SUMMARY The objectives of the present study were to determine whether coronary steal can occur in one-vessel coronary occlusion when other arteries are normally patent, and to determine the quantitative effect on the magnitude of coronary steal of a graded partial stenosis in the coronary circulation proximal to the origin of the collaterals. In 10 dogs, occlusion of the left anterior descending coronary artery (LAD) was induced and perfusion of the left main coronary artery was maintained with a special servo pump at constant pressure (measured distal to the cannula tip) during intracoronary adenosine infusion. Adenosine increased left main coronary blood flow from 85 + 15 to 229 42 ml/min, while mean pressure in the left main coronary artery was held constant at 83 + 3 mm Hg (mean + SD). Systemic hemodynamics did not change, but three indexes of collateral function decreased: peripheral LAD coronary pressure, from 14.0 2.7 to 12.5 2.9 mm Hg (p < 0.005); retrograde LAD flow, from 3.2 + 2.5 to 2.4 + 2.6 ml/min (p < 0.001);and microsphere blood flow to ischemic myocardium, from 0.055 + 0.029 to 0.040 ± 0.031 ml min-l g-1 (p < 0.03). Next, we added the resistance of the Gregg cannula to the perfusion system, i.e., in the same dogs we regulated constant pressure in the tubing proximal to the cannula tip. This added resistance created a 4.5-mm Hg pressure gradient (multivessel occlusion) and a larger drop in retrograde LAD flow (23% vs 36%, p < 0.05). Because a screwclamp caused graded increments in resistance of the left main coronary perfusion system and the resting pressure gradient, there was a linear increase in the magnitude of coronary steal (r = 0.71). Thus, arteriolar vasodilators increased flow velocity, which led to decreased pressure at the origin of the collaterals because there was some resistance in the native coronary circulation between the left main coronary artery and the origin of the collaterals (12% of total resting resistance). Further increments in vascular resistance proximal to the origin of the collaterals caused a linear increase in the magnitude of coronary steal. plied by a single partially occluded coronary artery.8-0 These studies and another study" showed associated decreases in aortic pressures, however, so they could not demonstrate coronary steal in the absence of partial occlusions on other coronary arteries. If the coronary vasodilator agent decreases aortic pressure, this fall in coronary perfusion pressure, per se, could explain the decreased collateral blood flow without invoking the specific mechanism described by Fam and McGregor. 2A previous study from this laboratory found that i.v. adenosine decreased collateral myocardial blood flow more than aortic pressure, yielding a decreased calculated value of conductance. The decreased value of conductance in that study did not prove the occurrence of coronary steal because the relationship between aortic pressure and collateral blood flow may not always be linear over the entire '4 range of aortic pressures.
VASODILATOR-INDUC...