SYSTEMIC HYPERTENSION

Maximal Hand Blood Flow in Hypertensive and Normal Subjects

DAVID HORWITZ, MD, and DALI J. PATEL, MD, PhD

The present study examined whether patients with systemic hypertension have evidence of structural vascular changes, whether such changes can be detected in early stages of hypertension and whether they are reversible with treatment. Hy- pertensive and normal subjects were studied under conditions of maximal vasodilation in which flow at a given driving pressure was considered to give an index of structural changes in resistance vessels. Thirty-two subjects were separated into 4 groups: 8 with sustained hypertension, 8 with intermittent hypertension, 8 treated hypertensive subjects maintained at systolic pressures of less than 125 mm Hg with drugs for 5 years, and 8 normal subjects. Flow was measured by venous occlusion plethysmography after 10 minutes of ischemia, using a water-filled plethysmograph at 43’C. Arterial blood pressure was measured by the arm cuff method. Transmural pressure, calculated as mean arterial minus external pressure, was varied by

Essential systemic hypertension in humans and spon- taneous hypertension in the rat are characteristically associated with increased peripheral resistance.1-3 This in turn may result from changes in smooth muscle tone or from structural changes in small blood vessels. The present study is concerned with whether hypertensive subjects show evidence of structural vascular changes when compared at equal transmural pressures, whether such changes can be detected in early stages of hyper- tension and whether they are reversible with treatment. Hypertensive and normal subjects were studied under conditions of maximal vasodilation wherein flow at a given driving pressure was considered to give an index of structural changes in resistant vessels.

From the Hypertension-Endocrine Branch, National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Maryland, and the Department of Physiology and Biophysics, College of Medicine, Howard University, Washington, D.C. Manuscript received June 29, 1984; revised manuscript received October 30, 1984, accepted October 31, 1984.

Address for reprints: Dali J. Patel, MD, PhD, Department of Physiology and Biophysics, College of Medicine, Howard University, Washington, D.C. 20059.

imposing varying external hydrostatic pressures. Flow at a transmural pressure of 85 mm Hg was calculated for each subject from the least-mean- square plot of transmural pressure vs flow. Mean flow for normal subjects was 41 ml/100 ml/min and differed significantly from that for sustained hyper- tensive patients (30 ml/100 ml/min), treated hy- pertensive patients (33 ml/100 ml/min), and inter- mittent hypertensive patients (32 ml/100 ml/min) (p <0.05). There was no overlap between sustained hypertensives and normal subjects, but half of the treated hypertensive patients were normal. Inter- mittent hypertensive patients usually showed evi- dence of pronounced arterial changes. Prolonged therapy appeared to reverse changes in some sus- tained hypertensives. This noninvasive method permits comparisons of hypertensive and normo- tensive subjects at equivalent transmural pres- sures.

(Am J Cardiol 1985;55:418-422)

Methods

Subjects: The subjects were 24 patients with uncompli- cated sustained or intermittent essential hypertension and 9 normal subjects. The subjects were separated into 4 groups:

Group I (sustained hypertensives): Each subject showed blood pressure (BP) readings of at least 140/90 mm Hg in supine and erect postures on each visit. Six subjects were untreated and 2 were hypertensive despite treatment.

Group II (intermittent hypertensives): Subjects were untreated and showed BP levels that varied above and below 140/90 mm Hg.

Group III (optimally treated hypertensives): Before treatment, each subject had shown sustained hypertension during 4 to 6 visits (average BP for all subjects 155 f 7/101 f 4 mm Hg supine, and 156 f 5/109 f 3 mm Hg standing). Thereafter, each subject showed systolic BP readings that did not exceed 125 mm Hg on all visits during at least 5 years of treatment before study (average of 118 f 2/80 f 2 mm Hg supine, and 118 f l/86 f 1 mm Hg standing).

Group IV (normal subjects): Each showed BP levels less than 125/80 mm Hg in supine and standing positions during each visit.

