Influence of Inhaled Nitric Oxide on Systemic Flow and Ventricular Filling Pressure in Patients Receiving Mechanical Circulatory Assistance
Background In patients with left ventricular (LV) dysfunction, inhaled nitric oxide (NO) decreases pulmonary vascular resistance (PVR) but causes a potentially clinically significant increase in left atrial pressure (LAP). This has led to the suggestion that inhaled NO may reach the coronary circulation and have a negative inotropic effect. This study tested an alternative hypothesis that LAP increases because of volume shifts to the pulmonary venous compartment caused by NO-induced selective pulmonary vasodilation.
Methods and Results The Thermo Cardiosystems Heartmate is an LV assist device (LVAD) that can be set (by controlling pump rate) to deliver fixed or variable systemic blood flow. Eight patients (between 1 and 11 days after LVAD implantation) were administered inhaled NO (20 and 40 ppm for 10 minutes), and LAP, systemic flow, and pulmonary arterial pressure were measured in both fixed and variable pump flow modes. In both modes, inhaled NO lowered PVR (by 25±6% in the fixed mode, P<.001, and by 21±5% in the variable mode, P<.003). With fixed pump flow, LAP rose from 12.5±1.2 to 15.1±1.4 mm Hg (P<.008). In the variable flow mode, LAP did not increase and the assist device output rose from 5.3±0.3 to 5.7±0.3 L/min (P<.008).
Conclusions A selective reduction in PVR by inhaled NO can increase LAP if systemic flow cannot increase. These data support the hypothesis that with LV failure, inhaled NO increases LAP by increasing pulmonary venous volume and demonstrate that inhaled NO has beneficial hemodynamic effects in LVAD patients.
Inhaled NO is an effective and selective pulmonary vasodilator in patients with primary pulmonary hypertension, congenital heart disease, valvular heart disease, post–heart transplantation, and adult respiratory distress syndrome.1 In pulmonary hypertension associated with LV dysfunction,2 3 4 we and others have shown that inhalation of NO decreases PVR at the expense of increasing LAP. Two possible explanations have been suggested for this observation. First, it has been proposed that inhaled NO may act as a negative inotrope.2 3 4 5 6 Endogenous NO has been shown to inhibit b-adrenergic positive inotropic responses,6 7 and if NO administered by inhalation reaches the myocardium, it could exert a negative inotropic effect.
An alternative explanation for elevation in LV filling pressure is that selective pulmonary vasodilation acutely augments pulmonary venous return to the failing LV that cannot increase forward cardiac output. At steady state, increased pulmonary venous return results in a shift of blood volume from pulmonary arterial to venous compartments.8 To test the latter hypothesis, we studied patients receiving circulatory assistance with the Heartmate LVAD.9 10 This device can be set to pump with either a fixed or an automatic rate, resulting in either a fixed (if stroke volume is maximal) or a variable pump flow, respectively. Accordingly, we administered inhaled NO to patients with the Heartmate LVAD and assessed their hemodynamic responses during both fixed and variable systemic blood flows.
The results of this study have therapeutic implications for the management of patients with LVADs, who often have concomitant pulmonary hypertension and RV failure that may limit LVAD filling and output.9 10 A selective pulmonary vasodilator could have significant benefits in the postoperative period by potentially alleviating the need for concomitant pharmacological or mechanical support of the RV.
This study was performed in 8 patients (5 men and 3 women; age, 49±11 years) receiving circulatory assistance with the Heartmate IP 1000 LVAD (Thermo Cardiosystems, Inc). Patient diagnoses were end-stage dilated cardiomyopathy (n=5) or ischemic cardiomyopathy (n=3). One of the patients with dilated cardiomyopathy had a history of Chagas disease (patient 6), and 1 patient had concomitant right heart failure (patient 2) and was 3 days status-postimplantation of an RVAD (ABIOMED). Patients were intubated and maintained on stable ventilator settings with fixed doses of inotropic or vasodilator drugs; they were studied between 1 and 11 days after LVAD implantation. The study protocol was approved by the Committee for the Protection of Human Subjects From Research Risks at the Brigham and Women's Hospital. Written informed consent was obtained from the patients.
