Impact of Choice of Risk Model in Perioperative Guidelines: Reply

We thank Drs. Cohn and Fernandez Ros for their careful review of our study examining the “Impact of the Choice of Risk Model for Identifying Low-risk Patients Using the 2014 American College of Cardiology/American Heart Association Perioperative Guidelines.”¹  The goal of our article was to determine the extent to which the three risk calculators recommended by the 2014 American College of Cardiology/American Heart Association Perioperative Guidelines agreed on the classification of patients at low risk (less than 1%) of major adverse cardiac event. Because we found wide variability in the proportion of patients assigned to the low-risk category depending on the choice of risk calculator, we recommended that future guidelines should select a best-in-class risk calculator to avoid the situation in which clinical care decisions would differ depending on the choice of one of several risk calculators recommended in the same guideline. However, our goal was not to “perform an independent validation of these three risk-prediction models”¹  or to define whether one calculator is better than the others (as mentioned by Drs. Cohn and Fernandez Ros), and our article did not recommend one risk calculator over the others. Although we present the results of a secondary analysis comparing the discrimination and calibration of these risk models, we did not discuss these findings in our article because we recognized the limitations of such a comparison. We read with interest the single-center retrospective study based on 663 patients²  by the authors and their comment in their letter to the editor stating that “all three calculators were similar in their classification of low versus elevated risk.”²  Although they do not quantify the level of agreement using kappa analysis, we note that in their study, the Revised Cardiac Risk Index identified more than three times the number of elevated risk patients as the Myocardial Infarction or Cardiac Arrest risk calculator. We also note that the findings of their single-center study may not be broadly generalizable. We again thank Drs. Cohn and Fernandez Ros for their thoughtful letter.

Supported by the Department of Anesthesiology, University of Rochester School of Medicine, Rochester, New York.

The authors declare no competing interests.