982 JACC Vol. 243. No. 4 October M94982-8

MARC J. SEMIGRAN, MD, BAR

B. TAYLOR THOMPSON, MD, WARREN M.

MRHAEL A. FIFER, MD, FACC

&tin, Mwsachusetls

As heart transplantation has evolved as a treatment for refractory heart failure, it has been recognized that early postoperative right heart failure is a common occurrence associated with substantial morbidity and mortality (l-3). Elevations of transpulmonaty pressure gradient and pulmo- nary vascular resistance are common in patients with chronic left heart failure, and their persistence through the peritrans- ptantation period can lead to right ventricular failure, partic- ularly if prolonged ischemia has occurred during harvesting and transportation of the donor organ. Severe, fixed elevation of transpuhuonary pressure gradient or pulmonaty vascular resistance is therefore considered a contraindication to ortho. topic heart transplantation. The immediate response of the pulmonary circulation to intravenous vasodilators, such as nitroprusside (4) amrinone (5) and prostaglandin E, (6), has

From the Cardiac and Pulmonary Units, Departments uf Medicine, Respi- ratory Therapy and Anesthesia, Massachusetts General Hospital and Harvard Medical Schuo!, Boston, Mass;\chusetts. This study was supported in part by Grant Hf.43297 to Dr. Zapol from the National Heart, Lung, and Blood institute. N-&mal fnstitutrs of Health, Bethesda, Maryland. ft was presented in art at the 42nd Annual scientific Win of the American College of Cardiol- ogy, Anaheim, California. March, 1993.

1 h(anusaipl received January 281% rev&d manuscript received April 14,

: Dr. Michael A. Firer, Cardiac Unit, Massachu- setts Qmsal Hospital, WACC 478 15 Parkman Street, Boston, Massachusetts 02114.

81998 by the American Colkge of Cardiology

decreased witb nitreprussid I patient who did not have vascular resistance with nit

of the 16 patients, incladia~ te dec~a~ In pM~rn~~a~ but did with nitric oxide,

been used to identify patients at acceptable risk for transplan- tation, but the effectiveness of these agents in predicting the reversibility of pulmonary vasoconstriction is limited by their systemic hypotensive effects.

Nitric oxide, an endothelium-derived relaxing factor (7) produced from L-arginine by endothelial nitric oxide syntbase (8) has an in viva half-life of 111 to 130 ms (9). Inhaled nitric oxide has been shown to reverse pulmonary vasoconstriction in hypoxic lambs (10) and pediatric patients with congenital heart disease complicated by pulmonary hypertension (11). In adults with pulmonary hypertension due to the adult respiratory distress syndrome, inhaled nitric oxide decreases pulmonary vascular resistance without altering systemic arterial pressure (l&13). It has been demonstrated that in patients with pulmo- nary vascular disease; inhaled nitric oxide also causes a reduc- tion in pulmonary vascular resistance (14,15), accompanied by an improvement in right ventricular performance, as indicated by an increase in stroke volume despite a decrease in right ventricular end-diastolic pressure (14).

To assess the potential role of inhaled nitric oxide in identifying reversible pulmonary vasoconstriction in patients with severe heart failure referred for heart transplantation, we compared its effects on transpulmonary pressure gradient, pulmonary vascular resistance and systemic arterial pressure with those of intravenous nitroprusside.

0735-1097/94i$7.00

JACC Vol. 24, No. 4 October 1994:949-8 I_ EMODYNAMK EFFECTS OF lNMALED NlTRlC OXIDE 983

Outcome

1 Siw DCM 1v 0.17 Died, not listed for transplantation 2 49/M DCM 111 0.25 Transplantation performed 3 .77/M DCM IV 0.12 Transplantation performed 4 50/M CAD IV 0.23 Died 4 weeks after transplantation

(renal, hepatic failure) 5 S4lM CAD IV 0.13 Transplantation performed 6 37/M CAD 111 0.18 Awaiting transplant 7 SO/M CAD 111 0.15 Awaiting transplant 8 W/M CAD 111 0.17 Awaiting transplant 9 S2iM DCM 111 0.10 Transplantation performed

IO 57/F HCMIDCM IV o.Kl Transplantation performed II 49/M DC 111 0. I7 Transplantation performed 12 t13ik4 CA 111 t1.14 CABG I3 4VM DCM 111 II.27 Awaiting transplant I4 45/M DCM SV 11SlY Died awaiting transplant 15 S’NF DCM 111 0.10 Awaiting transplant I6 53/M DCM III 0.17 Awaiting transplant

