Comparative Actions of Hydralazine, Nifedipine and Amrinone in Primary Pulmonary Hypertension

STUART RICH, MD, ROBERT GANZ, MD, and PAUL S. LEVY, ScD

The effects of 3 types of vasoactive agents, hy- dralazine, ntfedipine and amrinone, were evaluated in 7 patients wtth primary pulmonary hypertension (PPH). Hemodynamic values were measured before and afler drug administration in every patient. All drugs increased cardiac output and reduced both systemic and pulmonary resistance in the patients studied. Only nifedipine significantly reduced pul- monary artery (PA) pressure (6 f 5 mm Hg). In addition, it decreased pulmonary resistance to a greater degree than systemic resistance in 2 of the 7 patients, suggesting that nifedtpine can cause selective pulmonary vasodllation in some pattents. Hydralarine appeared to increase cardiac output and stroke volume by reducing systemic resistance.

Recent reports on the effectiveness of various vasodi- lators in patients with primary pulmonary hypertension (PPH) have been inconsistent, partly because of the lack of uniformity in testing these drugs and partly because of disagreement as to what constitutes a ben- eficial drug response. Three vasoactive agents, hydral- azine, nifedipine and amrinone, were tested to see if the acute drug responsiveness of patients with PPH could be characterized and valid comparisons made about how different drugs affect the same patients with PPH.

Methods

Patients: The study population consisted of 7 patients, 1 man and 6 women, aged 33 to 63 years (mean 48). All patients underwent extensive evaluation to determine the origin of pulmonary hypertension (as previously described), but none was apparent.’ All patients had symptoms related to pulmo- nary hypertension at the time of admission, were not receiving vasoactive medications at the time of study, and gave in- formed consent to drug testing.

From the Section of Cardiology, Department of Medicine, University of Illinois College of Medicine, and the Epidemiology and Biometry Program, University of Illinois, School of Public Health, Chicago, Illinois. Manuscript received May 16, 1963; revised manuscript received July 27. 1983. accepted July 29, 1983.

Address for reprints: Stuart Rich, MD. University of Illinois Hospital, Section of Cardiology, P.O. Box 6998, Chicago, Illinois 60680.

There was no evidence of direct pulmonary vaso- dilating effects; it decreased systemic resistance more than pulmonary resistance in every case. The increase in cardiac output from amrinone was secondary to a decrease in systemic artertal pres- sure with reflex tachycardia; stroke volume was unchanged. Amrinone had little pulmonary effect in all but 1 patient, in whom it substantially reduced PA pressure and pulmonary resistance. The mechanism of action of these 3 drugs in PPH differs. Nifedipine holds the most promise as an effective pulmonary vasodilator. A study of the effects of long-term ad- ministration of nifedipine in PPH is warranted.

(Am J Cardiol 1963; 52:1104-1107)

Measurements: Studies were performed in the coronary care unit with the patient in the supine position. Right-sided cardiac heart catheterization was performed with a 7Fr thermodilution Swan-Ganz flow-directed catheter, and the radial artery was cannulated with a short Teflon@ catheter. Pressures were monitored with a Gould-Statham p23dB pressure transducer and a 2-channel recorder (Hewlett Packard). An electrocardiographic lead was monitored con- tinuously throughout the study. Cardiac output was taken as the average of 3 serial determinations using a thermodilution computer (Edwards 9150 CO). Baseline measurements were made of the systemic arterial pressure (SAP) and pulmonary arterial pressure (PAP) and cardiac output (CO) before the administration of each drug. Mean intravascular pressures were determined by electronic integration. Values for vascular resistance were calculated (in Wood units) from the following formulas: total systemic resistance = SAP/CO; total pulmo- nary resistance = PAP/CO. Pulmonary capillary wedge and right atrial mean pressures were measured in most of the drug tests, but because of the inability to consistently obtain a hemodynamically adequate wedge pressure in every trial, total pulmonary resistance instead of pulmonary vascular resis- tance is being used for statistical analysis. When pulmonary wedge and right atria1 pressures were obtained sequentially during drug testing, no difference was noted in either the di- rection or the magnitude of change between total pulmonary resistance and pulmonary vascular resistance or total systemic resistance and systemic vascular resistance.

