The Relationship between Liver Volume, Clinicopathological Characteristics and Survival in Patients Undergoing Resection with Curative Intent for Non-Metastatic Colonic Cancer.

INTRODUCTION: The prognostic value of CT-derived liver volume in terms of cancer outcomes is not clear. The aim of the present study was to examine the relationship between liver area on a single axial CT-slice and the total liver volume in patients with colonic cancer. Furthermore, we examine the relationship between liver volume, determined using this novel method, clinicopathological variables and survival.

METHODS: Consecutive patients who underwent potentially curative surgery for colonic cancer were identified from a prospectively maintained database. Maximal liver area on axial CT-slice (cm2 ) and total volume (cm3 ), were obtained by the manual segmentation of pre-operative CT-images in a PACS viewer. The maximal liver area was normalized for body height2 to create the liver index (LI) and values, categorized into tertiles. The primary outcome of interest was overall survival (OS). Relationships between LI and clinico-pathological variables were examined using chi-square analysis and binary logistic regression. The relationship between LI and OS was examined using cox proportional hazard regression.

RESULTS: A total of 359 patients were included. A total of 51% ( n = 182) of patients were male and 73% ( n = 261) were aged 65 years or older. 81% ( n = 305) of patients were alive 3-years post-operatively. The median maximal liver area on the axial CT slice was 178.7 (163.7-198.4) cm2 . The median total liver volume was 1509.13 (857.8-3337.1) cm3 . Maximal liver area strongly correlated with total liver volume (R2 = 0.749). The median LI was 66.8 (62.0-71.6) cm2 /m2 . On multivariate analysis, age ( p < 0.001), sex ( p < 0.05), BMI ( p < 0.001) and T2DM ( p < 0.05) remained significantly associated with LI. On univariate analysis, neither LI (continuous) or LI (tertiles) were significantly associated with OS ( p = 0.582 and p = 0.290, respectively).

CONCLUSIONS: The simple, reliable method proposed in this study for quantifying liver volume using CT-imaging was found to have an excellent correlation between observers and provided results consistent with the contemporary literature. This method may facilitate the further examination of liver volume in future cancer studies.