MET Level Score Outperforms Comorbidities in Predicting Perioperative Open Aortic Surgery Morbidity

To compare the predictive value of the metabolic equivalent of tasks (METs) score and the Society for Vascular Surgery (SVS) comorbidity score, both individually and in combination, for perioperative outcomes in patients undergoing open aortic repair (OAR).

Retrospective analysis on 169 consecutive patients undergoing elective OAR from 2018 to 2023. Primary outcomes included perioperative major adverse events (MAEs), intensive care unit (ICU) admission, and length of hospital stay. Predictive performance was evaluated using receiver operating characteristic (ROC) curves with area under the curve (AUC) analysis, the DeLong test for ROC comparison, and bootstrap sensitivity analysis to evaluate robustness.

For MAEs, the SVS score showed acceptable discriminatory ability (AUC 0.71), whereas the METs achieved higher accuracy (AUC 0.89), and the combined SVS+METs model achieved the best performance (AUC 0.91). Bootstrap analysis confirmed a clinically relevant advantage of METs-based models (95% confidence interval (CI) 0.053-0.313). For ICU admission, SVS yielded an AUC of 0.65, improved by METs (AUC 0.73) and further by the combined model (AUC 0.75), which was significantly superior to SVS (P = 0.016; 95% CI 0.026-0.184). For prolonged hospital stay, SVS achieved an AUC of 0.61, while METs and the combined model improved discrimination (AUC 0.71 and 0.73, respectively), with a positive trend toward the combined model (P = 0.062; 95% CI 0.026-0.218).

The MET score outperforms the SVS comorbidity score in predicting major complications, ICU admission, and prolonged hospitalization. The integration of METs with the SVS score further enhances predictive accuracy.