Clonidine in congestive heart failure: A vasodilator with negative inotropic effects

CONGESTIVE HEART FAILURE r

Clonidine in Congestive Heart Failure: A Vasodilator With Negative Inotropic Effects JAMES B. HERMILLER,MD, RAYMOND D. MAGORIEN, MD, MARK E. LEITHE, MD, DONALD V. UNVERFERTH, MD, and CARL V. LEIER, MD

Fourteen patients with moderately severe congestive heart failure (CHF) were given clonidine orally (0.2 and 0.4 mg doses) to determine the hemodynamic effects of a typical centrally acting vasodilator. The 0.2 mg dose significantly reduced mean systemic ( 15 % ) and mean pulmonary artery (20 % ) pressure; the corresponding reductions in vascular resistance were not as great because of a diminished cardiac output. Pulmonary capillary wedge pressure decreased significantly (27 % ). Heart rate decreased 11% and stroke volume remained unchanged. At a higher dose (0.4 mg), clonidine augmented these reductions but increased stroke volume modestly (15 % ). Isovolumic developed pres-

sure/duration of isovolumic contraction and the duration of the preejection period were used as indexes of inotropy. After both doses, isovolumic developed pressure/duration of isovolumic contraction decreased dramatically ( > 3 3 % ) and the preejection period increased substantially ( > 18 % ) (both p <0.05). Compared with currently employed vasodilating agents, the centrally acting agent clonidine appears unique in that the drug-induced systemic and pulmonary arterial vasodilation are not accompanied by a commensurate improvement in ventricular systolic function. This lack of improvement appears to be a result of negative inotropic effects.

Peripherally acting vasodilators are now commonly used in the management of congestive heart failure (CHF). Many classes of agents (adrenergic blockers, nonspecific vasodilators, and angiotensin-converting enzyme inhibitors) have been extensively examined over the past decade. Numerous studies have demonstrated that these agents augment stroke volume and cardiac output, primarily by reducing ventricular afterload. ~,2 In contrast to the peripheral class of compounds, centrally acting vasodilators have received little attention in the setting of low output CHF. One typical centrally acting agent is clonidine. Clonidine is used predominantly in hypertensive patients, in whom it effectively reduces systemic blood pressure, systemic vascular resistance, and heart rate. 3-~ It induces its antihypertensive effect by attenuating sympathetic outflow by way of a complex interaction with the central alpha-adrenergic system.6,7 We postulated that clonidine may be an effective Vasodilator since it does reduce sympathetic activity,

a major cause of the elevated ventricular afterload accompanying CHF. This study was designed to determine the acute hemodynamic properties of oral clonidine in patients with chronic CHF.

Methods P a t i e n t population: Fourteen patients with moderately severe CHF were studied. The mean age was 53 years (range 19 to 73); 2 patients were women. The diagnosis of dilated cardiomyopathy was confirmed by cardiac catheterization performed in each patient within 2 months before the onset of study. Thirteen patients had dilated cardiomyopathy and I patient had severe occlusive atherosclerotic coronary artery disease with ischemic cardiomyopathy. Twelve patients were categorized as functional class III and 2 patients as functional class IV (New York Heart Association). All those studied were receiving a digitalis preparation (daily digoxin dose of 0.25 mg in 6 patients, 0.125 mg in 7 patients, and digitoxin 0.1 mg in 1 patient) and furosemide (daily dose range 40 to 120 rag). Several patients had previously received vasodilators (isosorbide dinitrate 20 mg every 6 hours in 5 patients and hydralazine 75 mg every 8 hours in 3 patients). All vasodilators were discontinued at least 48 hours before study. The digitalis and diuretics were continued throughout the study with afternoon doses to avoid peak effect. Measurements: On the first study day, a triple-lumen flow-directed catheter was introduced percutaneously into the subclavian vein and placed in the pulmonary artery for the measurement of pulmonary artery pressure, pulmonary

From the Division of Cardiology, Ohio State University College of ~,ledicine, Columbus, Ohio. This study was supported in part by the JamesD. Casto Cardiovascular Research Fund and the S. J. Roessler Foundation,Ohio State University,College of Medicine, Columbus,Ohio. NManuscriptreceived September 10, 1982; revised manuscript received OVember30, 1982, accepted December 1, 1982. Address for reprints: Carl V. Leier, MD, 653 Means Hall, 1655 Upham brive, Columbus, Ohio 43210. 791

