The DIGG risk score: A risk predictive model of perioperative mortality after elective treatment of intact abdominal aortic aneurysms in the DIGG register

Abstract

Objective: The aim of the study was to develop a specific risk score for the aortic register of the German Institute for Vascular Health Care Research (DIGG) of the German Society for Vascular Surgery and Vascular Medicine (DGG) for the prediction of postoperative mortality in elective treatment of intact abdominal aortic aneurysms (iAAA). The score should also enable a risk-adjusted presentation of the register results in the near future. Methods: The method of binary logistic regression analysis was used to calculate the model. The data from 10,404 patients were included in the analysis, of whom 7870 (75.6%) were treated by endovascular (EVAR) and 2534 (24.4%) by open (OR) aortic repair. It was examined which factors have an independent influence on hospital mortality and the effect size was determined as a score. Results: For EVAR, the influencing factors with their effect sizes (score in brackets) were: age >85 years (2), female gender (2), juxtarenal AAA (5), maximum diameter >65 mm (2), diabetes mellitus (2), American Society of Anesthesiologists (ASA) score >3 (2), cardiac comorbidities (3) and renal insufficiency stage >3 (5). For OR the factors were: age >80 years (2), female gender (2), juxtarenal AAA (2), ASA score >3 (3), previous myocardial infarction (2), renal comorbidities (3) and previous stroke (2). The estimated hospital mortality was calculated for the individual case from the sum of the risk factors (scores). The accuracy of the model (correlation between observed and expected results) was determined using the receiver operating characteristic (ROC) curve. An area under the curve (AUC) of 0.817 (confidence interval 0.789–0.844) demonstrated an excellent discrimination. In a validation group of 3831 patients, the good agreement between observed and calculated results was confirmed. Conclusion: The DIGG risk score can predict risk-adjusted hospital mortality after EVAR and OR of iAAA in the DIGG register. Improvements with respect to the prediction are desirable for OR and should be strived for by extending the model in the future.