hemodynamic effects of nifedipine versus hydralazine in primary pulmonary hypertension

Hemodynamic Effects of Nifedipine Versus Hydralazine in Primary Pulmonary Hypertension

JEFFREY FISHER, MD, JEFFREY S. BORER, MD, JEFFREY W. MOSES, MD,

HARVEY L. GOLDBERG, MD, ANDREAS P. NIARCHOS, MD,

HENDRICKS H. WHITMAN, Ill, MD, and MARGARET MERMELSTEIN, RN

The acute hemodynamic effects of both sublingual nifedipine (N) and intravenous hydralazine (Hy) were studied in 5 patients with primary pulmonary hypertension to ascertain whether the capacity for pulmonary vasodllatation was generalized or drug-specific, and to determine which of the 2 agents had preferential pulmonary vasodilatory effects. For the group as a whole, neither N nor Hy produced changes in heart rate, mean pulmonary capillary wedge or right atrial pressures. Both N and Hy reduced mean systemic arterial pressure (before N 90 f 8 mm Hg, after N 78 f 7 mm Hg, p <O.Ol; before Hy 92 f 11 mm Hg, after Hy 88 f 8 mm Hg, p <0.05), and decreased systemic vascular resistance (before N 1,558 f 845 dynes s cmm5, after N 1,192 f 430 dynes s cmW5, p <0.05; before Hy 1,700 f 415 dynes s cmV5, after Hy 957 f 285 dynes s cmm5, p <0.05). In addition, N administration resulted in an increased cardiac output (before N 4.5 f 2.0 liters/min, after N 4.8 f 2.0 liters/min, p <O.Ol); Hy administration was associated with a more varied effect on cardiac output (before Hy 4.0 f 1.0 liters/min, after Hy 5.3 f 1.8 liters/min, p <O.lO, difference not significant [NS]). Although for the group neither agent decreased mean pulmonary artery pressure (PAP) (before N 51 f 13 mm Hg, after N 44 f 13 mm Hg, NS; before Hy 50 f 15 mm Hg, after Hy 51 f 15 mm Hg, NS) or pulmonary vascular resistance (before N 873 f 458 dynes s cma5, after N 880 f 450 dynes s cmm5, NS;

before Hy 945 f 454 dynes s cmV5, after Hy 715 f 309 dynes s cm -5, NS), 4 of 5 patients had a decrease in PAP after N and 1 had no change, and only 1 of 5 patients had a decreased PAP after Hy administration. Moreover, compared with the changes after Hy administration, PAP declined after N administration (APAP after Hy 0.2 f 9 mm Hg, after N -7 f 9 mm Hg, p <0.05). The PA diastolic to mean PA wedge pressure gradient tended to decrease after N administration (after N -5.0 & 10 mm Hg, afler Hy +4.7 f 7 mm Hg, NS), suggesting more pulmonary vasodilatation after N administration. Moreover, the ratio of pulmonary to systemic vascular resistances was unchanged after N but increased after Hy administration (before N 0.55 f 0.2, after N 0.53 f 0.2, NS; before Hy 0.55 f 0.2, after Hy 0.74 f 0.3, p <0.02), indicating the more balanced vasodilatory effect of N. Two patients were treated chronically with Hy but had intolerable adverse effects; 1 was subsequently treated suc- cessfully with N. A third patient had pulmonary edema (presumably neurogenic) 30 minutes after Hy administration; this patient later died. Another patient has symptomatically improved with chronic N therapy. Thus, N appears to be a more specific pulmonary arterial vasodilator than Hy in acute drug testing; in this small group of patients with primary pulmonary hypertension, N appears to be more ef- ficacious when administered chronically.

(Am J Cardiol 1984;54:848-650)

Recent reports have suggested that hydralazine is clinically useful in reducing pulmonary vascular resistance in patients with primary pulmonary hypertension

From the Division of Cardiology, The New York Hospital-Cornell Medical Center, New York, New York. Manuscript received September 6,1963; revised manuscript received May 5, 1964, accepted May 11, 1964.

Address for reprints: Jeffrey S. Borer. MD. Professor of Medicine, The New York Hospital-Cornell Medical Center, 525 East 66th Street, New York, New York 10021.