792

CLONIDINEIN CONGESTIVE HEART FAILURE

SYSTEMIC

PULMONARY

BLOOD PRESSURE 0.4 m(j (n=7)

O. 2 mg (n=lO)

ARTERY

0.2 mg (n=lO)

PRESSURE 0.4 mg(n=7)

60 120

£

rE

E E

E

40 20

80

PULMONARY SYSTEMIC VASCULAR

VASCULAR

RESISTANCE

8O0

RESISTANCE

E 2O00

I

=,

I

iiil

400

" iooo

o

Q

HOURS

HOURS FIGURE 1. Both doses of clonidine significantly lowered systolic, mean, and diastolic systemic blood pressures. Only the 0.4 mg dose reduced

FIGURE 2. Pulmonary arterial pressure decreased significantly after clonidine administration. Pulmonary vascular resistance decreased as

vascular resistance. Data are presented as mean -4- 1 standard deviation. * p <0.05; B = baseline control.

well. Data are presented as mean 4- 1 standard deviation. * p <0.05; B = baseline control.

capillary wedge pressure, and cardiac output (thermodilution technique). The pulmonary artery and capillary wedge pressures were measured with Becton-Dickinson Electrodyne PR-18A pressure amplifiers and ST-149 and WR-4D recording systems. The cardiac outputs were computed with an Instrumentation Laboratory computer unit 601 and recorder unit 602, and were obtained in triplicate (5 measurements were made if variation exceeded 10%). Systemic blood pressure was measured by a cuff and a mercury column sphygmomanometer. Heart rate was obtained from continuous electrocardiographic monitoring. Systolic time intervals, derived from the simultaneous recording of a carotid pulse tracing, a precordial phonocardiogram, and an electrocardiogram, were obtained with an Electronics for Medicine Echo IV unit using specifications previously outlined, s Standard formulas were used to determine vascular resistances. Additional calculated variables included the following: diastolic time = R-R interval minus total electromechanical systole (QS2)9; preejection period index (milliseconds) = preejection period index corrected for heart rateS; and isovolumic developed pressure/duration of isov01umic contraction, 1°-12 where developed pressure during isovolumic contraction = systemic

diastolic blood pressure minus pulmonary capillary wedge pressure, and isovolumic contraction time = preejection period minus electromechanical delay (Q wave of the electrocardiogram to upstroke of left ventricular impulse on the apexcardiogram). The preejection period index and the isovolumic developed pressure/duration of isovolumic contraction were used as indexes of inotropy. 1°-12 Protoeoh The patients were studied over 3 days. Day 1 was utilized for catheter placement and served as an equilibration period. On Day 2, 0.2 mg of clonidine was administered, and on Day 3, the patients received 0.4 rag. Clonidine was administered orally in the postabsorptive state. In 10 patients, hemodynamic variables (cardiac output, heart rate, pulmonary capillary wedge pressure, systemic blood pressure, and pulmonary artery pressure) were obtained before and at hourly (X6) intervals after each clonidine dose. Two of these 10 patients and 4 additional ones also underwent hemodynamic, preejection period index, and is'ovolumic developed pressure/duration of isovolumic contraction measurements at baseline control and 3 hours after clonidine administration. S t a t i s t i c a l a n a l y s i s : The hemodynamic data were compared with control values for each dose with analyses of variance for repeated measures. The 3-hour isovolumic developed pressure/duration of isovolumic contraction and systolic time interval data were compared with control values using the t test for paired data.

O.2mg (n=lO)

0.4 mg (n=7)

20-

f:E

Results

I0"

E

0

I

i

i

HOURS

FIGURE 3, Pulmonary capillary wedge pressure was significantly reduced by clonidine, Data are presented as mean 4- 1 standard deviation, * p <0.05; B = baseline control.

