Arterial Compliance by Cuff Sphygmomanometer
Application to Hypertension and Early Changes in Subjects at Genetic Risk
Abnormalities of the arterial pulse waveform reflect changes in cardiovascular structure and function. These abnormalities may occur early in the course of essential hypertension, even before the onset of blood pressure elevation. Previous studies of cardiovascular structure and function have relied on invasive intra-arterial cannulation to obtain the arterial pulse wave. We evaluated arterial structure and function using a noninvasive cuff sphygmomanometer in hypertensive (n=15) and normotensive (n=36) subjects stratified by genetic risk (family history) for hypertension. Using a simple physical model in which the aorta was assumed to be a T tube and the brachial artery a straight tube, we determined vascular compliance and peripheral resistance by analyzing the brachial artery pulsation signal from a cuff sphygmomanometer. Essential hypertensive subjects tended to have higher peripheral resistance (P=.06) and significantly lower vascular compliance (P=.001) than normotensive subjects. Vascular compliance correlated with simultaneously determined pulse pressure in both groups (n=51, r=.74, P<.0001). Higher peripheral resistance (P=.07) and lower vascular compliance (P=.04) were already found in still-normotensive offspring of hypertensive parents (ie, normotensive subjects with a positive family history of hypertension) than in normotensive subjects with a negative family history of hypertension. Multivariate analysis demonstrated that both genetic risk for hypertension (P=.030) and blood pressure status (P=.041), although not age (P=.207), were significant predictors of vascular compliance (multiple R=.47, P=.011). However, by two-way ANOVA, genetic risk for hypertension was an even more significant determinant (F=7.84, P=.007) of compliance than blood pressure status (F=2.69, P=.089). Antihypertensive therapy with angiotensin-converting enzyme inhibitors (10 days, n=10) improved vascular compliance (P=.02) and reduced resistance (P=.003) significantly; treatment with calcium channel antagonists (4 weeks, n=8) tended to improve vascular compliance (P=.07) and significantly reduced peripheral resistance (P=.006). We conclude that arterial vascular compliance abnormalities detected by a noninvasive cuff sphygmomanometer reflect treatment-reversible changes in vascular structure and function. Early changes in vascular compliance in still-normotensive individuals at genetic risk for hypertension may be a heritable pathogenetic feature of this disorder.
- peripheral resistance
- angiotensin-converting enzyme inhibitors
- calcium channel antagonists
Guidelines for the diagnosis and treatment of essential hypertension are traditionally based solely on elevation of arterial blood pressure (BP). However, many factors may contribute to a rise in BP, including changes in arterial vascular compliance and peripheral resistance.1 2 3 4 5 6 7 8 9 Assessment of such vascular factors may provide crucial (and, currently, infrequently obtained) information about the severity and consequences of BP elevation. Peripheral resistance measurements gauge vascular function during diastolic blood flow.4 In contrast, measurement of vascular compliance, a function of the change in arterial volume in response to the pulsatile pressure cycle, is influenced by vascular structure.4 Assessment of both vascular compliance and peripheral resistance may serve to more precisely define the roles of vascular structure and function in hypertension and other cardiovascular diseases.
Several previous studies evaluated vascular structure and function by incorporation of pressure waveforms, derived by invasive catheterization or tonometry, into a Windkessel model,10 11 12 13 a simple electrical circuit–analogy model composed of resistors and capacitors used for the estimation of systemic compliance and resistance. We evaluated arterial vascular compliance and peripheral resistance using a physical model14 15 16 17 of the local vascular system patented by one of us (S.-S.C.). In contrast to the use of potentially hazardous invasive transducers or difficult-to-operate external tonometers, pressure waveforms are derived with a standard arm cuff sphygmomanometer coupled with pulse dynamic technology.14 15 16 17 We made noninvasive measurements on study subjects stratified by BP, antihypertensive treatment, and genetic risk for hypertension to evaluate the effects of hypertensive disease and treatment on vascular structure and function.
