Arterial tonometry: principles and clinical applications ...

High Blood Press. 1996; 5: 241-250

SPECIAL

ARTICLE

Arterial tonometry: principles and clinical applications in hypertension P.S. Saba*, C. Cavallini**, D. Scorzoni***, C. Longhini***, R. Pini** and A. G~nau*, on beh~lf of the study group on "Mechanics of Large Arteries", Italian Society of Arterial Hypertension *Istituto di Clinica Medica, llniverslta di Sassari; **Istituto di Geriatria e Gerontologia, Universita di Firenze; -*Cattedra di Medicina Interna, Universita di Ferrara, Italy

ABSTRACT.Non-invasive and accurate measurements of arterial pressure waveforms are very useful in assessing both cardiac and vascular complications of hypertension. Reliable peripheral pulse waveforms can be obtained by high fidelity applanation tonometry. Solid theoretical principles and modeling studies support this technique. When the flat and rigid surface of a pressure transducer is applied on the skin overlying the arterial pulse and applanates the curve surface of the artery the contact pressure balances the intraluminal circumferential stresses acting on the arterial wall and the transducer records the true pressure wave inside the artery. Validation studies have shown that under ideal conditions both the contour and amplitude of radial and carotid tonometric pressure waveforms are virtually identical to those recorded invasively at the same sites. Applanation tonometry has important clinical applications and is increasingly used in hypertension. Carotid tonometry provides reliable representations of the central pressure waveform and blood pressure values comparable to those recorded in ascending aorta. after calibration for the mean brachial sphygmomanometric pressure. Applanation tonometry has been r~cel"ltlyassociated with tWO-dimensionally-guided M-mode ultrasonography to estimate the visco.is-etastl ~properties of the carotid artery. This method allows assessment of regional cross-sectional compliance. elastic modulus and ~. an index of stiffness that takes into account the curvitineartty of the pressure-volume relationship: in addition, ultrasonography provides clinically relevant information about intimal-medial thickness and presence of atherosclerotic plaques. Analysis of the systolic portion of the carotid pressure waveform permits to obtain the augmentation index -a measure of intensity of the pressure wave reflection- and the transit time of pressure wave from the heart to the major reflection site. Finally, tonometric end-systolic pressure provides an estimate of effective arterial elastance (ratio of end-systolic pressure to stroke volume). a measure of external left ventricular afterload. Key words: Applanation tonometry, hypertension, ultrasonography, arterial compliance, pressure wave reflection.

INTRODUCTION Elevated levels of systolic or diastolic blood pressure are known to be associated with increased risk of coronary artery disease. stroke and cardiac failure. iCI

While these complications are related to high blood pressure in central arteries, in clinical practice blood pressure is measured in the brachial artery. However. relationships between sphygmomanometric blood pressure and target organ damage have been found

1996, EditriceKurtis

Conespondence: Dott. Antonello Ganau, Clinica Medica, Viale San Pietro 8, 07100 Sassari. Accepted October 31, 1996.

241

Applanation tonometry in hypertension

to be relatively weak in hypertension (1) tni th d f . . probably because IS me 0 ails.to .evaluate the reaI blood ·

pressure 1eveI In central districts. Althou h th . R' R ' ff h g e etasSIC iva- OCCI cu sp ygmomanometer and th Krotkoff auscultatory technique have pro 'd d Ie 0 I,.., I d . VI e a most all our nJlOWe ge on epidemiology , and . • prog nosis treatmen t 0f hypertenslon. this conventl I intn , . , IOna aph h proac as In rmsic itmltations that m t b id d ' us e carefu IIy consi ere . In fact. In the single S b' t id bl d'" u ~ec • conSl era e It Ierences may Occurbetwee n cent raI and , penpheral artery blood pressure. In addrtl · I Ion. th e sphyg~omanome t nc technique does not provide information about pressure waveform and cti hl h wave refI e Ion. w IC closely reflect the physical pro rti of the arteriai tree. pe res For this reason. research on hypertension has focused on alternative methods for measuring the pressure burden on heart and vessels, High fidelity applanation tonometry is a modern technique that records external arterial pressure waveforms and may help to overcome some limitations of the current non invasive methods to assesscardiac and vascular load, In this paper we will briefly report principles and major clinical applications of arterial tonometry in hypertension.