(PPH),rp2 although hydralazine, like earlier agents used for therapy in this group,a4 has been associated with adverse effects.7ps Calcium-channel blocking agents, vasodilators that are generally safe and well tolerated in patients with several types of cardiovascular disor- ders, have been shown experimentally and clinically to reduce hypoxic pulmonary vasoconstrictiongJO and, in small series, to reduce abnormal pulmonary vascular resistance in some patients without severe hypoxia.rlJs

848

September 1. 1984 THE AMERICAN JOURNAL OF CARDIOLOGY Volume 54 847

TABLE I Hemodynalmic Measurements Before and After Drua Administration

Control Hydralazine Control Nifedipine

HR (beats/min) 94f 17 (NS) 104 f 16 101 f 12 107 l 15 CO (liters/min) 4.0 f 1.0 po$O. 10) 5.3 f 1.6 4.5 f 1.5 4.8 f 1.6 MAP (mm Hg) 92f 11 68 f 8 90 f 8 76 f 7 PAP (mm Ho) 50f 15 51 f 15 51 f 13 44f 13 PCW‘P (mmTlg) RAP (mm Hg) APAD-PCW (mm ~g) SVR (dynes s cmm5) PVR (dynes s cm+) RJR, Art02sat(%)

9f5 9f3 iN$j 9f4 9f8 104~8 9f6

29f 11 26 f 9 I%{ 22f 11 957 f 285 1,558 f 645 {pNGO.O5) 1,192 f 430 715 f 309 a73 f 458 680 f 450

0.74 f 0.3 0.55 f 0.2 (NS) 91 l 5 90 f 7 (NS) OE 2 :.’ Mixed venous

02sat(%) 52f 11 (p <O.Ol) 62f 11 46 f 28 (NS) 51 f30

Art = arterial; CO = c,ardiac output; HR = heart rate; MAP = mean systemic arterial pressure; NS = not significant; PA = pulmonary arterial pressure; APAD-PCW = the gradient between the pulmonary artery diastolic and mean pulmonary capillary wedge pressures; PCW = pulmonary capillary wedge pressures; PVR = pulmonary vascular resistance; RA resistance; sat = saturation; SVR = systemic vascular resistance.

= right atrial pressure; RJR, = ratio of pulmonary to systemic vascular

We compared the acute hemodynamic effects of nifedipine with those of hydralazine in 5 patients with PPH to determine the relative hemodynamic effects of these agents in PPH and, as a corollary, to determine whether pulmonary vasoreactivity was generalized or drug specific.

Methods

Study group: Five patients with PPH and New York Heart Association functional class III symptoms, who were free from acute intercurrent medical illness, underwent hemodynamic and pharmacologic stuldy at The New York Hospital-Cornell Medical Center from July 1, 1981 through June 30, 1982. These patients had unchanged ventilation-perfusion pulmonary scintigraphy over several months; pulmonary function studies were within normal limits in all but 1 patient, who had a mild decrease in diffusing capacity (which itself was not of sufficient severity to explain the degree of pulmonary hypertension). All 5 patients had normal global left ventricular function indicated by radionuclide cineangiography or 2- dimensional echocardiography. No patient had evidence of intracardiac shunting by arterial and mixed venous blood gas analysis and indicator dilution curves.

Hemodynamic measurements: Medications (including all vasodilators, all P-blocking agents, aspirin, nonsteroidal antiinflammatory agents, narcotic or nonnarcotic analgesics and psychotropic medications) were held for at least 24 hours before the study, and no premeditations were given. The patients were studied after an overnight fast and were administered intravenous dextrose in water at 75 ml/hour. Systemic arterial pressure was continually monitored with a 5Fr can- nula inserted in the femoral artery. Right-sided cardiac catheterization was performed with fluoroscopy using the femoral vein and trip’le-lumen, flow-directed catheter for measurement of right atrial, pulmonary arterial (PA) and PA wedge pressures. Measurements of phasic and mean PA wedge pressure were reproducible within 2 mm Hg during the 15- to 30-minute recording periods. The 0 reference point for hemodynamic measurements was 5 cm deep to the anterior chest wall. Cardiac output was measured using the Fick technique, indocyanine green dye..dilution curve analysis or thermodi- lution; when performed at the same point in any study, these measurements corresponded to within 10% of each other.

