In Reply:—
We thank Drs. Smiler, Hariharan, and Teplick for their insightful comments stimulated by our article on prevention and treatment of perioperative myocardial ischemia. 1Dr. Smiler writes that we did not consider the fact that β-adrenergic blocking agents shift the oxyhemoglobin dissociation curve to the right, thus releasing a greater amount of oxygen to tissue at any given oxygen tension. Dr. Smiler suggests that “the clear reduction in perioperative infarction rates after β-blockade rests on this fundamental effect.” We completed early work in this area showing that both propanolol and nitroglycerin reduce the affinity of hemoglobin for oxygen. 2Unfortunately, these agents shift the P50(partial pressure of oxygen [Po2] at which hemoglobin is 50% saturated with oxygen) less than 3 mmHg. Myocardium is a high oxygen-extracting tissue under resting conditions, and coronary Po2 is relatively low. Thus, oxygen extraction in the heart operates at the lower portion of the sigmoid-shaped oxyhemoglobin dissociation curve. During ischemia, coronary sinus Po2 is reduced even further. Little additional oxygen could be extracted during a small rightward shift in P50. A shift in the P50produced by propanolol cannot explain the effectiveness of β-adrenergic blocking agents in the treatment of myocardial ischemia. In contrast, new drugs, such as RSR13, that shift the oxyhemoglobin dissociation curve by 10–15 mmHg, depending on dose, may be beneficial for ischemic myocardium. 3
Dr. Hariharan comments that we mentioned that high concentrations of volatile anesthetics can lead to increases in sympathetic nervous system activity. He suggests that we may have meant a reduction in sympathetic tone. Abrupt increases in the concentration of volatile anesthetics, such as isoflurane or desflurane, can cause dramatic increases in sympathetic tone, resulting in increases in heart rate and arterial pressure and, therefore, demand of the myocardium for oxygen. 4–6After such an initial stimulation or when used in lower concentrations, the volatile anesthetics ultimately decrease sympathetic nervous system activity. In patients with coronary artery disease, the former increase should be avoided, and the latter reduction in sympathetic tone is beneficial unless substantial decreases in arterial pressure in the presence of a critical coronary stenosis occur. If so, flow declines in direct proportion to diastolic aortic blood pressure.
Dr. Teplick thinks the concept that the upstream driving pressure for coronary blood flow is diastolic aortic pressure is “illogical.” He incorrectly states that “DBP [diastolic blood pressure] corresponds to the onset of ejection” and seems to suggest that systole and diastole in the ventricle occur at different times in the aorta. Ejection occurs during systole. Diastole occurs after ejection, its first phase being isovolumic relaxation. 7,8The contention by Dr. Teplick that diastolic aortic pressure as a determinant of coronary flow is not supported by data published in peer-reviewed journals is erroneous. Considerable work has been done defining this relation by a number of investigators, most notably Ronald Bellamy. 9,10A portion of the confusion may arise from the fact that different investigators calculate coronary vascular resistance with different driving pressures. This calculation can be performed with use of diastolic aortic pressure, and this is not unreasonable because coronary flow is highest during early diastole. In contrast, other investigators use mean arterial pressure because some coronary blood flow (albeit a small amount) occurs during systole. An interesting experiment completed by Downey and Kirk 11perfused the canine coronary circulation with blood from a shunt arising in the left ventricle. Coronary flow could occur only during systole in such a model. Little flow reached the coronary circulation, and essentially no perfusion of the subendocardium occurred, showing the dependence of myocardial perfusion on diastolic aortic pressure. The flow that occurred during systole was distributed preferentially to the subepicardium. Finally, left ventricular end-diastolic pressure is used as the opposing pressure to flow (DBP–left ventricular end-diastolic pressure), but this pressure considerably underestimates myocardial tissue pressure, which is the major determinant of extravascular resistance. Tissue pressure is difficult to measure, whereas pulmonary capillary wedge pressure, an index of left ventricular end-diastolic pressure, is readily obtainable clinically. Thus, this definition of coronary perfusion pressure (DBP–left ventricular end-diastolic pressure) can be found in any number of textbooks 12and is supported by data from the peer review literature.
The second comment of Dr. Teplick is interesting because he indicates that an important cause of myocardial infarction is plaque rupture. He suggests that prevention of infarction should “focus more on preventing plaque rupture or alterations in the coagulation system that might predispose to thrombosis at the sight of unstable plaques.” He contends that high-grade stenoses are related to stable angina, rate-related ischemia is not part of the acute coronary syndrome, and rate-related ischemia may be totally unrelated to the risk of perioperative infarction. Not all patients with myocardial infarction have plaque rupture. Plaque rupture is only one element of a continuum of multifactorial etiologies that cause irreversible tissue damage. Because of the multifactorial nature of myocardial infarction, it can be treated by any of a variety of means, including classic manipulations of oxygen supply and demand, as well as by interference with the coagulation cascade. No treatment should be considered in isolation from the others. β-Blockers have been proven to reduce the reinfarction rate after acute myocardial infarction, 13have been proven to decrease cardiac morbidity and mortality after surgery, 14and should be used in the perioperative period. New avenues for reduction of the incidence and severity of myocardial infarction are being explored, but this does not negate use of β-adrenergic blocking agents.