Abstract

Predicting post-hepatectomy liver failure (PHLF) after extended right hepatectomy following portal vein embolization (PVE) from serial gadoxetic acid-enhanced magnetic resonance imaging (MRI). Thirty-six patients who underwent hepatectomy following PVE were evaluated prospectively with gadoxetic acid-enhanced MRI examinations at predefined intervals during the course of their treatment, i.e., before and 14days and 28days after PVE as well as 10days after hepatectomy. Relative enhancement (RE) and volume of the left and right liver lobes were determined. The study population was divided into two groups with respect to signs of PHLF. Differences between the two groups were assessed using the Mann-Whitney U test, and predictive parameters for group membership were investigated using ROC and logistic regression analysis. RE of the left lobe prior to PVE versus 14days after PVE was significantly lower in patients with PHLF than in those without PHLF (Mann-Whitney U test p < 0.001) and proved to be the best predictor of PHLF in ROC analysis with an AUC of 0.854 (p < 0.001) and a cutoff value of - 0.044 with 75.0% sensitivity and 92.6% specificity. Consistent with this result, logistic linear regression analysis adjusted for age identified the same parameter to be a significant predictor of PHLF (p = 0.040). Gadoxetic acid-enhanced MRI performed as an imaging-based liver function test before and after PVE can help to predict PHLF. The risk of PHLF can be predicted as early as 14days after PVE. • To predict the likelihood of post-hepatectomy liver failure, it is important to estimate not only future liver remnant volume prior to extended liver resection but also future liver remnant function. • Future liver remnant function can be predicted by performing gadoxetic acid-enhanced MRI as an imaging-based liver function test before and after portal vein embolization. • A reduction of relative enhancement of the liver in gadoxetic acid-enhanced MRI after portal vein embolization of 0.044 predicts post-hepatectomy liver failure with 75.0% sensitivity and 92.6% specificity.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call