T h e results of the h e m o d y n a m i c studies are presented in Figures 1 through 5. T h e r e was no significant difference between a n y of the baseline control meas u r e m e n t s of the 2 doses. After b o t h doses t h e r e was a s u b s t a n t i a l reduction of systolic, m e a n , a n d diastolic systemic blood pressures (Fig. 1). All 10 p a t i e n t s who participated in the 6-hour protocol received the 0.2 mg dose, while only 7 were given 0.4 mg because hypotension (systolic blood pressure <90 m m Hg) developed in

March 1, 1983 THE AMERICAN JOURNAL OF CARDIOLOGY Volume 51

CARDIAC INDEX

793

DOUBLE PRODUCT

0.2 mg (n=tO)

12

0.2 mg (n-lO)

0.4mg ( n - 7 )

2.4 X

E

._= I0

E

2.0 ..J

E

E E

8

IIl

S'

S T R O K E V O L U M E INDEX

40

20 HEART RATE

I00

80

80

CORONARY PERFUSION

PRESSURE

E 60 E

60

40 HOURS HOURS

FIGURE 4. Clonidine did not alter cardiac index, Stroke volume increased slightly after the 0.4 mg dose, at 4 and 5 hours after drug administration. Both doses significantly lowered heart rate. Data are presented as mean 4- 1 standard deviation. * p <0.05; B = baseline control,

FIGURE 5. Double product, a calculated estimation of myocardial oxygen consumption, decreased substantially. Coronary perfusion pressure (diastolic blood pressure - pulmonary capillary wedge pressure) decreased slightly after 0.2 mg and was not altered by the higher dose. Data are presented as mean 4- 1 standard deviation. *p <0.05; B = baseline control.

3 patients after the lower dose. The decrease in systemic blood pressure was not accompanied by a comparable reduction of systemic vascular resistance. Pulmonary artery pressure decreased after clonidine administration; moreover, pulmonary vascular resistance decreased after both doses (Fig. 2). Pulmonary capillary wedge pressure decreased substantially (Fig. 3). Clonidine also evoked a significant negative chronotropic response; heart rate decreased 11 and 15% after the respective 0.2 and 0.4 mg doses (Fig. 4). Cardiac index was not significantly altered, and Stroke volume increased only modestly 3 and 4 hours after the 0.4 mg dose. Since heart rate and systolic blood pressure both decreased, the double product, an estimate of myocardial oxygen consumption, decreased (Fig. 5). Coronary perfusion pressure (diastolic blood pressure minus pulmonary capillary wedge pressure) was slightly reduced with 0.2 rag, and was not altered by the higher dose. Diastolic perfusion time increased significantly after both doses

TABLE I

(Table I). Isovolumic developed pressure/duration of isovolumic contraction and systolic time intervals were obtained in 6 patients. There was a dramatic reduction in isovolumic developed pressure/duration of isovolumic contraction for each patient after both doses (Fig. 6). Although the isovolumic developed pressure did not change significantly, duration of isovolumic contraction was prolonged substantially (Table I). The preejection period index lengthened significantly in each patient (Fig. 6). The ratio of the preejection period to left yentricular ejection time increased significantly after clonidine administration (Table I); this increase was secondary to a consistent prolongation of the preejection period and some shortening of the left ventricular ejection time (Table I). All patients had somnolence after both doses of clonidine. Four patients noted dry mouths. Three patients has systemic hypotension; 1 of these patients had

Systolic and Diastolic Time Intervals, Isovolumic Developed Pressure (Ap), and Isovolumic Contraction Time (At) Measurements Obtained at Baseline Control and 3 Hours After Clonidine (n = 6)

QS21 (ms) Diastolic time (ms) LVETI (ms) PEP/LVET ~P (mm Hg) t(ms)

Control

0.2 mg

Control

0.4 mg

528 4- 11 332 4- 63

530 4- 14 422* 4- 72

521 4- 18 399 4- 109

525 ± 18 528* 4- 128

368 0.56 63 91

351 0.71" 57 117"

356 4- 35 0.53 4- 0.08 59 4- 6 894- 14

4- 18 4- 0.09 4- 7 4- 14

4- 21 4- 0.16 4- 13 -t-21

348 0.74* 56 128"

4- 32 ~E 0.13 4- 7 4- 18

Values are presented as mean 4- 1 standard deviation. LVETI = left ventricular ejection time index; PEP/LVET = preejection period/left ventricular ejection time; QS21 = duration of electromechanical sYStole index.