Studies were approved by the Human Subjects Committee of the University of California, San Diego. The study population consisted of 51 subjects (34 men and 17 women; 47 whites and 4 blacks) ranging in age from 23 to 65 years (mean±SE, 40±2 years); 36 were normotensive and 15 had essential hypertension. A diagnosis of essential hypertension was established by the presence of a sustained elevation in BP (>140 mm Hg systolic, >90 diastolic, or both) on at least three determinations and the absence of clinical or laboratory evidence of secondary forms of hypertension.18 The family history of hypertension of each subject was assessed by personal interview and questionnaire as well as contact with parents by telephone or mail. Among the 36 normotensive subjects, 19 had a positive family history (FH+) and 17 a negative family history (FH−) of hypertension. Twelve of 15 hypertensive subjects were FH+, 1 of 15 was FH−, and 2 of 15 were indeterminate (parents not available). FH+ was defined as the presence of either treated or untreated hypertension (diastolic BP >90 mm Hg) documented in a first-degree relative (parent or sibling) before age 60 years.18 Previous antihypertensive medications were suspended for at least 2 weeks before a baseline BP reading was established for each subject.
Noninvasive arterial pressure signals were obtained for each subject with a DynaPulse 2000A (Pulse Metric) transducer. Three recordings were obtained for each subject after 5 minutes of seated rest. Systolic, diastolic, and mean arterial pressures were derived for each oscillometric pressure recording (Fig 1⇓) by a pattern-recognition algorithm.14 Each subject's BP was derived as the average of the three independent determinations. The arterial pressure signal with median systolic pressure was incorporated into a physical model of the brachial artery segment for determination of vascular compliance and peripheral resistance as described below. Data were recorded on the hard disk of a microcomputer (Cardinal DOS 386 dx) and backed up on floppy disk. BP, vascular compliance, and peripheral resistance were imported into a spreadsheet (Lotus 1-2-3, Lotus Development Corp) for evaluation.
To evaluate the effects of antihypertensive treatment on BP, we compared vascular compliance and peripheral resistance baseline (during placebo) arterial waveforms with arterial waveforms obtained after medication given once daily. Before antihypertensive treatment, subjects were treated with oral placebo for 10 to 14 days. The effects of angiotensin-converting enzyme inhibitor monotherapy were evaluated in 10 subjects (6 men and 4 women; all white; age, 44 to 71 years; mean±SE, 58±3 years) with either ramipril (10 mg/d, n=5) or enalapril (10 mg/d, n=5) for 10 days. Calcium channel antagonist monotherapy was also evaluated in 8 subjects (all men; 5 whites and 3 blacks; age, 38 to 66 years; mean±SE, 49±4 years) with either sustained-release verapamil (240 mg/d, n=4) or sustained-release felodipine (5 mg/d, n=4) for 4 weeks.
Vascular compliance and peripheral resistance measurements were derived by incorporation of noninvasive arterial pressure signals obtained from a standard cuff sphygmomanometer into a physical model of the brachial artery segment.15 16 17 18 In brief, the model assumes a straight tube brachial artery and T-tube central aortic system in which the systolic phase of the suprasystolic wave and the diastolic phase of the subdiastolic wave most closely approximate systolic and diastolic aortic pressures, respectively. The model derives vascular compliance from radial (perpendicular to the wall of the artery) compression and expansion of the brachial artery, resulting from the oscillating BP. In contrast, peripheral resistance measurements are derived from longitudinal (parallel to the wall of the artery) movement of blood through the brachial artery segment. For determination of pressure dynamics throughout the cardiac cycle, without damping from the cuff sphygmomanometer, a pseudoaortic waveform was derived from each oscillometric recording. Suprasystolic and subdiastolic signals from the oscillometric recording were normalized to systolic and diastolic pressures (Fig 2⇓) and combined for generation of the pseudoaortic waveform. Vascular compliance and peripheral resistance were calculated as:Vascular Compliance (mL/mm Hg)|<|=|>|\frac|<||<|\pi|>|^|<|2|>||<|\cdot|>|D^|<|2|>|_|<|o|>||<|\cdot|>|(D_|<|o|>||<|+|>|L_|<|c|>|)|>||<||<|[|>|(dP/dt)pp|<|\cdot|>|Tpp|<|]|>|SW|>|where SW is the systolic wave, DW is the diastolic wave, MAP is mean arterial pressure (millimeters of mercury), DBP is diastolic BP (millimeters of mercury), (dP/dt)pp is the amplitude from dP/dtMax (peak positive pressure derivative) to dP/dtMin (peak negative pressure derivative), Tpp is the interval between the peak positive and peak negative pressure derivatives, and (dP/dt)DW is the slope of the diastolic pressure decay. An estimate of brachial artery diameter, Do, was generated by a mathematical model in which the average size of the brachial artery at mean arterial pressure was scaled for body size (body surface area) with the use of the height and weight of each subject. The effective cuff width, Lc, was defined as the length of the inflated cuff in contact with the arm.
