Background: The lung is the most common site of distant metastases from hepatocellular carcinoma. Correct differentiation between metastatic hepatocellular carcinoma of the lung and primary lung cancer is sometimes difficult without biopsy.
Purpose: To evaluate the usefulness of measuring the attenuations of pulmonary nodules on early-phase contrast-enhanced computed tomography (CT) for the differentiation of pulmonary metastases from hepatocellular carcinoma and primary lung cancer.
Material and Methods: Thirteen patients with pulmonary metastases from hepatocellular carcinoma (nine men, four women; age 53.9±14.2 years, range 16–70 years) and 25 patients with primary lung cancer (14 men, 11 women; age 62.2±9.4 years, range 43–72 years) were retrospectively evaluated. Contrast-enhanced scans were obtained 35 s after commencing intravenous injection of contrast medium. Attenuation values and the size of the pulmonary nodules were measured on contrast-enhanced CT scans. CT and clinical features were analyzed with regard to age, sex, body surface area of the patients, the attenuation values and size of the nodules, and CT machines using univariate analysis (Fisher's exact test for binary data sets and the Mann-Whitney U test for continuous data sets). Multiple linear regression analysis was used to eliminate confounding factors.
Results: The mean attenuation value of metastatic pulmonary nodules from hepatocellular carcinoma (75.7±24.9 HU) was higher than that of primary lung cancer nodules (45.8±14.4 HU) (P<0.01). Other variables such as age, sex, body surface area of the patients, CT device, and nodule size were not significant variables on multiple regression analysis. When a cut-off value of 75 HU was applied, the positive predictive value for diagnosing metastatic nodules from hepatocellular carcinoma was 100%.
Conclusion: Pending confirmation in a large study, our findings suggest that there is a difference in contrast enhancement between pulmonary metastases from hepatocellular carcinoma and primary lung cancer.