Drug administration: After insertion of catheters, baseline hemodynamic measurements were obtained after an equili- bration period of 15 minutes. Baseline measurements were repeated 2 or 3 times at intervals of 15 to 30 minutes, before administration of pharmacologic agents. To avoid experi-

mental bias, the drug to be administered initially was alter- nated from study to study. The initial hydralazine dose was 5 mg, administered intravenously; after a 15-minute hiatus for assessment of the hemodynamic response, an additional 15 mg was administered over 30 minutes. Nifedipine was administered as a sublingual dose of 10 mg. The patient was observed, with continual hemodynamic monitoring to detect any untoward effect, for 15 to 30 minutes before a second lo-mg sublingual dose was administered. Intravenous hydralazine and sublingual nifedipine were used, rather than the clinically used oral forms, in order to shorten the time to peak drug effect, to reduce the time needed for drug washout, and shorten the period of patient immobilization relative to those times after oral administration. The desired pharmacologic effect was the development of a lo- to 20-mm Hg decrease in mean systemic arterial pressure (a clinically appropriate change for evaluation of each drug). To avoid an intolerable or disproportionate reduction in systemic arterial pressure with either drug, no more than 20 mg of each drug was used, and all patients had systemic systolic arterial pressure greater than 100 mm Hg at the initiation of either drug trial. Hemo- dynamic measurements were made every 15 minutes after the initial bolus dose until heart rate, arterial pressure and pulmonary artery pressures had returned to baseline, usually within 1 to 3 hours. The second agent was then administered. The hemodynamic effect for each drug was recorded at the nadir of systemic and pulmonary arterial pressures, which occurred 15 to 60 minutes (average 30) after both hydralazine and nifedipine administration.

Data analysis: Mean femoral artery, right atrial, PA and PA wedge pressures were determined by electronic filtration over 3 respiratory cycles. Derived hemodynamic variables were calculated according to the following formulas: Systemic vascular resistance (SVR) = 80 X (MAP - MRAP)/CO (dynes s cmm5); pulmonary vascular resistance (PVR) = 80 X (MPAP - PCW)/CO (dynes s cm-s); and Rpm, = ratio of PVR/SVR, where CO = cardiac output, MAP = mean arterial pressure, MPAP = mean pulmonary artery pressure, PCW = mean pulmonary capillary wedge pressure and MRAP = mean right atrial pressure.

Responses were evaluated by comparing hemodynamic findings during administration of each drug to the respective values before drug administration for each patient using the paired t test. Group data are expressed as mean f standard deviation.

Results Hemodynamic responses: During the observation

periods (before drug administration and after drug

640 NIFEDIPINE VERSUS HYDRALAZINE IN PRIMARY PULMONARY HYPERTENSION

washout), we observed little variation in the resting systemic and PA pressures. Occasionally, as with mic- turition, both systemic and PA pressures would increase 10 to 20 mm Hg. However, in general, when the patient was at rest and pharmacologic agents were not being administered, the intravascular pressures remained fairly constant.

The hemodynamic response to hydralazine is shown in Table I and Figures 1 and 2. Hydralazine administration produced no statistically significant changes in heart rate, mean PA, mean PA wedge or right atria1 pressures. Hydralazine caused a decrease in systemic vascular resistance and mean arterial pressure. Al- though some patients had a marked increase in cardiac output with concomitant increase in mixed venous oxygen saturation and associated with marked decrease in systemic vascular resistance, hydralazine had a dis- tinctly variable effect on systemic vascular resistance, and no statistically significant increase in cardiac output

e

C H C N

followed drug administration for the group as a whole.

The hemodynamic response to nifedipine is shown in Table I and Figures 1 and 2. Like hydralazine, nifedipine produced no statistically significant change in heart rate, mean PA wedge or right atrial pressures. Also as with hydralazine administration, systemic vascular resistance decreased with nifedipine administration, as did mean arterial pressure, whereas cardiac output increased. Although as with hydralazine, mean PA pressure and pulmonary vascular resistance did not decrease significantly after nifedipine therapy, a distinct trend was noted, with 4 of 5 patients manifesting a decrease in PA pressure after nifedipine (1 of 5, no change); in comparison, only 1 of 5 patients had a decreased mean PA pressure after hydralazine administration. In addition, when changes in mean PA pressure were compared after hydralazine and nifedipine administration, there was a statistically significant relative decrease in

e k \ e

PC.01

C H C N

r

1000 - e

600 -

e $ e

C H C N

FIGURE 1. Left panel, left column, cardiac output increased in 4 of 5 patients with primary pulmonary hypertension given intravenous hydralazine (H); however, there was no statistically significant increase in cardiac output for the group, although for individual patients there was often a marked increase in cardiac output. Left panel, right column, cardiac output increased in 5 of 5 patients after nifedipine (N), resulting in a statistically significant increase for the group, although for any individual the Increase was modest. Right panel, mean arterial pressure decreased after hydralazine (left) and nifedipine (right) administration. C = control: NS = not significant.

