BACKGROUND Percutaneous ethanol injection has been widely used as a non-surgical therapy for liver cancer, but it has some shortcomings such as local diffusion and unequal permeation. This study was designed to observe the volume, controllability and completeness of necrosis after injection of low concentration sodium hydroxide in the normal liver parenchyma so as to assess its possibility in treatment of liver cancer instead of ethanol. METHODS Twenty-seven New Zealand rabbits were divided randomly into 9 groups (Aa, Ab, Ac, Ba, Bb, Bc, Ca, Cb, and Cc) by a 3 x 3(three-by-three) factorial design, each consisting of 3 rabbits. Group A was given sodium hydroxide solution at a concentration of 5%, while B at 2.5% and C at 1% in liver parenchyma. Each group received three doses of the solution: a (0.2 ml), b (0.5 ml) and c (1.0 ml). Then another 3 rabbits as side-effect group were dropped with sodium hydroxide solution in their liver lobe space. Liver and renal function changes in all the rabbits were compared after injection with pre-injection. RESULTS All the lesions were localized. At the concentration of 2.5% and 5%, the lesion volume increased with the dose increased from 0.2 ml to 1.0 ml (P<0.05). No significant differences were found in the lesion volume of the groups receiving the same dose but different concentration. Changes in liver and renal function were not significant 7 days after injection, compared with those before injection. CONCLUSIONS 2.5% and 5% sodium hydroxide solution could control local complete necrosis in normal liver. With regard to safety, 2.5% alkali solution is considered promising as a new agent for intratumoral injection therapy instead of ethanol.