Of course we agree that the appearance of new lesions is a delayed phenomenon after radiofrequency ablation (RFA) or partial resection. We also agree that, being a delayed phenomenon, this factor is of little value in treatment planning and survival prediction. However, even in the presence of adverse prognostic parameters (high alpha-fetoprotein and des-gamma-carboxy prothrombin levels, low grading, and so forth), it is difficult to predict the individual natural history. For instance, our longest survivor presented at the baseline a 3.5-cm infiltrating hepatocellular carcinoma with alpha-fetoprotein level > 400 ng/mL. The patient, initially treated with percutaneous ethanol injection and subsequently with RFA when new lesions appeared, is currently free of desease after 19 years of follow-up. The result of our multivariate analysis simply means that even though the treatment is successful, the prognosis is dependent to the timing of appearance of new lesions.

On the basis of your suggestion, we performed a new multivariate analysis with the exclusion of this factor, but the other factors as operability (0.61 instead of 0.63) remained not significant.

Tito Livraghi*, * Department of Radiology, Ospedale Civil, Vimercate, Italy.