Introduction of applanation tonometry for clinical pulse waveform analysis in the 1980s [1,2] opened a hemodynamic version of Pandora's box. According to Greek legend, Pandora was the first woman on earth. Her secret box or jar, when opened, let loose a variety of evil spirits and mischievous gremlins, leaving only hope behind.
Brachial cuff pressure has been used almost exclusively over the last century for individual assessment, epidemiological studies and clinical trials. The Systolic Hypertension in the Elderly Program study , whose primary results were released in 1991, showing greater importance of systolic over diastolic pressure, had been commenced on the basis of Kannel et al.'s publication at Framingham  that related arterial rigidity, systolic pressure and finger plethysmographic waveform to the risk of stroke.
Evaluation of the accurate Millar tonometer became the MD thesis project of the late Ray Kelly in the mid-1980s . Kelly et al. [5,6] validated the device in the carotid and radial sites against intraarterial pressure, in which principles of arterial applanation could be satisfied, but could not do so for the brachial artery at the elbow where the artery is poorly supported beneath the bicipital aponeurosis. Kelly and coworkers [7,8] conducted studies with carotid tonometry but was dissatisfied by methods of calibration. When carotid tonometry only was available, he calibrated to brachial cuff pressure, assuming that mean pressure was diastolic pressure along with 33% of pulse pressure (PP) (i.e. that form factor  was 33%). Assuming that brachial mean and diastolic pressure were identical at brachial and carotid sites , he calculated carotid systolic pressure (CSP) from the formula: CSP = CDP + (CMP − CDP)/CFF, where carotid form factor (CFF) was determined from the shape of the carotid pressure wave, and CDP and CMP were carotid diastolic and mean pressures, respectively. A similar approach was used by Kelly and Fitchett  in calibrating the carotid pressure wave in later studies of noninvasive ascending aortic impedance.
Kelly et al.  were not comfortable with use of the carotid pressure wave. Apart from the uncertainty of calibration, carotid tonometry required considerable expertise, because the artery was deep in the neck and could not easily be applanated, because repeatability was indifferent, because artifact (from respiration) was frequent, because the technique was uncomfortable, because carotid manipulation was apt to stimulate baroreceptors and slow heart rate and because there was at least a theoretic possibility of dislodging plaque from the artery. Kelly et al.  noted the corresponding changes in aortic, brachial and radial pressure changes before and after use of nitroglycerine at cardiac catheterization and examined the possibility that the relationship between central aorta and radial–brachial arteries could be expressed as a transfer function. We had previously used transfer functions as vascular impedance in expressing the relationship between pressure and flow in the frequency domain . We found that a generalized transfer function could be used to generate an aortic pressure wave from the radial pressure waveform . In a prospective study, designed with the US Food and Drug Administration (FDA), and undertaken in association with Pauca et al.  at Wake Forest, we showed that the process accurately measured aortic pressure to within 1 mmHg and with standard deviation less than 5 mmHg when invasively recorded aortic and radial pressure waves were studied. Chen et al.  validated the techniques and observed that the variance of the calibrated aortic wave accounted for 96% of the measured aortic pressure wave when a generalized transfer function was used instead of the specific transfer function for the individual studied. Calibration was not a problem because there was less than 1-mmHg difference between mean pressure in the aortic and radial artery (measured arithmetically from digitized waves). When waves were recorded noninvasively by radial tonometry, form factor could be calculated individually for each radial wave and related to the brachial values of systolic and diastolic pressures in the brachial artery. The FDA had previously found differences between devices for measuring brachial and radial arteries to be well within the Association of Medical Instrumentation SP10 criteria and approved the use of the generalized transfer function for use in noninvasive radial tonometry as well as calibrated intraarterial manometry under K012487 and K002742, respectively.
This technique was incorporated into the SphygmoCor device and was used in a number of trials, which have calculated central aortic pressure waveforms, and showed that central pressure so measured was better than brachial cuff pressure for predicting outcomes. The studies include the pREterax in regression of Arterial Stiffness in a contrOlled double-bliNd (REASON) [17,18], the Conduit Artery Functional Endpoint (CAFE)  and the Strong Heart study [20,21]. Similar superior outcomes have been shown for central pressure measured invasively at cardiac catheterization .
