METHODS: We applied 2 risk prediction tools (Revised Cardiac Risk Index and Myocardial Infarction or Cardiac Arrest) included in current American College of Cardiology/American Heart Association guidelines to a multicenter prospective registry of patients undergoing surgery in the United States in 2009. We then calculated expected rates of preoperative cardiac stress testing if physicians were to follow American College of Cardiology/American Heart Association guidelines, expected nationwide direct medical expenditures that would result (in 2017 US dollars), and agreement beyond chance between the 2 risk prediction tools.
RESULTS: Current guidelines recommend considerable spending on preoperative stress testing. Guideline-recommended spending would differ substantially depending on the risk prediction tool used and the reliability of the functional status assessment. Rates of testing and resultant spending are likely much greater among patients at “elevated” risk, compared with patients at “low” risk. Two guideline-recommended risk assessment tools, Revised Cardiac Risk Index and Myocardial Infarction or Cardiac Arrest, have poor agreement beyond chance across the currently recommended risk threshold.
CONCLUSIONS: Preoperative stress testing is likely a considerable source of medical spending, despite unproven benefit. Which perioperative risk assessment tool clinicians should use, what risk thresholds are appropriate for patient selection, and the reliability of the functional status assessment all warrant further attention.
- Question: For what proportion of surgical patients would current guidelines recommend preoperative stress testing, and what direct medical spending would result?
- Findings: Two currently recommended testing tools (Revised Cardiac Risk Index and Myocardial Infarction or Cardiac Arrest) would lead to testing of a large number of nonoverlapping populations, with poor concordance across the 1% risk threshold currently recommended for discriminating between “low” and “elevated” risk patients.
- Meaning: Preoperative stress testing among patients above the 1% risk threshold is likely a much larger source of medical expenditures than testing among low-risk patients and is of uncertain value.