Systemic and regional hemodynamic effects of alpha-adrenoceptor blockade in chronic left ventricular dysfunction in the conscious dog In seven dogs with long-standing left ventricular dysfunction induced 16 weeks earlier by repetitive transmyocardial direct current (DC) shock, the acute hemodynamic effect of the or,-adrenoceptor antagonist urapidil was studied. Left ventricular end-diastolic pressure (LVEDP) was significantly increased from preshock levels at the time of study and cardiac output was reduced. Plasma norepinephrine was significantly increased from control levels and was not altered by urapidil infusion. The mean arterial pressure fell in response to at-blockade from 111 to 65 mm Hg, the LVEDP fell from 16 to 9 mm Hg, and cardiac output increased from 2.90 to 3.70 L/min (all p < 0.01). Regional blood flows measured by microsphere injection revealed an increase in blood flow to skeletal muscle, which had not been significantly decreased by the left ventricular dysfunction in this model, and further decreases in spianchnic flow, which was already depressed compared with that in normal dogs. Therefore acute cu-adrenoceptor blockade improves central hemodynamics in experimental heart failure but does not normalize the resting blood flow meldistribution in this model. (AM HEART J 1990;120:619.)

Peter F. Carlyle, BS, and Jay N. Cohn, MD. Minneapolis,

Vasoconstriction in the setting of left ventricular dysfunction contributes to an impedance to left ventricular ejection and a further impairment of cardiac 0utput.l The sympathetic nervous system is activated in most patients with left ventricular dysfunction,2p 3 and norepinephrine-induced vasoconstriction may account for at least some of the high vascular tone in this syndrome. This sympathetic vasoconstriction may also contribute to redistribution of peripheral blood flo~.~-~ Prazosin, an ai-adrenoceptor blocker, has been demonstrated to have a favorable short-term hemodynamic effect in patients with heart failure.7l * The regional redistribution of blood flow in response to a-blockade in the setting of left ventricular dysfunction has not been fully assessed. Long-standing left ventricular dysfunction can be induced by repetitive direct current (DC) countershock in the dog.g In previous studies, welo have demonstrated that the animals develop progressive left ventricular dilatation and elevated plasma nor-

From the Cardiovascular Division, Minnesota Medical School. Received Reprint 55455.

4/l/21995

for publication requests:

March

Jay N. Cohn,

Department 1, 1990; MD,

Box

of Medicine,

accepted

April

488 UMHC,

University

of

24, 1990. Minneapolis,

MN

Minn.

epinephrine levels similar to that observed in clinical heart failure. We have therefore utilized this model to evaluate the hemodynamic and regional blood flow effects of a a-adrenoceptor blockade. METHODS Experimental

myocardial

damage.

Seven mongrel dogs

weighing between 17and 21kg were usedin this study. The dogsunderwent repetitive DC countershockto produce left ventricular dysfunction (LVD) 16 weeks earlier. In this technique, as previously described,gthe dogs were anesthetized with pentobarbital sodium (30 mg/kg), intubated, and ventilated with a Harvard respirator (Harvard Apparatus Inc., S. Natick, Mass.). A pigtail catheter waspassed in a retrograde direction into the left ventricle from the femoral artery and a premeasuredsoft metallic guide wire was passed through the catheter with 5 mm of wire extending into the left ventricular cavity. One electrode paddle wasplaced at the left sternal border at the point of maximal impulse (fourth to sixth intercostal space)and the secondelectrodewasconnectedto the guide wire in the left ventricle. The dogswere given repetitive DC shocksat an energy level of 80 joules. The shockswere administered at lo-second intervals to a total of 1 shock per kilogram of body weight. The dogs were then returned to their cages and were allowed to recover. Hemodynemic studies. Hemodynamic studieswereperformed 16 weeksafter the shock procedure. The animals were lightly sedated with morphine sulfate (1 mgikg), a dosethat has little effect on resting hemodynamics,” and were placed in the right lateral position for study. Cathe619