Queries have arisen from the use of the generalized transfer function technique as used in SphygmoCor® for calculation of central arterial pressure. Sharman et al.  confirmed that this remained appropriate during exercise. Chen et al.  had shown that it effectively measured aortic pressures during Valsalva maneuvers. Similar queries surfaced when the concept of impedance was first used to describe pressure–flow relationships . A consensus is emerging for using generalized transfer function to generate aortic pressure . Chen et al. and Segers et al.  used a similar generalized transfer function to that described by us and they have provided similar values of aortic pressure as measured invasively. Karamanoglu et al. showed similar transfer function in a realistic model of the human upper limb arteries under different simulated conditions. The most critical group (in Melbourne) found results almost identical to our own when they determined modulus of a generalized transfer function from aortic pressure waves recorded by micromanometry . Their problems in application were caused by inappropriate registration of phase . Smulyan et al.  repeated the validation at cardiac catheterization, calibrating radial waveform to the cuff sphygmomanometer and comparing the calculated aortic waves to simultaneously recorded aortic waves. Authors found similar waveshape but poor correlation with directly recorded aortic pressure waveforms. They concluded that the greatest error was caused by the brachial cuff calibration . They directed attention to studies, which compared pressure waveforms in the brachial artery of one arm with those recorded in the other by sphygmomanometer [30–33]. Such errors are far greater than those that arise from use of a generalized transfer function. During this whole period of tonometer use (1989–present), other issues have been arising as well: the greater value of systolic over diastolic brachial pressure with age more than 40 years and the greater value of brachial pulse over systolic and diastolic pressure over the age of 60 years [34,35]. Present attention is focusing on even greater value of aortic or its surrogate, carotid systolic and PP at all ages .
These following issues: validity of transfer function, relationship of mean to PP in different pressure waveforms, calibration of tonometric to cuff pressures and calibration of radial to brachial cuff pressures might be seen as evil spirits and mischievous gremlins that have escaped from Pandora's box. They are all important and need to be understood and quantified if we are to make reasonable assumptions, and link new with existing technology, including the humble cuff sphygmomanometer.
The latest assumption has been use of the late systolic pressure surge in a radial or brachial artery to estimate central systolic pressure [37,38]. Estimates of aortic pressure correspond closely to those predicted by the generalized transfer function  and with invasive data under vasodilator challenge [38,39].
In this issue of Journal of Hypertension, Mahieu et al.  from Ghent examine implications of calibration methods to noninvasive assessment of central and peripheral arterial pressure waveforms. This group has initiated Asklepios, a large longitudinal prospective study of aging in humans aged 35–50 years . To date, they have confirmed the substantial increase in central augmentation with age, and the shape of ascending aortic impedance curves against those measured invasively in smaller human studies and in experimental animals . They have identified the importance of phase as well as pressure in synthesis of pressure waveform and have pointed out deficiencies in some of the previous estimations of ascending aortic pressure from peripheral waveforms . Their work is highly respected. In the accompanying article , they have pointed out the potential problems in using a set formula for generating mean pressure in a peripheral artery. Such an issue was raised by Bos et al. , and by Chemla et al. , both of whom questioned the value of assuming that mean pressure in the brachial artery is one-third of PP added to diastolic pressure (i.e. form factor is 33%). Bos et al.  and Chemla et al.  estimated that this was closer to 40% on the basis of digitizing and averaging the wave over one or a series of beats. Mahieu et al.  showed that substantial differences arise when form factor for the radial waveform is considered as 33% rather than 40%. They derived an intermediate number from their (undisclosed) generalized transfer function, applied to the radial artery.
We agree with such views (which rightly challenge our previous practice in calibrating carotid waveforms) and wonder why the customary form factor of 33% was agreed in the first place. We see this as another gremlin, which escaped from Pandora's box. But we disagree that 33% can be replaced by 40% when the form factor for each individual can be measured from the radial artery waveform. We have shown that form factor varies considerably in the radial artery from as low as 17 to 51% (Fig. 1) and is dependent among other things on the shape of the pressure wave, quantified as aortic–radial amplification (Fig. 1).