The Heartmate LVAD is a pneumatically powered blood pump with a pusher-plate design.9 10 Blood fills the LVAD passively through a valved inflow cannula inserted through the LV apex. The native LV preferentially decompresses into the LVAD, which has a low impedance to filling. Under usual resting conditions, native LV ejection into the aorta does not occur because of the higher aortic impedance. The LVAD pumps blood in a pulsatile fashion to the ascending aorta through a valved outflow graft, with a maximal forward stroke volume of ≈80 mL.
The LVAD pump rate can be set in either a fixed or an automatic mode. In the fixed rate mode of operation, the LVAD pumps a variable stroke volume that is determined by the relationship between pump rate, diastolic filling time, and native RV cardiac output. The rate of the device can be lowered to a point at which stroke volume is maximal. At this point of fixed LVAD rate and maximal stroke volume, the LVAD cannot increase its output even if pulmonary venous return to the LA and LV increases. In the automatic rate mode, the LVAD uses a computer algorithm to vary its pumping rate to maintain an average stroke volume of 78 mL. Provided sufficient diastolic filling and systolic ejection times are maintained, the LVAD rate and flow increase when pulmonary venous return to the native LA and LV rises.
Patients were studied in the cardiac surgical intensive care unit. All patients had pulmonary arterial catheters, and 6 of 8 patients had left atrial lines. The LVAD flow, stroke volume, and pump rate were read directly from the Heartmate console. Heart rate; pulmonary arterial systolic, diastolic, and mean pressures; and LAP, right atrial pressure, and PAWP were read at end-exhalation from a bedside strip-chart hemodynamic monitor (Marquette Tramscope, Marquette, Inc). All pressures were the mean of a minimum of 5 consecutive beats. PVR was calculated as mean pulmonary arterial pressure minus LV filling pressure divided by LVAD flow. In 6 patients, the direct LAP was used and in 2 the PAWP was used for LV filling pressure. RV stroke volume was calculated as LVAD flow divided by native heart rate.
In the fixed rate mode, the pump rate was progressively lowered from the maintenance automatic rate mode by decrements of 5 bpm until maximal stroke volume was achieved (≈80 mL). This setting provided a fixed systemic flow. To achieve a variable systemic flow responsive to increases in pulmonary venous return, the device was placed in an automatic rate mode. Hemodynamics were determined during both fixed and variable systemic blood flows at baseline and after inhaled NO (20 ppm for 10 minutes followed by 40 ppm for 10 minutes).
Inhalation of NO
NO gas (800 ppm) and N2 (Airco) were mixed with the use of a standard blender (Low Flow Microblender, Bird Products Corp) before introduction into the air inlet of the patient's ventilator (Puritan Bennett 7200B). The inhaled concentrations of NO and O2 were regulated separately. Inhaled O2 concentration was measured directly with an oxygen analyzer, and the inhaled concentrations of NO, nitrogen dioxide (NO2), and the higher oxides of nitrogen (NOX) were measured continuously by a chemiluminescence technique (Chemiluminescence NOX-NO2 Analyzer, Thermo Environmental Instruments, Inc), and the exhaled gases were scavenged by a vacuum system.
All data are presented as mean±SEM. Effects of NO on hemodynamic variables were measured by ANOVA (two-way, with an identification term for each patient). Testing for the effect at each inhaled NO concentration was performed with the Student-Newman-Keuls test. Baseline variables were compared by paired t test. Effects with a value of P<.05 were considered significant.
Baseline hemodynamics during LVAD support in both fixed and automatic rate modes are shown in Table 1.⇓ Patients had mild pulmonary hypertension, with an average PVR of 227±29 dyne·s·cm−5. Baseline PVR and pulmonary arterial pressures were not different in the automatic rate mode compared with the fixed mode. LAP was lower and LVAD pump rate and output were higher in the automatic mode; the pump rate had been lowered in the fixed mode by study design (Table 1⇓).