February 1, 1985 THE AMERICAN JOURNAL OF CARDIOLOGY Volume 55 419

TABLE I Patient Characteristics

Sustained Intermittent Treated Normal Hypertensives Hypertensives Hypertensives Subjects

Group I Group II Group Ill Group IV

BPJ;p;b)* 154 f 31101 f 3 138 f 3187 f 2 118&2/80f2 llOf3/66f2

Standing 156 f 41108 f 2 134 f 2197 f 2 108 f l/86 f 1 108 f 2175 f 1 ;igd 43(3;-$2) 45 (28-54)

4B/4W 493f,(3-$1) 41 (27-50)

2BI6W Sex 3Fl5M 2Fl6M 2Fl6M 2Fl6M

* Average of 6 visits for groups I, II and IV and of all visits during at least 5 years for group Ill. Blood pressures (BP) before initiation of therapy in group Ill averaged 155 f 71101 f 4 (supine) and 156 f 5/109 f 3 (standing) during 4 to 6 visits. BP values are mean f standard error of the mean; age values are mean and range (years).

Assignments to various groups were based on BP recordings during 6 different visits over 4 to 6 weeks preceding experi- mental studies for groups I, II and IV and all BP recordings (at least 12 per subject) during at least 5 previous years of continual observations for group III. Patient characteristics and BP for each group before entering the study are shown in Table I. The various groups were reasonably matched, though not exactly, for age and sex. All subjects received no medication for at least a week before plethysmography.

Procedures: Studies were performed in a room maintained at a temperature of 22 to 26°C. BP was measured in 1 arm with a sphygmomanometer. Hand blood flow was measured in the other arm by venous occlusion plethysmography immediately after 10 minutes of arterial occlusion intended to produce maximal vasodilation. A water-filled plethysmograph was used as shown in Figure 1. The hand was enclosed in a loosely fitting thin rubber glove and placed in a rigid chamber, and maintained above heart level in all studies. The chamber was filled from a constant temperature bath maintained at 43%; water in the chamber was exchanged with that in the bath for 3 minutes before each run. The plethysmograph chamber was connected by Tygon@ tubing to a cylinder containing a float that displaced a linear motion transducer as the water level rose. The linear displacement was calibrated to measure vol- ume changes by injecting known amounts of water into the chamber. The cylinder was placed on a platform that was raised or lowered on a pole to give varying levels of hydrostatic pressure in the chamber.

Before each run the veins of the hand were emptied by el- evating the pressure in the chamber. Thereafter, the cuff on the forearm was abruptly filled with air and maintained at suprasystolic levels for 10 minutes to induce ischemia (Fig. 2). Flow during postischemic vasodilation was measured after

FIGURE 1. The experimental apparatus.

BLOOD PRESSURE lCKl/63 BLOOD PRESSURE 100168

FIGURE 2. Curves of displacement of fluid from the plethysmograph after abrupt reduction of cuff pressure from occlusive levels. Studies are of the same subject (S.L.) at 2 plethysmograph pressures. The more rapid displacement of fluid with a lower external pressure in the plethysmograph reflected the greater hand flow ac- companying a higher transmural pressure.

VOLUME OF FLUID 15

DISPLACED ml

0

CUFF PRESSURE

mm Hg

TIME IN SECONDS

420 HAND BLOOD FLOW IN HYPERTENSION

abruptly reducing the cuff pressure to between 40 and 60 mm Hg, which was higher than the chamber pressure but lower than the diastolic BP. Under these circumstances, the arterial inflow to the hand resumed in the absence of venous outflow, resulting in an increase in the volume of the hand. This in turn displaced fluid from the chamber. Hand blood flow (ml/s) was measured from the initial slope of the curve of volume dis- placement vs time. Such measurements were based on a line drawn tangent to 3 complexes (excluding the first complex which often included a cuff artifact) as shown in Figure 2.

In a typical study, the subjects underwent 6 to 8 runs sep- arated by intervals of 3 minutes, during which the temperature in the chamber was equilibrated with the constant tempera- ture bath. A series of different levels of chamber pressure (5 to 40 mm Hg) was imposed on the hand by varying the height of the meniscus in the cylinder above the chamber (Fig. 2). Chamber pressure and cuff pressure were recorded contin- uously by Statham gauges (P23Db). Measurements and de- cisions about suitability of runs were made by an observer blinded to the clinical status of the subject.

Blood flow increased in successive postischemic runs and stabilized by the third run. Because of this, the first 2 runs of a series of 6 to 8 runs were discarded and a curve of flow vs transmural pressure was computed for each subject using a least-mean-square plot of the remaining runs. The predicted flow for a transmural pressure of 85 mm Hg was obtained from this curve for each subject.

