Institutional actual utilization of postoperative mortality using nationwide survey based risk calculator in patients who underwent major hepatectomy

Background: Relationship between outcomes of major hepatectomy and the mortality rate predicted by National Clinical Database risk calculator (NCD-RC) was examined . Methods: Patient demographics and postoperative morbidity and mortality were compared between 30-day and in-hospital mortality rates among 55 patients who underwent major hepatectomies . The cut-off value for high-risk mortality was set at 5%. Patients were divided into four groups: A) no severe complications and low predictive mortality rate (woML) , B) severe complications or mortality, and low mortality rate (wML) , C) no severe complications and high mortality rate (woMH) , and D) severe complications or mortality, and high mortality rate (wMH) . Results: Morbidity higher than CD III occurred in 17 patients (28%) and 30-day and in-hospital mortality in none and two (3%), respectively. The in-hospital mortality rate was significantly higher for male patients (p<0.01). Age, elderly patients, diseases, and co-morbidity did not significantly differ among groups. Although bile leakage was common in group wML , there were no in-hospital deaths. All surgical procedures performed in group wMH were right hepatectomy with bile duct resection (RH-BDR) for biliary malignancy, and two died of hepatic failure; however, the incidence of RH-BDR was not significantly higher than those in other groups. Conclusions: Preoperative mortality rate predicted by NCD-RC was not always consistent with outcomes in actual clinical settings and further improvements are needed. In case of RH-BDR for biliary malignancy with high predictive mortality rate, careful decision making for liver function and perioperative management are required.