Hemodynamic Effects of Inhaled NO
In both flow settings, inhaled NO lowered PVR (by 25±6% in fixed mode, P<.001, and by 21±5% in variable mode; P<.003; Table 2⇓). With fixed systemic flow, the decrease in PVR was associated with a rise in LAP from 12.5±1.2 to 15.1±1.4 mm Hg (P<.008) and no significant change in LVAD flow (see Figure⇓ and Table 2⇓). In contrast, with variable systemic flow, reduction in PVR was not associated with an increase in LAP (10.3±0.8 and 10.6±0.9 mm Hg, respectively). In this setting, the Heartmate flow increased by 0.4±0.1 L/min (P<.008) because of an increase in the rate of the device from 73±5 to 79±5 bpm (P<.003).
Effect of NO on the Right Heart and Systemic Hemodynamics
Inhaled NO had no effect on right atrial pressure or RV heart rate in either the fixed or automatic mode. RV stroke volume was unchanged in the fixed mode but rose from 59.2±5.1 to 64.1±5.0 mL (P<.01) in the variable mode in response to the increase in LVAD flow. In the one patient receiving RVAD support (patient 2), RVAD flow was maximal (4.9 to 5.1 L/min) and did not change with inhaled NO. In the fixed mode, NO did not lower systemic blood pressure or systemic vascular resistance (data not shown). In the automatic mode, mean arterial pressure rose from 83±3 to 88±2 mm Hg (P<.01). This appeared to be due to the increase in LVAD flow, as systemic vascular resistance remained unchanged (1017±67 versus 1008±57 dyne·s·cm−5).
In this study we have shown that in patients with pulmonary hypertension associated with LV failure, inhalation of NO caused selective pulmonary vasodilation during LVAD support. With the LVAD set to deliver a fixed pump output, reduction in PVR was associated with an increase in LAP. In contrast, with the LVAD set to allow an increase in LVAD output if venous return increased, reduction in PVR led to an increase in LVAD flow without an increase in LAP.
These data address an important controversy regarding the mechanism responsible for the rise in LV filling pressure observed in patients with pulmonary hypertension associated with LV dysfunction during the inhalation of NO.2 3 4 5 In these patients, inhaled NO has been shown to increase the PAWP, which is associated with unchanged pulmonary arterial pressures and a cardiac output and/or stroke volume that was either unchanged3 4 or tended to fall.2 These hemodynamic changes can be interpreted as possibly caused by a depression of the Frank-Starling curve. Bocchi and colleagues5 have reported the development of pulmonary edema associated with inhaled NO in patients with severe congestive heart failure, illustrating the potential clinical significance of these pressure increases. Two explanations have been proposed for these observations: (1) NO delivered by inhalation can reach the myocardium and produce a negative inotropic effect, and (2) the LAP rises because of volume shifts from pulmonary arterial to pulmonary venous compartments8 ; despite the increased LAP volume, the failing LV cannot increase forward cardiac output.
Support for a negative inotropic effect of NO comes from observations that inhibitors of NO synthase augment β-adrenergic positive inotropic responses.6 7 While it is generally thought that inhaled NO is rapidly inactivated by the heme moiety of hemoglobin,11 administration of NO may increase levels of circulating nitrosylated compounds.12 Hemoglobin itself has recently been appreciated to form S-nitroso-hemoglobin under oxygenated conditions and may serve as an NO donor in the periphery.13 In this regard, inhaled NO has been reported to have hemodynamic effects in the renal vasculature.14 15 An unchanging measurement of LV peak +dP/dt during inhaled NO, however, argues against a negative inotropic effect.2 Additionally, studies in dogs with pacing-induced failure16 and humans with normal LV function17 have shown that inhaled NO does not depress the end-systolic pressure-volume relation.
An alternative explanation for increased LAP caused by NO in patients with LV dysfunction is that selective pulmonary vasodilation can lead to volume shifts from the pulmonary arterial to the pulmonary venous compartments. The present data support the contention that filling pressures rise with increased pulmonary venous volume if the LV cannot augment cardiac output and that rising LV filling pressures do not necessarily indicate a negative inotropic effect. With LV dysfunction, the heart may not be able to utilize a Starling mechanism18 and has reduced compliance, both of which could contribute to rises in LAP with increased venous return.