BLOOD PRESSURE IN PERiPHERAL AND CENTRAL ARTERiES In the common practice. radial or brachial sphygmomanometric blood pressure values are considered to be representative of aortic pressure, Nevertheless. both invasive (2) and non-invasive studies (3-5) have shown that this assumption is not correct. The non-uniform elasticity of the arterial tree as well as the pressure wave reflection from the periphery (6) have important implications: a) there is a progressive increase in both systolic and pulse pressures moving from the center to the periphery of the arterial tree. so that pulse pressure is usually higher in the brachial and radial artery than in ascending .aorta (wave amplification): b) despite significant differences in pulse pressure and pressu,re wave~orm between central and peripheral artenes. plammetered areas under the different pressure curves are statistically similar and thus mean pressure is the same within the large capacitance vessels: c) there is a measurable delay (transmission time) in wave propagation. so that the time interval between the

foot of two waves combined with the distance between measurements permits to estimate the pulse wave velocity.that increaseswith the stiffness of the arterial tree; d) moving from the heart toward the periphery. the arterial pressure waveform tends progressivelyto losethe diastolic or dicrotic component, Thesefeatures of pressure wave propagation can be observed in a work by Hamilton and Dow published in 1939 (7). where arterial pulse contours were recorded simultaneously by intrarterial catheter in points at increasing distances from the aortic arch. The arterial pressure waveform is composed by a forward component directed from the heart to the periphery and a backward reflected component returning to the heart (8). When ventricular-vascular coupling is optimal. as usually happens in young people. the pressure wave transmission time is normal and the reflected wave matches the forward component in early diastole at the level of central arteries. determining an increase in aortic diastolic pressure that contributes to coronary perfusion. When arterial tree becomes stiffer because of aging (9). atherosclerosis (10.11). hypertension (6. 9). or high salt intake (1Z). pulse wave velocity increases and the impact between forward and backward pressure waves occurs in systole. increasing both pulse and systolic peak pressure and decreasing early diastolic pressure in ascending aorta. Clinical observations from the U.S. National Survey in 1977 and the Framingham Study demonstrate that either systolic and diastolic blood pressures increase gradually with age. with a plateau or even a decrease in diastolic pressure after 50 years of age. Moreover. the increase in central systolic pressure due to higher pulse wave velocity reduces the pressure wave amplification from the center to the periphery. so that the difference between central and peripheral systolic pressure attenuates with aging. Kelly et al. have shown that in elderly the increase in systolic pressure occurring in central arteries is underestimated when blood pressure is evaluated at the radial or femoral artery level (3. 13). In patients aged 30 to 60 years studied by catheterization. Murgo et a\. (13) and Takazawa (14) have reported that in ascending aorta the end-systolic pressure rise due to age was about 40% of pulse pressure. This phenomenon depends on increased stiffness of the entire arterial system -that determines amplification of the arterial pressure waveform generated by a given stroke volume- and earlier pressure wave re-

242

P.S.Saba, C. Cavallini, D. Scorzani,

at al.

flection from the periphery. While the first mechanism increases systolic pressure in both central and peripheral arteries. the second mechanism elevates systolic pressure exclusively in the central districts (15). Thus. the evaluation of arterial pressure in central rather peripheral arteries allows more accurate estimates of effective left ventricular hydraulic load (13. 16) and its effects on left ventricular performance and structure (17.18). Analysis of central pressure waveform has long been limited to selected groups of patients by the need of using invasive methods to measure blood in ascending aorta. The recent development of high fidelity applanation tonometers allows accurate and non invasiverecordings of central pressurewaveform in larger groups of population (3) and has greatly expanded both clinical and research investigations in this field.