Drug administration protocol: In every patient the Swan-Ganz catheter was placed 6 to 12 hours before drug

November 1. 1983 THE AMERICAN JOURNAL OF CARDIOLOGY Volume 52 1105

TABLE I Individual Hemodynamic Responses to Hydralazine

PAP SAP co TSR TPR HR (mm Hg) (mm Hg) (liters/min) (Wood Units) (Wood Units)

Pt C RX C Rx C Rx C Rx C Rx C Rx C Rx % TSR %TPR

1 90 105 65 2.44 3.66 25 27 33

3 104 94 1:; :: !Z

;: 120 I: 123 z: 3.96 1.80 5.46 1.84 ;: z :: 36 :: :: 17 ;: 18 z: 30 0 4 :

1:: 104 100 68 5

zl zz :s 2.43 4.52 :; :: 23 :“9 26 43 5: 43 29 31 9 0

;1 :: z: ::

86 2.88 3.49 3.13 4.21 33 27 20 19 15

7 103 42 2.79 3.09 24 14 23 :t ;: :: tX 2:

Mean 93 97 62 63 94 87 2.83 3.70 36 27 23 19 31 40 28 SD 13 10 6.9 9.7 24 16 1.17 0.72 15 <o.oi” 7.9 6.8 15 <o.o:”

18 :7” P NS NS NS <0.05 <0.05 <0.05

C = control; CO = cardiac output; HR = heat-l rate; NS = not significant; PAP = pulmonary artery pressure; Rx = after treatment; SAP = systemic arterv oressure; SV = stroke volume; TPR = total pulmonary resistance; %TPR = percent reduction in TPR; TSR = total systemic resistance; %T$d = percent reduction in TSR.

testing, which allowed the patient to remain stable in the coronary care unit for baseline hemodynamic monitoring. Before each drug test, baseline hemodynamic measurements were reestablished with the patient supine. At least 6 hours (in the trials after amrinone) or 12 hours (after hydralazine or nifedipine) were allowed between serial drug intervention to prevent overlapping of drug effects. Nifedipine was ad- ministered as a 20-mg oral dose, with hemodynamic responses measured after 90 minutes. Hydralazine was administered as an 0.3-mg/kg intravenous injection over 10 minutes, with measurements of pressure and cardiac output after 45 min- utes. Amrinone was administered as a 1-mg/kg injection over 5 minutes, and pressures and cardiac output were measured after 15 minutes. The sequence of drug administration was varied from patient to patient.

Statistical methods: For each of the 3 drugs tested, the mean difference between control and treatment levels of each of the hemodynamic variables was evaluated for statistical significance by means of the Student t test for paired data. Comparisons among the 3 drugs with respect to differences between the control and treatment levels of each hemody- namic variable were made by means of 2-way analysis of variance tests.

Results

Effect of hydralazine (Table I): Hydralazine sig- nificantly increased both cardiac output and stroke volume and significantly reduced total systemic resis- tance and total pulmonary resistance. The average percent reduction in vascular resistance was greater in

TABLE II Individual Hemodynamic Responses to Nlfedipine

the systemic circulation (28%) than in the pulmonary circulation (19%). Hydralazine failed to significantly change PA pressure.

Effect of nifedipine (Table II): Nifedipine signif- icantly increased cardiac output and stroke volume, with a concomitant decrease in systemic and pulmonary resistance and a significant reduction in PA and sys- temic artery pressure. Although 5 patients had a de- crease in systemic resistance that exceeded the decrease in pulmonary resistance, in 2 patients the decrease in pulmonary resistance exceeded that of the systemic resistance. PA pressure was decreased in every case (mean 6 f 5 mm Hg).

Effect of amrinone (Table III): As with nifedipine, amrinone significantly reduced both systemic and pulmonary resistance. It caused a small but significant increase in cardiac output that appeared to be secon- dary to a reflex tachycardia, because heart rate in- creased but stroke volume remained unchanged. Am- rinone caused a significant decrease in mean systemic arterial pressure but not in mean PA pressure. However, in 1 patient (Patient 2), a reduction in pulmonary re- sistance occurred that was manifest almost solely by a reduction in PA pressure, because the cardiac output was changed very little, suggesting a pulmonary vaso- dilator response.

