Cardiocirculatory responses to muscular exercise in congestive heart failure

SPECIAL

ARTICLE

Cardiocirculatory Responses to Muscular in Congestive Heart Failure Dean T. Mason, Robert Zelis, John Longhurst,

T

HE PURPOSE of this review is to delineate current knowledge of the cardiocirculatory adjustments that accompany physical exercise in clinical congestive heart failure. To provide the necessarybackground information for understanding such a discussion,it is first important to concisely consider the cardiac and peripheral circulatory responsesoperative in normal individuals during exertion. Thus, the first section describes the normal responsesof the heart to muscular exercise and those of the systemic vasculature to physicial stress. Then, proceeding from this fundamental information, the principal portion of this review is focused on the seriesof systematic studies,largely carried out in our laboratories, to elucidate the cardiocirculatory mechanismsin the congestiveheart failure state. NORMAL CARDIAC RESPONSES TO DYNAMIC EXERCISE

The cardiac responsesin normal subjectsto exercise are now acknowledgedto involve interactions of alterations in heart rate, contractility, preload, and impedance. The relative roles of eachof these factors in regulating cardiac output are, to a great extent, dependent upon the conditions comprising the manner of physical exertion. It is the current belief that rise in cardiac output occurring during moderate exercise in the supine position results principally from an increasein heart rate.’ Conversely, improvement of cardiac output during heavy exertion in the erect posture is accompanied by marked elevation of stroke volume.’ In addition, myocardial contractility is increased and From

the Section of Cardiovascular Medicine, University at Davis, School of Medicine, Davis. Calif., and Sacramento Medical Center, Sacramento, Calif. Supported in part by Research Program Project Grant HL-14780 from the National Heart, Lung and Blood Institute, National Institutes of Health, Bethesda, Md. and research grants from California Chapters of the American Heart Association, Dallas, Texas. Reprint requests should be addressed to Dean T. Mason, M.D., Professor and Chief Department of Cardiovascular Medicine. University of California, School of Medicine, Davis, Calif: 95616. 019 77 by Grune & Stratton, If7c.

of California

Progress

in Cardiovascular

Diseases,

Vol.

XIX,

No.

6 (May/June),

Exercise

and Garrett

Lee

aortic impedance is reduced, thereby facilitating ventricular emptying under all condition:s of dynamic exercise. The increasein heart rate is primarily due to adrenergic stimulation, while parasympathetic withdrawal is also influential. .Augmentation of the sympathetic nervous system accounts for the increasein inotropism, while the elevation of stroke volume is attributable to operation of the Frank-Starling mechanismin consort with the increasedcontractile state. The reduction in ventricular impedanceis due to decreasein total peripheralvascular resistance. While the normal heart usually decreasesin size during exertion, the Frank-Starling mechanismis still operative but is masked by the opposing effects on cardiac diameter of tachycardia, elevated myocardial contractility, and lowered impedanceto ventricular ejection.3 Thus, the normal responseof the heart to exercise constitutes the integrated effects of all four of the aforementioned determinants of cardiac output: heart rate, contractility, preload, and impedance. Although mild exertion can be accomplishedby the action of only one or two of these variables, the comsiderable rise in cardiac output required during maximal levels of muscular activity necessitatesthe operation of all four of these determinants of ventricular performance. NORMAL PERIPHERAL CIRCULATORY RESPONSES TO DYNAMIC EXERCISE Systemic Arterial

System

Although the many factors that govern blood flow in the regionalcirculations are not necessarily similar in quality or degree,certain physical principles apply to each of the vascular beds4 Thus, blood flow to an organ is determined by the ratio of the driving pressure(the difference between the arterial and venous pressures)to the resistanceto flow offered by the vessels.In general, regional flow is regulated by alterations in vasomotor tone, which in turn are controlled by intrinsic humoral and neural influences. It is important to remember that calculated vascular resistancedoesnot necessarily equate with vasomotor tone, but rather is a 1977

475

476

complex function of the cross-sectional area of the arteriolar bed. Furthermore, changes in arteriolar resistance may be either active or passive. The fact that resistance to blood flow is related critically to the radius of the arterioles is expressed in the Poiseuille relationship, which indicates that blood flow is proportional to the fourth power of the radius. Thus, relatively large adjustments in vascular resistance may be brought about by small changes in the radius of the arterioles. Since the arterioles in the peripheral circulation are connected in parallel rather than in series, the total resistance in a single organ is approximately equal to the resistance in a single vessel divided by the number of vessels in parallel. In view of the fact that blood vessels are distensible and not rigid tubes, increasing the transmural pressure across the arteriolar wall will reduce the resistance opposing flow. The distending force pushing the arteriole outward is balanced by the restraining force of tension developed within the vessel wall. Expressed in terms of the Laplace equation, this tension is proportional to the product of the transmural pressure and the vessel radius. At very low intraluminal pressures, usually about 20 mm Hg for arterioles, the elastic and muscular forces in the vessel wall exceed the distending force so that the vessel collapses and flow ceases. The local control of vasomotor tone of the arterioles is achieved principally by vasodilation produced by metabolic products and anoxia. The cerebral circulation is particularly sensitive to changes in carbon dioxide tension, while the coronary and skeletal muscle beds adjust to alterations in oxygen tension of the blood. It is of interest that the pulmonary arterioles respond in an opposite manner to changes in these blood gases. The resistance of the arterioles is also regulated by sympathetic vasoconstrictor fibers. Vasoconstriction is produced by augmentation of reflex sympathetic nervous activity, and vasodilation occurs as a result of increased local metabolic vasodilator. influences and decreased sympathetic vasoconstrictor impulses. It should be remembered that in the peripheral vascular beds, receptors for the adrenergic nervous system situated at the effector cell are separated into two types: activation of the alpha-adrenergic receptors results in arteriolar constriction, while activation of the beta-adrenergic receptors induces arteriolar dilatation. In the regional circulations, these adren-

MASON

ET

AL.