620

September lQQ0 American Heart Journal

Carlyle and Cohn

ters wereplacedpercutaneouslyunder local anesthesiainto the left ventricle (LV) and aorta (Ao) for pressuremeasurement. A thermodilution flow-directed catheter was passedinto the pulmonary artery (PA) for pressuremeasurementand cardiac output determinations. Left ventricular systolic, left ventricular end-diastolic, aortic, aortic mean,pulmonary, pulmonary mean,and pulmonary wedge pressureswere recorded. Cardiac output was determined by the thermodilution indicator dilution method, and lead II wasrecorded to monitor heart rate and the electrocardiogram. The pressures,electrocardiogram, and cardiac outputs were recorded on a Hewlett-Packard 7758B eightchannel physiologic recorder (Hewlett-Packard Co., Medical Products Group, Andover, Mass.) usingStatham P23B pressuretransducers(SpectramedInc., Critical Care Division, Oxnard, Calif.). A Gould cardiac output computer (model SPl435, Gould, Inc., Oxnard, Calif.) was used for cardiac output determinations. The first derivative of pressure(dp/dt) signal was recorded through a HewlettPackard model 8802A medium gain amplifier (HewlettPackard Co.). After completion of the catheterization procedure the animalswere allowed to rest for a minimum of 30 minutes or until their pressuresstabilized. Control hemodynamic measurementswere then recorded and blood samplesfor plasma norepinephrine and plasma renin activity were drawn. An injection of 15pm radioactive microsphereswas then given. A test bolus of phenylephrine HCl (Neo-synephrine, Winthrop Pharmaceuticals,Division of Sterling Drug Inc., New York, N.Y.), 2 pglkg, was then administered intravenously to ensurethat the dogswere responsiveto cui-stimulation. When all hemodynamic variables returned to baseline, urapidil (2 mg/kg) was slowly injected over a 5-minute period. At the end of the injection, another bolus of phenylephrine was given to ensure a al-blockade. Twenty minutes following the urapidil intervention, hemodynamic measurementswere performed and another injection of radioactive microsphereswasgiven. Blood was again drawn for determination of plasmanorepinephrine and plasmarenin activity. The animalswere put to death at the end of the study using a lethal doseof pentobarbital sodium. Organ blood flow studies. Microspheres were injected through a pigtail catheter located in the left ventricle. The arterial line was connected to a constant rate withdrawal pump. Measurements of regional tissue flow were made using injections of microspheres15pm in diameter labeled with gamma-emitting radionuclides (lz51 and s5Sr, 3M Company, St. Paul, Minn.), diluted in 10% low molecular weight dextran. Before injection, the microsphereswere mixed for at least 15minutes in an ultrasonic bath. At the end of the control hemodynamic measurementsand at peak responseto urapidil, approximately 3 x lo6 microsphereswere injected through the pigtail catheter. Beginning with the injection, a referencesampleof arterial blood waswithdrawn from the aortic catheter at a constant rate of 15 ml/min for 90 seconds. After the animalswere put to death, the hearts were ex-

cisedand fixed in 10% buffered formalin. Tissue samples weighing 1 to 2 gm each were obtained from the following organs: brain, lungs, skin, skeletal muscle, liver, spleen, pancreas,renal cortex, stomach,smallintestine, bowel, and diaphragm. After fixation, the atria and great vesselswere removed from the heart. The right ventricle, septum, and left ventricle were then sliced into four circumferential piecesand were then separatedinto left ventricular (LV), right ventricular (RV), and septal sections.The LV and septum were then sliced transversely into epicardial and endocardial sections. Tissue and blood reference sampleswere counted in a Packard model S912gammacounting system(Packard Instrument Co., Meriden, Conn.) at window settings corresponding to the peak energiesof each radionuclide. The counts per minute recorded in each energy window were corrected for background activity and for overlapping counts contributed by the accompanying isotopeswith a digital computer. Blood flow to each specimen(Qs) was computed using the formula Qs = Qr x Cs/Cr, where Qr = reference blood flow (in milliliters per minute), Cs = counts per minute of specimen,and Cr = counts per minute of reference blood specimen.The blood flow from each specimen (in milliliters per minute) was divided by sampleweight and wasexpressedas milliliters per minute per gram of tissue sample. Plasma norepinephrine sampleswere measuredwith a radioenzymatic technique (Cat-a-Kit, Amersham, Arlington Heights, Ill.). Duplicate determinations in our laboratory have a coefficient of variation of 8.3%. Plasmarenin activity was measuredby radioimmunoassay. Statistical analysis of hemodynamic and microsphere data was performed using paired t tests for within-group comparisonsand unpaired t tests for group comparisons. Data are summarizedas the mean + SEM. RESULTS Hemodynamics

(Table I). The hemodynamic changes over the 16 weeks after LVD are shown in Table I. There were significant increases in pulmonary artery wedge pressure (7 +- 1 to 13 +- 1 mm Hg, p < 0.05), right atria1 pressure (3 + 1 to 9 + 1 mm Hg, p < O.OOl), and left ventricular end-diastolic pressure (8 + 1 to 16 st 2 mm Hg, p < 0.01). Cardiac output fell from 3.9 _+0.9 to 2.9 2 0.6 Llmin. In response to the urapidil infusion, mean arterial pressure fell significantly from 111 -I- 7 to 85 -t 6 mm Hg (p < 0.01) and pulmonary wedge pressure decreased from 13 + 1 to 8 ? 1 mm Hg @ < 0.05). Left ventricular end-diastolic pressure, right atria1 pressure, and systemic vascular resistance also all decreased significantly (p < 0.01). Cardiac output rose from 2.9 +- 0.2 to 3.7 f .03 L/min (p < 0.01) and heart rate was significantly increased from 74 + 10 to 95 f 10 beats/min @ < 0.01). Therefore the 2 mg/kg dose of urapidil restored systemic hemodynamics to preshock control levels (Fig. 1).

