A new formula for estimation of standard liver volume using computed tomography-measured body thickness

Abstract

The objective of this article is to derive a more accurate and easy‐to‐use formula for finding estimated standard liver volume (ESLV) using novel computed tomography (CT) measurement parameters. New formulas for ESLV have been emerging that aim to improve the accuracy of estimation. However, many of these formulas contain body surface area measurements and logarithms in the equations that lead to a more complicated calculation. In addition, substantial errors in ESLV using these old formulas have been shown. An improved version of the formula for ESLV is needed. This is a retrospective cohort of consecutive living donor liver transplantations from 2005 to 2016. Donors were randomly assigned to either the formula derivation or validation groups. Total liver volume (TLV) measured by CT was used as the reference for a linear regression analysis against various patient factors. The derived formula was compared with the existing formulas. There were 722 patients (197 from the derivation group, 164 from the validation group, and 361 from the recipient group) involved in the study. The donor's body weight (odds ratio [OR], 10.42; 95% confidence interval [CI], 7.25‐13.60; P < 0.01) and body thickness (OR, 2.00; 95% CI, 0.36‐3.65; P = 0.02) were found to be independent factors for the TLV calculation. A formula for TLV (cm3) was derived: 2 × thickness (mm) + 10 × weight (kg) + 190 with R2 0.48, which was the highest when compared with the 4 other most often cited formulas. This formula remained superior to other published formulas in the validation set analysis (R2, 5.37; interclass correlation coefficient, 0.74). Graft weight/ESLV values calculated by the new formula were shown to have the highest correlation with delayed graft function (C‐statistic, 0.79; 95% CI, 0.69‐0.90; P < 0.01). The new formula (2 × thickness + 10 × weight + 190) represents the first study proposing the use of CT‐measured body thickness which is novel, easy to use, and the most accurate for ESLV.