Background: Malnutrition is recognised as a preoperative risk factor for patients undergoing hepatic resection. It is important to identify malnourished patients and take preventive therapeutic action before surgery. However, there is no evidence regarding which existing nutritional assessment score (NAS) is best suited to predict outcomes of liver surgery. Methods: All patients scheduled for elective liver resection at the surgical department of the University Hospital of Heidelberg and the Municipal Hospital of Karlsruhe were screened for eligibility. Twelve NASs were calculated before operation, and patients were categorised according to each score as being either at risk or not at risk for malnutrition. The association of malnutrition according to each score and occurrence of at least one major complication was the primary endpoint, which was achieved using a multivariate logistic regression analysis including established risk factors in liver surgery as covariates. Results: The population consisted of 182 patients. The percentage of patients deemed malnourished by the NAS varied among the different scores, with the lowest being 2.20％ (Mini Nutritional Assessment) and the highest 52.20％ (Nutritional Risk Classification). Forty patients (22.0％) had a major complication. None of the scores were significantly associated with major complications. Conclusions: None of the twelve investigated NAS defined a state of malnutrition that was independently associated with postoperative complications. Other means of measuring malnutrition in liver surgery should be investigated prospectively.