Comparison of radiographic liver volume and the 3D volumetry data based on CT examination in canine patients

Abstract

The liver is regarded as one of the most important organs in the body, thus diseases affecting the liver can be potentially fatal. The size and volume of the liver are highly valuable parameters in the diagnosis of hepatic disease processes and are further used as prognostic indicators for preoperative, operative and postoperative considerations. The aim of this study was to determine if the radiographic assessment of liver size significantly correlates with the real liver volume, as measured by CT volumetry. The study included 26 patients from whom we obtained lateral and ventrodorsal abdominal radiographs and contiguous abdominal CT scans. Through performing manual slice by slice segmentation of the abdominal CT scans, extrapolated to further attain a 3D model reconstruction within the 3D Slicer software, the liver volume was concluded. Several parameters, including liver length and liver area, were computed from the radiographs in order for comparison and correlation with the determined liver volume, as calculated from the CT scans with 3D Slicer. The age, body weight and body surface area of each individual were also recorded and included in our analysis. All parameters obtained from the radiographic images had a statistically significant correlation with liver volume (p < 0,05). Moreover, body weight and body surface area also had a high correlation with liver volume. With these results in mind, manual volume reconstruction of the liver with 3D Slicer is a reliable method, provided that the operator has an expansive knowledge of hepatic anatomy and its relationship to adjacent organs. However, it should be noted that the manual segmentation process, applied in this study, is highly time consuming and thus not suitable for daily clinical use. Despite a positive relationship occurring between parameters obtained from radiographs and the 3D reconstructed liver volume from 3D Slicer, radiographic assessment of liver volume is an unreliable method due to the numerous influential factors which do not allow, in most cases, a wholly objective assessment. The body weight and the body surface area are guidelines for the estimation of liver volume, but their value should be considered solely as a tendency auxiliary to additional measurements. Therefore, we can conclude that achievement of a timely and precise liver segmentation protocol is an ongoing issue which requires further investigation. With sophisticated technology, automated volumetry may replace manual volumetry for precise assessment of liver volume.