Mahieu et al.  in this and previous articles criticize our assumption that values of brachial systolic and diastolic cuff pressure can be applied to the radial artery. We agree that this is an assumption, but we believe that it is well based on currently available data, and especially that recorded invasively from humans during cardiac catheterization. Wood et al.  and Rowell et al.  showed little or no amplification between the brachial artery and radial artery (Fig. 2). We believe that there is less error in doing this than in attempting to applanate the brachial artery and apply the cuff calibration to this. The brachial artery cannot be applanated with confidence on account of its lack of support and uncertain orientation to the humerus, together with the fact that the stiff brachialis aponeurosis intervenes between sensor and artery .
The gremlins from Pandora's hemodynamic box continue to create confusion and debate. The saving grace is that hope remains. This will help set applanation tonometry in its rightful place. The Ghent studies are very strong, but the group may need to reconsider pressure calibration from brachial form factor, which at their recommended value of 40%  (or 42.4% ) is virtually the same as CFF (44.0) . Near identity of form factors for central and upper limb arteries  explains identical systolic and PP in carotid and brachial arteries and lack of benefit for central over brachial pressure in prediction of cardiovascular events [48,49].
The present problem with calibration arises from a known error in reliance on brachial tonometry . There is no need to assume a set value of form factor in any artery if mean pressure is measured in each individual through integration of an accurate waveform in the radial artery and calibration of this to brachial systolic and diastolic pressures. This is justifiable on the basis of invasive data (Fig. 2). Hope remains in Pandora's box!
M.F.O'R. is a founding director of AtCor Medical, manufacturer of pulse wave analysis system.
1 Kannel WB, Wolf PA, McGee DL, Dawber TR, McNamara P, Castelli WP. Systolic blood pressure, arterial rigidity, and risk of stroke. The Framingham study. JAMA 1981; 245:1225–1229.
2 Kelly R, Hayward C, Avolio A, O'Rourke MF. Noninvasive determination of age-related changes in the human arterial pulse. Circulation 1989; 80:1652–1659.
3 SHEP Cooperative Research Group. Prevention of stroke by antihypertensive drug in older persons with isolated systolic. Final results of the Systolic Hypertension in the Elderly Program (SHEP). JAMA
4 O'Rourke M, Kass DA. In memoriam: Ray Kelly. Hypertension 2001; 37:e23.
5 Kelly RP, Hayward CS, Ganis J, Daley JE, Avolio AP, O'Rourke MF. Noninvasive registration of the arterial pressure pulse waveform using high-fidelity applanation tonometry. J Vasc Med Biol 1989; 1:142–149.
6 Kelly RP, Karamanoglu M, Gibbs HH, Avolio AP, O'Rourke MF. Noninvasive carotid pressure wave registration as an indicator of ascending aortic pressure. J Vasc Med Biol 1989; 1:241–247.
7 Kelly RP, Daley JE, Avolio AP, O'Rourke MF. Arterial dilation and reduced wave reflection: benefit of dilevalol in hypertension. Hypertension 1989; 14:14–21.
8 Hayward CS, Kelly RP. Gender-related difference in the central arterial pressure waveform. J Am Coll Cardiol 1997; 30:1863–1871.
9 Chemla D, Hebert JL, Aptecar E, Mazoit JX, Zamani K, Frank R, et al
. Empirical estimates of mean aortic pressure: advantages, drawbacks and implications for pressure redundancy. Clin Sci 2002; 103:7–13.
10 Pauca AL, Wallenhaupt SL, Kon ND, Tucker WY. Does radial artery pressure accurately reflect aortic pressure? Chest 1992; 102:1193–1198.
11 Kelly RP, Fitchett D. Noninvasive determination of aortic input impedance and external left ventricular power output: a validation and repeatability study of a new technique. J Am Coll Cardiol 1992; 20:952–963.
12 Kelly RP, Gibbs HH, O'Rourke MF, Daley JE, Mang K, Morgan JJ, Avolio AP. Nitroglycerine has more favourable effects on left ventricular afterload than apparent from measurement of pressure in a peripheral artery. Eur Heart J 1990; 11:138–144.
