Author Affiliations: Department of Medicine, Emory School of Medicine, Atlanta, Georgia (Dr Goyal) (firstname.lastname@example.org); and St Luke's Mid America Heart Institute, Kansas City, Missouri (Drs Spertus and Kosiborod).
In Reply: Dr Perren and colleagues raise several concerns about our study. First, history of dialysis was included as a covariate in the mortality models and was not simply a baseline characteristic. Second, although rates of certain therapies and comorbidities differed across subgroups of potassium levels, we did adjust for all of these factors in an attempt to account for these differences. Third, causes of death are not captured in the HealthFacts database. As we stated in our article, the greater number of deaths than ventricular arrhythmia and cardiac arrest events may be due to misclassification bias. However, the rate of ventricular arrhythmia events has decreased substantially during the β-blocker1 and reperfusion era2 and is not as clinically meaningful as mortality.
Goyal A, Spertus JA, Kosiborod M. Potassium Levels After Acute Myocardial Infarction—Reply. JAMA. 2012;307(15):1578-1580. doi:10.1001/jama.2012.487