S nsor

Tonometer-

Radial arte

---

Applanated area Hone

Fig. 1 - Principles of applanation tonometry. ~onometer has a piezoelectric sensor positioned in the center of a flat ceramic plane. When the h~ld-down pressure applied to the tonometer a.chleves the optimal applanation of the arterial wall, radial forces acting inside the arter~ ar~ exac!ly b~lan~edand transmitted witllout distortions to the sensor. Thls Yields high-fidelity srterial pulse recordings and. when tonometer is internally calibrated, direct measures of blood pressure.

ARTERIAL TONOMETRY Principles of applanation tonometry Applanationtonometry. a techniquewidelyusedin ophthalmology for measuring the intra-ocular pressure, was introduced in 1963 by Pressmanet al. (19) in the study of the arterial system to obtain non-invasiveand accurate arterial pressure waveforms at several peripheral arterial sites. The flat. rigid surface of a pressure transducer is appliedon the skin overlyingthe arterial pulse to applanate the curved surface of the artery. Whenapplanationis achieved.the contact pressure balancesthe circumferential intraluminal stresses acting on the arterialwall andthe transducerrecords the actual pressure insidethe artery (19) (Fig. 1).This technique is basedon solid theoretical principlesthat have been developedoveryears by accurate modeling studies (19. 20). According with these studies. stringent criteria are neededto achievethe optimal applanation of the arterial wall.The artery must be flattened by a planar.rigid and small surface.in order: a) to minimize the deformational stress of the vesselwall: b) to measureonlythe localstresses.avoidingto includethe force externallyappliedto the surface:c) to analyzeexclusivelythe signal originated by the applanated surface of the arterial wall, assuming the transmission of shear and bending stresses from the peripheral sections of applanated vesselwall, to be negligible. Optimization of the signalis revealedby the maximal pulse amplitude. representing the radially directed intrarte-

243

;_

..._--------

rial pressure. Under ideal conditions of applanation. the pressure wave measured non invasivelyis virtually identicalto that recordedwith an intra-arterial highfidelity transducer (4. 5). Sincein the clinicalpractice the 'hold down pressure' required to applanate the artery may not be constant. only the contour of the tonometric pressurewaveand its excursion(pulsepressure) are reliable.whereas assessmentof the absolute valuesof both systolicand diastolic pressuresrequires an external calibration. For the radial artery. calibration is performed by setting diastolic tonometric pressure to the value of brachial diastolic pressure determined by sphygmomanometer. assuming as negligible the pressure wave amplification between brachial and radial artery (3). The same method of calibration is not appropriate for the carotid artery. becausethe pressure wave amplification increases markedlyfrom central to peripheral arteries and affects both systOlic and diastojc pressure. However,mean blood pressure does not vary significantly in the capacitance vessels (21.22). Thus, the value of brachial mean bloodpressure measured by cuff and mercury sphygmo,:,anometer can be aSSignedto the planimetrically-derived mean blood pressure of the carotid waveform (18.23).

Instruments At present. two systems of applanation tonomet~ are commercially available: the single element Mll-

