Factors predicting complete necrosis rate after ultrasound-guided percutaneous laser thermoablation of small hepatocellular carcinoma tumors in cirrhotic patients: a multivariate analysis.

PURPOSE: To assess factors affecting the effectiveness of percutaneous laser ablation (PLA) under ultrasound (US) guidance in terms of complete ablation achievement. MATERIAL AND METHODS: The clinical records of 86 hepatocellular carcinoma (HCC) tumors (mean diameter 23.7 mm) in 60 cirrhotic patients (mean age 68.3 years; 36 males; 57 HCV+; 53 Child's class A, seven Child's class B) treated by means of PLA were reviewed. PLA was performed with a continuous-wave Nd:YAG laser by a single operator who positioned two to four 300-microm optic fibers advanced in 21-gauge needles into target lesions under US guidance. Triphasic computed tomography (CT) studies were used to verify treatment effectiveness 1 month after PLA completion. The association between characteristics of the lesion and outcome (complete or incomplete ablation) was evaluated by logistic regression, taking into account the following predictive factors: tumor size, pattern of growth (infiltrating or not) at imaging, location, first diagnosis of HCC (naïve tumors vs. non-naïve tumors), number of sessions (1/ > 1), total delivered energy, and years of treatment in 2001-2002 (first period) vs. 2003-2004 (second period). RESULTS: Complete ablation was obtained in 62 nodules (72%). Statistically significant predictors of incomplete ablation after the first PLA course at both univariate and multivariate analysis included: infiltrating growth pattern (odds ratio (OR) 12.3, P<0.002), non-naïve tumors (OR 8.7, P<0.001), and first period of treatment (OR 10.3, P<0.002). CONCLUSION: The effectiveness of US-guided PLA for HCC tumors < or =4 cm turned out to be negatively affected by both operator-related (the beginning of the operator's experience with the technique) and tumor-related factors (non-naïve, infiltrating HCC tumors).

Multislice CT with single-phase technique in patients with suspected pancreatic cancer.

PURPOSE: The purpose of this study was to evaluate the role of multislice computed tomography (MSCT) with a single-phase technique in patients with suspected pancreatic cancer (PC). MATERIALS AND METHODS: Seventy-eight patients underwent MSCT with the following technical parameters: collimation: 4x1 mm; pitch 1; 120 kVp; 260 mAs. The pre-contrast scan was followed by a single acquisition phase in the caudocranial direction from the inferior hepatic margin to the diaphragm with a 60-s delay after IV administration of 150 ml of iodinated contrast material at a rate of 3 ml/s. Two radiologists assessed the single images independently. Receiver operating characteristics (ROC) curves were obtained for each of the two observers. RESULTS: The final diagnosis was pancreatic cancer in 46 cases and chronic pancreatitis in 32 cases. Areas under the curve (AZ) for diagnosis and evaluation of disease resectability were 0.97 and 0.93 for the first observer (p=ns), and 0.97 and 0.90 for the second observer (p=ns). The mean difference in tissue attenuation values between the cancer and normal pancreas was 72 +/- 3 Hounsfield units (HU). No statistically significant differences were observed in the degree of opacification between the peripancreatic arteries and veins. CONCLUSIONS: MSCT with a single-phase technique is an accurate and reproducible method for diagnosis and evaluation of disease resectability in patients with suspected PC, ensuring optimal tumour-to-pancreas contrast and maximal opacification of the main peripancreatic arterial and venous structures.