Mean Sf 0. I6

SEM 2 0.02

CABS = coronary artery bypass grafting: CAD = curmutry artery disease; KM = dilated cardiomyopathy; Diagn =

diagnosis; F = female; HCM = hypertrophic cardiomyopathy: LVEF = left ventricular ejection fraction; M = male;

NYHA class = New York Heart Association functional class: PI = patient.

atients. The study group included 13 men and 3 I 2 years) with chronic New Yor

A.ssociation class III or IV heart failure who were referred to the Massachusetts General Hospital for consideration of heart transplantation (Table 1). The cause of heart failure was dilated cardiomyopathy in nine patients, coronary artery dis- ease in six and progression of hypertrophic cardiomyopathy to ventricular dilation and systolic dysfunction in one patient. No patient had a history of primary pulmonary disease, and the results of pulmonary function testing were consistent with chronic left heart failure, showing a mild restrictive pattern and a mild decrease in diffusion capacity (16). All patients were treated with digoxin, diuretic drugs and vasodilators, and three (Patients 1,9 and 15) were treated with amiodarone for atrial or ventricular arrhythmias. Radionuclide left ventricular ejec- tion fraction was 0.16 f 0.02 (range 0.09 to 0.27). Patients 1,2 and 14 had chronic atrial fibrillation, and Patient 15 had a ventricular paced rhythm; the remaining 12 patients were in sinus rhythm. The study protocol was approved by the Sub- committee on Human Studies of the Massachusetts General Hospital, and written informed consent was obtained from all patients.

emodynamic measurements. Digoxin, diuretic drugs and vasodilators were discontinued 12 to 24 h before catheteriza- tion, and amiodarone therapy was continued. No premedica- tion was administered. Right heart catheterization was per- formed by way of the internal jugular vein with a triple-lumen balloon-tipped catheter. Systemic arterial pressure was mea- sured from a radial artery cannula or from the side arm of a femoral artery introducer.

The following hemodynamic variables were recorded: heart rate, right atrial pressure, pulmonary artery pressure, pulmo- nary capillary wedge pressure and systemic arterial pressure. Cardiac output was determined by the Pick oxygen technique. Q,xygen consumption was measured with an consumption monitor (Waters Associates). Derived hemody- namic variables were calculated using the following formulas: Cardiac index (liters/min per m”) = Cardiac output/Body surface area; Transpulmonary pressure gradient (mm Hg) = ivlean pulmonary artery pressure - Pulmonary capillary wedge pressure; Stroke volume index (ml/m2) = Cardiac index/Heart rate; Right ventricular stroke work index (g-m/m2) = (0.0136) (Mean pulmonary artery pressure - Right atrial pressure)(Stroke volume index); Left ventricular stroke work index (g-m/m’) = (O.O136)(Mean systemic arterial pressure - Pulmonary capillary wedge pressure)(Stroke volume index); Systemic vascular resis- tance (dynesscm-‘) = 8O(Mean systemic arterial pressure - Right atrial pressure)/Cardiac output; Pulmonary vascular resis- tance (dynesscm-“) = 80(Transpulmonary pressure gradient)/ Cardiac output.

Nitric oxide delivery. Nitric oxide gas (800 ppm in nitrogen [NJ) (Airco) was mixed with room air with the use of a standard low flow blender (Bird Blender) and then titrated with 100% oxygen using standard high flow rotameters (Timeter Instruments) to achieve a mixture of >90% oxygen with 20,40 or 80 ppm nitric oxide and the remhinder Np The mixture was then introduced into a non-rebreathing circuit consisting of large-bore aerosol tubing and a modified con- tinuous positive airway pressure mask (Respironics Inc.). The inspired concentration of nitric oxide and nitrogen dioxide was measured by chemiluminescence (17) (model 141%

984 SEMIORAN ET AL. HEMODYNAMIC EFFECTS OF INHALED NITRIC OXIDE

JACC Vol. 24, No. 4 October 1994:952-8

T&le 2 Acute Hcme@amic Effects of Oxygen, Inhaled Nitric Oxide and Nitroprusside

NO (aa ppm) Nitroprussidc Baseline 02 +02 0, to2

Heart rate (bcats/min) 93 + 6 9026 96 t 6 9926 98 ‘t 7 h(ean systemic arterial pressure (mm Hg) PAZ2 87 t 2 89 z 2 YO’2 68 _c I’?$