Results are expressed as mean±SE. Statistical analyses were performed with Excel (Microsoft Corp), Systat (Systat Inc), Statworks (Cricket Software), or InStat (GraphPad Software). Groups were compared by t tests for unpaired data or by two-way ANOVA, whereas t tests for paired (before and after treatment) data evaluated the effects of antihypertensive medications.
Fig 1⇑ illustrates a typical oscillometric pulsation signal obtained during brachial sphygmomanometric cuff deflation, and Fig 2⇑ shows components of the oscillometric signal used in the computation of compliance and resistance.
In addition to higher BP (P<.001, Table 1⇓), hypertensive subjects had significantly lower vascular compliance (P=.001) and a tendency toward greater peripheral resistance (P=.06, Table 1⇓) than normotensive subjects.
A significant correlation was observed between vascular compliance and simultaneously determined pulse pressure in the total subject group (n=51, r=.74, P<.0001) as well as in both normotensive (n=36, r=.72, P<.0001) and hypertensive (n=15, r=.70, P=.0025) subgroups.
After antihypertensive angiotensin-converting enzyme inhibition (n=10 subjects) for 10 days, mean arterial pressure decreased by 7 mm Hg (P=.02), vascular compliance improved (increased, P=.02), and peripheral resistance declined significantly (P=.003, Table 2⇓). Similar results were obtained after antihypertensive treatment with calcium channel antagonists (n=8 subjects): Mean arterial pressure declined significantly (from 117.4±4 to 112.6±4 mm Hg, P=.01), vascular compliance tended to improve (P=.07), and peripheral resistance declined significantly (P=.006) after 4 weeks of therapy (Table 2⇓).
In the still-normotensive offspring of hypertensive parents (FH+, Table 3⇓), mean arterial pressure was already higher (P=.008) by 9 mm Hg than in the FH− group. In the FH+ subgroup, vascular compliance was lower (0.158±0.01 mL/mm Hg) than in the FH− subgroup (0.181±0.01 mL/mm Hg, P=.04). In contrast, FH+ peripheral resistance tended to be higher (557.2±39.9 versus 484.1±24.4 mm Hg/L per minute, P=.07).
In the entire study population (n=51), compliance did not vary by age (r=.204, P=.15). Since 47 of 51 subjects were white, we did not stratify hemodynamic parameters by race. In the 36 normotensive subjects, men had somewhat higher mean arterial pressures than women (92.1±2.4 versus 80.1±2.1 mm Hg, t=3.53, P=.001), although compliance did not vary by sex (0.165±0.0063 versus 0.174±0.014 mL/mm Hg, t=−0.66, P=.51).