Using 2-way analysis of variance we could not detect significant differences (p <0.05) between the drugs with

PAD SAP co TSR TPR HR (mm Hg) (mm Hg) (litersjmin) (Wood Units) (Wood Units)

Pt C Rx C Rx C Rx C Rx C Rx C Rx C Rx % TPR %TPR

Mean SD P

91 90 58 52 95 87 13 9 8

NS <0.02 g 10 12

<O.Ol

3.63 3.74 4.24 5.51 1.95 2.55 2.39 2.86 2.52 2.98 4.46 5.68 2.65 2.85

3.12 3.73 1.32 0.98

<0.02

fX 21 17 :; ::

:; x: 32 :: ;: 32 ;: 20

26 :: 2: 22:

33 25 20 15 8 11 5 7 -<0.01 -<0.02

38 44 16 47 58

:: 19 %; z 29

2: 41 25 24

:: :“2 3x 8 :3 13

36 43 23 23 19 16 9 10

<O.Ol NS

Abbreviations as in Table I.

1108 HYDRALAZINE, NIFEDIPINE AND AMRINONE IN PULMONARY HYPERTENSION

TABLE III Individual Hemodynamlc Responses to Amrinone

PAP SAP co TSR TPR

HR (mm WI (mm Hg) (liters/min) (Wood Units) (Wood Units)

Pt C Rx C Rx C Rx C Rx C Rx C Rx C Rx %TSR % TPR

1 85 95 63

::

95

:8

2.16 3.28 44 27 29 21 25 35 39

3’ 101 95 107 72 51 100 4.47 2.68 4.72 3.52 z: 17 9”: 20’ g”f ssoo

:: :: ;t :: z 3’: 16 4 82 t: 2.24 2.47 42 ;; 21 16 27 26 24

: 106 109 95 24 22

22:

7 t: ;: 5: 40 69 101 90

:: 2.55 3.99 3.10 3.71 40 23 31 26

:: :A :: :; -13 -1:

61 72 3.02 3.30 24 20 32 33 17 10

90 96 12 13

(0.01

56 56 92 84 12 11 10 10

NS <0.05

3.01 3.44 0.68 0.89

co.05

33 25 6 10 <0.05

20 16 4 6 <0.02

34 36 9 13

NS

24 18 8 9

NS

Abbreviations as in Table 1.

respect to the degree of change in the pulmonary and systemic resistance. However, when each drug was an- alyzed separately, significant differences between them were apparent with respect to their mode of action on the pulmonary and systemic circulations (Fig. 1).

Discussion

There is no agreement as to what type of drug effect is beneficial in patients with PPH. We have shown that vasodilators decrease systemic and pulmonary vascular resistance in patients with PPH primarily by increasing the cardiac output, as opposed to reducing arterial

Hydralozlne

PRESSURE

45 STROKE ” t ^, VOLUME

m t SYSTEMIC

g 35 i&L RESISTANCE z 25

P

RESISTANCE

Nlfedlpme Amrinone

x

_I& ‘,

*

DL *

dl- *

Ih *

_Ih

x

Ikl

m ‘,

al- *

Ih i6

0

* - p.co.05 0 CONTROL q AFTER DRUG

FIGURE 1. Similarities and differences of the effects of hydralazine, nifedipine and amrinone. All 3 drugs caused similar reductions in sys- temic and pulmonary resistance. Nifedipine was the only drug to sig- nificantly decrease pulmonary artery pressure, hydralazine the only drug that had no significant effect on systemic pressure, and amrinone the only drug that did not increase stroke volume.

pressure, and that an active reduction in the resistance of 1 circulation is often accompanied by a passive re- duction in the other.2 Predominant pulmonary vaso- dilator responses are manifest by a reduction in pul- monary resistance greater than the reduction in sys- temic resistance and is associated with a concomitant decrease in PA pressure.2