ergic receptors are not equally partitioned, either as to receptor type or as to absolute number of receptors. Thus, the arterioles of the coronary and cerebral beds appear to contain relatively few adrenergic receptors, while the vascular beds in most other areas possess abundant receptors. Importantly, in the arteriolar bed of skeletal muscle, humorally transported norepinephrine acts on both alpha- and beta-receptors, while neuronally released norepinephrine stimulates only alphareceptors. Although certain regional circulations exhibit little reflex activity, the afferent pathways for the reflex control of arteriolar tone are situated in the carotid and aortic arch baroreceptors5 and chemoreceptors, the ventricular chambers, in the low-pressure areas of the intrathoracic vascular bed, and in somatic nerve fibers in exercising skeletal muscle. Resting Regional Blood Flow The distribution of the cardiac output in the major organs of the body in normal subjects is shown in Fig. 1.6 The largest of these vascular beds is the splanchnic bed. By infusing bromsulphthalein and applying the Fick principle, it has been shown that hepatic blood flow approximates 25% of the total cardiac output in fasting subjects at rest, while the oxygen uptake of this region makes up a similar fraction of the total oxygen consumption of the body. This proportionally equal uptake of oxygen to blood flow is characteristic of the splanchnic circulation; all other organs exhibit a disproportionate relation between oxygen consumption and regional flow. The second largest of the peripheral circulations is the renal circulation, which receives approximately 20% of the cardiac output. The study of renal blood flow has been based on the renal clearance of a substance, such as sodium paraaminohippurate, removed in one passage through the kidney. Since the oxygen saturation of blood from the renal vein is only slightly less than that of arterial blood, the oxygen consumption of the kidney is small compared to its blood flow. The cerebral circulation has been investigated by measuring the rate of uptake by the brain of an inert foreign gas, such as nitrous oxide. From these studies, it has been determined that the brain receives about 12% of the cardiac output but consumes a greater proportion of oxygen (20% of the total oxygen consumption of the body). Coronary

EXERCISE

IN

HEART

FAILURE

477

normal

subject

Cl = 3.0 IfminIm

skeletal

Cl = 6.0 Vminlm’

muscle

splanchmc El pE!J

renal cerebral

heart

Fig. 1. Regional distribution of blood flow at rest and during exercise in normal subjects and in patients with heart failure. Cl, cardiac index. (Reproduced by permission.16)

Cl = 1.5 Ilminlm

rest

blood flow has been measured by a similar technique, utilizing the myocardial uptake of inhaled nitrous oxide or injected radioactive krypton8’ Although the heart is supplied with only 4% of total body flow, it consumes nearly three times this percentage of the oxygen utilized by the body. Thus, the coronary arteriovenous oxygen difference is greater than that occurring in any vascular bed, and coronary sinus blood normally contains less oxygen than any other venous effluent. Knowledge of the blood flow to skeletal muscle has been obtained from studies on the forearm and leg employing the plethysmographic technique.7 In this method, obstruction to venous out-

failure

Cl = 2.0 Ilminlm’

exercise

flow from the limb is produced suddenly, and the change in volume of the extremity, which is proportional to arterial inflow, is measured. Since epinephrine iontophoresis can eliminate the cutaneous circulation temporarily, the amount of blood flow partitioned between muscle and skin can be calculated. Further, the determination of deep and superficial venous oxygen content of the limbs also has been used to follow relative changes in muscle and skin flow. Skeletal muscle blood flow is approximately 20% of the cardiac output, while these tissues utilize about 30% of the total oxygen uptake of the body. Thus, the musculature consumes a greater share of oxygen than any other organ system in the body and, like the

478

cardiac and cerebral circulations, extracts a large amount of oxygen relative to blood flow. Since the primary function of the cutaneous circulation is the regulation of body temperature, the rate of blood flow to the skin is very labile. However, in terms of percentage for the body as a whole, this regional bed receives a flow that is nearly five times that of its oxygen consumption. Thus, the low metabolic requirements relative to blood flow of the skin are similar to those of the kidney, and it is these circulations that are constricted selectively when the cardiac output falls.

Regional Blood Flow During Exercise The distribution of the cardiac output in normal subjects during muscular exercise has received considerable attention recently (Fig. 1).6 It is now recognized that the augmented blood flow to exercising muscles is accomplished not only by an increase in the total cardiac output, but also by redistribution of blood flow. With moderate to severe exercise, blood flow to active skeletal muscles rises, accompanied by an elevation of coronary blood flow, while splanchnic and renal flow remain at precontrol levels despite increased vasoconstriction in these visceral beds.8 Blood flow to nonexercising skeletal muscles is diminished during moderate to strenuous activity.’ The cerebral circulation is maintained during moderate activity, but when maximal exercise is performed, there is a tendency for this flow to become slightly reduced as a result of the fall in arterial carbon dioxide tension that accompanies hyperventilation. The circulation in the skin varies with the intensity and duration of exercise.’ At the onset of exercise, flow declines, but in order to eliminate heat, the cutaneous flow rises as activity continues. However, with maximal exercise, this augmented flow to the skin is often delayed until after the exercise is terminated.’ Thus, vasodilation occurs in the arteriolar beds of exercising skeletal muscle and the heart, while vasoconstriction takes place in the gut and kidney and, to a lesser extent, in the skin and resting muscle. The mechanisms by which the redistribution of regional flow is accomplished are not completely understood, but it is not unreasonable to assume that, with the marked increase in cardiac output occurring during severe exertion, local flow in exercising muscles is augmented in response to the accumulation of vasodilator metabolites, perfusion is maintained in certain

MASON

ET

AL.

visceral organs despite vasoconstriction, and blood flow to some areas is reduced with the overall result being reduction of total peripheral vascular resistance accompanied by moderate elevation of blood pressure. This view implies that the generalized reflex sympathetic discharge occurs in all of the regional circulations, but in certain organs, this action is overridden by vasodilator influences.