Both BP (Table 1⇑) and a family history of hypertension (Table 3⇑) seemed to have an effect on compliance, yet the BP groups differed in age (Table 1,⇑ 32±1 versus 55±3 years, P<.0001). To better understand the relative contributions of these three independent variables (age, family history, and BP status) to vascular compliance (as a dependent variable), we undertook a multivariate analysis (multiple linear regression) on the entire study population (Table 4⇓). Significant effects on compliance were found for two of the independent variables tested—BP status (P=.041) and family history of hypertension (P=.030)—although not for age. The combination of three independent variables used in this multiple regression analysis predicted compliance (multiple R=.47, P=.011, n=49). Two-way ANOVA, evaluating the effects of BP status (hypertensive or normotensive) and family history of hypertension (FH+ or FH−) on vascular compliance, demonstrated a more significant effect for family history status (F=7.84, P=.007) than for BP status (F=2.69, P=.089), without significant interaction between these two independent variables (F=0.022, P=.882).
Several recent studies have directed attention toward vascular determinants of BP, of which abnormalities of compliance or distensibility may be of primary importance.1 2 4 7 Evolving from invasive methods, several noninvasive techniques have been developed for the determination of vascular compliance.10 11 13 The objective of the present study was to evaluate the effect of hypertension (and genetic risk for hypertension) on vascular compliance with a new technology (Fig 2⇑) based on a simple cuff sphygmomanometer.14 15 16 17
In agreement with previous studies,1 2 3 4 7 19 20 vascular compliance was lower (by 17.9%, P=.001; Table 1⇑) in hypertensive compared with normotensive control subjects. Compliance improved (increased, Table 3⇑) during antihypertensive therapy with either angiotensin-converting enzyme inhibitors (P=.02) or calcium channel antagonists (P=.07). Thus, diminished compliance in hypertension may be reversible even over a relatively short period of time. Ting et al9 and De Luca et al21 previously showed that compliance improved after antihypertensive angiotensin-converting enzyme inhibition, and de Cesaris et al3 found that compliance improved after calcium channel antagonist treatment in hypertension.
Although pulse pressure is not directly involved in the derivation of vascular compliance by our method,15 16 17 we found a significant correlation between pulse pressure and vascular compliance in this study sample (n=51, r=.74, P<.0001). Li et al22 observed a similar correlation during drug-induced hypertension in dogs.
Girerd et al19 previously reported abnormalities of the arterial pulse wave very early in the course of hypertension, and Weber et al,1 using an invasive technique, found diminished arterial compliance even in the still-normotensive offspring of essential hypertensive individuals. It is evident from our results (Table 3⇑) that in young (34±2 years old) still-normotensive offspring of hypertensive subjects (FH+), compliance was already lower (by 12.6%, P=.04) than in FH− subjects. Since a positive family history of hypertension confers an increased relative risk of future development of hypertension in a currently normotensive individual,24 we construed a positive family history of hypertension to be a genetic risk factor for hypertension. Although BPs were already somewhat higher in the FH+ group (P=.012 to .005, Table 3⇑), both normotensive groups had average BP values well within the normotensive range. Multivariate (multiple) linear regression analysis (Table 4⇑) demonstrated that family history was a significant (P=.030) predictor of compliance, even after controlling for the effects of age and BP status; this result documents a hereditary influence on compliance. Indeed, Benetos et al24 have recently demonstrated an association between aortic stiffness and particular polymorphic DNA alleles at the vascular (type 1) angiotensin II receptor locus.
Although vascular compliance may decrease with age,8 20 the results of the present study suggest that vascular compliance depends more on BP and a family history of hypertension than on age (Table 4⇑). However, this conclusion may be tempered by our relatively young subjects (age, 23 to 71 years; mean±SE, 39±2 years).
In summary, the results of this study suggest that vascular compliance abnormalities in hypertension can be noninvasively derived from a brachial artery pulsation signal with the use of a simple cuff sphygmomanometer and novel pulse dynamic technology. During antihypertensive treatment, improvements in vascular compliance (Table 2⇑) suggest that such abnormalities may be reversible. Finally, our results demonstrate that vascular compliance changes in hypertension may be at least partly genetically determined. The early (prehypertensive) appearance of such compliance deficits may represent a presymptomatic trait or “intermediate phenotype,”25 providing clues to very early (and perhaps pathogenetic) vascular structure alterations in this disorder.
