The effect of virtual reality and training on liver operation planning.

Abstract

The three-dimensional relation of a liver tumour to the intrahepatic vascular trees is basis of operation planning in liver surgery. Yet it has not been proven whether 3D reconstruction and further computerised processing will enhance precision of operation planning in liver surgery which has been based on the liver segment classification of Couinaud up to now. Our interdisciplinary group (department of Surgery, German Cancer Research Center and Department of Radiology) has developed a new interactive computer-based quantitative 3D operation planning system for liver surgery which is being introduced into the clinical routine. The system quantifies the organ structures semiautomatically, defines resection planes depending on safety margins and the vascular trees, and presents the data in digital movies as well as in quantitative reports. We conducted a clinical trial to evaluate whether 3D reconstruction will lead to an improved operation planning. Data of 7 virtual patients were presented to a total of 81 surgeons in different levels of training. The tumours had to be assigned to a liver segment and subsequently drawn together with the operation proposals into a liver model. The precision of both was measured quantitatively for each surgeon and stratified concerning 2D and different types of 3D presentations. The 3D anatomy can be visualised in high quality which results in good perception of the third dimension (depth). Tumour assignment to liver segments was significantly correlated to the level of training (p < 0.05). There was a significant increase (p < 0.001) in the precision of tumour localisation by 51% and resection proposal from 2D through 3D reconstructions by 13%-21%. Quantitative differences of the simplified Couinaud's classification of the liver segments compared to the true vascular anatomy of up to 40% were found. The impact of individual 3D-reconstruction on surgical planning has been proven to be significant and increases precision quantitatively. The merit of Couinaud's classification may be enhanced by individualisation of the segment borders in future.