An additional normal subject underwent studies to deter- mine whether flow during postischemic vasodilation was al- tered by the reflex sympathetic discharge resulting from

60

Hywemic State I/

. 0

01 I I I I 60 I 70 80 90 100 110

Trammural Pressure in mm Hg

Resting state

I Temperature of Contralat 25% I WC 1 37%

FIGURE 3. Effect of reflex vasoconstriction

em II Hand I BC /

1 blood flow in the hand during and in the absence of postischemic hyperemia. Top, during the hyperemic state the relation of blood flow and the transmural pressure was similar when the contralateral hand was at room temperature (solid circles) and when it was immersed in ice water (open circles). Bottom, in the absence of postischemic hyperemia, blood flow decreased strikingly in response to immersing the contralateral hand in ice water despite higher systemic blood pressures.

placing the contralateral hand in ice water for up to 3 minutes. The findings are shown in Figure 3. Flow during the hyperemic state was unchanged when the contralateral hand was placed in ice water and followed the linear relation of flow vs transmural pressure observed when the contralateral hand was at 25°C. In the absence of hyperemia, flow decreased markedly after immersion of the contralateral hand in ice water, despite increased BP.

Theoretical considerations and data analysis: The he- modynamics of blood vessels subjected to external pressure is governed by the Starling resistor, or waterfall, phenomenon4 provided that the external pressure (PO) lies between inflow pressure (P) and the outflow pressure (PO). In this model, flow (Q) is independent of the exit pressure. Flow is then governed by the driving pressure, which in this case is equal to the transmural pressure, i.e., inflow pressure (P) minus external pressure (PO) and is not influenced by variations of Pc.4~5

In this study, the mean value of the lumen pressure (P) at the arterial end was determined using a sphygomanometer on the opposite arm and calculated as P = diastolic blood pressure + l/3 pulse pressure. With the hand in the plethys- mograph chamber, PO was considered to be the chamber pressure as measured by a Statham gauge. PO, the pressure in the veins outside of the chamber, was not measured; however, it was assumed to be low in the initial filling phase.

The transmural pressure at the arterial end was calculated as?

TMP=P-Pg.

The resistance (R) of the vascular bed within the chamber is given as:

R-P-P0 TMP N-N-. ($ b!

In Figure 4, flow is plotted against transmural pressure. The slope of this curve gives the reciprocal of resistance (i.e., conductance) of the vascular bed of the hand.

Results Figure 4 is a plot of hand blood flow vs transmural

pressure of individual runs of normal subjects (Group IV) and patients with sustained hypertension (Group I). The regression lines for each group differed signifi- cantly in position (p < O.OOl), but not in slope (p <0.2). The flow increased with an increase in transmural pressure in both groups but at a given value of transmural pressure, flow was higher in the normal subjects.

Figure 5 shows the hand blood flow at 85 mm Hg transmural pressure for each subject. The mean and standard error of the mean for each group are also shown. Flow in this case is inversely proportional to vascular resistance because the driving pressure, equal to transmural pressure, is constant. Mean flow of nor- mal subjects differed significantly from mean flow of each of the hypertensive groups (p <0.05) when ana- lyzed by analysis of variance. Analysis by a nonpara- metric method (Kruskal-Wallis statistic with Miller approximation) similarly showed a significant differ- ence between normal subjects and patients with sus- tained or intermittant hypertension (p <0.05), but significance was not achieved for the comparison of normal subjects and treated hypertensive patients.

Values of computed flow for individual untreated sustained hypertensive patients did not overlap values

February I, 1985 THE AMERICAN JOURNAL OF CARDIOLOGY Volume 55 421

FIGURE 4. Relation of hand blood flow g and transmural pressure during all = studies of untreated sustained hyper- z tensives and normal subjects. Re- 3 gression lines are plotted for the 2 f3 groups. u. 30

20

10

4 Sustained Hypertensives

l Normals

/ a

l *

50 60 70 I I I I / I I I I

80 90 100 110 120 130 140

for normal subjects, whereas half of treated hyperten- sive patients resembled normal subjects. Analysis by Fisher’s exact test showed significant differences be- tween the treated and sustained hypertensive groups (p <0.03). There was no obvious relation between the types of drugs received (diuretics, propranolol, methyldopa or hydralazine) and failure of flow to reach control levels.

The only intermittent hypertensive patient with high hand blood flow in Figure 5 clinically appeared to have a hyperdynamic circulation with tachycardia.

