Supine bicycle ergometer exercise was performed when the patients were in the postabsorptive state. Two or 3 days before the study, supine bicycle exercise was performed to familiarize the patients with a bicycle ergometer (Monark 881E ergometer, Sweden). On the day of study, a Swan-Ganz catheter was inserted through the internal jugular vein and advanced to the pulmonary artery. An arterial catheter was inserted in the radial artery. After 30 min of supine rest, the bicycle exercise test was performed with ECG monitoring (leads aVF, V4, and Vg). A full ECG was obtained at resting control and at peak exercise. Exercise began at a work load of 25 W with the pedal speed maintained at 60 revolution per minute and was increased by 25 W every 3 min until a symptom-limited maximum was reached. The exercise test was judged positive when the S-T segment in any recorded lead had 1 mm or more of horizontal or downsloping S-T segment depression 60 ms after the QRS complex in three consecutive beats or when the patient experienced typical anginal pain. If the S-T segment was already depressed on the resting ECG, 1 mm or more of additional depression was considered as a positive test.
The patients with a positive exercise test were not included in this study. To evaluate the responses of DT and systolic time intervals during exercise, recordings were made at control and at peak exercise. Data for systolic time intervals were recorded from the ear densitograph pulse derivative and the ECGA Ear densitograms were obtained with a photoelectric earpiece (Hewlett Packard 780-10) modified and filtered through a polygraph (Fukuda Denshi MIC 6600). Recordings were made on a thermal recorder (Fukuda Denshi RF-85) at a paper speed of 100 mm/s.
Expired gases were analyzed with an oxycon-4 (Mijnhardt Company, Holland). Instruments were calibrated at the beginning of each study. From these data, systemic oxygen consumption was measured at supine rest on the bicycle and continuously during exercise.
Right atrial, pulmonary arterial, and systemic arterial pressures were monitored continuously, and pulmonary artery wedge pressure was recorded at rest and at peak exercise with DS-3300 system (Fukuda Denshi, Japan) at a paper speed of 25 mm/s. Blood samples were drawn simultaneously from the radial and pulmonary artery at rest, within the last 30 s of exercise. The blood samples were used for the immediate measurements of oxygen tension (Radiometer Company ABL2) as well as oxygen saturation and hemoglobin concentration (Radiometer Company OS М2). Cardiac index was determined by the Fick principle with use of systemic arteriovenous oxygen difference and directly measured systemic oxygen consumption. The methods employed for measuring HR, pre-ejection period, and left ventricular ejection time were as described elsewhere. Diastolic time was calculated as cycle length minus electromechanical systole (QSJ. Measurements of HR and systolic time intervals were aided by a digitizer coupled to a computer (Fujitsu FM-8) and were calculated from the average of five consecutive beats for each of the recordings.