794

CLONIDINEIN CONGESTIVE HEART FAILURE

AP/At

PEPI

!,

800

260

220 c~ z

O

bJ

180 -,I .-I

200

140

0

|

,

!

,

CONT. 0.2 CONT. 0.4 mg mg

0

,

i

,

I

CONT. 0.2 CONT. 0.4 mg mg

symptoms of postural hypotension. Intermittent right bundle branch block and first degree atrioventricular block was noted in 3 patients. Discussion Clonidine evoked significant hemodynamic effects in these patients with CHF. This centrally acting vasodilator reduced systemic blood pressure, pulmonary artery pressure, and pulmonary capillary wedge pressure. Heart rate decreased significantly after clonidine administration. Cardiac and stroke volume indexes did not change substantially. Clonidine appeared to favorably affect the myocardial oxygen supply-demand ratio by minimally altering coronary perfusion pressure, increasing diastolic perfusion time, and reducing the double product and inotropic state of the ventricle. Since stroke volume changed little after clonidine administration, the decrease in systemic blood pressure was probably secondary to systemic arteriolar vasodilatation. The systemic arteriolar vasodilatation after clonidine administration was not reflected in changes in calculated systemic vascular resistance since cardiac output either did not change or tended to decrease. Systemic vascular resistance, the major contributing factor to the state of ventricular afterload, is generally used as an approximation of afterload. Ventricular wall tension during systole, formulated by the Laplace equation as ventricular systolic pressure X ventricular volume (at the designated pressure) + wall thickness, is another expression (and probably a more precise one) of ventricular afterload. Ctonidine substantially decreased systolic pressure. Its effects on ventricular volume were not measured; however, since ventricular systolic and diastolic pressures decreased, ventricular volume probably either decreased or remained unaltered. Therefore, using the ventricular wall tension formulation for afterload, clonidine probably elicited a reduction in ventricular afterload despite minimal to

FIGURE 6. The preejection period index (PEPI) decreased substantially while the developed pressure during isovolumic contraction/isovolumic contraction time (Ap/At) decreased dramatically after clonidine. These changes in inotropic indexes occurred in each patient after both doses. Individualdata (n = 6) are presented with brackets indicating 4- 1 standard deviation and * p <0.05; CONT. = control.

no changes in calculated systemic vascular resistance. Certainly, clonidine effected systemic arteriolar dilatation. The inability of the failing ventricle to increase stroke volume in response to systemic arteriolar vasodilatation (or afterload reduction) may be caused by 1 or more factors, such as, excessive preload reduction, end-stage ventricular dysfunction, provocation of ventricular ischemia, or drug-related negative inotropic effects. The degree of preload reduction (approximated by the decrease in pulmonary capillary wedge pressure) was not excessive in this population, but quite similar to the reduction effected by some of the peripheral vasodilators. 12-14 None of the patients demonstrated end-stage ventricular dysfunction and, when subsequently tested, most responded favorably (augmentation of stroke volume and cardiac index) to various peripheral vasodilators. None of the patients demonstrated clinical evidence of "rebound" after discontinuation of their vasodilator therapy (>48 hours prestudy), although more subtle hemodynamic changes cannot be excluded. It is unlikely that a mild "rebound" effect would account for the lack of augmentation of ventricular function; in this setting, the response to clonidine might be even more favorable. It is unlikely that clonidine elicited myocardial ischemia, since all but I of the subjects had angiographically normal coronary anatomy and, as previously indicated, clonidine tended to favorably alter the determinants of myocardial oxygen supply-demand. Clonidine did, however, decrease isovolumic developed pressure/duration of isovolumic contraction and increase the preejection period and preejection period/left ventricular ejection time in all subjects tested at both doses indicating that this compound evoked negative inotropic effects. The decrease in isovolumic developed pressure/ duration of isovolumic contraction with clonidine was secondary to lengthening Of the isovolumic contraction time. The isovolumic contraction time and the pre-

