External Validation of the ACTA-PORT Transfusion Risk Score in Older Cardiac Surgery Patients at Risk of Frailty

Abstract

OBJECTIVES: Accurate preoperative transfusion risk stratification may serve to better manage older patients undergoing cardiac surgery. Therefore, the aim of the present study was to externally validate the existing Association of Cardiothoracic Anesthetists perioperative risk of blood transfusion (ACTA-PORT) score in a population ≥70 years of age scheduled for cardiac surgery. Furthermore, the study authors investigated the additional prognostic value of individual frailty variables to this transfusion risk score. DESIGN: A retrospective analysis. SETTING: At a tertiary-care hospital. PARTICIPANTS: Five hundred seven patients aged ≥70 years undergoing elective cardiac surgery from July 2015 to August 2017. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: The primary outcome was the administration of a perioperative blood transfusion. Frailty domains were assessed in a preanesthesia geriatric assessment, and a priori selection of biomarkers derived from blood was determined. The original ACTA-PORT score resulted in a c-statistic of 0.78 (95% confidence interval 0.74-0.82), with moderate calibration in predicting perioperative allogeneic transfusion in older patients undergoing cardiac surgery. Model updating, using the closed testing procedure, resulted in model revision with a higher discriminatory performance (c-statistic of 0.83, 95% confidence interval 0.79-0.86) and corrected calibration slope. Iron deficiency, impaired nutritional status, and physical impairment were associated with perioperative transfusions. The addition of individual frailty variables to the updated ACTA-PORT model did not result in improved predictive performance. CONCLUSIONS: External validation of the original ACTA-PORT score showed good discrimination and moderate calibration in older patients at risk of frailty undergoing cardiac surgery. Updating the original ACTA-PORT improved the predictive performance. Careful evaluation of additional frailty domains did not add prognostic value to the ACTA-PORT score.