Benchmarking a Center of Excellence in Vascular Surgery: Using Acute Physiology and Chronic Health Evaluation II to Validate Outcomes in a Tertiary Care Institute.

Abstract

The Society of Vascular Surgery (SVS) has made it a top priority to implement verification of vascular "centers of excellence". Our institutional aortic network was established in 2008 in order to standardize care of patients with suspected acute aortic pathology. The implementation and success of this program has been previously reported. We sought to use our experience as a benchmark for which to develop prognostic modeling to quantify clinical status upon admission and help predict outcomes. Our objective was to validate the Acute Physiology and Chronic Health Evaluation (APACHE) II scoring system using a cohort of aortic emergencies transferred by an organized transfer network. This was a retrospective, single institution review of patients transferred through an institutional aortic network for acute aortic pathology from 2017-2018. Demographics, comorbidities, aortic diagnosis, APACHE II score, as well as 30-day mortality were recorded. Associations with 30-day mortality were evaluated using two-sample t-tests, ANOVA models, Pearson chi-square tests and Fisher exact tests. Receiver operating characteristic (ROC) curves were fit overall and by pathology to predict 30-day mortality by Apache II total score. There were 395 consecutive transfers were identified. The mean age was 64.7years. Diagnoses included Type A Dissection (n = 134), Type B (n = 81), Aortic Aneurysm (n = 122), and PAU/IMH (n = 27). Mean APACHE II score on arrival was 12. Overall there were 53 deaths (13.4%) in the cohort. Patients that died had significantly higher Apache II total scores (11.3 vs 16.5, P < .001). The area under the receiver operator characteristic (ROC) curve (AUC) was .66 for the full cohort, indicating a poor clinical prediction test. APACHE II score is a poor predictor of 30-day mortality in a large transfer network accepting all aortic emergencies. The authors believe further refining a prognostic model for diverse population will not only help in predicting outcomes but to objectively quantify illness severity in order to have a basis for comparison among institutions and verification of "centers of excellence".