Systemic Venous System Although it has been appreciated for many years that the basic function of the venous system is the return of blood to the heart, only recently have satisfactory methods become available for the assessment of changes in tone of the capacitance vessels. lo These methods for the study of venous tone or distensibility are based on establishing the pressure-volume characteristics of the capacitance bed in the forearm, calf, or hand. It is now established that the veins participate actively, through reflexes mediated by the sympathetic nervous system, in maintaining normal circulatory function and can constrict in response to physiologic stimuli to preserve venous pressure and to augment venous return. Thus, in response to emotion, cold environment, norepinephrine, hyperventilation, muscular exercise, and the assumption of upright posture, venoconstriction occurs in certain regional beds, l1 accompanied by a shift of blood in the systemic venous reservoir toward the central circulation. In general, the veins respond to sympathetic and humoral effects less rapidly and quantitatively than do the arterioles. Little information is available concerning the location of the afferent limb of the reflex that is capable of initiating changes in venomotor tone. There is some evidence to suggest that such receptors might exist in low-pressure vascular compartments within the chest. The effects of stimulation of the carotid baroreceptors on venous tone indicate that skeletal muscle is only minimally innervated, while the cutaneous circulation demonstrates marked venoconstriction.12

Additional Compensatory Mechanisms During Exercise Besides the aforementioned cardiocirculatory adjustments during exercise, certain other normal mechanisms help in augmenting oxygen delivery to metabolizing tissues. These additional systems include: (1) increased oxygen extraction from

EXERCISE

IN

HEART

FAILURE ASVI

,479

(+)

- ALVEDP

ALVEDPI-)-

II

I ASVI

(+I

Ix t-j

Fig. 2. The patterns of left ventricular response to supine muscular exercise. Normal left ventricular (LV) function (quadrants I and II, hatched area) includes a variable change in stroke volume, usually an increase, and a fall or no change in LV end-diastolic pressure (LVEDP). Abnormal left ventricular dynamics (quadrant Ill, stippled area) is associated with an increase in stroke volume index (SVI) and an increase in LVEDP. Depressed LV function (quadrant IV, hatched area) is characterized by no change or a fall in SVI and an increase in LVEDP. (Reproduced by permission.‘)

blood perfusing exercising skeletal muscleicausing widening of the arteriovenous oxygen difference; (2) augmentation of the lower extremity, thoracic, and respiratory muscles by pumping increased venous return to the heart; (3) some enhanced oxygen-carrying capacity of the arterial blood resulting from splenic contraction by increasing circulatory red cell mass; and (4) rightward displacement of the hemoglobin-oxygen dissociation curve thereby facilitating unloading of oxygen from red blood cells to the peripheral tissues. CARDIAC RESPONSES TO DYNAMIC EXERCISE IN CONGESTIVE HEART FAILURE

Although studies in which cardiac output is related to total oxygen consumption indicate the degreeto which the output of the heart is capable of satisfying increasedmetabolic demandsof exercise, they are not helpful in differentiating loading from inotropic factors that may limit the cardiac output response.A useful method for qualitative evaluation of these factors is the study of left ventricular performance by determination of the effects of supine exercise on cardiac output, stroke volume or stroke work, and left ventricular end-diastolic pressure.’ The integrity of ventricular performance in responseto the stressof muscular exercise can be analyzed within the

framework of alterations of the position of ventricular function curves.r3 Thus, augmentation of sympathetic activity during exercise normally increases myocardial contractility and thereby alters the shape and position of the ventricular function curve, so that its ascending slope is steeper and elevated compared to the control curve. In contrast, patients with ventricular failure exhibit a depressedand flat curve with little changeor a fall in cardiac performance despitethe development of high levels of end-diastolic pressure. Thus, the failing ventricle doesnot appearto be able to increase its contractile state appropriately to the positive inotropic stimulation of the myocardium accompanying muscularexercise. A simplified and valid meansof assessment of these exercise data hasbeenthe comparisonof the change from resting levels of stroke volume and end-diastolic pressure’*14-16(Fig. 2). In normal subjectsduring exercise, the end-diastolicpressure does not exceed 12 mm Hg and usually falls or risesby no more than 2 mm Hg accompaniedby a rise in stroke volume. In contrast, patients with abnormal myocardial contractility exhibit a fall in stroke volume despiteexcessiveincrementsin lenddiastolic pressurethat reach total levels of greater than 12 mm Hg.’ An intermediate responseIconsisting of a rise in stroke volume accompaniedby excessiveelevation of end-diastolicpressureis consideredto be due to abnormal ventricular compliance or to a lesser impairment of ventricular contractility with increased use of the Frank-Starling principle to achieve an appropriate rise of cardiac output. PERIPHERAL CIRCULATORY RESPONSES TO EXERCISE IN CONGESTIVE HEART FAILURE

Systemic Arterial System It is clear that a state of arteriolar (Fig. 3) and venous constriction is characteristic of human congestive heart failure. 7*16-Z’ This vasoconstriction compensatesfor the reducedperformance of the heart in support of central and peripheral circulatory function. The increasein total systemic vascular resistance provides a peripheral circulatory mechanism by which arterial pressureis maintained in the face of a low cardiac output. The effects of heart failure on the dynamics of the arteriolar bed in the forearm have been char-

MASON

480

FBF = 5.0 m,,,M) g/rain

ET

AL

FBF = 2 0 ml/lOO p/min

m

humoral

norepinephrine

c.

FBF = 40 ml/lOO g/min

FBF = M ml/l00 g/min

normsI subject

00 0 0

c 3

I 40

I 80

I I20

FOREARM VASCULAR RESISTANCE mm.HglML IMII.

