Interstudy variability of coronary flow reserve. Influence of heart rate, arterial pressure, and ventricular preload.

To define the long-term variability of serial coronary flow reserve (CFR) measurements in humans and to evaluate the influence of changes in heart rate, mean arterial pressure, and left ventricular preload on CFR, 45 patients with normal left ventricular function (38 cardiac allograft recipients, five patients with normal coronary arteries, and two patients with minimal coronary artery disease less than 50% diameter stenosis) were studied. CFR (ratio of peak hyperemic h to resting r coronary blood flow velocity CBFV) was measured with a 3F coronary Doppler catheter and intracoronary papaverine. Initial CFR measurements were highly correlated with repeat measurements obtained 11 +/- 0.6 months later (r = 0.95; mean absolute difference, 0.3 +/- 0.1; n = 17). Differences in CFR between studies were related to changes in heart rate (r = 0.61, p = 0.01) but not to changes in mean arterial pressure (r = 0.25, p = 0.33). To define the effects of rapid changes in heart rate, mean arterial pressure, and preload on CFR, these variables were altered by atrial pacing, handgrip exercise, and volume expansion, respectively. Atrial pacing produced a rate-related increase in rCBFV but did not change hCBFV. Consequently, CFR was significantly reduced as heart rate was increased progressively from 76 +/- 2 in sinus rhythm (4.5 +/- 0.2) to 100 (3.8 +/- 0.2, p less than 0.05, n = 32) to 120 beats/min (3.2 +/- 0.1, p less than 0.05, n = 7). Despite a 19 +/- 2 mm Hg rise in mean arterial pressure during handgrip exercise, CFR was unchanged from baseline (3.7 +/- 0.3 vs. 3.7 +/- 0.4, p = NS, n = 7) because rCBFV rose proportionally with hCBFV. When pulmonary capillary wedge pressure was increased from 9 +/- 1 to 16 +/- 1 mm Hg after volume expansion, CFR was significantly decreased (from 3.8 +/- 0.2 to 2.9 +/- 0.2, p less than 0.05, n = 9) because rCBFV was increased while hCBFV remained unchanged. Hence, serial CFR measurements in humans are highly reproducible in the absence of conditions known to affect resting or hyperemic coronary blood flow. Increases in heart rate or preload reduced CFR because rCBFV was increased while hCBFV was unchanged. In contrast, changes in mean arterial pressure did not alter CFR. Proper interpretation of CFR measurements should take into account the hemodynamic conditions at which they are obtained.