|
|
|
Clin. Cardiol. 23, 187–194 (2000)
Phyllis K. Stein, Ph.D., Peter P. Domitrovich, Ph.D.,* Robert E. Kleiger, M.D., Kenneth B. Schechtman, Ph.D.† Jeffrey N. Rottman, M.D.,‡ for the CAST Investigators
Division of Cardiology, Department of Medicine, Washington University School of Medicine; *Barnes-Jewish Hospital; †Division of Biostatistics, Washington University School of Medicine, St. Louis, Missouri; ‡Departments of Medicine (Cardiology) and Pharmacology, Vanderbilt University School of Medicine, Nashville, Tennessee, USA
Summary
Background: Clinical and demographic determinants of heart rate variability (HRV), an almost universal predictor of increased mortality, have not been systematically investigated in patients post myocardial infarction (MI).
Hypothesis: The study was undertaken to evaluate the relationship between pretreatment clinical and demographic variables and HRV in the Cardiac Arrhythmia Suppression Trial (CAST).
Methods: CAST patients were post MI and had >=6 ventricular premature complexes/h on pretreatment recording. Patients in this substudy (n = 769) had usable pretreatment and suppression tapes and were successfully randomized on the first antiarrhythmic treatment. Tapes were rescanned; only time domain HRV was reported because many tapes lacked the calibrated timing signal needed for accurate frequency domain analysis. Independent predictors of HRV were determined by stepwise selection.
Results: Coronary artery bypass graft surgery (CABG) after the qualifying MI was the strongest determinant of HRV. The markedly decreased HRV associated with CABG was not associated with increased mortality. Ejection fraction and diabetes were also independent predictors of HRV. Other predictors for some indices of HRV included beta-blocker use, gender, time from MI to Holter, history of CABG before the qualifying MI, and systolic blood pressure. Decreased HRV did not predict mortality for the entire group. For patients without CABG or diabetes, decreased standard deviation of all NN intervals (SDANN) predicted mortality. Clinical and demographic factors accounted for 31% of the variance in the average of normal-to-normal intervals (AVGNN) and 13–26% of the variance in other HRV indices.
Conclusions: Heart rate variability post MI is largely independent of clinical and demographic factors. Antecedent CABG dramatically reduces HRV. Recognition of this is necessary to prevent misclassification of risk in patients post infarct.
Key words: heart rate variability, postmyocardial infarction, coronary artery bypass graft surgery (CABG), diabetes
This study was supported by NHLBI RO-3 Grant # HL53776.
Address for reprints:
Phyllis K. Stein, Ph.D.
Division of Cardiology
Barnes-Jewish Hospital
216 S. Kingshighway Blvd.
St. Louis, MO 63110, USA
Received: May 4, 1999
Accepted with revision: June 22, 1999