Applanation tonometry in hypertension

tar transduce~.in~roducedby H. Millar in 1989 (4. S). and the multl-unit arrays 'Jentow' instrument. developed more r~cently by Nihon Kohden Company (24. 25). T~e Millar transducer consists of a handheldprobe Incorporating a tip micromanometer composed by.s~iff ceramic material with piezoelectric characteristics(the ~e.nSitivearea is 0.5 x 1 mrn) (Fig. 1). The tonome~er ISmternally calibrated (1 mV=10 mmHg) and registers absolutechangesin bloodpressure over a range of 300 mmHg. Everyforce applied at t~e free end o~t~e pressure sensorinducesa proportional change In Its electricalreslstancs,converted to a voltage change and then amplified. The Millar transducer can be applied to virtually any palpable artery. although the radial and carotid arteries are preferred becausethe firm tissues behindthese vessels allow the principles of tonometry to be satisfied. The 'Jentow' instrument is a continuousnon-invasive tonometric blood pressure monitoring system (25) basedon an array of 30 piezoresistivepressuretransducers. embedded in a tonometric sensorstraddling the radial artery and pressedagainst it by hold-down pressure in an air chamber. A servo-controlled positioning system determines automaticallythe best position of the sensor.The optimal hold-downpressure is given automatically to maximizethe pulsepressure measured by the transducers locatedoverthe artery. Arterial pressure is measured by the transducer centered over the flattened portion of the arterial wall. The signal of the tonometric sensoris calibratedbyoscillometric measurementsof systolicand diastolicblood pressure. The 'Jentow' system hasdefinite advantages (continuousnon-invasivemonitoringof pressurewaves. automatic operation of the sensor.and optimal placement of the tonometer on the artery). but its use is limited to the radial artery. To date. most data reported in literature have been obtained with the single unit hand-held Millar system. In our experience. we use the Micro-TIp SPT301 applanation tonometer (Millar Ins~ruments In~.. Houston.TX). a high fidelity externalsolid-statestralngauge pressure transducer. A control u~it (Millar T~B 500) provides eXCitation voltage and Signal amplification (15). The transducer requires external calibration to obtain the actual blood pressure valuesat a specific arterial site. Sincemean arterial blood pressure does not vary significantly in the capacitance vessels (21. 22). we measure brachial artery blood pressure by cuff and mercury sphygmomanometer at the end of the study. and assign mean blood pres-

sure [diastolic blood pressure + (pulse pressure/3)] to the planimetrically computer-derived me~n bl?od pressure of the carotid waveform. After calibratlon. carotid peak systolic and end-diastolic pressure are automatically calculated by computer. Validation studies Validationstudies have compared. either in the time andfrequency(Fourieranalysis)domain. arterial waveforms and modulus and phase of harmoniC components obtainedby applanationtonometers with those derived from intrarterial recording. An initial study showed excellent correlations between waveforms recorded invasivelyin the femoral artery and those recordedby applanationtonometry both percutaneously andonthe exposedcontralateralfemoral artery in dogs (4).In the samereport. a wide age-range(17-79 years) human population undergoing invasive radial artery pressuremonitoring was studied by arterial tonometry of the contralateralradialartery. Arterial waves obtamedwith the two methodswere comparablein shape and spectral analysiscontents (4). Moreover.close relationshipshavebeenfound either in the time and frequencydomainbetweenthe pressurewaveform recorded in ascendingaorta and the carotid pressure waveform obtained simultaneouslyby external tonometer. in patients undergoingcardiaccatheterization (5). The strength of these relationswas not affected by the patient ageor nitroglycerin administration that reduced late systolic peak in both carotid and aortic pressure waves(5). Accordingly.the carotid pulse can be used in substitution of the ascending aortic pressure. although tonometry slightly underestimates aortic pressure late systolicpeak and augmentation index (3. 5. 26). Thus.arterial applanationtonometry provides accurate non-invasiverecordings of the peripheral arterial pressurewaveforms and reliable estimates of the ascendingaortic pressure and left ventricular load. This technique has definite advantages over the current methods of non invasive assessment of arterial pressure.Unlikesphygmomanometric. oscillometric or plethysmographictechniques. arterial tonometry does not require cuffs or intermittent vessel occlusions -which may alter the hemodynamic status of the underlying artery- and permits the continuous beat-tobeat monitoring of the pressure wave contour. Compared to invasivearterial pressure monltormq, tonometry avoidsproblems related to arterial trauma. infections. and management of fluid-fuled tubing associated with the use of intra-arterial catheters.

244

c P.S. saba, C. Gavalllni, D. Scorzoni, et aI.

CLINICAL AND RESEARCH APPLICATIONS OF ARTERIAL TONOMETRY 1.Assessment of arterial mechanicalproperties Elasticity is the property by which a body resists to deformation when a force is applied to its surface. Arterial elasticity can be evaluated by the physiological changes in diameter and pressure occurring during the cardiac cycle in a given district. Since arterial wall thickness is not negligible compared to radius. the radial force acting on the examined arterial segment is better estimated in terms of arterial wall stress. Present availability of high fidelity arterial tonometry and vascular ultrasonography allows reliable and simultaneous measures of pressure, diameter and wall thickness to be obtained in explorable arterial segments.

