In this work, we present a flexible method for explaining, in human readable terms, the predictions made by decision trees used as decision support in liver transplantation. The decision trees have been obtained through machine learning applied on a dataset collected at the liver transplantation unit at the Coru\~na University Hospital Center and are used to predict long term (five years) survival after transplantation. The method we propose is based on the representation of the decision tree as a set of rules in a logic program (LP) that is further annotated with text messages. This logic program is then processed using the tool xclingo (based on Answer Set Programming) that allows building compound explanations depending on the annotation text and the rules effectively fired when a given input is provided. We explore two alternative LP encodings: one in which rules respect the tree structure (more convenient to reflect the learning process) and one where each rule corresponds to a (previously simplified) tree path (more readable for decision making).