Prediction of major adverse cardiac events in vascular surgery: are cardiac risk scores of any practical value?

Abstract

Background: Disease-specific preoperative scoring systems are often used to predict postoperative cardiac complications. We retrospectively evaluated the accuracy of 2 cardiac risk scores in the prediction of major adverse cardiac events (MACE) after vascular surgery.

Methods: Consecutive procedures were ''scored'' according to the Revised Cardiac Risk Index (RCRI) and the Eagle criteria. Two ''generic'' risk scoring systems, ASA (American Society of Anesthesiology) grade and the physiology Portsmouth POSSUM (P-POSSUM) score, were also documented for comparison.

Results: After 344 surgical procedures, 18 patients suffered a MACE (5.2%; 95% CI = 2.8-7.58-4 fatal). The RCRI (AUC 95% CI = 0.68 [0.57-0.83], P = .009) and the Eagle criteria (AUC 95% CI = 0.73 [0.57-0.8], P = .001) were no better than P-POSSUM (AUC 95% CI = 0.82 [0.73-0.91], P < .001) and ASA grade (AUC 95% CI = 0.67 (0.56-0.78), P = 0.016) in predicting MACE. Of the variables included in the 2 cardiac scoring systems, only age and history of ischemic heart disease were associated with MACE in our patients.

Conclusion: Cardiac risk scores were no better than generic risk scoring systems in predicting MACE. Poor performance may be due to differences between our patient population and those in which the scores were developed and to improved perioperative management of cardiac risk.