Discussion The present study shows that hand blood flow and

vascular conductance of hypertensive subjects during postischemic vasodilation was less than that of normal subjects. Similar results have been obtained by others in the forearm133 and hand.7

Based on observations of others using infusions of norepinephrinerJ or phenylephrine* or using sym-

FIGURE 5. Computed hand blood flow at a transmural pressure of 85 mm Hg for each subject in the 4 groups. Mean flow and standard error of the mean for each group are also plotted.

. T

TRANSMURAL PRESSURE mm Hg

.

z E

1 .

6

.

0 I I I J Sustained Intermittent Treated Normals

Hypertensives Hypertensives Hypertensives

422 HAND BLOOD FLOW IN HYPERTENSION

pathectomized subjects,g it is reasonable to assume that the postischemic blood.flow during maximal vasodila- tion is relatively free from the influence of sympathetic nervous tone. In the present study we could not evoke reflex vasoconstrictive responses to a cold stimulus during postischemic vasodilation, whereas pronounced responses occurred during control observations in the same subject (Fig. 3).

Assuming an absence of sympathetic neurohumoral or other vasoconstrictor influences during postischemic vasodilation, the caliber of the blood vessel would be governed by the transmural pressure of the vessel and structural changes in the vessel wall. Since comparisons between various groups were made at comparable values of transmural pressure,6 the results of the present study, i.e., a decrease in hand blood flow and an increase in vascular resistance in hypertensive subjects (Fig. 4 and 5), suggest structural changes in the small blood vessels of the hand. Our findings in the treated hypertensive patients suggest that such structural changes may be reversed by prolonged therapy. Similar results were obtained by Sivertsson and Hanson10 in patients restudied after 5 years of therapy.

The method was sensitive enough to pick up early changes in blood vessels of intermittent hypertensives. Similarly, Takeshita et a18J1 showed a lower vasodilator capacity of forearm resistance vessels in borderline hypertensive patients and normotensive subjects with a family history of hypertension. Thus, this simple noninvasive method may be useful in following the

adaptation of blood vessels in the early states of hy- pert.ension and during various therapies.

Acknowledgmenti We thank Joan Folio and Gary Fried- lander for technical assistance and Dr. Harry Keiser for his support throughout this research. Thanks are also due to Claudette Williams for typing the manuscript.

1.

2.

References

Folkow 8, Grimby G, Thulesius 0. Adaptive structural changes in the vascular walls in hypertension and their relation to the control of peripheral resistance. Acta Physiol Stand 1958;44:255-272. Folkow 6, Hallback M, Kundgren Y, Sivertsson R, Weiss L. Importance of adaptive changes in vascular design for establishment of primary hy- pertension studied @ man and in spontaneously hypertensive rats. Circ Res .__^ __ ._ 1Yrs;sn:suppl 1:1-2-l-13.

3. Conway J. A vascular abnormality in arterial hypertension: a study of blood flow in the forearm. Circulation 1963;27:520-529.

4. Fry DL, Thomas LJ, Greenfield JC. Flow in collapsible tubes. In: Pate1 DJ, Vaishnav RN, eds. Basic Hemodynamics and its Role in Disease Processes. Baltimore: University Park Press, 1980: 407-423.

5. Button AC, Yamada S. Relation between blood pressure and flow in human forearm. J. Appl. Physiol 1951;4:329-339.

6. Pate1 DJ, Mallos AJ, Defreitas FM. Importance of transmural pressure and lung volume in evaluating drug effect on pulmonary vascular tone. Circ Res 1961;9:1217-1222.

7. Silver&on R. The hemodynamic importance of structural vascular changes in essential hypertension. Acta Physiol Stand 197O;Suppl. 343, 6-13.

6. Takeshita A. lmaizumi T. Ashikara T. Yamamoto K. Hoka S. Nakamura N. Limited maximal vasodilator capac’lty of forearm resistance vessels in normotensive young men with a familial predisposition to hypertension. Circ Res 1982:50:671-677. Eichna LW, Wilkins RW. Blood flow to the forearm and calf. II. Reactive hypermia: factors influencing the blood flow during the vasodilatation fol- lowing ischemia. Bull J Hopkins Hosp 1941;68:450-476. Sivertsson R, Hanson L. Effects of blood pressure reduction on the structural vascular abnormality in skin and muscle vascular beds in human essential hypertension. Clin Sci Mol Med 1976;51:77s-79s Takeshita A, Mark AL. Decreased vasodilator capacity of forearm resis- ;;F vessels in borderline hypertension. Hypertension 1980;2:610-

9.

10.

11.