Fig. 4. Relation between forearm forearm vascular resistance in normal patients with congestive heart failure. permission.t7)

venous tone and subjects and in (Reproduced by

Fig. 3. Diagram represents the components of vascular resistance in the arterioles of the forearm at rest (A and 6) and during maximal vasodilation (C and D) in a normal patient (A and Cl and in a patient with heart failure (B and D). Components expressed as percentage of radius of arteriolar lumen, value estimated from changes in blood flow induced by certain interventions, and based on the Poiseuille relationship that indicates Row varies directly as the fourth power of the inner radius of the vessel. FBF, forearm blood flow. (Reproduced by permission. le 1

acterized in recent studies employing plethysmographic techniques. In patients exhibiting heart failure, the forearm vascular resistance is significantly greater than in normal subjects [Fig. 4).17 Since the total peripheral vascular resistance is elevated abnormally in heart failure, a state of arteriolar constriction exists in the entire systemic vascular bed, including the forearm. This arteriolar constriction in heart failure is produced mainly by increased sympathetic nervous activityI and by a stiffness component in the arteriolar wall (Figs. 5-Qz1 In contrast to the failing myocardium,22 there are increased labile stores of norepinephrine in the arteriolar beds of skeletal muscle in patients with heart faihrre (Fig. 9).23 The altered mechanical properties of the arterioles in heart failure result from increased sodium and water content in the vessel itself (Figs. 10 and 1 1)24P25 and are responsible, in part, for the decreased arteriolar dilator capacity that has recently been demonstrated in these patients,21 as well as the increased

EXERCISE

IN

HEART

FAILURE

481

DURATION

OF ARTERJAL OCCLUSION * I Mlnures 0 3 Mfiutes 1. 5 hlmurcs

0

MINUTES

AFTER

RELEASE

OF

10 Minutes

ARTERIAL

OCCLUSION -

Fig. 5. Forearm blood flow measured for 1 min prior to and sequentially following release of arterial occlusion of various durations (A). The total integrated area under each reactive hyperemia curve is presented in (6). The peak reactive hyperemia blood flow following release of arterial occlusion of various durations is presented in (C). The data in (6) and (C) were calculated from the curves depicted in (A). RHBF, reactive hyperemia blood flow. (Reproduced by permission.21)

DURATION

NORMAL co&90-91

HEART REACTIVE

‘PEAK RHBF 45 i-ml / n-l” / IOOgr

n

ACTIVE

RHBF m m / IOOgm

-FOREARM I

D

EXERCISE

I 9 8 loLL /EXERCISE

1

HYPEREMIA

c

#FOREARM

(mmulcs)

( 5 m,n )

PEAK 200ml/

Z30r, 8 d m-

OCCLUSION

‘9

L!

I SCHEMIA

ARTERIAL

FAILURE F w X06-89-9

HYPEREMIA

OF

,

Iv,,

MINUTES

,y-y-y-y

I

,

Fig. 6. Forearm blood flow measured in a normal individual (A, C, El and a patient with congestive heart failure (6, D, F). following relief of 5 min of ischemia (reactive hyperemia A, B); following 30 set of forearm dynamic exercise (active hyperemia C, D): and during direct heating of the forearm with a heat lamp (E, F). Arrows indicate release of arterial occlusion (A, B) and completion of forearm exercise (C, D). RHBF, reactive hyperemia blood flow. (Reproduced by permission.21 )

MASON

i:F O MINUTES

OF

c.01<01 L-pe.01

ARTERIAL

I 35-OCCLUSION

tissue pressure in heart failure that offers increased resistance to flow through the capillary bed (Fig. 12)? Resting Regional Blood Flow in Congestive Heart Failure With failure of the heart as a pump, there is a decrease in blood flow to most regions of the body. However, since a uniform reduction of flow to all areas might result in a critical decline of tissue oxygen tension in some organs, in severe heart Cl-IF NERVE

,.A.

BLOCK

IO

AL.

Fig. 7. Total reactive hyperemia blood flow (A) and peak reactive hyperemia blood flow (6). Following release of 1, 3, 5, and 10 min of arterial occlusion in normal subjects (solid circles, solid lines) and patients with congestive heart failure (CHF; open circles, dashed lines). Numbars in parentheses refer to the number of subjects studied. Brackets indicate SEM. Total reactive hyperemia blood flow was calcutated as described in Fig. 4. (Reproduced by permission.21 j

failure this reduction of flow is not uniform and redistribution of flow occurs (Fig. l).” In advanced heart failure, the rate of blood flow to the renal and cutaneous circulations is reduced disproportionately to that of other areas. At the same time, coronary blood flow remains normal at all stages of heart failure in the absence of coronary disease, and blood flow to skeletal muscle tends to be preserved except with advanced heart failure. The blood flow to the splanchnic and cerebral circulations is reduced in proportion to CHF PHENTOLAMINE

IV

CALF

FOREARM

20-

C.01

ET

NORMAL NOREPINEPHRINE FOREARM

A

5-

MINUTES

Fig. 8. (A) Peak reactive hyperemia blood flow (open circles) deep nerve blockade of the forearm blood flow in the calf of patients with congestive trol; shaded bar, after intraarterial phentolamine. ischemia in normal subjects before (solid circles) (Reproduced by permission.2l j

OF

ARTERIAL

OCCLUSION

after various periods of forearm ischemia before (solid circles) and after in patients with congestive heart failure. iB) Peak reactive hyperemia heart failure after release of 5 min of arterial occlusion. White bar, con(C) Peak reactive hyperemia blood flow after various periods of forearm and during (open circles) intravenous (i.v.) norepinephrine infusion.

