Predicting postoperative morbidity and mortality by model for endstage liver disease score for patients with head and neck cancer and liver cirrhosis.

BACKGROUND The purpose of this study was to evaluate the Model for Endstage Liver Disease (MELD) scoring system for predicting the morbidity and mortality of patients with head and neck cancer with liver cirrhosis undergoing tumor resection with microsurgical free-tissue transfer. METHODS Between January 2000 and December 2008, 3108 cases were retrospectively reviewed. RESULTS There were 59 men and 2 women enrolled in this study. Preoperatively, 31 and 30 patients were classified as having lower (<9.73) and higher (>9.73) MELD scores, respectively. Patients with higher MELD scores had significantly more postoperative medical morbidities including pulmonary complications and gastrointestinal bleeding. The mortality rate was also significantly higher for higher MELD scorers (23.3% vs 3.2%; p = 0.026). By logistic regression model, preoperative MELD score was a significant predictive factor for morbidity and mortality in multivariate analysis. CONCLUSION MELD score could be used to predict morbidity and mortality for patients with head and neck cancer with liver cirrhosis undergoing tumor resection with microsurgical free tissue transfer.