The Psychophysiological Investigations of Myocardial Ischemia (PIMI) study: objective, methods, and variability of measures.

OBJECTIVE: This study evaluated physiological, neuroendocrine, and psychological status and functioning of patients with coronary artery disease in order to clarify their role in the expression of symptoms during myocardial ischemia (MI), and to establish repeatability of responses to mental stress. Design and methods of the study are presented.
METHODS: One hundred ninety-six coronary artery disease patients were examined during physical and mental stress tests in four hospitals. Eligibility criteria included narrowing of at least 50% in the diameter of at least one major coronary artery or verified history of myocardial infarction, and evidence of ischemia on an exercise treadmill test. Psychological, biochemical, and autonomic function data were obtained before, during, and after exposure to mental and exercise stressors during 2 or 3 half-days of testing. Ventricular function was assessed by radionuclide ventriculography, and daily ischemia by ambulatory electrocardiography. Sixty patients returned for a short-term mental stress repeatability study. Twenty-nine individuals presumed to be free of coronary disease were also examined to establish reference values for cardiac responses to mental stress.
RESULTS: Study participants were 41 to 80 years of age; 83 (42%) had a history of MI, 6 (3%) of congestive heart failure, and 163 (83%) of chest pain; 170 (87%) were men; and 90 (46%) had ischemia accompanied by angina during exercise treadmill testing. Ischemia during ambulatory monitoring was found in 35 of 90 (39%) patients with and 48 of 106 (45%) patients without angina during exercise-provoked ischemia. Intraobserver variability of ejection fraction changes during bicycle exercise and two mental stress tests (Speech and Stroop) was good (kappa = 1.0, .90, and .76, respectively; percent agreement = 100, 97.5, and 93.8%, respectively). Variability of assessed wall motion abnormalities during bicycle exercise was better (kappa, agreement = 85%) than during Speech or Stroop kappa and .57, percent agreement = 70% and 82.5%, respectively).
CONCLUSIONS: Study design, quality control data, and baseline characteristics of patients enrolled for a clinical study of symptomatic and asymptomatic myocardial ischemia are described. Lower repeatability of reading wall motion abnormalities during mental stress than during exercise may be due to smaller effects on wall motion and lack of an indicator for peak mental stress.