EXERCISE

IN HEART

FAILURE

483

Fig. 9. Norepinephrine dose-response curves in patients without heart failure (class I) and patients with heart failure (class IV). The average peak values of vascular resistance (2 SEM) elicited by intraarterial injections of graded doses of norepinephrine are shown. The probability values indicate the significance of the difference of the responses between the two groups. The mean maximal resistance in response to tyramine (open circles and vertical broken lines) is positioned on the appropriate curve, providing an expression of the tyramine response in terms of norepinephrine-equivalent dose. (Reproduced by permission.23)

f I” k looE$j -i

I I

p<.o1

I basal

I

I/ ,r

the fall in cardiac output. It is suggested that vasoconstriction occurs in the kidneys and skin, and thus blood is diverted to the heart and skeletal muscle that have high metabolic requirementsrelative to flow. As a consequenceof these regional adjustmentsin heart failure, there is deterioration of renal function and impairment of the dissipation of heat generated by the metabolic processeswithin the body. Perhaps it is this latter disorder that underlies the heat intolerance and body temperature elevations experienced by some patients suffering from congestive heart failure (Fig. 13).9

I I I I

< 01

< .Ol

:: .Ol

< .Ol

I

I

I 1 ,

I

,

025

0.5

IT-

I I

0.125 0.0625 norepinephrine dose. 189/m BSA

< .I)5

RegionalBlood Flow During Dynamic Exercise in CongestiveHeart Failure In patients exhibiting heart failure, the cardiac output is relatively fixed during exercise and fails to respond normally to the increased oxygen

FEMORAL ARTERY

I 3

, DURATION

aI

CONTROL

CHF

CONTROL

OF ARTERlAL

5 OCCLUSION

IO (MINUTESI

CHF

Fig. 10. Artercal sodium content (A) of the aorta and (6) of a tertiary branch of the femoral artery in normal animals compared to animals with congestive heart failure (mean ? SEM). (Reproduced by permission.24j

Fig. 11. The peak reactive hyperemia blood Row t SEM of six subjects as a function of the duration of arterial occlusion, before (control, solid line) and during treatment with fludrocortisone acetate (fludrocortisone, broken line). (Reproduced by permission.25)

484

MASON

CONTROL

NORMAL

IM/LDI

%

NOD. ISJR

- 15

PERCENT CHANGE

-12

P

> .05

/

<.05

Fig. 12. Reactive hyperemia blood flow measured in the vascularly isolated forelimb of a dog before (A) and after venous congestion (B) of the limb at a venous pressure of 70 mm Hg for 4 hr, and following the intravenous infusion of Dextran 60 (Cl. Tracings represent restoration of flow (at the arrow) following 5 min of ischemia produced by termination of flow. Prior to ischemia and at a point where flow is reinstituted, flow was maintained constant. The initial plateau that the perfusion pressure reaches represents the level of vascular resistance in the &hernia dilated forelimb. (Reproduced by permission.%)

requirements of the metabolizing tissues. Certain adjustments, therefore, are necessary in the peripheral vascular beds in order to supply more blood flow to exercising muscle (Fig. 1).6 Present evidence suggests that exercise leads to a less than normal augmentation of total flow to the exer-

AL.

CHF

1

-28

-54

< 05


Fig. 14. Decreases in blood flow in during the first 3 min of supine leg moderate; Str., strenuous; CHF, congestive (Reproduced by permission.9)

After 5 Minutes lschemia

ET

forearm exercise. heart

muscle Mod., failure.

cising limb, but that the flow to skeletal muscle in these areas is increased to a normal extent at the expense of flow to the skin. At the same time, the blood flow to the splanchnic and renal circulations and to resting skeletal muscle is reduced. Coronary blood flow is elevated while cerebral blood flow remains unchanged. In this manner, blood flow is directed from the resting areas of the body to the exercising parts. Thus, vasodilation occurs in exercising skeletal and cardiac muscles, and vasoconstriction takes place in the gut, kidney, skin, and resting muscles where blood flow declines (Figs. I4 and 1 5).9 This redistribution occurs earlier and to a greater extent than that observed in normal subjects performing similar levels of exercise, although these adjustments during mild to moderate exercise in patients who have heart failure probably are comparable to those in normal subjects carrying out strenuous exertion.’ NORMAL

zii

1 Yl‘D

I

1 MOD. 1 SJR. 1

+05r

CHF

I SJR

I

,' =: -cr.5 z; 3 .z -1.0 g 5 -1.5 w z -2.0 I* -2.5 E g, -3.00 P-T PERCENT CHANGE t82

P

< 01

+96

+I08

+4

< 01

c 05

) .5

Fig. 13. Maximum increase in blood flow in forearm skin during 10 min following supine leg exercise. The severity of exercise is shown: Mod., moderate; Str., strenuous; CHF, congestive heart failure. (Reproduced by permission.g)

PERCENT CHANGE

-49

-47

- 74

-98

P

< 05

( 01

< 05

cot

Fig. 15. during the exercise is congestive

Decreases in blood flow in forearm skin first 3 min of supine exercise. The severity of shown: Mod., moderate; Str., strenuous; CHF, heart failure. (Reproduced by permission.9)

EXERCISE

IN

HEART

FAILURE

LEVEL

OF EXERCISE

Fig. 16. Forearm blood rest (control) and during exercise in normal subjects patients with congestive dashed line). (Reproduced

485

Imm Hgl

flow ( fSEM) measured at three levels of rhythmic grip (solid circles, solid line) and heart failure (open circles, by permission.%)

Skeletal Muscle Metabolic Alterations in Congesh’ve Failure One of the problems in evaluating regional blood flow during forearm dynamic exercise is the marked changes in forearm geometry that make it difficult to evaluate flow by a plethysmographic technique. However, with slow rhythmic grip exercise, forearm blood flow can be measured during the last half of the relaxation period by the plethysmographic technique. It has been determined clinically that blood flow measured in this manner correlates well with that simultaneously measured with a flowmeter on the brachial artery.27 Therefore, it was possible to compare the blood flow response of the forearm during dynamic exercise at three different intensities in normal subjects and in patients with congestive heart failure. In these studies, it was observed that heart failure patients responded to this exerciseinduced metabolic vasodilator stimulus less well than normal subjects (Fig. 16).28 Blood flow was lower in the heart failure patients at rest and at each level of increasing intensity of exercise. When the arterial-venous oxygen difference across the forearm was evaluated, it was noted that the heart failure patients extracted more oxygen at rest and at each level of exercise. Whereas the oxygen extraction at rest was sufficient to maintain forearm oxygen consumption within normal limits, it was not sufficient to provide enough oxygen to the forearms of the heart failure patients during exercise (Fig. 17). 28 This was true despite the fact