FIGURE 2. Left, mean pulmonary artery pressure for the group did not change after either hydra&be (H) (left) or nifedipine (N) (rl9ht) administration. However, 4 of 5 patients had a decrease in pulmonary artery pressure after nifedipine administration (vs only 1 of 5 after hydralazine). However, nifedipine led to a decrease in pulmonary artery pressure compared with hydralazine (mean pulmonary artery pressure after nlfedipine -7 f 9 mm Hg, mean pulmonary artery pressurs after hydralazine 0.2 f 9 mm Hg, p <0.05). Rl@t, there were no statistlcally significant changes in pulmonary vascular resistance after either hydralazine (left celumn) or nlfedipine (rl9ht column) administration. C = control; NS = not slgnlflcant.

September 1, 1984 THE AMERICAN JOURNAL OF CARDIOLOGY Volume 54 649

mean PA pressure after nifedipine administration (postnifedipine change -7 f 9 mm Hg, posthydralazine change +0.2 f 9 mm1 Hg, p cO.05). In addition, the gradient between the PA diastolic and mean PA wedge pressure tended to decrease after administration of nifedipine, but to increase with hydralazine (postnifedipine change -5.0 f 10 mm Hg, posthydralazine change +4.7 f 7 mm Hg, NS), indicating relatively greater pulmonary vasodilatation associated with nifedipine.

The more balanced vasodilatory effect of nifedipine than of hydralazine on the systemic and pulmonary vascular beds was reflected in the Ri,& ratio after drug administration (Table I), which increased significantly after hydralazine administration, indicating preferential systemic vasodilatation, but was unchanged after nifedipine.

Clinical response: Surviving patients have been followed for 18 to 30 months since the initial catheterization study. One patient had clear relief of dyspnea within 15 minutes after the administration of sublingual nifedipine. In this pat&t, mean PA pressure decreased 22 mm Hg after nifedipine, and pulmonary vascular resistance decreased !j3%. This patient also had pulmonary vasodilatatio:n (although less marked) with hydralazine, but has been treated chronically with oral nifedpine and has maintained marked symptomatic improvement during a a-year observation period. Two other patients initially were treated with oral hydralazine, but they either lacked symptomatic improvement (1 patient who subsequently died) or had intolerable adverse effects (dizziness in 1 patient); the latter patient then noted mild symptomatic improvement with oral nifedipine. The fourth patient was not treated with vasodilators and died approximately 1 year after the initial study of progressive right-sided cardiac failure. Repeat catheterization study was performed in this patient 13 months after the initial study reported herein, and revealed increased PA pressure and resistance as compared with the initial study (PA pressure 60 vs 42 mm Hg, PVR 2,455 vs 675 dynes s cmm5); although this patient initially responded to nifedipine with a 16% decrease in PVR, no response was noted during the second study. The fifth patient had pulmonary edema after the administration of hydralazine. The PA wedge pressures before the development of pulmonary edema was normal, although the R, to R, ratio exceeded unity after thle administration of hydralazine. The cause of the puhnonary edema is unclear, but in view of the wedge pressure reading, it was most likely neurogenic.

Ubiscussion

These data suggest that, during assessment of acute hemodynamic and symptomatic drug effects in the laboratory in patients with PPH, nifedipine has a more pronounced effect on the pulmonary arterial or arteri- olar vasodilatory capacity than does hydralazine. In addition, nifedipine appears to have a more balanced effect on pulmonary a.nd systemic vascular beds than does hydralazine.

Although a clinically important reduction in PA pressure (as assessed by symptom relief with acute administration) occurred only in 1 of 5 patients after nifedipine administration, chronic nifedipine administration was clinically effective in an additional patient who had intolerable adverse effects with hydralazine. The study group clearly is too small to permit definitive conclusions about the presence of generalized pulmonary vasoreactivity or drug specificity. However, be- cause only 1 patient had pronounced hemodynamic benefit from both agents and the acute reduction in pulmonary vascular resistance was significantly greater after nifedipine than after hydralazine, it seems likely that the effect of nifedipine may be more specific for the pulmonary vascular bed than is that of hydralazine in PPH. Our findings with hydralazine contradict those of Rubin and Peter,1 who found hydralazine to be an effective pulmonary vasodilator in patients with PPH, but are consistent with those of Lupi-Herrara et al,” who found more variable results. The small size of patient groups in all these studies may account for apparent differences in results but, taken together, suggest, at best, a modest therapeutic response to hydralazine in patients with PPH.

