Accurate preoperative risk-assessment tools have been developed by the American College of Cardiology/American Heart Association (ACC/AHA) and by the American College of Physicians (ACP). These guidelines make recommendations that may result in differing noninvasive preoperative cardiac evaluations. Gordon and Macpherson studied the two most commonly used guidelines for preoperative cardiac evaluation of patients scheduled for elective noncardiac surgery.

A retrospective review of a large hospital Medical Preoperative Evaluation Clinic (MPEC), where internists evaluated preoperative risks and arranged preoperative testing, collected information about decision-making in this setting. Recommendations of the MPEC physicians were noted and compared with the recommendations of the ACC/AHA or the ACP. Recommendation categories included surgery without testing, noninvasive stress testing, cardiac catheterization, and cancellation or delay of surgery.

More stress tests were ordered by the MPEC physicians than would have been recommended by the ACC/AHA or ACP guidelines. Although guideline recommendations agreed with the decisions for 115 (83 percent) of the 138 patients evaluated, the guidelines did not agree on a patient's need for a noninvasive test. Among patients with an ischemic cardiac history, the guidelines were more concordant in their recommendations.

The authors conclude that the ACC/AHA and ACP guidelines potentially decrease cardiac testing, but that there is substantial disagreement in the guidelines about which patients require further noninvasive testing. The major reason for this discordance appears to be the ACP evidence-based recommendation that no tests are needed in nonvascular surgical patients; the ACC/AHA guideline recommends stress testing in patients with low functional capacity. There is inadequate evidence to resolve this difference. This disagreement negatively affects guideline credibility and may discourage physicians from following guidelines. Physicians ordered more non-invasive stress tests than were recommended by either guideline. Areas of guideline discordance should be resolved by expert groups that develop recommendations.