Which Mechanisms Determine Blood Pressure?

Blood pressure (BP) can be conceptually viewed as oscillations between two extremes (systolic blood pressure [SBP] and diastolic blood pressure [DBP]), the axis being mean BP (MBP) and the amplitude being pulse pressure (PP). MBP becomes the so-called steady component of BP, assuming laminar flow in arteries (which occurs under physiological conditions only in the microcirculation). MBP has been traditionally used to describe hemodynamics and the relationship between cardiac work and peripheral resistance. Pulsations around MBP, generated by the pulsatile pump (the heart), lead to PP and SBP. The determinants of PP are more complex, and include aortic properties in early systole (characteristic impedance), the timing and magnitude of wave reflections from peripheral sites, aortic stiffness, and cardiac function. Brachial SBP increases by 25 mm Hg (22%) between ages 20 and 80 years, whereas central SBP and PP increase by 40% and >100%, respectively.1 In the literature, there is controversy about the relative contribution of aortic impedance/antegrade wave2 vs wave reflections1, 3 to the age-associated increase in PP. Against this background, in this issue of the Journal, Tanaka and colleagues4 report correlates of steady and pulsatile BP components in 3762 elderly adults (70–89 years) participating in the Atherosclerosis Risk in Communities (ARIC) study. Brachial BP was measured with an oscillometric automated sphygmomanometer. MBP was calculated with the formula MBP = DBP + 1/3 PP. Central (carotid) SBP was derived from carotid tonometry, calibrated with brachial BP. Participants with arrhythmias and major cardiovascular disease were excluded. After adjustments for age, sex, race, body mass index, diabetes, and smoking status, hemodynamic determinants of brachial SBP were aortic stiffness (carotid-femoral pulse wave velocity [cfPWV], measured with tonometry), stroke volume (measured with two-dimensional echocardiography), arterial wave reflections (augmentation index [AIx], from carotid pressure waveforms), left ventricular ejection time (ET), and upstroke time, with cfPWV, AIx, and ET having the strongest influence, explaining 13.4%, 6.7%, and 4% of the variability, respectively. For central SBP, findings were similar, with cfPWV, AIx, and ET explaining 8.9%, 4.5%, and 7.2% of the variability, respectively. Determinants of PP were roughly the same, with the largest contribution by ET, followed by cfPWV. In the oldest age tertile (≥80 years), the influence of ET increased both for SBP and PP. Determinants of MBP were cardiac index and peripheral resistance, explaining 0.3% and 15.1% of the variance, respectively. The authors have to be applauded for this study. Its major strength is the large sample size and the careful acquisition of tonometric and echocardiographic data. Shortcomings include the cross-sectional design, the performance of BP measurements/tonometry and echocardiography on separate days, and the use of the DBP + 1/3 PP rule to estimate MBP. The latter is not accurate, when compared with invasive measurements.5 Instead of using estimates, one can take advantage of the fact that oscillometric sphygmomanometers are quite good in determining MBP, because the maximum of pressure oscillations during cuff deflation occurs at the MBP level.6 Recently, we have observed that this directly measured MBP should be used to calibrate waveforms, when accurate noninvasive determination of central SBP is intended.7 Clinical studies published since that confirmed the superiority of this approach.8, 9 Measurements of characteristic impedance or antegrade wave were not included in the study. Therefore, the study does not add new data to the controversy mentioned above. The relationship between cardiac function (stroke volume, cardiac output, ET) and PP should be discussed in more detail. In the study by Tanaka and colleagues,4 ET contributed much more to SBP and PP than left ventricular ejection fraction, which is the most popular measurement for left ventricular systolic function and is widely used in clinical routine. A decrease in ET, however, seems to be one of the hallmarks of an impaired systolic function,10 probably due to the fact that the failing left ventricle is not able to cope with the load of the incoming reflected wave and stops ejection of the blood prematurely, leading to a lower PP. Consequently, in systolic heart failure—in contrast to the situation with normal systolic function11—a higher PP is associated with a better prognosis.12 We recently observed a direct relationship between ET and more conventional measurements of systolic function (ejection fraction, stroke volume, cardiac output), which was stronger in patients with systolic heart failure, but also present in patients with normal ejection fraction.10 Obviously, SBP and PP are the result of the interaction between cardiac and vascular function. Can we dissect which one is more important? In addition, might this be different in different age groups? In young relatively healthy untreated individuals, sex, stroke volume, wave reflections (AIx), and aortic stiffness (cfPWV) explained 37%, 6%, 2%, and 2% of the variability of brachial PP, respectively.13 In the same group of patients, the influence of cardiac output on SBP was more important in normal-weight adults, as compared with overweight patients, where peripheral resistance was more important.14 In the elderly ARIC population reported by Tanaka and colleagues,4 the increase in ET contributed most to the model predicting PP, outnumbering aortic stiffness and wave reflections. This effect is most pronounced in the oldest age group (≥80 years) and is somewhat surprising, challenging the traditional view that the increase of PP with aging is mostly related to a combined effect of unfavorable alterations of the arterial system.15 We need to keep in mind, however, that a longer ET is not necessarily a sign of better cardiac performance. We have observed previously that a prolongation of ET can be found in heart failure with preserved ejection fraction.16, 17 Moreover, a shortened as well as a prolonged ET is associated with an impaired prognosis.18 Taken together, ET may be an interesting marker of ventriculovascular interaction rather than a simple marker of cardiac function. Studying these relationships may further advance our understanding of poor old brachial BP. The authors have no conflicts of interest to disclose.