Right atrial pressure (mm Hg) 922 10r2 852 9t2 -I-+ I”?$ Mean putmonary artery pressure (mm Hg) 38 -c 3 37 2 3 39 ‘+ 2 38 t 4 19 5 3*+* Pdrnonaty capillary wedge pressure (mm Hg) 2523 26 t 2 32 r 2*$ 26 5 3 9 ? 2*‘rl

Transpuhnonary pressure gradient (mm Hg) 14 4 2 11 t I* 7%-l** II ?2* IO c I*? Cardiac in&x (IiWmin per m’) 2.1 t 0.2 2.1 c 0.1 2.3 r 0.2 “._ 77+0? - ._ 2.Y t 0.3’1% Stroke volume index (ml/m’) 23 -c 2 24 + 2 24 I? 2 *_ 73 * P _ 7 30 2 4 RV stroke work index (g-m/m’) 921 921 IO c 1 921 72 Ii LV struke work index (g-m/m”) 19 ” 2 20 ” 2 I9 22 ?I) _+ 2 23 rt 2*? Systemic vascular resistance (dynes%*m -s) 1,648 + 1,493 1,649 + 119 1,625 t I21 I .692 + 144 1,121 t 14lW$

Pulmonary vascular &stance (dynes-sem “) 323 + 64 256 + 4l* I39 ?I I.$*$ 264 C! 49” I69 ? 31)4t$ PVWSVR 0.19 L 003 0. IS f 0.02 0.09 It 0.01 “I 0. IS I? 0.03 0.15 L o.ot Oxygen saturation (a) 94 c 2 9YzI I* 992 I* lot) 2 1* YY”_ 1’

*p < 0.05 wsus baseline. tp < O.(1S nitroprussidc versus nitric oxide (NO). $p < 0.05 versus previous period uloxygcn (0,). Data are prcscnted irs nlcnn value -C SEM. LV 1 left ventricular: PVR = pulmomuy vascular rcsistancc; RV = right ventricular; SVR = systemic v;rsWlar resistnncc.

ThermoEnvironmental Instruments Inc.) and that of oxygen by polarimetry (Hudson Oxygen Meter). The total gas flow rate was maintained at 45 liters/min, a level that reduced the nitric oxide residence time within the breathing circuit and, hence, the time for the oxidation of nitric oxide to nitrogen dioxide.

Nitrogen dioxide was not measurable in the expired gas at any dose of nitric oxide. Exhaled gases were scavenged and dis- carded to the atmosphere. Blood methemoglobin levels were determined spectrophotometrically (18) at baseline and during the inhalation of 80 ppm nitric oxide.

Study pmtucol. All measurements were made with the patient supine and wearing a face mask. Hemodynamic data were obtained during a baseline period with the patient inhaling room air; during inhalation of >90% oxygen; during inhalation of 20, 40 and 80 ppm nitric oxide in addition to >!#I% oxygen; during another period of >90% ovgen; and during the intravenous administration of nitroprusside in ad- dition to the inhalation of z+JO% oxygen. We administered >9O$G oxygen during nitroprusside infusion to avoid hypox- emia due to intrapulmonary shunting. To allow valid compar- isons among baseline, nitroprusside and nitric oxide periods, we administered >90% oxygen throughout the study protocol. The dosage of nitroprusside was titrated upward until a systolic arterial pressure of 85 mm Hg, a mean pulmonary artery pressure of 20 mm Hg or a maximal dose of 500 @min was achieved. Hemodynamic measurements were obtained 5 min after the dose of nitric oxide was altered or 5 min after the desired dose of nitroprusside was reached. The total duration of nitric oxide administration to each patient was -40 min.

Statistics. Results are expressed as mean value -C SEM. A two-way analysis of variance with subsequent comparisons of group means by the Newman-Keuls test was used for compar- isons among the l.atment periods of room air, oxygen, 80 ppm nitric oxide with oxygen and nitroprusside with oxygen, as well as for evaluation of the dose response to 0 to 80 ppm of nitric oxide. Preoperative and postoperative hemodynamic variables

in those patients who underwent transplantation were com- pared by paired Student t testing. A linear regression was used to compare the changes in pulmonary vascular resistance produced by nitric oxide with those observed after transplan- tation. A p value < 0.05 was considered sigtlificant.

Mrememts, Right atrial pressure was elevated as were mean pulmonary artery pressure at

38 -C 3 mm Hg and pulmonary capillary wedge pressure at 25 2 3 mm Hg (Table 2). Cardiac index was depressed at 2.1 Ir 0.2 liters/min per m’. Pulmonary vascular resistance was elevated at

323 t 64 dynesscm-“.