March 1, 1983

ejection period by themselves correlate well with other indexes of inotropy, although independent changes of entricular filling ana systemic pressure can profoundly v. these va riables.8'l°-12'~5-17 That is, without a al~er change in contractility, an isolated reduction of ventricular filling pressure can increase the preejection period and the isovolumic contraction time, and an isolated reduction of diastolic blood pressure would shorten them. While ventricular filling pressure did decrease with c]onidine administration, this drug effected a parallel reduction in systemic diastolic pressure; the resultant isovolumic developed pressure remained unchanged. Therefore, the increase in preejection period and isovolumic contraction time and the consequent decrease in isovolumic developed pressure/duration of isovolumic contraction noted after clonidine administration indicates that this drug elicited a negative inotropic effect. Clonidine's central site of action 6,%18'19probably accounts for most of the differences between the hemodynamic effects of this drug and those of the peripheral vasodilating agents. Clonidine's central effects elicit an overall reduction in sympathetic nervous system outflow to cardiovascular target organs including vascular smooth muscle (generalized vasodilation), specialized conduction tissue (negative chronotropy), and myocardium (resultant negative inotropy). Clonidine has been shown to reduce circulating cathecholamines and also retard renin release (resultant decrease in angiotensin activity).2°'~ The vasodilation is not likely to be a consequence ofdirect peripheral action, because its peripheral effects are predominantly vasopressor, z2 The negative chronotropic effects may be partly related to heightened parasympathetic activity, since clonidine appears to exert some parasympathomimetic effects.23 The negative inotropic effects are, in large part, mediated indirectly through reduced myocardial adrenergic tone; clonidine itself does not directly reduce inotropy. 24 Elevated sympathetic activity is an important compensatory mechanism in maintaining adequate cardiac function in moderate to severe CHF. 25 Although norepinephrine levels of neurons innervating the heart are reduced and the number and sensitivity of betal-adrenergic receptors are attenuated, cardiac function remains quite dependent on heightened sympathetic support. 25-27 Unlike clonidine, the peripheral vasodilators act primarily at vascular or peripheral vascular innervation sites and, therefore, do not elicit myocardial depression; for example, hydralazine induces vasodilation by acting directly on arterioles, and prazosin, an alphal-antagonist, postsynaptically blocks norepinephrine at vascular innervation sites. It is possible that clonidine's overall hemodynamic effects may be more favorable in patients with heart failure of lesser severity and with less dependence on sympathetic nervous system support. The clinical role of clonidine in CHF is likely to be restricted because of its inability to improve ventricular systolic function. Since it reduces double product, inotropy, and pulmonary capillary wedge pressure and lacreases diastolic perfusion time, it may be useful in Patients with ischemic cardiomyopathy who have high

THE AMERICAN JOURNAL OF CARDIOLOGY

Volume 51

795

ventricular filling pressure with mild ventricular systolic dysfunction. While its role in the treatment of CHF is probably limited, clonidine is an unusual vasodilator which has provided valuable insights into the role of the central sympathetic nervous system in human CHF.

Acknowledgment: We thank Dawn Orth, RN, Max Bacher, and Mitzi Prosser for their technical assistance and the nurses and house officers of the Coronary Care Unit for their professional assistance in the care of these patients. References 1. Cohn JN. Vasodilator therapy for heart failure: the influence of impedance on left ventricular performance. Circulation 1973;48:5-8. 2. Mason DT. Afterload reduction and cardiac performance: physiologic basis of systemic vasodilators as a new approach in the treatment of congestive heart failure. Am J Med 1978;65:106-125. 3. Onesti G, Schwartz AM, KIm KE, Schwartz C, Brest AN. Pharmacodynamic effects of a new antihypertensive drug, Catapres (ST-155). Circulation 1969;39:219-228. 4. Muir AL, Burton JL, Lawrie DM. Circulatory effects at rest and exercise of clonidine, an imidazoline derivative with hypotensive properties. Lancet 1969;2:181-184. 5. McRaven DR, Kroetz FW, Kioschos JM, Kirkendall WM. The effect of clonidine on hemodynamics in hypertensive patients. Am Heart J 1971; 81:482-489. 6. Sharma JN, Sandrew B, Wang SC. CNS site of clonidine induced hypotension: a micro iontophoretic study of cardiovascular neurons. Brain Res 1978;151:127-133. 7. Tadepalli AS, Mills E. Contribution of supracollicular structures of the brain to the central depression of cardiovascular function by clonidine. J Pharmacol Exp Ther 1978;205:693-701. 8. Lewis RP, Leighton RF, Forester WF, Weissler AM. Systolic time intervals. In: Noninvasive Cardiology. New York: Grune & Stratton, 1974:301-314. 9. Boudoulas H, Lewis RP, Rittgers SE, Leier CV, Vasko JS. Increased diastolic time: an important factor in the beneficial effect of propranolol in patients with coronary artery disease. J Cardiovasc Pharmacol 1979;1: 503-513. 10. Diamond G, Forrester JS, Chatterjee K, Wegner S, Swan HJC. Mean electromechanical Ap/At. Am J Cardiol 1972;30:338-342.

