Central Blood Pressure and Hypertension
During the 20th century, the brachial artery auscultatory method of blood pressure (BP) measurement has been the major basis for our understanding of clinical hypertension. First, the height of systolic (S) and diastolic (D) BP was used to define cardiovascular (CV) risk. Second, BP reduction by drug treatment was shown to be associated with a substantial decrease of CV risk in several varieties of high BP, including systolic hypertension in the elderly.1,2 However, drug treatment often reduced DBP more than SBP, resulting in increased pulse pressures (PP) and the related arterial stiffness. An associated increase in coronary risk was also observed.1,3 Because PP and arterial stiffness did not respond readily to drug treatment and was made even less responsive by aging, it has been suggested and shown that mechanical factors such as PP, arterial stiffness, and wave reflections, preferably measured by central (aortic or carotid) BP, should be fully investigated. Because the majority of the studies used noninvasive peripheral BPs the need arose for invasive methods to verify and validate central BP measurement in the evaluation of CV risk.4
In the present issue of this journal, Jankowski et al5 established, in a 4.5-year follow-up of 1109 patients undergoing coronary angiography, that central BP was a significant and independent predictor of CV risk, as potent as cardiac ejection fraction and more powerful than brachial sphygmomanometric measurements. For these coronary patients, the pulsatile component of BP was the most important factor for CV risk prediction and was more closely related to CV morbidity and mortality than mean BP (MBP), the steady BP component (Table). In this study, central BP was determined using a standard, low-compliance fluid-filled catheter/amplification system. The major difficulty in the interpretation of these findings results from the selection of patients, all of whom required coronary angiography, and therefore were not representative of the large population of hypertensive subjects.
There is evidence that central pressure is more relevant than peripheral pressure for the prediction of CV risk. The pressure wave generated by the left ventricle travels down the arterial tree and is reflected from multiple peripheral sites, mainly at resistance arterioles.1,2 Consequently, the pressure waveform recorded at any arterial site is the sum of the forward traveling waveform and the reflected wave, the “echo” of the incident wave. The difference between brachial and central SBP defines “pressure wave amplification.” Typically DBP and mean BP change little across the arterial tree. However, as a consequence of transmitted and reflected waves, SBP and PP are amplified as much as 10 to 14 mm Hg when moving from the aorta to brachial artery.1,2 This phenomenon, seldom considered in published clinical guidelines, may have 3 major consequences: CV complications of hypertension, choice of antihypertensive agents, and regulation of heart rate.
Regarding CV complications, when the large conduit arteries are healthy and compliant (young subjects), the reflected wave merges with the incident wave in the proximal aorta mostly during diastole, thereby augmenting DBP and supporting coronary perfusion.1,2 In contrast, when the arteries are stiff (elderly), wave travel is faster and the reflected wave merges earlier with the incident wave, augmenting aortic systolic rather than diastolic pressure. As a result, left ventricular afterload is increased and coronary filling compromised. Such pathophysiological mechanisms explain the results of Jankowski et al5 and support the belief that central BP is superior to brachial BP in the prediction of CV risk.
The choice of drug treatment in hypertension is as well influenced by the amplitude of wave reflections (ie, the proportion of incident wave which is reflected), as by their timing. Acutely, vasoactive agents may modify the amplitude of wave reflections, and hence the SBP. This process is observed with some nitrates, but also occurs with calcium blockade or angiotensin blockade.1,2 Long-term antihypertensive therapy with angiotensin blockade or calcium blockade remodels the arterioles and modifies their baseline characteristics (geometry, distensibility) and reflection sites,6–8 thus consistently reducing central SBP.