This work was supported by the Department of Veterans Affairs, the National Institutes of Health, the American Heart Association, and Pulse Metric Inc.
- Received December 11, 1996.
- Revision received April 25, 1996.
- Accepted May 14, 1996.
O'Rourke MF, Brunner HR. Introduction to arterial compliance and function. J Hypertens. 1992;10S:S3-S5.
Hayoz D, Rutschmann B, Perret F, Niederberger M, Tardy Y, Mooser V, Nussberger J, Waeber B, Brunner HR. Conduit artery compliance and distensibility are not necessarily reduced in hypertension. Hypertension. 1992;20:1-6.
Safar ME. Hemodynamic changes in elderly hypertensive patients. Am J Hypertens. 1993;6:20s-23s.
Ting CT, Yang TM, Chen JW, Chang MS, Yin FC. Arterial hemodynamics in human hypertension: effects of angiotensin converting enzyme inhibition. Hypertension. 1993;22:839-846.
Watt TB, Burrus CS. Arterial pressure contour analysis for estimating human vascular properties. J Appl Physiol. 1976;40:171-176.
Goldwyn RM, Watt TB. Arterial pressure contour analysis via a mathematical model for the clinical qualification of human vascular properties. IEEE Trans Biomed Eng. 1967;14:11-17.
Simon AC, Safar ME, Levenson JA, London GM, Levy BI, Chau NP. An evaluation of large arteries compliance in man. Am J Physiol. 1979;237:H550-H554.
Liu Z, Brin KP, Yin FC. Estimation of total arterial compliance: an improved method and evaluation of current methods. Am J Physiol. 1986;251:H588-H600.
Chio SS. Method and apparatus for determining blood pressure and cardiovascular condition. US patent No. 4,880,013; November 14, 1989.
Chio SS. A method of diagnosing, monitoring and treating hypertension and other cardiac problems. US patent pending; December 19, 1993.
Kailasam MT, Brinton TJ, Wu RA, Hu G, Chio SS, Cervenka JH, Parmer RJ, O'Connor DT. Non-invasive arterial pulse waveform analysis in hypertension: development of a method, and early compliance changes in subjects at genetic risk of hypertension. Am J Hypertens. 1994;7:86A. Abstract.
Brinton TJ, Cotter B, Brown DL, Baddour P, Voung A, Chio SS, Calisi C, DeMaria AN. A new non-invasive method for obtaining arterial pressure waveforms: assessment of vascular compliance and validation with catheter data. Circulation. 1994;90(suppl I):I-445. Abstract.
Parmer RJ, Cervenka JH, Stone RA. Baroreflex sensitivity and heredity in essential hypertension. Circulation. 1992;85:497-503.
Girerd X, Chanudet X, Larroque P, Clement R, Laloux B, Safar M. Early arterial modifications in young patients with borderline hypertension. J Hypertens. 1989;7:S45-S47.
Li J, Ying Z, Drzewiecki G. Pulse pressure is a significant determinant of arterial compliance in hypertension and vasodilation. Circulation. 1994;90:166. Abstract.
Hunt SC, Stephenson SH, Hopkins PN, Williams RR. Predictors of an increased risk of future hypertension in Utah: a screening analysis. Hypertension. 1991;17(part 2):969-976.
Benetos A, Topouchian J, Ricard S, Gautier S, Bonnardeaux A, Asmar R, Poirier O, Soubrier F, Safar M, Cambien F. Influence of angiotensin II type 1 receptor polymorphism on aortic stiffness in never-treated hypertensive patients. Hypertension. 1995;26:44-47.
Williams GH, Dluhy RG, Lifton RP, Moore TJ, Gleason R, Williams R, Hunt SC, Hopkins PN, Hollenberg NK. Non-modulation as an intermediate phenotype in essential hypertension. Hypertension. 1992;20:788-796.