Arterial ultrasonography Ultrasound exploration of capacitancearteries is widely used to assess vessel anatomy and detect atherosclerotic lesions (27-30) At carotid level. both intimal-medial thickening and discrete atherosclerotic plaque are related to risk factors and predict cardiovascular morbid events (31-36). 8-mode ultrasonographic characterization of carotid wall layers and measurements of intimal-medial thickness have been validated in anatomic correlation studies (27). More recently. two-dimensionally guided Mmode carotid ultrasonography has been proposed to obtain arterial internal diameter and intimal-medial thickness measurements gated with EKGor pressure waveform (23, 37). In fact, significant dynamic variations of these parameters occur in response to the pulsatile pressure changes (23. 37-39). with the internal lumen reaching the largest dimension at peak-systole and the smallest at end-diastole, and the intimal-medial complex being 5.3% thicker at end-diastole than at peak-systole (23, 37). In addition. M-mode ultrasonography has greater axial resolution (0.2 mm with the 7.5 MHZ probe) than Smode technique. Several methodological recommendations must be followed in performing clinical or investigational arterial ultrasonographic studies: 1) use of 7.5-10 MHz ultrasound probes is recommended for carotid artery studies (31-34,40). while higher frequencies are required to evaluate smaller arteries (38. 39); b) both intimal-medial thickness and internal diameter must be measuredat levelof common carotid artery, where

245

Rg. 2 - Simultaneous recordings of carotid artery M-mode ultrasonography and pressure waveform in a ~oung .norm~1subJ~ct (I~~) and in a middleaged hypertensive (right). The inflection point (PJ Identifies the return of the reflected waves to the central districts. In young normal subjects, the reo flected waves merge with the incident wave after the peak systolic pressure is reached (left). In hypertensive or elderly subjects, the earlier retum of the reflected waves determines an inflection on the ascending branch of the carotid pulse and increases peak systolic pressure.

measurements are more accurate and discrete atherosclerotic lesions less frequent compared to carotid bulb and internal or external carotid arteries (32-40); c) measurements of carotid intimal-medial thickness should be performed on the arterial far wall, that is usually better visualized than the near wall (33); d) arterial wall thickness and dimensions should be measured at end-diastole, as averageof severalcardiaccycles,using the leading edge-to-lead· ing edge method (37). It must be considered that carotid waii thinninq and internal diameter dilation independent of cardiac cycle may occur becauseof additional mechanisms (use of vasodilator drugs, increased flow. increase in internal pressure unrelated to changes in arterial smooth muscle tone) (37).

Association of tonometry and M-mode ultrasonography Drs. M.J. Roman and R.B. Devereuxat the CornellUni· versity Medical Center, New York, have recentlyasso· ciated applanation tonometry with M-mode ultrasonographyto obtain simultaneous measuresof carotid anatomy and pressure and assessthe viscous-elastic characteristics of this artery (18. 23, 41). The studY subject is examined in the supine position with the neck in slight hyperextension. Using a commercially available ultrasound system equipped with 7.5 or 10 MHz probe and input for external pulse transducer,

.,.. "'

Applanation tonometry in hypertension "

