Background: Cardiac complications after noncardiac surgery are a serious cause of illness and death. Echocardiography is being used before noncardiac surgery to assess risk for cardiac complications, but its role remains undefined.
Objective: To examine the prognostic value and operating characteristics of transthoracic echocardiography for assessing cardiac risk before noncardiac surgery.
Design: Prospective cohort study.
Setting: University-affiliated Veterans Affairs medical center.
Patients: 339 consecutive men who were known to have or were suspected of having coronary artery disease and were scheduled for major noncardiac surgery.
Measurements: Information from detailed histories, physical examinations, and electrocardiographic and laboratory studies was routinely collected. Transthoracic echocardiography was done before surgery to assess ejection fraction, wall motion abnormalities (reported as the wall motion score [range, 5 to 25 points]), and left ventricular hypertrophy.
Main outcome measures: Postoperative ischemic events (cardiac-related death, nonfatal myocardial infarction, and unstable angina), congestive heart failure, and ventricular tachycardia.
Results: 10 patients (3%) had ischemic events; 26 (8%) had congestive heart failure; and 29 (8%) had ventricular tachycardia. No echocardiographic measurements were associated with ischemic events. In univariate analyses, an ejection fraction less than 40% was associated with all cardiac outcomes combined (odds ratio, 3.5 [95% CI, 1.8 to 6.7]), congestive heart failure (odds ratio, 3.0 [CI, 1.2 to 7.4]), and ventricular tachycardia (odds ratio, 2.6 [CI, 1.1 to 6.2]). In multivariable analyses that adjusted for known clinical risk factors, an ejection fraction less than 40% was a significant predictor of all outcomes combined (odds ratio, 2.5 [CI, 1.2 to 5.0]) but not congestive heart failure (odds ratio, 2.1 [CI, 0.7 to 6.0]) and ventricular tachycardia [corrected] (odds ratio, 1.8 [CI, 0.7 to 4.7]). Wall motion score was a univariate predictor of all cardiac outcomes (odds ratio for each 3-unit increase, 1.6 [CI, 1.3 to 2.1]) and ventricular tachycardia (odds ratio, 1.6 [CI, 1.2 to 2.2]) but was only a multivariable risk factor for all events (odds ratio, 1.3 [CI, 1.0 to 1.7]). An ejection fraction less than 40% had a sensitivity of 0.28 to 0.31 and a specificity of 0.87 to 0.89 for all categories of adverse outcomes. Likelihood ratios for ejection fraction had poor operating characteristics. Adding echocardiographic information to predictive models that contained known clinical risk factors did not alter sensitivity, specificity, or predictive values in clincally important ways.
Conclusions: The data did not support the use of transthoracic echocardiography for the assessment of cardiac risk before noncardiac surgery. Echocardiographic measurements had limited prognostic value and suboptimal operating characteristics.