Volume Number

120 3

Alpha

blockade in LV failure

621

1. Changesin the central hemodynamic pattern after left ventricular damage(LVD) and after infusion of urapidil into the dogswith LVD. MAP, Mean arterial pressure;LVEDP, left ventricular end-diastolic pressure;CO, cardiac output. SVR, systemic vascular resistance.

Fig.

A 2 pg/kg bolus of phenylephrine given to the group of seven dogs prior to urapidil infusion increased mean arterial pressure from 110 f 3 to 127 + 3 mm Hg (p < 0.001). Rechallenging the dogs with phenylephrine after completion of the urapidil infusion produced no change in mean arterial pressure (82 f 2 versus 80 + 2 mm Hg). Plasma norepinephrine was significantly higher 16 weeks after shock than in the preshock control period (Fig. 2). Urapidil did not change plasma norepinephrine concentration. Plasma renin activity was not altered by LVD or by urapidil. Blood flow distribution (Table II). In comparison with a group of 11 normal dogs, the group of dogs that underwent the shock procedure had significant decreases in blood flow to the small intestine and a reduction in the endocardial-to-epicardial myocardial blood flow ratio. In response to urapidil in the LVD dogs, blood flow to the epicardium tended to decrease with a decrease in flow to the endocardium reaching a nearly significant level (1.204 & 0.21 to 0.883 t- 0.13 ml/min/gm, p = 0.08). The endocardiakepicardial flow ratio was unchanged. Right ventricular blood flow fell significantly from 1.133 + .23 to 0.706 + 0.13 ml/min/gm 0, < 0.05). Blood flow to the splanchnic bed diverged, with hepatic arterial flow increasing significantly in response to urapidil (0.333 -t 0.12 to 0.467 + 0.13 p < 0.05), and blood flow to the stomml/min/gm, ach and small intestine decreasing significantly

Table I. Hemodynamic effect of left ventricular dysfunction and responseto urapidil (2 mg/kg) in seven dogs Hemodynamic parameter AOm (mm Hg) PAm (mm Hg) PAW (mm Hg) RA (mm Hg) LVEDP (mm Hg) Heart rate

Control

LVD

99 + 6 17 ?z 1

111 k I 21 ? 2 13 k 1t 9 f 1*

71tl

321 8+1 104 * 5

Peak urapidil 85 f 6$ 17 + 1% 8 f 1%

16 + 2t 74 k 10*

5 rt 11 9k 1% 95 f 10%

3.8 k 0.4

2.9 + 0.2

3.7 f 0.31

31 -+ 3

43 k 6

42 k I

(beats/min) Cardiac output (L/min) Stroke volume (ml/beat) SVR (dynes . set . cme5) PVR (dynes . set cmm5)

2127 + 285 2912 rt 307 368 r 31

583 k 43t

1903 * 3151 379 f 30%

Values expressed as mean +_ SEM. AOm, Mean aortic pressure; LVD, left ventricular dysfunction; LVEDP, left ventricular end-diastolic pressure; PAm, mean pulmonary artery pressure; PAW, pulmonary artery wedge pressure; PVR, pulmonary vascular resistance; RA, right atrial pressure; SVR, systemic vascular resistance. *p < 0.05 versus control. tp < 0.002 versus control. tp < 0.05 versus LVD.

(1.295 -+ 0.28 and 0.492 f 0.04 to 0.63 + 0.13 and 0.406 + 0.06 ml/min/gm, respectively) (p < 0.05). Blood flow to skeletal muscle increased an average of 33 % , but the increase did not quite achieve statisti-

September1990 622

Carlyle and Cohn

American

Heart

Journal

2. Plasmarenin activity and plasmanorepinephrine concentration measuredin the control period, 16 weeksafter left ventricular damage(LVD), and after the infusion of urapidil.

Fig.

