Background: Algorithms for preoperative cardiac evaluation prior to noncardiac surgery use indices of the metabolic equivalent of activities of daily living (METs). We evaluated METs as a predictor of cardiac complications following elective, noncardiac surgery.
Methods: A study was performed in an outpatient university preadmission center METs were estimated prospectively for 5,939 inpatients admitted for elective, noncardiac surgery who underwent a preanesthetic assessment within two months prior to surgery. Cardiac outcomes were retrieved retrospectively from relational databases. Outcomes included death, myocardial infarction, acute congestive failure, arrhythmias, cardiac arrest, acute ischemia, acute renalfailure, stroke, respiratory failure, severe hypertension, peripheral vascular occlusion, and pericardial effusion. Adverse outcomes were correlated with age, gender, surgical procedure, activities, and the American Society of Anesthesiologist's Physical Status (ASA-PS) using receiver operator characteristic curve analysis.
Results: 94 of 5,939 (1.6 percent) patients had cardiac complications; 16 died, six from their cardiac complication. 38.3 percent of complications occurred following vascular surgery. Using a multinomial logistic regression analysis, both age and physical status were highly significant predictors (p < 0.001) but METs was not (p = 0. 793). Receiver operator characteristic (ROC) curves were usedfor predictive value of variables. Area of the curves for age versus cardiac complications and death were 0.814 and 0.782; for physical status, 0.744 and 0.803; for METs, 0.664 and 0.524.
Conclusions: METs are not a reliable index for the prediction of adverse cardiac events following elective, noncardiac surgery. Age and physical status are more predictive. Adverse cardiac outcomes are most frequent following vascular surgery.