:' ": ~

two-dimensional images of left and . h carotid artery. carotid bulb. internal and rextelg t comm~n . bt . mal carotid art enes are 0 amed. 'TWo-dimensi II . mode tracings of the distal ona Y gUided M. common carotid art f one side. about 1 cm proximal to the b Ib ery ed Sllmulta neous I'y With an electrocardiu • are record. and the pressure waveform obtained bograPhlC I~ad . et f th 'Y applanation t onom ry 0 e contra-lateral carotid rt ~ '. a ery. Carotid pr~ure waverorm IS obtained using the SPT-301high fidelity external pressure transducer (Millar Instruments. H~uston. USA)and its TC8 500 control unit that provides excitation voltage and Signal a I·ti · B h' rnpnncatIon. rac lal artery blood pressure is measured at the end of the study by C~ff and mercury sphygmomanomet~r. Pressu~e tracings are displayed simultaneously with ca~otld M-mode images (Fig. 2) and recorded on O. ~-mches S-VHSvideotape. Recordings are carefully reviewed. and suitable frames for measurements are acquired by a frame-grabber (Variable Scan. Imaging Technology lnc .. Woburn. MA, USA)interfaced with a high-resolution video monitor and stored on diskettes. The value of brachial mean blood pressure (diastolic blood pressure + 1/3 pulse pressure) is assigned to carotic mean blood pressure obtained planimetrically by computer. After pressure calibration. pressures at each point of carotid waveform can be automaticaliy calculated by computer. Excellent intra- and inter-observer reproducibility has been reported for both peak systolic and end-diastolic carotid pressures determined by this method (r=0,99. SEE=1 mmHg for all comparisons) (23). After calibration for depth and time, measurements of carotid anatomy and pressure are performed over several cardiac cycles and averaged by using a dedicated mouse-driven software (©ARTSS,Arterial Structure and Stiffness, Copyright 1989-1995. Cornell University Research Foundation. Ithaca. NY, USA) (18, 41).

iI

°

:X.

Indices of carotid stiffness The following parameters are measured ?n ca:o~id M-mode ultrasonographic tracings: end-.dlastollc I~tlrnal-medial thickness of the far wall; Intern~1 diameters at end-diastole and peak-systo~e, obtam.ed by continuous tracing of the intima lea~lng edge interfaces of the near and far wall. Carotid measurements are gated with the cardiac cycle, based on t~e simultaneous superimposed pressure ~av~form (Fig. 3). Previous studies have reported hl~h Intra- ~nd inter-observer reproducibility for carotid wall thickness and internal diameters measurements (23. 42).

Carotid atherosclerotic plaques, defined as the presence of wall thickening at least 50% greater than the surrounding wall (28), are bilaterally investisgated and quantified by two-dimensional images. Since the human arterial tree is composed by elastic and muscular vessels with different functional characteristics. arterial compliance of a single district may not be representative of systemic arterial compliance. Nevertheless, a less compliant carotid artery is likely to reflect a more extensive damage of the arterial tree, as suggested by the high rate of cardiovascular morbid events occurring in the presence of carotid plaques or mia-intimal thickening (31-34, 37, 43). Arterial compliance is defined as the absolute volume (diameter. or area) change for a given pressure increment at fixed vessel length (i.e., dV/dP. the slope of the VOlume-pressure relationship). The relationship between volume and pressure is not linear. because arterial wall has viscous-elastic rather than pure elastic properties that reflect its structural composition: elastine is highly extendible even at low pressure levels. whereas collagen is stiff and virtually not extendible. CUrvilinearity of the pressure-volume relationship implies that dV/dP declines when arterial pressure increases, thus distending pressure itself may affect arterial compliance independently of structural changes of the vessel wall (44). Indices of arterial stiffness can be determined from the tangent of the curve at every given distending pressure or diameter. Elastic modulus (Ep' Peterson's modulus) is the pressure increment required for (theoretical) 100% stretch of the resting diameter at fixed length (Ep=(Ps-Pd)/(Ds-Dd)xDd) (45). Young's modulus (E) is the wall tension per centimeter of wall thickness required for (theoretical) 100% stretch of the resting length (E=(ps-P d)/(DsDd) Dth} (46). Ps.Pd, e are systolic and diastolic pressures and diameters. respectively. and D and h are the carotid mean diameter and wall thickness. respectively. These indices can be estimated in vivo or in vitro, measuring changes in diameter due to a given change in pressure. at every level of pressure or diameter (Fig. 3). In contrast with Peterson's modulus, Young's modulus incorporates structural adaptive changes in arterial wall thickness. Since both indices do not take into account differences in distending pressure. they are dependent on blood pressure level and are usually calculated at a given mean arterial pressure and diameter.

o, o,

246

J

P.S.Saba. C. Cavallini. O. Scorzoni. et al.

