Novel three-dimensional imaging technique improves the accuracy of hepatic volumetric assessment

Background: With pre-operative prediction of liver volume becoming increasingly important to safely carry out complex hepatic resections, the aim of the present study was to validate the accuracy of a three-dimensional (3-D) liver surgery operative planning software in performing hepatic volumetry. Methods: Between 1999 and 2007, we performed 29 live donor liver resections for transplantation. Eleven patients had pre-operative volumetry performed by radiologists from either computed tomography (CT) or magnetic resonance (MR) imaging with documentation of the corresponding specimen weight. Retrospectively, images were uploaded into Scout (TM) where 3-D models of each case were generated to perform volumetry. A correlational analysis was performed followed by an accuracy comparison. Results: Estimations by both radiologists and Scout (TM) were significantly correlated with the specimen weights, P <= 0.0001. Compared with radiologists' volumetry, Scout (TM) significantly improved overall accuracy [per cent error (PE) 20.0% +/- 5.3 vs. 32.9% +/- 5.7, P = 0.005], accuracy of CT-based estimations (PE 23.2% +/- 6.7 vs. 37.2% +/- 6.9, P = 0.023) and accuracy of the left lateral section (PE 11.1% +/- 3.9 vs. 26.6% +/- 6.8, P = 0.027). Discussion: This 3-D planning software is a valid tool for use in volumetry. Significance is greatest for CT-based models of the left lateral section. This approach gives surgeons the ability to assess volumetrics and actively plan resections.