Effects of > oxygen. During inhalation of >90% oxy- gen, arterial oxygen saturation increased from 94 Z!I 2% to 99 I- i% (p < 0,05), transpulmonary pressure gradient decreased from 14 2 2 to I1 2 1 mm Hg (p < O.OS), and pulmonary vascular resistance decreased from 323 + 64 to 256 2 41 dynesscm-” (p < 0.05) (Table 2). There was no effect on mean systemic arterial pressure or systemic vascular resistance. There were no significant differences in hemodynamic variables between the two >90% oxygen periods.

Effects of nitric oxide. With the Jddition of 80 ppm nitric oxide to >90% oxygen, mean pulmonary artery pressure was unchanged, but pulmonary capillary wedge pressure increased from 26 ,C 2 to 32 2 2 mm Hg (p < 0.05) (Table 2). An increase in pulmonary capillary wedge pressure ~5 mm Hg was observed in 9 of the 16 patients (Patients 1 to 3,5 to 8, 10, 16). Transpulmonary pressure gradient decreased from 11 2 1 to 7 +- 1 mm Hg (p < 0.05), and pulmonary vascular resistance decreased from 256 + 41 to 139 I!I 14 dynesscan-” (p e 0.05, Fig. 1A). Cardiac output, mean systemic arterial pressure and systemic vascular resistance were unchanged. The ratio of pulmonary vascular resistance to systemic vascular resistance

i

l@we 2. Relation between dose of nitric oxide and pulmonary vascular rcsistancc (PVR) IA) and pulmonary capillary wedge pressure (PCWP) (B). Data are exprcsscd as mean value 5 SEM. ‘rp c 0.05 versus oxygen alone. #p < 0.05 versus 20 and 40 ppm nitric oxide. PPM = parts per million hy volume.

1. A, Pulmonary vascular resistance (PVR) during oxygen itdministrarion and after the addition of 80 ppm nitric oxide in each of the 16 patients. B, Pulmonary vascular resistance during oxygen administrA_m and with the maximal dose al nitroprnsside in each uf the I6 patients. The mean value ( and SEM are also shown.

decreased from 0,15 t 0.02 to 0.09 t 0.01 with nitric oxide (p ==I 0.01).

The effects of nitric oxide dosage on pulmonary vascular resistance and pulmonary capillary wedge pressure are shown in Figure 2, A and B, respectively. The maximal reduction of pulmonary vascular resistance was seen at 80 ppm, although an effect was observed with doses of 20 and 40 ppm. Pulmonary capillav wedge pressure increased with a dose of 20 ppm of nitric oxide; there was no further significant increase with doses of 40 or 80 ppm.

Effects uf nilropmsside. With the addition of nitroprusside (214 t, 43 Fg!min) to >90% oxygen, mean pulmonary artery pressure decreased from 38 f 4 to 19 2 3 mm Hg (p -C O.Ol), and pulmonary capillary wedge pressure decreased from 26 4 3 to 9 2 2 mm Hg (p -=c 0.01) (Table 2). Transpulmonary pressure gradient did not change, but cardiac index increased

frown 2.2 2 0.2 to 2-Y It 0.3 litcrs/min per n? (p < 0.05), and pulmonary vascular resistance decreased from 264 4 49 to 169 2 30 dynesscm ” ’ (p -C 0.05) (Fig. 1B). Mean systemic arterial pressure decreased from 90 + 2 to 68 + I mm Hg (p < 0.01). The dosage of nitroprusside was limited to <SOtI pgfmin by a decrease in systemic arterial pressure in 7 of the 16 patients (Patients I,5 and 10 to 14). The ratio of pulmonaq vascular resistance to systemic vascular resistance did not

change. Comparison of the erects of nitric oxide and nitruprusside,

Kight atrial, mean pulmonary artery and pulmonary capillary wedge pressure were all lower during administration of nitro- prusside than during administration of nitric oxide. However, both transpulmonary pressure gradient and pulmonary vascu- lar resistance decreased to a greater extent with nitric oxide than with nitroprusside. Mean systemic arterial pressure was lower with nitroprusside, as was systemic vascular resistance. The ratio of pulmonary vascular resistance to systemic vascular resistance was lower with nitric oxide than with nitroprusside.