Fig. 17. Forearm oxygen consumption (k SEMI measured at rest (control) and during three levels of intermittent grip exercise in normal subjects (solid circles, solid line) and in patients with congestive heart failure lopen circles, dashed line). Forearm oxygen consumption was calculated as the product of forearm blood flow and the forearm arterial venous oxygen difference. (Reproduced by permission.281

that the greater tissue acidosis and increased red blood cell 2, 3-diphosphoglycerate content in the heart failure state tended to facilitate oxygen transport from the capillaries to the tissues. Thus, oxygen consumption of the dynamically exercising forearm in heart failure did not increase normally, and it appeared that there was a metabolic shift to anaerobiosis with consequent lactic acidosis during exercise in heart failure patients. Although it has been suggested that the principal reason in heart failure for the shift to anaerobiosis during exercise is a failure of skeletal muscle blood flow to rise appropriately, it is also possible that local structural abnormalities might exist that would retard diffusion of oxygen from the capillaries to the skeletal muscle fibers. Thus, !interstitial edema might be an important factor in retarding diffusion. More recently, an increase in skeletal muscle capillary basement membrane thickness has been observed in heart failure.2”

Pathophysiologic Role of Systemic Arteriolar Constriction in Heart Failure It appears that, in the peripheral circulation of symptomatic patients with congestive heart failure, there is a greater level of stimulation of alpha-adrenergic receptors than in normal subjects (Fig. 18). This may play a more major role than the heart failure stiffness component in reducing limb blood flow at rest. However, with a maximal

486

MASON

r

NORMAL

CHF

RESTING ARTERIOLE

DILATED ARTERIOLE

Fig. 18. Diagrammatic representation of the components determining arteriolar tone at rest and with metabolic vasodilatation (horizontal rows) in normal subjects and in patients with congestive heart failure (CHF) (vertical columns). The internal cross-sectional areas have been drawn to scale on the basis of blood flow data. At rest there is increased sympathetic arteriolar tone in symptomatic patients with congestive heart failure. With metabolic vasodilatation, sympatholysis occurs, and the stiffness component in congestive heart failure becomes important in limiting excessive regional blood flow. (Reproduced by permission.18)

metabolic vasodilator exercise stimulus, sympathetic tone plays less of a role in determining the regional resistance of the skeletal muscle vascular bed. However, the increased vascular sodium content and tissue edema would tend to limit vasodilation. This is a protective effect that appears to be an important factor in maintaining blood pressure during brief bursts of exercise. In such a situation, if the muscle vessels dilated normally and the cardiac output of the heart failure patient were inadequate, then a decline in blood pressure might occur during exercise rather than the increase in blood pressure observed in normal individuals. Indeed, there is failure of blood pressure to rise normally in patients with congestive heart failure. Furthermore, following diuresis, substantial fall in blood pressure may occur during exercise. This exercise hypotension is partially the result of loss of the salt-induced vascular stiffness as well as a reduction in circulating blood volume. This stiffness abnormality inherent in the resistance vessels in heart failure appears to be of possible physiologic importance from another viewpoint, During exercise, a generalized sympathetically induced vasoconstriction occurs in the systemic arterioles. However, in active skeletal muscle, vasodilator metabolites accumulate and override the augmented sympathetic discharge to

ET

AL.

these vessels, resulting in pronounced local vasodilatation. Furthermore, it has been found in normal subjects that this metabolic vasodilatation and enhanced blood flow is excessive in relation to the true metabolic requirements of the exercising limbs.30 The fact that metabolic vasodilatation is excessive in active skeletal muscle appears to be fortunate for patients with heart failure, who fortuitously appear to conserve the portion of total blood flow partitioned to the exercising limbs. Thus, their resistance vessels in the active limbs do not dilate to a normal extent during exercise and, therefore, for the same amount of external work performed, the patient with heart failure provides less blood flow to the exercising extremities, resulting in an increased extraction of oxygen and a widened arterio-venous oxygen difference in the regional bed. In addition, since the cardiac output cannot be augmented sufficiently during exercise, arterial perfusion pressure is maintained, because the arteriolar stiffness factor prevents total peripheral vascular resistance from decreasing inordinately. Mecharlism of Increased Sympathetic Activity in Congestive Failure With systemic dynamic exercise, the mechanism of increased vascular tone in regions other than exercising muscle is an important question that has not been completely answered. It has been suggested that somatic afferent-nerve fibers from exercising skeletal muscle probably play an important role in this reflex response to exercise. Recent studies appear to have delineated the manner of stimulation of the afferent limb of this reflex arc. When the hindquarters of a dog are perfused with an extracorporeal pump-oxygenator circuit, and the perfusion medium is hypoxemic, a significant tachycardia occurs and an increase in total systemic and regional limb, splanchnic, and renal vascular resistance is observed. It was noted that hypercapnia and acidosis did not stimulate this reflex arc; however, 2,4-dinitrophenol and sodium cyanide did induce the reflex.31 Since reduction in hindquarter oxygen consumption was seen with hypoxia, and an increased oxygen consumption was observed with dinitrophenol, it was initially thought that tissue oxygen content might be the major determinant for activation of the reflex. The finding that sodium cyanide also stimulated the reflex suggests that local hypoxemia was not the

EXERCISE

IN

HEART

FAILURE

primary stimulus. When cyanide was introduced into the extracorporeal circuit, tissue oxygen content of the hindquarters was increased. However, the availability of oxygen to the respiratory chain was blocked, and intracellular oxygen utilization was depressed. The common consequence of all three stimuli (hypoxia, dinitrophenol, and cyanide) is reduction in tissue high-energy phosphate stores and an accumulation of the breakdown products of adenosine triphosphate. It is possible that these metabolic products may play an important role in stimulating the afferent limb of this reflex. The possibility of defective adrenergic control of heart rate in heart failure patients has been suggested. Thus, a defect exists in the adrenergic component of baroreceptor-mediated reflex heart rate control in congestive failure.32 Importantly, the possibility of reduced responsiveness of myocardial beta-receptors is not a factor.32 Heart failure is associated with abnormal parasympathetic function as well. Thus, the degree of vagal restraint of sino-atrial function is diminished in heart failure.33 This aberrancy in autonomic control of heart rate in congestive failure is of considerable importance during exercise in such patients whose ability to elevate stroke volume is impaired together with a defective mechanism for providing tachycardia to raise cardiac output with exertion.