13 O'Rourke MF. Vascular impedance in studies of arterial and cardiac function. Physiol Rev 1982; 62:570–623.
14 Karamanoglu M, O'Rourke MF, Avolio AP, Kelly RP. An analysis of the relationship between central aortic and peripheral upper limb pressure waves in man. Eur Heart J 1993; 14:160–167.
15 Pauca A, O'Rourke M, Kon N. Prospective evaluation of a method for estimating ascending aortic pressure from the radial artery pressure waveform. Hypertension 2001; 38:932–937.
16 Chen CH, Nevo E, Fetics B, Pak PH, Yin FC, Maughan WL, Kass DA. Estimation of central aortic pressure waveform by mathematical transformation of radial tonometry pressure: validation of generalized transfer function. Circulation 1997; 95:1827–1836.
17 Asmar RG, London GM, O'Rourke MF, Safar ME, for the REASON Project Coordinators and Investigators. Improvement in blood pressure, arterial stiffness and wave reflections with a very-low-dose perindopril/indapamide combination in hypertensive patients; a comparison with atenolol. Hypertension 2001; 38:922–926.
18 De Luca N, Asmar RG, London GM, O'Rourke MF, Safar ME. Selective reduction of cardiac mass and central blood pressure on a low-dose combination of perindopril/indapamide in hypertensive subjects. J Hypertens 2004; 22:1623–1630.
19 Williams B, Lacy PS, Thom SM, Cruickshank K, Stanton AV, Collier D, et al
, The CAFE Investigators. Differential impact of blood pressure-lowering drugs on central aortic pressure and clinical outcomes. Circulation 2006; 113:1213–1225.
20 Roman MJ, Devereux RB, Kizer JR, Lee ET, Galloway JM, Ali T, et al
. Central pressure more strongly relates to vascular disease and outcome than does brachial pressure. The Strong Heart Study. Hypertension 2007; 50:197–203.
21 Roman MJ, Devereux RB, Kizer JR, Okin PM, Lee ET, Wang W, et al
. High central pulse pressure is independently association with adverse cardiovascular outcome. J Am Coll Cardiol 2009; 54:1730–1734.
22 Jankowski P, Kawecka-Jaszcz K, Czarnecka D, Brzozowska-Kiszka M, Styczkiewicz K, Loster M, et al
, on behalf of the Aortic Blood Pressure Survival Study Group. Pulsatile but not steady component of blood pressure predicts cardiovascular events in coronary patients. Hypertension 2008; 51:848–855.
23 Sharman JE, Lim R, Qasem AM, Coombes JS, Burgess MI, Franco J, et al
. Validation of a generalized transfer function to noninvasively derive central blood pressure during exercise. Hypertension 2006; 47:1203–1208.
24 Agabati-Rosei E, Mancia G, O'Rourke MF, Roman MJ, Safar ME, Smulyan H, et al
. Central blood pressure measurements and antihypertensive therapy. A consensus document. Hypertension 2007; 50:154–160.
25 Segers P, Rietzschel E, Heireman S, De Buyzere M, Gillebert T, Verdonck P, Van Bortel L. Carotid tonometry vs synthesized aorta pressure waves for the estimation of central systolic blood pressure and augmentation index. Am J Hypertens 2005; 18:1168–1173.
26 Karamanoglu M, Gallagher D, Avolio AP, O'Rourke MF. Functional origin of reflected pressure waves in a multibranched model of the human arterial system. Am J Physiol 1994; 267:H1681–H1688.
27 Hope SA, Meredith IT, Tay D, Cameron JD. ‘Generalizability’ of a radial–aortic transfer function for the derivation of central aortic waveform parameters. J Hypertens 2007; 25:1812–1820.
28 Segers P, Mahieu D, Kips J, Van Bortel LM. The use of a generalized transfer function: different processing, different results! J Hypertens 2007; 25:1783–1787.
29 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.
30 Breit SN, O'Rourke MF. Comparison of direct and indirect arterial pressure measurements in hospitalized patients. Aust N Z J Med 1974; 4:485–491.
31 Cohn JN, Finkelstein S, McVeigh G, Morgan D, LeMay L, Robinson J, Mock J. Noninvasive pulse wave analysis for the early detection of vascular disease. Hypertension 1995; 26:503–508.