Side effects. No side effects were observed during nitric oxide or nitroprusside administration. Despite the increased pulillo~dry capillary wedge pressure with nitric oxide, arterial oxygen saturation remained >95% in all patienrs, in part

986 SEMIGRAN ET AL. HEMODYNAMIC EFFEC’l5 OF INHALED NITRIC OXIDE

lO@O

rr 3. Pubnonery vascular reristancc (PVR) at the time of initial evaluation (PRE) and 1 week postopcrativcly (1 WK POST) in the

tients who have undergone heart wuwplantation. The mean ) and SEM are also shown.

because of concomitant administr;ltion of XH.l% oxygen. Mct- hemoglobin Ievvcls did not increase to >1.5% in any patient

during nitric oxide administration. Clfnfenl outcome. In 13 of the 16 patients studied, pulmo-

nary vascular resistance decreased to <200 dynesscm-’ with both nitroprusside and nitric oxide. In Patients 1 and 14, neither agent lowered pulmonaty vascular resistance to <200 dynessunnms. These patients were thought not to be

suitable transplant recipients on the basis of these data, and they died of progressive heart failure. Patient 9, whose pulmo- nary vascular resistance decreased to 240 dynesscm-’ with nitroprusside and to 200 dynes-srm’-s with nitric oxide, under- went successful transplantation; his pulmonary vascular rcsis- tance was 146 dynes-s-cm-” 1 week postoperatively.

In all, seven study patients have undergone transplanta- tion. Pulmonary vascular resistance decreased from 328 -c 104 dynes%*m-s preoperatively to 157 z!z 18 dyness-cm+ 1 week after transplantation (p C 0.05) (Fig. 3). There was no

I change 4 weeks after transplantation. The decrease in pulmonary vascular resistance observed during inhalation of fJ0 ppm nitric oxide predicted the decrease measured 1 week after transplantation (Fig. 4A), as did the decrease in pulmo- nary vascular resistance observed with nitroprusside (Fig. 4B). ‘Lvo patients (Patients 10 and ll), who had the highest preoperative pulmonary vascular resistance among those undergoing transplantation, were treated for early postopera- tive right heart failure with intravenous prostaglandin E, for the 1st 48 and 30 h, respectively. There were no deaths due to postoperative right heart failure,

iscussion Elevation of left atrial pressure in chronic heart failure is

associated with an obligatory increase in pulmonary artery pressure to maintain a pressure gradient for forward flow in the pulmonary circulation. Further elevation of pulmonary artery pressure results from pulmonary vasoconstriction. The

JACC Vol. 24, No. 4 October 1994:982-B

dl 0

0 20 40 80 80 100

% DECREASE IN PVR WllM NRROPRUSSIDE

@we 4. Correlation between the decrease in pulmonary vascular resistance (PVR) observed during inhalation of nitric oxide (A) or infusion of nitroprussidc (B) and that seen I week postoperatively in the seven patients who have undergone heart transplantation. PPM = parts per million by volume.

presence of pulmonary hypertension is of importance to pa- tients undergoing heart transplantation because it is a risk factor for premature death in the postoperative period (1). The right ventricle of the donor heart undergoes ischemic injury during the harvesting and implantation procedures, making it particularly vulnerable to acute dysfunction and failure when confronted with an increase in afterload (19). Thus, patients with a fixed elevation of pulmonary vascuilar resistance are generally excluded as candidates for heart transplantation because of a high early postoperative mortality rate (2,3,20).

Pulmonary hypertension caused by either an elevation in

JACC Vol. 24, No. 4 October 199.1:982-8

SEMlOM ET XL. NEMODYNAMiC EFFECTS OF INWSD NITRIC OXIDE 987

ssure or arteriolar vasoconstr~ctio~

his observation 91 utilize vasodilators such as nitro

at patients whose pulmo- nary vascular resistance decreased to S-200 dynespcmm5 dur- ing nitroprusside infusion before transplantation had a mor- tality rate of oniy 4% at 3 months after tra~sp9a~tat~o~. Bn

contrast. those whose pulmonary vascular resistance could not be reduced to ~200 dynepscm-’ or could be so re at the expense of systemic ~~ypoteas~o~ bad a 3-mon ity of 41% and 28%, r

er vas~~d~9ators have ate pulmonary vasoreactivity hypotension. ~rostag~a~di~ E, also caused ~i~~~~t~i~g systemic 9iy~ote~~s~on at dosages required for adequate pulmonary v~is(~di9at~o~ ia patients with heart failure (6). Adcnosine decreased pulmonary vascular resistance witbout decreasing systemic arterial pressure in patients M~~der~oi~~8 evasuation for heart transplantation, but the degree of baseline pu9monary vasoconstriction of patients in this study was not in the range cons

