCC 7th Edition TNM Stage was not significantly different between NAFLD and non-NAFLD patients (p = 0.56). In univariate analysis, predictors of decreased OS were larger tumor size (p < 0.01), node positive disease (p < 0.01), and presence of multiple tumors (p = 0.02). Pathologic evidence of NAFLD did not predict decreased OS (HR 1.5, 95% CI 0.8-3.0, p = 0.20). In subgroup analysis of node-negative patients (n = 77), NAFLD was a predictor of decreased OS (HR 3.7, 95% CI 1.6-8.5, p < 0.01). Other predictors of worse OS in node-negative patients were tumor size, ASA, and positive resection margin (all p ≤ 0.04). After adjusting for other significant covariates in the node-negative cohort, NAFLD was an independent predictor of decreased OS (HR 2.7, 95% CI 1.1-6.6, p = 0.03).
CONCLUSIONS: Although NAFLD is not a predictor of OS for all patients undergoing hepatic resection for ICC, NAFLD is associated with decreased OS in patients with node-negative ICC. This adverse correlation of NAFLD and OS in patients with ICC suggests that treatment strategies should include management of NAFLD.