We observed 1 major untoward event, pulmonary edema, in response to intravenous hydralazine administration, which probably was neurogenic in origin. Several other investigators also have reported deleterious hemodynamic responses with hydralazine.‘s These findings indicate the need for extreme caution when testing the efficacy of intravenous hydralazine in patients with pulmonary hypertension, with aware- ness of the potential for sudden cardiovascular catastrophe.

Our results with nifedipine suggest that, as compared with other agents thus far assessed, this drug may be relatively useful in the treatment of patients with PPH. The extent to which our results with nifedipine can be extrapolated to other calcium-channel blocking agents is unclear. Experimental data suggest that among such agents, nifedipine has the most specific pulmonary vasodilatory effects. l3 Clinically, in the small series available, verapamil did not have beneficial effects in the majority of patients treated,14 and untoward adverse effects attributable to the negative inotropism of verapamil were noted.14 The clinical experience with diltiazem is limited,15 and conclusions regarding clinical efficacy in pulmonary hypertension cannot as yet be drawn.

The results of the present investigation, indicating the relative efficacy of nifedipine in mitigation of PA hypertension in patients with PPH, must be evaluated with recognition of certain aspects of study design. Our protocol involved the administration of intravenous and sublingual drugs over several hours and, although baseline hemodynamic measurements were reproducible, the potential confounding effects of drug-drug interactions cannot be ruled out. In addition, evaluation of these drugs was performed with resting hemodynamic measurements only, so that the comparative effects during exercise, when symptoms are most marked, are

650 NIFEDIPINE VERSUS HYDRALAZINE IN PRIMARY PULMONARY

unknown. Further study of the effects of nifedipine therapy on PA hypertension, taking into account the limitations of the present assessment, is indicated.

1.

2.

a.

4.

6.

6.

References

Rubin LJ, Peter RH. Oral hydralazine therapy for primary pulmonary hypertension. N Engl J Med 180;302:69-73. Lupl-Herrara E, Sandoval J, Seoane M, Bialostozky D. The role of hydralazine therapy for pulmonary arterial hypertension of unknown causes. Circulation 1982;65:645-650. Belman MJ Tlep BL, Westinghousa WD. lsoproterenol in pulmonary hypertension [letter]. N Engl J Mad 1978;298:51-52. Elkayam U, Frishman WH, Yoran C, Strom J, Sonnenbllck EH, Cohen MN. Unfavorable hemodynamic and clinical effects of isoproterenol in primary pulmonary hypertension. Cardiovasc Med 1978;3:1177-1180. Rubino JM, Schroeder JS. Diazoxide in treatment of primary pulmonary hypertension [letter]. Br Heart J 1979;42:362-363. Cohen ML, Kronzon I. Adverse hemodynamic effects of phentolamine in

7.

8.

9.

10.

11.

12.

13.

14.

15.

primary pulmonary hypertension. Ann Intern Med 1981; 95591-592. Packer M, Greenberg B, Massle B, Dash H. Deleterious effects of hydralazine in patients with pulmonary hypertension. N Engl J Mad 1982;306: 1326-1331. Kronron I, Cohen M, Wlner HE. Adverse effect of hydralazine in patients with primary pulmonary hypertension. JAMA 1982;247:3112-3114. Slmmonneau G, Escourrou P, Duroux P, Lockhart A. Inhibition of hypoxic pulmonary vasoconstriction by nifedipine. N Engl J Med 1981; 304: 1.5A7-1!iR5 - - _ _ _ _. Kennedy T, Summer W. Inhibition of hypoxic pulmonary vasoconstriction bv nifedioine. Am J Cardiol 1982:50:864-868. damerlil F, Albert1 E, Klugmanh S, Salvl A. Primary pulmonary hypertension: effects of nifedipine. Br Heart J 1980;44:352-356. McLeod AA, Wise JR Jr, Daly K, Jewltt DE. Nifedipine in primary and secondary pulmonary hypertension (abstr). Circulation 1981;64:180. Young TE, Lundqulst LJ, Chesler E, Weir EK. Comparative effects of ni- fadipine, verapamil, and diltiazem on experimental pulmonary hypertension. Am J Cardiol 1983;51:195-200. Landmark K, Refsum AM, Slmonsen S, Storstein 0. Verapamil and pulmonary hypertension. Acta Med Stand 1978;204:299-302. Kambara H, Fujlmoto K, Wakebayashi A, Kawai C. Primary pulmonary hypertension: beneficial therapy with diltiazem. Am Heart J 1981;lOl: 230-231.

hemodynamic effects of nifedipine versus hydralazine in primary pulmonary hypertension

Documents