Prognostic implications of atrial fibrillation in patients undergoing myocardial perfusion single-photon emission computed tomography

OBJECTIVES The aim of this research was to determine whether presence of atrial fibrillation (AF) provides incremental prognostic information relative to myocardial perfusion single-photon emission computed tomography (MPS) with respect to risk of cardiac death (CD). BACKGROUND The prognostic significance of AF in patients undergoing MPS is not known. METHODS A total of 16,048 consecutive patients undergoing MPS were followed-up for a mean of 2.21 +/- 1.15 years for the development of CD. Of those, 384 patients (2.4%) had AF. Cox proportional hazards method was used to compare clinical and perfusion data for the prediction of CD in patients with and without AF. RESULTS Atrial fibrillation was a significant predictor of CD in patients with normal (1.6% per year vs. 0.4% per year in non-AF patients), mildly abnormal (6.3% per year vs. 1.2% per year), and severely abnormal MPS (6.4% per year vs. 3.7% per year) (p < 0.001 for all). By multivariable analysis, AF patients had worse survival (p = 0.001) even after adjustment for the variables most predictive of CD: age, diabetes, shortness of breath, use of vasodilator stress, rest heart rate, and the nuclear variables. In the 4,239 patients with left ventricular ejection fraction evaluated by gated MPS, AF demonstrated incremental prognostic value not only over clinical and nuclear variables, but also over left ventricular ejection in predicting CD (p = 0.014). CONCLUSIONS The presence of AF independently increases the risk of cardiac events over perfusion and function variables in patients undergoing MPS. Patients with AF have a high risk of CD, even when MPS is only mildly abnormal. Whether patients with AF and mildly abnormal MPS constitute a group more deserving of early referral to cardiac catheterization is a question warranting further study. (C) 2004 by the American College of Cardiology Foundation.