Table

II. Tissue blood flow (ml/min/gm) before and after urapidil Organ Skeletal muscle Skin Diaphragm Liver Stomach Small intestine Bowel Renal cortex Epicardium Endocardium Endo/Epi Right ventricle

Normal 0.108 0.112 0.382 0.196 1.392 0.781 0.881 5.70 1.211 1.549 1.242 1.07 0.858

Values expressed as mean i: SEM. End&pi, Endocardial/epicardial ratio; other abbreviations *p 5 0.05 versus normal dog. tp or 0.05 versus LVD. $pI 0.1 versus LVD.

dog + & 2 f rt i i -f * + i f t

0.03 0.03 0.06 0.08 0.28 0.13 0.14 0.54 0.22 0.37 0.04 0.16 0.09

LVD 0.103 0.060 0.325 0.333 1.295 0.492 0.584 6.799 1.141 1.204 1.102 1.133 0.928

* + + rt f * + k 2 + + I k

Urapidil 0.03 O.Ol* 0.07 0.12 0.28 0.04* 0.07* 0.78 0.25 0.21 0.05* 0.23 0.07

0.144 0.068 0.269 0.467 0.63 0.406 0.579 6.306 0.813 0.883 1.126 0.706 0.777

+ 0.04 It 0.02 ?c 0.051 ‘- 0.13t + 0.13t 2 0.067 2 0.1 AC 1.03 + 0.141 + 0.131 + 0.05 t 0.13t 2 0.12

as in Table I.

cal significance (p = 0.16). After urapidil, the regional blood flow distribution appeared to be more disturbed compared with that in the normal dogs than it was during the untreated LVD state (Fig. 3). DISCUSSION

The clinical syndrome of heart failure is characterized by systemic vasoconstriction and activation of the sympathetic nervous system.lm3 Although it has been suggested that the vasoconstriction may relate predominantly to sympathetic nervous system stim-

ulation, clinical studies have revealed only a weak correlation between calculated systemic vascular resistance and plasma norepinephrine levels.3 The animals subjected to myocardial damage from repetitive DC shock in this study were not in overt heart failure at the time of study, but they did exhibit abnormal LV function, low cardiac output, and high circulating norepinephrine levels. Elevated plasma norepinephrine has also recently been reported in patients with long-standing LVD but without signs or symptoms of clinical heart failure.12

Volume

120

Number

3

Alpha blockade in LV failure

623

Fig. 3. Percent changefrom normal blood flow distribution (11 normal dogs)after left ventricular damage (LVD) and after the infusion of urapidil into dogswith LVD.

These animals also provided us an opportunity to assess the regional distribution of the reduced cardiac output in LVD. Resting flow to the kidney, brain, and skeletal muscle, and arterial flow to the liver were well maintained, but flow to the splanchnic viscera and skin were reduced, and there was a fall in the endocardiakepicardial ratio of myocardial blood flow. This redistribution of myocardial flow may result from the elevated LV diastolic pressure, which can compress the subendocardial vessels during diastole.i3 To assess a possible role of sympathetic-mediated vasoconstriction in the regional redistribution of blood flow, a-adrenoceptor blockade was produced over the short term by the administration of urapidil. Although this drug may have additional central effects, especially during long-term administration,14 the short-term response to the drug appears to be largely through al-receptor blockade.15 Inhibition of a phenylephrine-induced rise in arterial pressure after urapidil in these dogs confirmed the presence of cu-adrenoceptor blockade. a-Blockade in these dogs produced central hemodynamic effects identical to those previously observed in patients with heart failure given single doses of prazosin, trimazosin, or urapidil.7y 8, 16,l7 The mean arterial pressure fell, and there was a reduction in the elevated left and right heart filling pressures, with an increase in cardiac output and a modest reflex tachycardia. The net effect was a return toward

normal of the disturbed LV performance (Fig. 1). Despite the fact that cardiac output was restored to normal levels, the regional blood flow distribution remained abnormal. Resting blood flow to skeletal muscle, which was not underperfused at rest in the dogs with LVD, increased to more than 30% above normal levels, whereas flow to skin and the splanchnit viscera remained depressed. Myocardial blood flow fell because of the fall in arterial pressure, but the endocardiahepicardial ratio was not restored to normal (Fig. 2). Symptoms from heart failure occur predominantly during exercise, so changes in resting blood flow distribution may not reflect the abnormal flow patterns that contribute to disability in heart failure. Nonetheless, correction of the resting ventricular dysfunction with vasodilating drugs is usually assumed to result in restoration of regional blood flows to normal. These data indicate that regional flows are not necessarily corrected when the cardiac output rises, and the specific vasodilator drug employed may have a profound effect on these regional flows. Although q-blockade normalized the impaired central hemodynamics, it did not correct the resting regional maldistribution of peripheral blood flow in this syndrome. It therefore appears that sympathetic-mediated vasoconstriction through a-adrenoceptor of the activation may not be a critical determinant regional blood flow alterations in heart failure. Furthermore, the failure of a-blockade to normalize