Hydralazine has been used in several studies of pa- tients with PPH, with conflicting results. Although initial reports suggested beneficial effects,3*4 recent reports reveal short- and long-term deleterious ef- fects.“p6 Careful scrutiny of the early reports shows that the reduction in pulmonary vascular resistance rarely exceeded the reduction in systemic resistance, suggesting that the major effect of hydralazine was seen in the systemic circulation.3,4 In our series the decrease in systemic resistance from hydralazine exceeded the decrease in pulmonary resistance in all patients and was not associated with a significant decrease in arterial pressure. Thus, the decrease in pulmonary resistance was probably related to the increased cardiac output caused by the reduction in systemic resistance.7

Nifedipine appeared to work differently from hy- dralazine. Although it significantly reduced both sys- temic and pulmonary resistance, it did decrease pul- monary resistance more than systemic resistance in 2 of the 7 patients and decreased PA pressure in all pa- tients. Thus, some patients with PPH may have a pre- dominant pulmonary vasodilator effect from nifedipine. We could not identify any unique characteristics in the 2 patients who had a pulmonary vasodilator effect with regard to baseline PA pressure, cardiac output and systemic or pulmonary resistance.

The effect of amrinone was also distinctive in these patients. Although we anticipated an inotropic effect, the increase in cardiac output that occurred was sec- ondary to an increase in heart rate, as there was no sig- nificant change in stroke volume. Although amrinone might have been more effective if given in larger doses, we are not aware that amrinone exhibits different vasoactive effects that are dose dependent.8 In 1 patient amrinone reduced both PA and systemic artery pressure and resistance without an associated increase in cardiac output, suggesting that it was acting primarily as a

November 1, 1983 THE AMERICAN JOURNAL OF CARDIOLOGY Volume 52 1107

vasodilator in this case. Other studies suggest that amrinone may have strong systemic and pulmonary vasodilator properties. 9~~ The lack of effectiveness of amrinone as an inotropic agent in this group of patients raises the possibility that the chronically pressure- overloaded right heart may not respond to inotropic agents unless accompanied by a substantial reduction in afterload.

Because nifedipine was the only drug to consistently decrease PA pressure, we believe it holds the most promise as an effective vasodilator in patients with PPH, even though beneficial drug responses are only modest, at least acutely. A study of the long-term effects of nifedipine in PPH is warranted. The use of amrinone in PPH needs further investigation.

Acknowledgmenti We thank the Sterling-Winthrop Research Institute, Rennselaer, New York, for supplying amrinone for this study.

References

1. Rich S. Primary pulmonary hypertension. Cardiac Impulse 1981;2:1-6. 2. Rich S, Marilnez J, Lam W, Levy PS, Rosen KM. Reassessment of the

effects of vasodilator drugs in primary puknonary hypertension: guidelines for determinina a oulmonarv vasodilator resoonse. Am Heart J 1983: 105:119-127.- . .

3. Rubln LJ, Peter RH. Oral hydralazine therapy for primary pulmonary hy- pertension. N Engl J Med 1980;302:69-73.

4. Luai-Herrera E. Sandoval J. Seoane M. Blalostorkv D. The role of hv- dralazine therapy for pulmonary arterial hypertension of unknown cause. Circulation 1982;65:645-650.

5. Kronzon I, Cohen M, Winer HE. Adverse effect of hydralazine in patients with primary pulmonary hypertension. JAMA 1982;247:3112-3114.

6. Packer M, Greenberg B, Massle B, Dash H. Deleterious effects of hydral- azine in patients with pulmonary hypertension. N Engl J Med 1982;306: 1326-1331.

7. Franclosa JA, Fischer HA. Right ventricular unloading: lessons from the left. Ann Intern Med 1981;95:647-648.

8. Wynne J, Yalacoff RF, Benottl JR, Curfman GD, Grossman W, Holman BL, Smfth TW, Braunwakf E. Oral amrinone in refractory congestive heart failure. Am J Cardiol 1980;45:1245-1249.

9. Benotti JR, Grossman W, Braunwald E, Davolos DD, Alousl AA. Hemo- dynamic assessment of amrinone. A new inotropic agent. N Engl J Med 1978;299:1373-1378.

10. Hill NS, Rounds S. Amrinone is a pulmonary vasodilator and blunts hypoxic vasoconstriction (abstr). Am Rev Resp Dis 1982;125:270.