E circulations. Nitric oxide, an cndothelium-derived relaxation factor, causes vascular smooth muscle cell relaxation by acti- vating guanylate cyclasc, leadiag to an increase in ~ntracel~ulaF cyclic guanosine monophosphate (GMP) and reduction of intracellular calcium concentration (24,25). Because nitric oxide is rapidly bound to (26) and inactivated by (27) hemo- globin, inhaled nitric oxide might be expect be a selective pulmonary vasodilator. Previous studies in ic sheep (IO). patients with primary pulmonary vclscular disease (15) and patients with pulmonary hypertension that was secondary to congenital heart disease (91) or chronic obstructive lung disease (28) or that persisted after cardiopulmonary bypass (29,30) have shown significant reductions in pulmonary vascu- lar resistance, without alterations in systemic arterial pressure, in response to nitric oxide at dosages ranging from 20 to 80 ppm.

The results of the present study extend to patients with chronic left heart failure the observation that nitric oxide decreases pulmonary vascular resistance without reducing sys- temic vascular resistance or mean systemic arterial pressure. The major effect of nitric oxide was a decrease in transpulmo- nary pressure gradient associated with an increase in pulmo- naly capillary wedge pressure without an equivalent increase in pulmonary artery pressure. In contrast, nitroprusside, a nitric oxide donor (31), caused a decrease in pulmonary vascular resistance associated with an increase in cardiac output with- out a change in transpulmonary pressure gradient; although pulmonary artery pressure decreased with nitroprusside, there was a corresponding decrease in pulmonary capillary wedge pressure.

Furthermore, nitroprusside had, as expected, systemic he-

reducing mean syst

ic vascular ~~~~sta~~e

n, while having a be tr~c~~ar performance through afterload reduction of the left ventricle, may have limited the dose of nitropmsside adminis. tered to the p~~rn~~a~ c~rc~lat~oa. Ths, the ratio of pulmo- nary vascular reSiStanCe to systemic vascular resistance did not change with nitroprusside, whereas it decreased by 40% with nitric oxide, demonstrating the selectivity of ~~baled nitric

sure observed with doubtedly reflects an in-

crease in left VentricM~ar end-diastolic pressure. Similarly, Haywood et al. (23) observed an increase in pulmonary

ressure associated with a decrease in pulmo- ring infusion of adellosine in pa- here are several possible mecha-

nisms for our observation: 9) There may have been increases in left ventricular end-systolic and end-diastolic volumes due to a negative inotropic effect of nitric oxide. The demonstration by Finkel et al. (32) that inhibitors of nitric oxide synthase reverse the ability of pro-inflammatory cytokines to decrease contrac- tility in hamster papillary muscle supports such a negative inotropic etiect. 2) Left ventricular preload may have increased in response to improved right ventricular pump function. These cllanges would be associated with an increase in left ventricular end-diastolic volume and pressure. Because strok.e volume was unaltered by nitric oxide in our study, this expla- nation is uniikeiy. 3) Left ventricular diastolic function may have been impaired without a change in Icft ventricular end-diastolic volume. It seems unlil<ely that an agent that increases intracellular cyclic GMP and decreases intracellular calcium would have a direct adverse effect on myocardial relaxation. However, if inhalation of nitric oxide causes dila- tion and increased turgor of the coronary circulation, it could impair left ventricular compliance by this mechanism (33). Because nitric oxide is rapidly inactivated by hemoglobin (27), any effect of nitric oxide on the left ventricular myocardium or the coronary circulation must be mediated by a metabolite or carrier molecule that preserves its biologic effect: Evidence has been presented for the existence of such molecules (34). Whatever the mechanism, the increase in left ventricular filling pressure seen during nitric oxide administration may limit its role to that of a diagnostic rather than a therapeutic agent in patients with severe heart failure.

