Despite refinements in surgical techniques for liver transplantation, liver size disparity remains one of the most common problems in pediatric patients. Optimal liver graft size remains unknown and the volume of diseased liver in the recipient is not indicative of the volume (standard liver volume [LV]) optimal for the recipient's metabolic demands. To establish a formula for calculating the standard LV in the pediatric and adult populations for liver transplantation, whole LVs were measured using computed tomography (CT) in 96 patients (65 pediatric and 31 adolescent or adult subjects) with normal liver whose disease conditions did not seem to affect body weight (BW) or LV. In the 96 subjects, the ratio of estimated LV to BW decreased gradually as age increased until approximately 16 years, when it started to level off. On the other hand, there seemed to be a directly proportional relationship between the estimated LV in vivo and body surface area (BSA) (r = .981; r2 = .962; P < .0001) in the subjects as a whole, and the formula, LV (mL) = 706.2 x BSA (m2) + 2.4, was established from the measured data by simple regression analysis. Another predicting equation, LV (mL) = 2.223 x BW (kg)0.426 x body height (BH) (cm)0.682, was produced by multiple regression analysis (r2 = .969; P < .0001). Considering its simplicity of use, we adopted the first formula for predicting standard LV in an individual patient.