Finally, central BP is highly influenced by heart rate. Lowering of heart rate by beta blockade attenuates the reduction of central SBP and PP, decreases SBP and PP amplification, and increases ventricular afterload, whereas tachycardia has the opposite effects. These bradycardic liabilities are not shared by calcium blockade or angiotensin blockade.7,8
The development and widespread use of oscillometric cuff blood pressures measurement has changed the landscape (Table). The accuracy of the auscultatory method was initially validated by comparing its results with intraarterial measurements. But these data cannot be applied to the oscillometric method that measures pressures differently. Although details of the algorithms vary among manufacturers, the principles are similar. When the deflating cuff pressure reaches the systolic pressure, cuff oscillations increase and the SBP is identified. When the cuff pressure reaches the DBP, the magnitude of the oscillations return to baseline. The MBP is selected as the minimum cuff pressure at maximum cuff oscillation and is separate from the estimates of the SBP and DBP. This eliminates the errors of SBP and DBP measurement, when the MBP is calculated using the 1/3 rule. Validation of the oscillometric brachial artery SBP and DBP has been described largely by comparing them to auscultatory pressures rather than intraarterial pressures. MBP comparisons by either method are uncommon, as are comparisons of either auscultatory or oscillometric pressures with aortic pressures.9 Aortic pressures and wave forms have been calculated from applanated radial pulses, using Fourier transforms and transfer functions by the Syphygmocor device.1 Unfortunately, these pulses must be calibrated by the poorly validated and relatively inaccurate cuff pressures.9
The MBP is known to be nearly constant throughout the arterial tree. Therefore, the oscillometric MBP, free of associated error from SBP and DBP measurement, might be a satisfactory representative of aortic MBP. Pulse amplification of the SBP is related to age, heart rate, and perhaps other individual characteristics such as height. Multiple regression equations using oscillometric cuff MBP and SBP and age are already available to better predict aortic SBP.10 Using a reliable MBP and SBP, an applanated carotid pulse (representative of the aortic pulse) could be calibrated to provide the DBP and the PP. Of the 4 available BPs, the MBP and PP (or some measure of pulsatility) should be the best CV risk predictors because the MBP reflects the status of the peripheral vascular resistance and the PP reflects aortic stiffness.
Studies to determine which measurement, or combinations of measurements, best predict aortic pressures are sorely needed but are difficult to design. Patients undergoing routine aortic catheterization, as in the study by Jankowski,5 are a select group, not entirely representative of the hypertensive population. Aortic pressures should be measured using pressure sensor tip catheters or fluid filled systems, insuring that their frequency responses are capable of accurate SBP and DBP recording. But only with additional studies will we learn how best to estimate aortic pressures noninvasively.
The opinions expressed in this editorial are not necessarily those of the editors or of the American Heart Association.
Nichols WW, O’Rourke MF. McDonald’s Blood Flow in Arteries. Theoretical, Experimental and Clinical Principles, V ed. London: Arnold, 2005.
Safar ME, O’Rourke ME. Handbook of Hypertension,vol. 23. Edinburgh: Elsevier, 2005: 3–62, 75–136, 459–501.
Blacher J, Evans A, Arveiler D, Amouyel P, Ferrieres J, Bingham A, Yarnell J, Haas B, Montaye M, Ruidavets JB, Ducimetiere P; PRIME Study Group. Residual coronary risk in men aged 50–59 years treated for hypertension and hyperlipidemia in the population: the PRIME study. J Hypertens. 2004; 22: 415–423.
Agabiti-Rosei E, Mancia G, O’Rourke MF, Roman MJ, Safar ME, Smulyan H, Wang Ji-Guang, Wilkinson IB, Williams B, Vlachopoulos C. Central blood pressure measurements and antihypertensive therapy: a consensus document. Hypertension. 2007; 50: 154–160.
Jankowski P, Kawecka-Jaszcz K, Czarnecka D, Brzozowska-Kiszka M, Styczkiewicz K, Loster M, Kloch-Badelek M, Wiliński J, Curylo AM, Dudek D. Pulsatile but not steady component of blood pressure predicts cardiovascular events in coronary patients. Hypertension. 2008; 51: 848–855.
Agabiti-Rosei E, Rizzoni E. The effects of hypertension on the structure of human resistance vessels. In Lip GYH, Hall JE, eds. Comprehensive Hypertension. Philadelphia, PA: Mosby Elsevier; 2007: 579–590.
London GM, Asmar RG, O’Rourke MF, Safar ME, on behalf of the REASON Project Investigators. Mechanism(s) of selective systolic blood pressure reduction after a low-dose combination of perindopril/indapamide in hypertensive subjects: comparison with atenolol. J Am Coll Cardiol. 2004; 43: 92–99.
Williams B, Lacy PS, Thorm SM, Cruickshank K, Stanton A, Collier D, Hughes AD, Thurston H, O’Rourke M. Differential impact of blood pressure-lowering drugs on central aortic pressure and clinical outcomes: principal results of the Conduit Artery Function Evaluation (CAFE) study. Circulation. 2006; 113: 1213–1225.
Smulyan H, Siddiqui DS, Carlson RJ, London GM, Safar ME. Clinical utility of aortic pulses and pressures calculated from applanated radial artery pulses. Hypertension. 2003; 42: 150–155.
Smulyan H, Sheehe PR, Safar ME. A preliminary evaluation of the mean arterial pressure as measured by cuff oscillometry. Am. J Hypertens. 2008, Jan 3 [Epub ahead of print].