Patients with severe LV dysfunction supported with LVADs could benefit greatly from selective pulmonary vasodilators, since pulmonary hypertension and right heart failure are major limitations to LVAD implantation. Because the Heartmate LVAD responds to increased venous return with an augmented pump flow, increases in LV filling pressure can be avoided in LVAD patients administered inhaled NO. It is therefore possible that inhaled NO could have a role in improving RV function by providing selective RV afterload reduction.19 Future study will be needed to determine whether inhaled NO improves the clinical outcome of LVAD patients. Inhaled NO may serve as a useful therapeutic modality in the perioperative period, reducing the need for concomitant mechanical or pharmacological support of the RV, and may allow expansion of the pool of candidates eligible for LVAD support.
Selected Abbreviations and Acronyms
|LAP||=||left atrial pressure|
|LVAD||=||LV assist device|
|PVR||=||pulmonary vascular resistance|
|PAWP||=||pulmonary arterial wedge pressure|
|RVAD||=||RV assist device|
Dr Hare is the recipient of National Institutes of Health (NIH) grant K08-HL-03228-03 and a Grant-in-Aid from the American Heart Association. This work was supported in part by NIH grant HL-52320 (Dr Colucci). The authors wish to acknowledge Patricia Sullivan for expert assistance in manuscript preparation, the nursing staff of the cardiac surgical intensive care unit, and members of the respiratory care unit, Brigham and Women's Hospital.
Presented in part at the 68th annual Scientific Sessions of the American Heart Association, Anaheim, Calif, November 13-16, 1995.
- Received November 20, 1996.
- Revision received March 6, 1997.
- Accepted March 13, 1997.
- Copyright © 1997 by American Heart Association
Loh E, Stamler JS, Hare JM, Loscalzo J, Colucci WS. Cardiovascular effects of inhaled nitric oxide in patients with left ventricular dysfunction. Circulation. 1994;90:2780-2785.
Hare JM, Loh E, Creager MA, Colucci WS. Nitric oxide inhibits the contractile response to β-adrenergic stimulation in humans with left ventricular dysfunction. Circulation. 1995;92:2198-2203.
Keaney JF Jr, Hare JM, Kelly RA, Loscalzo J, Smith TW, Colucci WS. Inhibition of nitric oxide synthase potentiates the positive inotropic response to β-adrenergic stimulation in normal dogs. Am J Physiol. 1996;271:H2646-2652.
McCarthy PM, James KB, Savage RM, Vargo R, Kendall K, Harasaki H, Hobbs RE, Pashkow FJ, and the Implantable LVAD Study Group. Implantable left ventricular assist device: approaching an alternative for end-stage heart failure. Circulation. 1994;90(suppl II):II-83-II-86.
Rimar S, Gillis CN. Selective pulmonary vasodilation by inhaled nitric oxide is due to hemoglobin inactivation. Circulation. 1993;88:2884-2887.
Stamler JS, Jaraki O, Osborne J, Simon DI, Keaney J, Vita J, Singel D, Valeri CR, Loscalzo J. Nitric oxide circulates in mammalian plasma primarily as an S-nitroso adduct of serum albumin. Proc Natl Acad Sci U S A. 1992;89:7674-7677.
Szwarc RS, Graves CL, Flemming F, Ball HA, Bryan CA, Bohn D, Benson LN. Evidence for systemic effects of inhaled nitric oxide in the anesthetized minipig. Circulation. 1995;92(suppl I):I-701. Abstract.
Troncy E, Vinay P, Francoeur M, Salazkin I, Blaise G. Nitric oxide by inhalation influences renal function in pigs. Endothelium. 1995;3(suppl):S-114. Abstract.
Polidori D, DiPierro F, Lankford E, Doering E, Hanson W, Acker M, Loh E. Cardiovascular effects of nitric oxide in a canine model of cardiomyopathy. Circulation. 1995;92(suppl I):I-184. Abstract.
Hayward C, Kalnins W, Rogers P, Feneley M, Macdonald P, Kelly R. Evaluation of the effects of inhaled nitric oxide on normal human left ventricular function. Circulation. 1995;92(suppl I):I-791. Abstract.
Schwinger RHG, Bohm M, Koch A, Schmidt U, Morano I, Eissner HJ, Uberfuhr P, Reichart B, Erdmann E. The failing human heart is unable to use the Frank-Starling mechanism. Circ Res. 1994;74:959-969.