Clinical implications. Although controversy remains with regard to the level of pretransplantation pulmonary vasocon- striction beyond which the risk of right heart failure after transplantation is excessive, the study of Costard-llckle and Fowler (4) has established an acceptably low risk in patients whose pulmonary vascular resistance is 1200 dynespcm ’ either at baseline or during administration of a dose : nitroprusside that does not induce systemic hypotension. in our study, 13 of 16 patients had this level of pulmonV vasodilation without systemic hypotension with either p:‘~ic

988 SEMIGRAN ET AL. HEMODYNAMK EFFECR3 OF INHALED NITRIC OXIDE

JACC Vol. 24, No. 4 October 1994:98,?-8

oxide or nitroprusside; 2 patients had pulnionary vasoconstric- tion that was not reversed by these agents; 1 patient (Patient 9) did not achieve the specified degree of pulmonary vasodilation with nitroprusside because administration of this agent was limited by systemic hypotension. His pulmonary vascular resis- tance did decrease satisfactorily with nitric oxide; he under- went heart transplantation without postoperative right heart failure and had a subsequent early decrease in pulmonary vascular resistance. Nitric oxide may thus have an important role in identifying potential heart transplant recipients in whom reversal of pulmonary vasoconstriction by nonselective vasodilators such as nitroprusside is limited by systemic hypo tension. Further studies enrolling greater numbers of patients with pulmonsry vasoconstriction that reverses with nitric oxide who ultimately undergo heart transplantation are necessary to test this hypothesis. In the seven patients who have thus far received a heart transplant, the decrease in pulmonary vascular resistance observed with inhaled nitric oxide during their initial evaluation did predict the decrease observed after transplan- tation.

Summary. Inhaled nitric oxide is a selective pulmonary V ilator in patients with severe chronic heart failure. The selectivity of inhaled nitric oxide for the pulmonary circulation offers a potential advantage over nonselective vasodilators such as nitroprusside in the identification of reversible pulmo- nary vasoconstriction in potential heart transplant recipients. Nitric oxide increases left ventricular filling pressure in pa- tients with severe heart failure by an unknown mechanism.

ferences I. Griepp R, Stinson E, Dung E, Clark D, Shumway N. Determinants of

operative risk in human heurt transplantation. Am J Surg 1971;122:192-7.

2. KirkIn 3, Naftel D, Kirkln J, Blackstone E. White-Williams C. Bourge R. Pulmonary vascular reaistanc%a and the risk of heart transplantation. J Heart Transplant 198&?:331-6.

3. Addoniiio L, Gersony W, Robbiw R, et al. Elevated pulmonaty vascular resistance and cardiac transplantation. Circulation 1987;76 Suppt V:V-52-S.

A Cc&ardJitckle A, Fowler hi, Influence of preoperative pulmonary artery pressure on mortality after heart transplantation: testing of potential revers- ibility of pulmonary hypertension with nitroprusside is useful in defining a hiih risk group. J Am Coil Cardiol1992@%-54.

5. Decb G, Boiling !& The role of amrinone in potential heart transplant patients with pulmonary hypertension. J Cardiothorac Anesth 1989$33-7.

6. Murali S, Uretsky B, Armitage J. et al. Utility of prostaglandin E, in the pretrunsplantation evaluation-of heart failure -patients w&h significant pul- monary hvnertension. J Heart Lunn Transnlant 1992:11:716-23. - . .

7. lgnarro L, Lippton H, Edwards J: et al.‘Mechanism of vascular smooth muscle relaxation by organic nitrates, nitrites, nitroprussidt and nitric oxide: evidence for the involvement of Snitmso’hiils as active intermediates. J Pharmacor Exp Titer 1981;218:739-49.

8. Palmer R, Rces 4 Ashton D, Moncada S. Larginine is the physiological precursor for the formation of nitric oxide in endothelium-depcndent relaxation. Biihem Biiphys Res Commun 19t&153:1251-6.

9. Kelm M, Schrader J. Contml of comnary vascular tone by nitric oxide. Circ Res 199@66zl561-75,

10. Roberts J. Chen TY, Kawai N, et al. Inhaled nitric oxide reverses oulmonanr vaso#mtriction in the hypoxic and acidotic newborn lamb.-Circ Rks 1993;72z246-54.

13. Roberta J, Land P, Bigatello L, Vlahakes G, Zapol W. Inhaled nitric oxide in congenital heart di. Circulation 1993;87:447-53.

12. Rossaint R, Falke K, Lonez F, Slama K, Pison U, Zapol W. Inhaled nitric oxide for the adult respiiatory distress syndrome. N Engl J Med 1993328: 399-45.

13. Bigatello LM, Hurford WE, Kacmarek RM, Roberts JD, Zapol WM. Prolonged inhalation of low levels of nitric oxide in patients with severe ARDS: effects on pulmonary hemodynamics and oxygenation. Anesthesiol- ogy 1994,80:761-70.