624

Carlyle and Cohn

regional blood flow may be a factor in the lack of effectiveness of chronic prazosin therapy in previous clinical heart failure trials.18 REFERENCES

1. Cohn JN, Mashiro I, Levine TB, Mehta 3. Role of vasoconstrictor mechanisms in the control of left ventricular performance of the normal and damaged heart. Am J Cardiol 1979;44:1019-22. 2. Thomas JA, Marks BH. Norepinephrine in congestive heart failure. Am J Cardiol 1978;41:233-43. 3. Levine TB, Francis GS, Goldsmith SR, Simon A, Cohn JN. Activity of the sympathetic nervous system and renin-angiotensin system assessed by plasma hormone levels and their relationship to hemodynamic abnormalities in congestive heart failure. Am J Cardiol 1982;49:1659-66. 4. Higgins CB, Vatner SF, Franklin D, Braunwald E. Pattern of differential vasoconstriction in response to acute and chronic low output states in the conscious dog. Cardiovasc Res 1974; 8:92-8. 5. Flaim SF, Minter SJ, Nellis SH, Clark DP. Chronic arteriovenous shunt: evaluation of a model for heart failure in rats. Am J Physiol 1979;5:H698-704. 6. Zelis R, Nellis SH, Longhurst J, Lea G, Mason DT. Abnormalities in the regional circulation accompanying congestive heart failure. Prog Cardiovasc Dis 1975;18:181-99. 7. Miller RR, Awan NA, Maxwell KS, Mason DT. Sustained reduction of cardiac impedance and preload in congestive heart failure with the antihypertensive vasodilator, prazosin. N Engl J Med 1977;297:303-7. 8. Awan NA, Miller RR, Mason DT. Comparison of effects of nitroprusside and prazosin on left ventricular function and peripheral circulation in chronic congestive heart failure. Circulation 1978;57:152-9. 9. Mehta J, Runge W, Cohn JN, Carlyle P. Myocardial damage after repetitive direct current shock in the dog: correlation

American

10. 11. 12.

13. 14.

15.

16. 17.

18.

September 1990 Heart Journal

between left ventricular end-diastolic pressure and extent of myocardial necrosis. J Lab Clin Med 1978;91:272-9. Carlyle PF, Cohn JN. A non-surgical canine model of chronic left ventricular myocardial dysfunction. Am J Physiol 1983; 244:H769-74. Priano LL, Vatner SF. Morphine effects on cardiac output and regional blood flow distribution in conscious dogs. Anesthesiology 1981;55:236-43. Francis G, Benedict C, Johnson D, Kirlin P, Neuberg G, Hosking J, Liang C, Yusuf S. Differences in neurohormonal activation in patients with left ventricular dysfunction with and without heart failure [Abstract]. J Am Co11 Cardiol 1989;13:246A. Salisbury PF, Cross CE, Rieban PA. Acute ischemia of inner layers of ventricular wall. AM HEART J 1963;66:650-6. Kellar KS, Quest JA, Spera AC, Buller A, Conforti A, Souza SD, Gillis RA. Comparative effects of urapidil, prazosin, and clonidine on ligand binding to central nervous system receptors, arterial pressure, and heart rate in experimental animals. Am J Med 1984;77(suppl4A):87-95. Zeigler DW, Shebuski RJ, Zimmerman BG. Central and peripheral cardiovascular actions of urapidil in normotensive and Goldblatt hypertensive animals. Am J Med 1984;77(suppl 4A):81-6. Franciosa JA, Cohn JN. Hemodynamic effects of trimazosin in aatients with left ventricular failure. Clin Pharmacol Ther i978;23:11-18. Wang RYC, Chow KH, Chan HY, Pan HYN, Wang RPY. Acute hemodynamic and myocardial metabolic effects of intravenous urapidil in severe heart failure. Eur Heart J 1984;5:745-51. Cohn JN, Archibald DG, Ziesche S, Franciosa JA, Harston WE, Tristani FE, Dunkman WB, Jacobs W, Francis GS, Flohr KH. Goldman S. Cobb FR. Shah PM. Saunders R. Fletcher RD,’ Loeb HS, Hughes Vd, Baker B: Effect of vasodilator therapy on mortality in chronic congestive heart failure. Results of a Veterans Administration Cooperative Study (VHeFT). N Engl J Med 1986;314:1547-52.