History repeats itself.

To the Editor: In a recent issue of Circulation, Zaman et al1 report a model for the preoperative risk stratification of patients at risk for developing atrial fibrillation (AF) after isolated coronary artery bypass grafting (CABG). They studied 326 patients prospectively and demonstrated that prolonged signal-averaged P-wave duration (SAPD), advanced age, and male sex identified patients at a high risk for post-CABG AF. Moreover, there was an incremental increase in risk for patients with .1 of these 3 variables. We agree with the authors that a predictive model for patients at risk for post-CABG AF could be useful for the management of this common complication. Advanced age was identified as an important predictor, and this has been consistently reported as a risk factor in prior studies.2,3 Male sex was identified as a risk in one prior report but not in the majority of other existing studies.4 The authors also highlight the usefulness of a prolonged SAPD as a means of identifying patients at risk. However, there is an important and simple clinical variable that is not included in their model. The highest rates of post-CABG AF have been encountered in patients with a prior history of AF. In a retrospective analysis of 183 consecutive patients undergoing isolated CABG at our institution, patients with a prior history of AF were 19 times more likely to develop postoperative AF compared with patients without a prior history of AF.2 In a larger series by Mathew et al,3 a prior history of AF was associated with an odds ratio of 2.28, which was again the highest among all the variables tested. The tendency of patients to develop post-CABG AF may well be related to slow atrial conduction and, in turn, a prolonged P-wave duration and SAPD.5 In that regard, prolonged SAPD is an electrocardiographic marker for patients with the appropriate electrophysiologic substrate who are vulnerable for post-CABG AF. However, it is conceivable that a prior history of AF is a clinical marker that also identifies the same vulnerable patients on the basis of their tendency for recurrent episodes of AF. We are not aware of any reports that have evaluated prior history of AF and SAPD in the same model. Until then, it is hard to justify the routine measurements of SAPD before CABG. Although SAPD measurements may be accomplished quickly at the bedside, a simple history may provide just as much information. After all, history repeats itself.