Corrected age-adjusted Charlson Comorbidity Index (CaCCI) for patients undergoing ALPPS: A tool for preoperative patients selection.

Abstract

Background: Risk stratification is of extreme importance for selecting patients for Associating Liver Partition and Portal vein Ligation for Staged hepatectomy (ALPPS). Comorbidities are a well known risk factor in liver surgery, however, when analyzed either singularly or using a comorbid score as the Charlson Comorbidity Index (CCI), it seems they do not influence the outcome in ALPPS. Despite the Age-adjusted-CCI (aCCI) is the most used tool to weight comorbidities in ALPPS Literature, it does not correlate to postoperative outcome, too. A possible explanation may be a wrong estimation of the weight of the neoplastic diagnosis in this type of surgery, considering that secondary tumors, as colorectal liver metastases, are weighted with 4 points more than primary tumor of the liver, as hepatocellular- and cholangiocarcinoma.We here propose a Corrected Age-adjusted Charlson Comorbidity Index (CaCCI) to predict postoperative 90-days mortality in patients undergoing ALPPS. Methods: Thirteen centers participated to this retrospective multicentric study. A risk analysis based on patients' characteristics and underlying disease was performed to recalibrate CCI on the basis of the relative size of logistic model regression coefficients. A nonparametric receiver operating characteristic (ROC) analysis was performed to estimate the predictive ability of CCI, aCCI and CaCCI. A good predictivity ability was defined as a C-statistic equal or greater than 0,7. Results: 449 patients were included. Mortality was observed in 65 cases (14,2%). CaCCI was generated on CCI corrected for age, weight of comorbidities and tumor type, with primary tumors weighted more that secundary. The predictive ability of the different scores measured was 0,445 for CCI, 0,523 for aCCI and 0,751 for CaCCI. Conclusions: The proposed CaCCI, unlike CCI and aCCI, has a good prediction capacity of postoperative mortality and can be easily integrated in the preoperative workout to improve patient selection.