Arterial stiffness can also be estimated as Stiffness B index or 13(13=ln(P/Pd)/(Ds-Dd)/Do)' according with A Hayashi. Hirai et aJ.(41. 47-49). where Dois arterial dimension at standardized blood pressure. Sincethis method expresses the pressure/diameter ratio as a logarithm and standardizes the distending pressure by using Do(Le.arterial diameter at a givendistending pressure). it takes into account the non-linearity of Pd the pressure/diameter relationship and provides an estimate of arterial compliance independent of changes in distending pressure. The method is inFig. 3 - Schematic representation of ?arotid artery pulse and M-mode Ultrasonography in a young .norm.al subject (A) and. In a mlddle-age~ hyper. dependent of distending pressure even when Ddrepatient (8). ~d= dl~stollc blood pres.sure. ps.= peak sys_toll~blOOd places Do. resulting in 13'[13'=In(PsiPd)/(Ds-Dd)/DdJ tensive pressure; Pj= inflection POl~t; ~otch= dicrotiC n?t~h •.Od= car~tId ~Iastolic (48). An isobaric regional arterial stiffnes (1315) can be diameter; Os= carotid systolic diameter; h= carotid intima-media thickness. Augmentation index is calculated as (P;-PJ/pulse pressure when Ps OCcurs estimated calculating 13 for fixed intervals of systolic in early systole (A), and as (Ps-P1)/pulsepress~re when P~?Ccurs in late sys. and diastolic blood pressures; i.e.. for the interval tole (S). Indices of arterial st.iffness are obtained ~ombmlng pressure data 120 to 100 mmHg.l3is= In(P12c1P100)/(D120-D100)/D100' with carotid diameter and thickness (for explanation. see text). This approach provides the most accurate non-invasive estimate of stiffness in a visualized arterial segment. allowing the impact of arterial distending presmentation index is then calculated as (Ps-Pj)/pulse sure and structural adaptation to be taken into acpressure (when Psoccurs in late systole). or as (Pi" count and data to be compared between groups of Ps)/pulsepressure (when Psoccurs in early systole) subjects whose range of pressure includes the in(13). Interobserver reproducibility for this indexhas terval of pressure used for calculation. been reported to be high (18. SO).A report from our 2. Assessment of pressure wave group (18) has shown that the amplitude of the rereflection flected pressure wave evaluated as augmentation index is an independent determinant of left ventricAmplitude and timing of the backward component ular mass in normotensive subjects. of the arterial pressure waveform recorded in the More recently. the combined use of carotid tonocentral districts depend on pulse wave velocity. dismetric pressure waveform and carotid-femoral pulse tance of the reflection site(s). and coefficient of reflection (15). In turn. pulse wave velocity depend on wave velocity has been proposed to estimate the arterial stiffness at the level of aorta and its major functional site of pressure wave reflection. a major branches. Analysis of the systolic portion of the presdeterminant of augmentation index (51). Prelimisure waveform provides useful information about nary data from our laboratory suggest that the maseveral aspects of wave reflection. jor site of pressure wave reflection is not affected by Augmentation index is commonly considered a relihypertension (51) while it becomes more proximal able estimate of precocity and intensity of the wave in severe hypercholesterolemia (52). reflected from the periphery of the arterial tree toTIming of wave reflection is potentially of great relwards the heart. It measures the increase in peak evancefor myocardial function and perfusion. In norsystolic pressure occurring in late systole -which is mal conditions. the reflected pressure wave usually due to early return of the reflected waves- as a promerges with the incident wave when left ventricuportion of pulse pressure (3. 13. 16. SO)(Fig.3). Nonlar ejection has been completed. contributing to ininvasive measures of augmentation index obtained crease diastolic blood pressure. Conversely. in conat the carotid levelare closelyrelated to those recordditions of increased arterial stiffness and higher pulse ed invasivelyin ascending aorta (r=0.70; p