11. Agress CM, Wegner S, Forrester JS, Chatterjee K, Parmley WW, Swan

HJC. An indirect method for evaluation of left ventricular function in acute myocardial infarction. Circulation 1972;46:291-297.

12. Leier CV, Desch CE, Magorien RD, Triffon DW, Unverferth DV, Boudoulas

13,

14,

15, 16.

17. 18. 19. 20.

21. 22. 23, 24. 25. 26. 27.

H, Lewis RP. Positive inotropic effects of hydralazine in human subjects: comparison with prazosin in the setting of congestive heart failure. Am J Cardiol 1986;46:1039-1044. Levine TB, Franciosa JA, Cohn JN. Acute and long-term response to an oral converting enzyme inhibitor, captopril, in congestive heart failure. Circulation 1980;62:35-42. Leier CV, Bambach D, Thompson MJ, Cattaneo SM, Goldberg RJ, Unverferth DV. Central and hemodynamic effects of intravenous ~sosorbide dinitrate, nitroglycerin and nitroprusside in patients with congestive heart failure. Am J Cardiol 1981 ;48:1115-1123. Talley RC Meyer JF, McNay JL. Evaluation of the preejection period as an estimation of myocardial contractility n dogs. Am J Cardiol 1971;27: 384-391. Metzger CC, Chough CB, Kroetz FW, Leonard JJ. True isovolumic contraction time. Am J Cardiol 1970;25:434-442. Boudoulas H, Karayannacos PE, Lewis RP, Leier CV, Vasko JS. Effect of afterload changes on left ventricular performance. Eur Congress NonInvasive Cardiovasc Dynamics 1980; 12:89-91. Kobinger W, Walland A. Investigations into the mechanism of the hypotension effect of 2- (2,6-dichlorophenyiamino)-2-imidazoline-HCI. Eur J Pharmacol 1967;2:115-162. Schmitt H, Schmitt H. Localization of the hypotensive effects of 2(-2,6dichlorophenylamino)-2-imidazoline-hydrochloride. Eur J Pharmacol 1969;6:8-12. Hokfelt B Hedeland H Hansson BG. The effect of clonidine and penbutolol respectively on cathecho am nes in b ood and urine, plasma renin activity and urinary aldosterone in hypertensive patients. Arch Int Pharmacodyn Ther 1975;213:307-321. Reid IA, MacDonald DM, Pachnis B, Ganong WF. Studies concerning the mechanism of supppession of renin secretion by clonidine. J Pharmacol Exp Ther 1975;192:713-721. Haeusler G. Studies on the possible contribution of a peripheral presynaptic action of clonidine and dopamine to their vascular effects under in vivo conditions. N Sch Arch Pharmacol 1976;295:191-202. Lobson RD, Kaplan HR, LaForce S. An investigation into the bradycardiac effects of St-155 2-(2,6-dichlorphenylamino)-2-imidazoline HCI in the anesthetized dog. J Pharmacol Exp Ther 1969; 169:120-131. Motz W, Ippisch R, Strauer BE. Clonidine exerts no direct myocardial depressant effect. Chest 1982;82:214. Vogel JHK, Chidsey CA. Cardiac adrenergic activity in experimental heart failure assessed by beta receptor blockade. Am J Cardiol 1969;24:198208. Chidsey CA, Sonnenblick E, Morrow AG, Braunwald E. Norepinephrine stores and contractile force of papillary muscle from the failing human heart. Circulation 1966;33:43-51. Bristow MR, Ginsburg R, Minobe W, Cubicciotti RS, Sageman WS, Lurie K Billingham ~E, Harrison DC Stinson EB. Decreased catecholamine sensitivity and ~'-adrenergic receptor dens ty n fail ng human hearts. N Eng J Med 1982;307:205-211.