Coronary ischemic disease, arterial stiffness, and pulse pressure

There is now ample evidence that brachial artery pulse pressure (PP) is an independent determinant of cardiovascular outcomes, principally in the population more than 59 years of age.1,,–4 More specifically, it has been shown that brachial artery PP is an independent predictor of cardiac (primarily coronary) events, whereas mean blood pressure (MBP) is a stronger predictor of cardiovascular risk affecting cerebral, cardiac, and renal events.4 However, some reports5,–7 have noted that the predictive value of PP is lower than that of MBP or systolic blood pressure (BP). Although these studies reported large numbers of patients, they used mainly retrospective data where PP was not the primary goal and was evaluated only at baseline in most cases. In addition, the statistical analysis often did not consider MBP and PP as independent variables. When prospective BP changes were used and PP analyzed independently of MBP, PP emerged as a major predictor of cardiovascular risk.1,,,–8 Identification of PP as an independent risk factor has prompted several investigators, including the present ones,9 to identify in subjects with ischemic coronary disease, the links between brachial artery PP, aortic PP and atherosclerotic lesions of coronary arteries. For a review of such links, it is important to recall that a significant statistical association between increased aortic stiffness and atherosclerotic coronary heart disease (CHD) has been previously reported.10,11 In 1989, Hirai et al10 found that increased arterial stiffness was related to the number of atherosclerotic stenoses of the coronary arteries. However, in this study of patients with previous myocardial infarction, differences in systolic BP or PP were not observed among groups with CHD of varying severity. This lack of association might have been related to the prior myocardial infarctions that could have reduced left ventricular ejection and affected the BP. Because the detection of CHD by coronary angiography requires catheterization of the aorta, most of the studies relating increased PP and the presence of stenotic coronary arteries have been performed using invasive methods. Direct recording of aortic pressure, against which the heart must pump, is clearly superior to noninvasive brachial artery pressures, which are only inaccurate surrogates of the aortic pressure. Aortic pressures have a number of advantages over brachial artery pressures. First, the aortic MAP is directly measured and not calculated as diastolic BP + k.PP. Traditionally, “k” is believed to be constant and equal to 1/3. In fact, “k” varies with the known age-related change in the shape of the BP pulse. This change in the elderly makes “k” less than 1/3 and the calculated MBP from the brachial artery higher than the measured aortic MBP. Second, intra-aortic diastolic BP, which is closely approximated by invasive brachial diastolic BP, differs markedly from diastolic BP measured by the forearm cuff. Published comparisons of cuff and direct brachial diastolic BP measurements have shown wide variations and systematic over estimations by the cuff pressures, particularly in the elderly (Fig. 1).12 In addition to the inaccuracy of the cuff diastolic BP and MBP, the systolic BP is well-known to be higher in the arm than in the aorta—a difference that diminishes with aging. The noninvasive calibrated carotid or aortic pulses, although more difficult to record, have been shown to be superior to brachial BP measurements in the evaluation of cardiovascular risk.13 Aside from assessing risk, the cuff overestimates of diastolic BP in large therapeutic trials, have led to an underestimation of the incidence of isolated systolic hypertension, before and after drug treatment. In such trials, the more central (carotid) noninvasive measurements of systolic BP and PP14 should be used in place of brachial cuff pressures to avoid diagnostic and interpretive errors regarding therapy. For these many reasons, the statistical associations reported in the literature between PP and coronary artery stenoses relate better to invasive, rather than noninvasive, methods of BP measurement. Previous studies reporting significant associations between intra-aortic PP and coronary atherosclerosis have been performed in a relatively small number of subjects, mostly without major myocardial dysfunction, who are undergoing cardiac catheterization (see review in Jankowski et al9). These studies report on highly selected populations in which PP, independent of MBP, has been found to be higher in atherosclerotic than in nonatherosclerotic subjects and positively related to the number of stenotic coronary arteries. Similar results have been obtained when PP is calculated as a percentage of MBP or when PP is replaced by the augmentation index, a classic marker of wave reflections. No comparable findings were observed when aortic systolic or diastolic BP alone were used in place of PP. In the present publication by Jankowski et al,9 low fractional diastolic and systolic pressures in the aorta were related to the extent of coronary artery disease and thus represent an important improvement in identifying links between BP, aortic stiffness, and CHD. In this publication, once again, the overestimation of diastolic BP, MBP, and underestimation of PP by the arm cuff when compared to direct measurement in the aorta are quite evident. To understand the connections between aortic PP and coronary atherosclerosis, it is important to recall that coronary blood flow is physiologically autoregulated, that is, it remains unchanged within well-established upper (140 mm Hg) and lower (90 mm Hg) MBP limits as defined by in vitro experiments.15 These limits are maintained by changes in the diameter of small coronary arteries, implying a significant reserve of arteriolar vasodilation and constriction. However, the normal mechanism of coronary autoregulation is complex, and made more complex by disease. The driving pressure of the coronary circulation is not MBP (as it is for the brain and the kidney), but rather the diastolic BP. During ventricular ejection, the large increase of intramyocardial pressure leads to minimal systolic coronary blood flow.15 Furthermore, in subjects with CHD, the coronary reserve is diminished, particularly in subjects with cardiac hypertrophy or hypertension. Therefore, in patients with advanced CHD, the overriding hemodynamic factor that controls coronary blood flow is aortic diastolic BP. Normally, aortic diastolic BP is influenced by two different hemodynamic factors: systemic vascular resistance and aortic stiffness (distensibility). Because subjects with CHD are treated with multiple vasodilating agents (nitrates, converting enzyme inhibitors, AT1 blockers, and calcium entry blockers) to reduce afterload, treated patients are continuously subjected to a reduced diastolic BP, mainly due to a decrease of the systemic vascular resistance. On the other hand, if these agents are not used, the diastolic BP and MBP will be higher, increasing cardiovascular (mainly cerebral) risk, particularly in hypertensive subjects. The diastolic and systolic BP are both affected by changes in aortic stiffness. The increasing aortic stiffness of aging, largely uninfluenced by current drug treatment, acts to increase the systolic BP, to decrease diastolic BP, and to increase PP. Thus, in older subjects, increased PP, more than a reduction in diastolic BP alone, should be expected to reduce coronary perfusion, increase left ventricular end-systolic stress, and therefore increase coronary risk. Ideal therapy to significantly decrease aortic stiffness alone would be lower systolic BP, increase or maintain diastolic BP, and reduce risk, but such therapy is not yet available. For the methodological reasons discussed, BPs that vitally affect cardiac function are difficult to measure using noninvasive methods, but are easily measured using intra-aortic BP techniques. The study by Jankowski et al9 stresses the inadequacy of cuff versus direct aortic pressures in assessing the important interactions between PP, arterial stiffness, and atherosclerotic coronary stenoses. Although the association of aortic pressures with aortic stiffness contributes to the understanding of cardiovascular risk and its possible prevention, the requirement for aortic catheterization continues to limit its widespread clinical utility. We thank Anne Safar, MD, for her skillful technical help.