14. Cockriil BA, Kacmarek R, Thompson BT, Fifer MA, Zapol W, Semigran MJ. Nitric oxide is a selective puln.~naty vasodilator in patients with pulmonary vascular disease [abstract]. J Am Coil Cardiol 199321 Suppl A:368A.

15. Pepke-Zaba J, Higenbottam TW, Dinh-Xuan AT, Stone D. Wallwork J. Inhaled nitric oxide as a cause of selective pulmonary vasodilation in pulmonary hypertension. Lancet 1991;338:1173-4.

16. Singh T, Dinda P, Chattcrjce S, Riding W, Pate1 T. Pulmonary function studies before and after closed mitral valvotomy. Am Rev Rcsp Dis 1970;101:62-6.

17. Font:jn A, Sabadell A, Ronco R. Homogeneous chcmiluminescent measure- ment of nitric oxide with ozone. Implications for continuous selective monitoring of gaseous air pollutants. Anal Chem 1971);4>575-9.

tH. Fairbanks E, Klee G. Biochemical aspects of hcmatolagy. In: Tietz NW, edttor. Textbook of Clinical Chemistry. Philadelphia: Saunders, 198631536- 40.

19. Fonger J, Borkon A, Baumgartner W. Achuff S. Augustine S, Reita B. Acute right ventricular failure following heart transplantation: imprcwement with prostagiandin El and right ventricular assist. J Heart Transplant 1986;5:317- 21.

20. Thompson ME. Selection of candidates for cardiac transpluntation. J Heart Transplant 1983;3:65-9.

21. Bhatia S, Kirshembaum J, Shcmin R, et al. Time course of resolution of pulmonary hypertension and right ventricular remndcling after orthotopic cardiac transplantation. Circulation 1987;76:819-26.

22. CostardJiicklc A, Hill I, Schroeder J. Fowler M. The intluence of preopcr- ative patient characteristics on early and late survival following cardiac transplantation. Circulation 1991;84 Suppl 111:111-329-37.

23. Haywood G, Sneddon 3, Bashir Y, Jennison S. Gray H. McKenna W. Adenosinc infusion for the reversal of pulmonary vasoconstriction in biven- tricular failure, a good test but a poor therapy. Circulation 1992$h:9%-92.

24. Moncada S. Palmer R, Higgs E. Nitric oxide: physiology, pathophysiology, and pharmacology. Pharm Rev 1991;43:109-42.

25. Lowenstein CJ, Dinerman JL Snyder SH. Nitric oxide: a physiologic messenger. Ann Intern Mcd 1994;120:227-37.

26. Cassoly R, Gibson QH. Conformation, coopcrativity. and ligand binding in human hemoglobin. J Mol Biol 1975;91:301-13.

27. Rimar S, Gillii CN. Selective pulmonary vasodilation by inhaled nitric oxide is due to hemoglobin inactivation. Circulation 1993$8%84-7.

28. Adnot S, Kouyoumdjian C, Dcfouilloy C, et al. Hemodynamic and gas exchange responses to infusion of acetylcholine and inhalation of nitric oxide in patients with chmnic obstructive lung disease and pulmonary hypertension. Am Rev Respir Dis 1993;148:310-6.

29. Rich GF, Murphy GD, Roos CM. Johns RA. Inhaled nitric oxide: selective pulmonary vasodilation in cardiac surgical patients. Anesthesiology 199378: 1028-35.

30. Girard C, Lehot JJ, Pannetier J, Filley S, French P, Estanove S. Inhaled nitric oxide after mitral valve replacement in patients with chronic pulmo- nary artery hypertension. Anesthesiology 1992;77:880-3.

31. Feelisch M, Noack E. Correlation between nitric oxide formation during degradation of organic nitrates and activation of guanylate cyclase. Eur J Pharmacol 1987;139:19-30.

32. Finkel M, Gddis C, Jacob T, Watkins S, Hattler B, Simmons R. Negative inotmpic effects of cytokines on the heart mediated by nitric oxide. Science 1992;257:387-9.

33. Vogel W, Apstein C, Brings L, Gaasch W, Ahn J. Acute alterations in left ventricular diastolic chamber stiffness, role of the ‘erectile” effect of coronary arterial pressure and flow in normal and damaged hearts. Circ Res 1982:51:465-78.

34. Stamler JS, Simon Df, Osborne JA. S-Nitmsylation of proteins with nitric oxide: synthesis and characterization of biologically active compounds. Proc Natl Acad Sci U S A. 1992;89:444-8.