Preoperative risk score for 90-day mortality following major liver resection

Abstract

Background: Major liver resection is frequently performed for primary and secondary liver tumors. However, 90-day mortality rates can exceed 10% in high-risk patients. This study aimed to develop a preoperative risk score for postoperative mortality after major liver resection. Methods: All major liver resections between 2014 and 2019 in 2 Dutch tertiary referral centers were identified. A validation cohort consisted of all consecutive patients who underwent a major liver resection in the nationwide Dutch Hepato Biliary Audit from 2014 to 2020. Multivariate logistic regression was used to identify prognostic factors and develop a mortality risk score. Results: Major liver resection was performed in 513 patients, of whom 238 (46.4%) had a primary liver cancer, and in 148 patients (28.8%), a hepaticojejunostomy was performed; 90-day mortality occurred in 56 patients (10.8%). Mortality was independently predicted by 5 risk factors: age ≥ 65 years, diabetes mellitus type 2, diagnosis of primary liver cancer, American Society of Anesthesiologists ≥ 3, and extended hemihepatectomy. A risk score with 1 point assigned to each risk factor showed good discrimination (area under the curve [AUC], 0.77; 95% CI, 0.71–0.83). The predicted 90-day mortality was 3.5% for low-risk (0 or 1 points; 53.8% of all patients), 11.1% for intermediate-risk (2 points; 25.3%), and 29.7% for high-risk patients (3–5 points; 20.9%). External validation in the nationwide cohort with 1617 patients showed similar concordance (AUC, 0.69; 95% CI, 0.64–0.75). Conclusion: The proposed and validated risk score can aid in shared decision making.