Risk prediction model for major complication after hepatectomy for malignant tumour - A validated scoring system from a university center

Abstract

Objective To derive and validate a scoring system for major complication after hepatectomy. Background Complications after hepatectomy significantly compromise survival outcomes, method to predict such risk is lacking. A reliable scoring system is therefore awaited. Methods Consecutive adult patients receiving hepatectomy for primary or secondary liver malignancy from 1995 to 2014 were recruited. After randomization, patients were allocated to derivation and validation group respectively. A scoring system predicting occurrence of major complication was developed. Results There were 2613 patients eligible for the study. The overall complication rate for the series was 10%. Impaired performance status (p = 0.014), presence of pre-existing medical illness (p = 0.008), elevated ALP (p = 0.005), urea (p < 0.001), and hypoalbuminemia (p = 0.008), and major hepatectomy (p < 0.001) were found to be independently associated major complications. A score was assigned to each of these factors according to their respective odd ratio. A total score of 0–17 was calculated for all patients. This score was shown to discriminate well with complication rate in both derivation and validation group (c-statistic: 0.71, p < 0.001 and 0.74, p < 0.001 respectively). The complication rate for low (score 0–5), moderate (score 6–10) and high (score 10 or above) risk group were respectively 5%, 16% and 28%. This risk stratification model was tested and confirmed in the validation group using Chi-square goodness-of-fit test (p = 0.864). Conclusion A validated risk stratification model provides an accurate and easy-to-use reference tool for patients and clinicians during the informed consent process.