Clinical significance and risk factors of postembolization fever in patients with hepatocellular carcinoma

Aim: To investigate tumor response and survival in patients with postembolization fever (PEF) and to determine the risk factors for PEF.

Methods: Four hundred forty-three hepatocellular carcinoma (HCC) patients who underwent the first session of transcatheter arterial chemoembolization (TACE) between January 2005 and December 2009 were analyzed retrospectively. PEF was defined as a body temperature greater than 38.0 °C that developed within 3 d of TACE without evidence of infection. The tumor progression-free interval was defined as the interval from the first TACE to the second TACE based on mRECIST criteria. Clinical staging was based on the American Joint Committee on Cancer tumor, node, metastases (TNM) classification of malignant tumors. All patients were admitted before their 1(st) TACE treatment, and blood samples were obtained from all patients before and after treatment. Clinicoradiological variables and host-related variables were compared between two groups: patients with PEF vs patients without PEF. Additionally, variables related to 20-mo mortality and tumor progression-free survival were analyzed.

Results: The study population comprised 370 (85.4%) men and 73 (14.6%) women with a mean age of 62.29 ± 10.35 years. A total of 1836 TACE sessions were conducted in 443 patients, and each patient received between 1 and 27 (mean: 4.14 ± 3.57) TACE sessions. The mean follow-up duration was 22.23 ± 19.6 mo (range: 0-81 mo). PEF developed in 117 patients (26.41%) at the time of the first TACE session. PEF was not associated with 20-mo survival (P = 0.524) or computed tomography (CT) response (P = 0.413) in a univariate analysis. A univariate analysis further indicated that diffuse-type HCC (P = 0.021), large tumor size (≥ 5 cm) (P = 0.046), lipiodol dose (≥ 7 mL, P = 0.001), poor blood glucose control (P = 0.034), alanine aminotransferase (ALT) value after TACE (P = 0.004) and C-reactive protein (CRP) value after TACE (P = 0.036) served as possible risk factors correlated with PEF. The ALT value after TACE (P = 0.021) and lipiodol dose over 7 mL (P = 0.011) were independent risk factors for PEF in the multivariate analysis. For the 20-mo survival, poor blood sugar control (P < 0.001), portal vein thrombosis (P = 0.001), favorable CT response after TACE (P < 0.001), initial aspartate aminotransferase (P = 0.02), initial CRP (P = 0.042), tumor size (P < 0.001), TNM stage (P < 0.001) and lipiodol dose (P < 0.001) were possible risk factors in the univariate analysis. Tumor size (P = 0.03), poor blood sugar control (P = 0.043), and portal vein thrombosis (P = 0.031) were significant predictors of survival in the multivariate analysis. Furthermore, the tumor progression-free interval was closely associated with CRP > 1 mg/dL (P = 0.003), tumor size > 5 cm (P < 0.001), tumor type (poorly defined) (P < 0.001), and lipiodol dose (> 7 mL, P < 0.001).

Conclusion: PEF has no impact on survival at 20 mo or radiologic response. However, the ALT level after TACE and the lipiodol dose represent significant risk factors for PEF.