Systemic Venous System in Congestive Failure Although it has been appreciated for many years that the venous system participates actively in alterations of venous return to the heart, recent evidence suggests that the capacitance bed plays somewhat less of a responsive role in the regulation of circulatory function. Thus, the sympathetic nervous system does not appear to innervate the veins in skeletal muscle, and moderate stimulation of the carotid baroreceptors does not produce venoconstriction in the skin. Nevertheless, the overall capacitance bed is relatively indistensible in heart failure (Fig. 4) and thereby affords another peripheral circulatory mechanism for aid of the failing heart. 34 Thus, adrenergic-induced increase of venous tone is accompanied by a shift of blood in the systemic venous reservoir toward the central circulation, which augments venous return to the heart and assists in utilization of the Frank-Starling mechanism to maintain cardiac output by elevating ventricular diastolic filling. In

487

addition to these alterations in venous tone at rest in heart failure, during exercise there is an excessive rise in venous tone and central venous pressure, mediated by increased sympathetic nervous activity and, thereby, enhanced return of blood to the impaired ventricle. However, in contrast to the arterial system, local factors appear to be the major determinant of limb venous volume. Since limb venous congestion can produce a reduction in measured venous volume, it is possible that edema may be the most important factor in reducing the compliance of the venous system in the limbs of patients with congestive heart failure.35 CONCLUSIONS

In congestive heart failure, patients appear to have a limited ability to dilate their resistancevesselsin skeletal muscle in responseto a metabolic stimulus. This appearsto be the casewhether the metabolic stimulus is ischemia or dynamic or static exercise. The mechanismfor this limited arteriolar capacity is at least twofold: (1) increased sodium content of the vesselsand (2) increasedtissue pressurethat is seenin edematousstates.This can be considereda positive compensatory mechanism, in that it helps to maintain systemicarterial pressureduring exercise when the cardiac output fails to increasenormally. If the resistancevessels were to dilate normally, then in the face of a limited cardiac output exercise syncope would be expected to occur. The price paid by heart failure patients for the maintenance of arterial pressure during exercise is an earlier shift to anaerobic metabolism. Becauseof the limited blood flow to exercising muscle, there is an increased oxygen extraction as the blood traversesthe metabolically active muscular beds. This increased oxygen extraction, however, is not sufficient to ,allow aerobic metabolism to proceed normally. IIence, patients with heart failure are more subje’ct to developing metabolic acidosisduring exercise and at lesserlevelsof exertion than normal subjects. Another abnormality present in the peripheral circulations of patients with congestivehear.t failure is increasedsympathetic alpha-adrenergictone, as well as increased level of circulating catecholamines. This is most true in symptomatic patients at rest and during exertion in most patients with heart failure who are limited in their exercise capacity. During exercise, the sympathetic nervous system appearsto be activated to

488

MASON

a greater extent in patients with congestive heart failure, and, as expected, the circulations with the greatest population of alpha-receptors (the cutaneous, renal, and splanchnic circulations) appear to suffer the most from inadequate blood flow. The failure to dilate the cutaneous circulation in response to thermal stress makes these patients particularly susceptible to cardiac failure in hot and humid environment. The failure to handle even the normal thermogenesis of mild exertion helps to explain the slight temperature elevations commonly seen when patients are hospitalized with acute cardiac failure. In the systemic venous beds of the limbs of heart failure patients, there is increased venous tone. This is minimally related to increased sympathetic efferent impulses to the cutaneous veins and increased circulating catecholamines, but it is primarily explained on the basis of local factors that determine venous tone. The most important

ET

AL.

local factor that appears to reduce the venous volume is edema. Since the venous volume of the limbs is reduced only slightly by increased sympathetic tone in congestive heart failure, it might be expected that drugs, such as morphine, that seem to work by producing a central-nervoussystem-induced sympatholysis would increase venous volume of the limbs only minimally.36 Thus, in acute pulmonary edema, if morphine induces peripheral pooling of blood, it would appear to act by its arteriolar dilator mechanism and the perfusion of previously unperfused vascular beds, leading to a passive filling of the venous system. The circulation in which this is most likely to occur is the splanchnic circulation.37 ACKNOWLEDGMENT The authors gratefully tance of Robert Kleckner

acknowledge the technical and Leslie Silvernail.

assis-

REFERENCES 1. Ross J Jr, Gault JH, Mason D, et al: Left ventricular performance during muscular exercise in patients with and without cardiac dysfunction. Circulation 34: 597, 1966 2. Epstein SE, Robinson BF, Kahler RL, et al: Effects of beta-adrenergic blockade on the cardiac response to maximal and submaximal exercise in man. J Clin Invest 44: 1745, 1965 3. Sonnenblick EH, Braunwald E, Williams JF Jr, et al: Effects of exercise on myocardial force-velocity relations in intact unanesthetized man: Relative roles of changes in heart rate, sympathetic activity, and ventricular dimensions. J Clin Invest 44: 2051, 1965 4. Mason DT: Control of the peripheral circulation in health and disease. Mod Concepts Cardiovasc Dis 36: 25, 1967 5. Beiser GD, Zelis R, Mason DT, et al: The role of skin and muscle resistance vessels in reflexes mediated by the baroreceptor system. J Clin Invest 49: 225-231. 1970 6. Wade OL, Bishop JM: Cardiac Output and Regional Blood Flow. Oxford, Blackwell, 1962 7. Mason DT, Braunwald E: Studies on digitalis X: Effects of ouabain on forearm vascular resistance and venous tone in normal subjects and in patients with heart failure. J Clin Invest 43: 532-543, 1964 8. Vatner SF, Higgins CB, White S, et al: The peripheral vascular response to severe exercise in untethered dogs before and after complete heart block. J Clin Invest 50: 1950, 1971 9. Zelis R, Mason DT, Braunwald E: Partition of blood flow to the cutaneous and muscuiar beds of the forearm at rest and during leg exercise in normal subjects and in patients with heart failure. Circ Res 24: 799, 1969 10. Mason DT, Braunwald E: Effects of guanethidine,