32 Watson S, Wenzel RR, diMatteo C, Meier B, Luscher TF. Accuracy of a new wrist cuff oscillometric blood pressure device: comparisons with intra-arterial and mercury manometer measurements. Am J Hypertens 1998; 11:1469–1474.
33 Lane D, Beevers M, Barnes N, Bourne J, John A, Malins S, Beevers DG. Inter-arm difference in blood pressure: when are they clinically significant? J Hypertens 2002; 20:1089–1095.
34 Franklin SS, Gustin W 4th, Wong ND, Larson MG, Weber MA, Kannel WB, Levy D. Hemodynamic patterns of age-related changes in blood pressure. The Framingham Heart study. Circulation 1997; 96:308–315.
35 Franklin SS, Khan SA, Wong ND, Larson MG, Levy D. Is pulse pressure useful in predicting risk for coronary heart disease? Circulation 1999; 100:354–360.
36 Safar ME, O'Rourke MF. Arterial stiffness in hypertension. In: Birkenhager WH, Reid JL, editors. Handbook of hypertension, vol. 23. London, UK: Elsevier; 2006.
37 Adji A, O'Rourke MF. Determination of central aortic systolic and pulse pressure from the radial artery pressure waveform. Blood Press Monit 2004; 9:115–121.
38 Takazawa K, Kobayashi H, Shindo N, Tanaka N, Yamashina A. Relationship between radial and central arterial pulse wave and evaluation of central aortic pressure using the radial arterial pulse wave. Hypertens Res 2007; 30:219–228.
39 Pauca AL, Kon ND, O'Rourke MF. Benefits of nitroglycerine on arterial stiffness is directly due to effects on peripheral arteries. Heart 2005; 91:1428–1432.
40 Mahieu D, Rietzschel ER, De Buyzere ML, Verbeke F, Gillebert TC, De Backer G, et al
. Noninvasive assessment of central and peripheral arterial pressure (waveforms): implications of calibration methods. J Hypertens
41 Segers P, Rietzschel ER, De Buyzere ML, Vermeersch SJ, De Bacquer D, Van Bortel LM, et al
, on behalf of the Asklepios investigators. Noninvasive (input) impedance, pulse wave velocity, and wave reflection in healthy middle-aged men and women. Hypertension 2007; 49:1248–1255.
42 Bos WJW, Verrij E, Vincent HH, Westerhof BE, Parati G, van Montfrans GA. How to assess mean blood pressure properly at the brachial artery level. J Hypertens 2007; 25:751–755.
43 Chemla D, Brahimi M, Nitenberg A. Thumb-rule for the proper assessment of mean blood pressure at the brachial artery level: what should be changed? J Hypertens 2007; 25:1740–1741.
44 Kroeker EJ, Wood EH. Comparison of simultaneously recorded central and peripheral arterial pressure pulses during rest, exercise and tilted position in man. Circ Res 1955; 3:623–632.
45 Rowell LB, Brengelmann GL, Blackmon JR, Bruce RA, Murray JA. Disparities between aortic and peripheral pulse pressures induced by upright exercise and vasomotor change in man. Circulation 1968; 37:954–964.
46 O'Rourke MF, Adji A, Hoegler S. Calibration of noninvasively recorded upper-limb pressure waves. Hypertension 2005; 46:e15.
47 Segers P, Mahieu D, Kips J, Rietzschel E, De Buyzere M, De Bacquer D, et al
, for the Asklepios investigators. Amplification of the pressure pulse in the upper limb in healthy, middle-aged men and women. Hypertension 2009; 54:414–420.
48 Dart AM, Cameron JD, Gatzka CD, Willson K, Liang UL, Berry KL, et al
. Similar effects of treatment on central and brachial blood pressures in older hypertensive subjects in the Second Australian National Blood Pressure Trial. Hypertension 2007; 49:1242–1247.
49 Mitchell GF, Gudnason V, Launer LJ, Aspelund T, Harris TB. Hemodynamics of increased pulse pressure in older women in the community-based age, gene/environment susceptibility: Reykjavik study. Hypertension 2008; 51:1123–1128.