reserpine and methyldopa on reflex venous and arterial constriction in man. J Clin Invest 43: 1449-1463, 1964 11. Braunwald E, Ross J Jr, Kahler RL, et al: Reflex control of the systemic venous bed: Effects on venous tone of vasoactive drugs, and of baroreceptor and chemoreceptor stimulation. Circ Res 12: 539-552, 1963 12. Zelis R, Mason DT: Comparison of reflex reactivity of skin and muscle veins in the human forearm. J Clin Invest 48: 1870-1877, 1969 13. Mason DT, Miller RR, Vismara LA, et al: Clinical assessment of heart disease and ventricular performance by cardiac catheterization, in Mason DT (ed): Congestive Heart Failure. New York, Yorke, 1976, pp 225-272 14. Ross J Jr, Morrow AG, Mason DT, et al: Left ventricular function following replacement of the aortic valve: Hemodynamic responses to muscular exercise. Circulation 33: 507, 1966 15. Mason DT, Ross J Jr, Gault JH, et al: Combined prosthetic replacement of the mitral and aortic valves: Pre- and post-operative hemodynamic studies including left ventricular responses to muscular exercise. Circulation 35 (Suppl I): 15, 1967 16. Mason DT, Spann JF Jr, Zelis R, et al: Alterations of hemodynamics and myocardial mechanics in patients with congestive heart failure: Pathophysiologic mechanisms and assessment of cardiac function and ventricular contractility. Prog Cardiovasc Dis 12: 507-557, 1970 17. Braunwald E, Chidsey CA, Pool PE, et al: Congestive heart failure: Biochemical and physiological considerations. Ann Int Med 64: 904, 1966 18. Mason DT, Zelis R: The function of the arterial and venous beds in congestive heart failure. Heart Bull 17: 109, 1968 19. Zelis R, Mason DT: Compensatory mechanisms in

EXERCISE

IN HEART

FAILURE

congestive heart failure: The role of the peripheral resistance vessels. N Engl J Med 282: 962, 1970 20. Zelis R, Longhurst J, Capone RJ, et al: Peripheral circulatory control mechanisms in congestive heart failure, in Mason DT (ed): Congestive Heart Failure. New York, Yorke, 1976, pp 129-142 21. Zelis R, Mason DT, Braunwald E: A comparison of the effects of vasodilator stimuli on the peripheral resistance vessels in normal subjects and patients with congestive heart failure. J Clin Invest 47: 960, 1968 22. Braunwald E, Chidsey CA, Mason DT, et al: Effects of reserpine and of congestive heart failure on the myocardial norepinephrine concentration in man. Trans Assoc Am Physicians 76: 254-261,1963 23. Kramer RS, Mason DT, Braunwald E: Augmented sympathetic neurotransmitter activity in the peripheral vascular bed of patients with congestive heart failure and cardiac norepinephrine depletion. Circulation 38: 629, 1968 24. Zelis R, Delea CS, Coleman H, et al: Arterial sodium content in experimental congestive heart failure. Circulation 41: 213, 1970 2.5. Zelis R, Mason DT: Diminished forearm arteriolar dilator capacity produced by mineralocorticoid-induced salt retention in man: Implications concerning congestive heart failure and vascular stiffness. Circulation 41: 589, 1970 26. Zelis R, Lee G, Mason DT: The influence of experimental edema on metabolically determined blood flow.,Circ Res 34: 482-490,1974 27. Longhurst J, Capone RJ, Mason DT, et al: Comparison of blood flow measured by plethysmograph and flowmeter during steady state forearm exercise. Circulation 49: 535, 1974 28. Zelis R, Longhurst J, Capone RJ, et al: A comparison of regional blood flow and oxygen utilization during dynamic forearm exercise in normal subjects and

489

patients with congestive heart failure. Circulation 50: 137-143, 1974 29. Longhurst J, Capone R, Amsterdam E, et al: A microcirculatory defect in congestive heart failure -Etiology of depressed oxygen consumption during exercise? Clin Res 20: 208, 1972 30. Blair DA, Glover WE, Roddie IC: The abolition of reactive and postexercise hyperaemia in the forearm by temporary restriction of arterial inflow. J Physiol (L.ond) 148: 648-658, 1959 31. Tabaie H, Mangseth G, Tremain S, et al: Stimulation of somatic afferent receptors by divergent mechanisms: Implications concerning exercise reflexes. Circulation 50 (Suppl III): 208, 1974 32. Goldstein RE, Beiser GD, Stampfer M, et al: Impairment of autonomically mediated heart rate control in patients with cardiac dysfunction. Circ Res 36: 571,1975

33. Eckberg DL, Drabinsky M, Braunwald E: Defective cardiac parasympathetic control in patients with heart disease. N Engl J Med 285: 877, 1971 34. Zelis R, Capone R, Amsterdam EA, et al: The concept of local determinants of venous volume: The risle of nonadrenergic factors in the elevated venous tone of congestive heart failure. J Clin Invest 50: 102A, 1971 35. Zelis R, Mansour E, Capone R, et al: The role of edema in regulating arteriolar and venous tone in congestive heart failure. Clin Res 20: 406, 1972 36. Zelis R, Mansour EJ, Capone RJ, et al: The cardiovascular effects of morphine: The peripheral capacitance and resistance vessels in human subjects. J Clin Invest 54: 1247-1258, 1974 37. Vismara LA, Miller RR, Mason DT: The cardiocirculatory actions of morphine in pulmonary edema: Importance of splanchnic blood flow during systemic vasodilation. Clin Res April 1977