Lack of Benefit From Nitric Oxide Synthase Inhibition in Patients With Cardiogenic Shock: Title and subTitle BreakLooking for the Reasons

Cardiogenic shock complicating acute myocardial infarction (AMI) is one of the most serious and challenging conditions in cardiovascular medicine, with up to two thirds of patients dying within a few weeks.^{1 - 2} According to data from a national registry of nearly 300 000 patients with ST-segment elevation AMI, the overall incidence of cardiogenic shock (diagnosed at both presentation and after admission) was 8.6%.³

Even though the incidence of cardiogenic shock has remained stable over the time,^{2 - 4} mortality rates among patients with cardiogenic shock have decreased over the last decade. Some of the major factors contributing to this decline are early revascularization with percutaneous coronary intervention or coronary artery bypass graft surgery^{1 ,5} and interventions that increase cardiac output, including inotropic agents and supportive mechanical devices such as intra-aortic balloon counterpulsation and left ventricular assist devices.⁶ Babaev et al³ reported that percutaneous coronary intervention rates in patients with cardiogenic shock increased from 27.4% in 1995 to 54.4% in 2004 and that this increase was associated with a reduction in overall in-hospital mortality rates from 60.3% to 47.9% during that time.

Even with application of these therapies, mortality rates remain unacceptably high. Additional and novel therapeutic interventions are badly needed. One possible new approach stems from recent work demonstrating important pathophysiological roles of inflammatory cytokines, the complement system, and increased levels of nitric oxide due to enhanced expression of inducible nitric oxide synthase (NOS).⁷ These intriguing findings led to the Tilarginine Acetate Injection in a Randomized International Study in Unstable MI Patients With Cardiogenic Shock (TRIUMPH) trial, reported in this issue of JAMA.⁸

The TRIUMPH trial was a randomized, multicenter, double-blind, placebo-controlled trial of patients with refractory cardiogenic shock despite successful coronary revascularization. Patients were randomized to receive the nonspecific NOS inhibitor tilarginine (L-N^G-monomethylarginine [L-NMMA]), 1-mg/kg bolus plus 1-mg/kg per hour 5-hour infusion, or matching placebo. Although the investigators planned to enroll 658 patients, the study was terminated prematurely after enrollment of 398 patients based on a prespecified analysis of futility. Despite a significant increase in systolic blood pressure at 2 hours among patients in the tilarginine group vs the placebo group (12 vs 7 mm Hg; P = .001), there were no differences in the primary end point of 30-day mortality or in secondary outcomes of shock resolution and duration and 6-month mortality. The authors concluded that tilarginine, 1-mg/kg bolus plus 5-hour infusion, does not reduce mortality in patients with refractory cardiogenic shock complicating myocardial infarction.

In contrast to prior small studies that have suggested a mortality benefit from NOS nonspecific inhibition,^{9 - 10} the TRIUMPH trial found no such benefit. Although the exact reasons for this disappointing outcome are not clear, there are several tentative explanations. Nitric oxide is produced by 3 isoforms of NOS: neuronal, endothelial, and inducible NOS. These 3 isoforms are products of different genes and have different localizations, kinetics, regulation properties, and roles in vascular and myocardial biology.¹¹ The first 2 are constitutive enzymes (ie, enzymes for which the intracellular concentration is constant and not influenced by substrate concentration) and nitric oxide production by these 2 is calcium dependent. On the other hand, inducible NOS (ie, the enzyme for which production is induced by substrate concentration or other stimuli) is a slow-kinetic, high-output, calcium-independent enzyme. Activated inducible NOS generates large amounts of nitric oxide by a sustained production for up to 10 hours.¹² All 3 isoforms of NOS are expressed in myocardium.

It has been suggested that constitutively produced nitric oxide is protective, whereas the large amount of nitric oxide generated by inducible NOS may contribute to tissue damage.¹³ Nitric oxide produced by constitutive enzymes is a primary determinant of blood vessel tone, regulating the coronary microcirculation,¹⁴ preventing platelet adhesion and aggregation,¹⁵ and promoting ventricular relaxation.¹⁶ Shear stress and inflammatory cytokines lead to excessive inducible NOS–stimulated nitric oxide production, resulting in profound vasodilatation, myocardial depression, resistance to vasopressor agents, and vascular leak syndrome.¹³ Thus, any non–isoform-specific inhibition of NOS, such as that achieved in the TRIUMPH trial with tilarginine, together with attenuation of alterations caused by excessive nitric oxide generation by inducible NOS may adversely affect the protective actions of constitutively generated nitric oxide as well.

In cardiogenic shock, there is a critical balance between vasoconstrictor stimuli (catecholamines, angiotensin, endothelin, and vasopressin) and vasodilator stimuli (nitric oxide, prostaglandins, and natriuretic peptides). Non–isoform-specific inhibition of nitric oxide production in the presence of excessive amounts of endogenous and exogenous vasoconstrictor agents potentiates vasoconstrictor effects of catecholamines¹⁷ and may result in excessive vasoconstriction, compromising myocardial and systemic tissue perfusion. Numerous experimental studies have demonstrated peripheral and coronary vasoconstriction, reduced cardiac output, and myocardial dysfunction by nonselective NOS inhibition.^{18 - 21} In experimental models of AMI, nonselective NOS inhibition with N^ω-nitro-L-arginine (L-NNA) was associated with exaggeration of myocardial dysfunction due to reduction of regional blood flow in the surviving myocardium.¹⁸ Elimination of endogenous nitric oxide synthesis has been shown to exacerbate myocardial stunning¹⁹ and reduce the percentage of perfused capillaries in normal and stunned myocardium.²⁰ Patients with cardiogenic shock may be particularly sensitive to even slight reductions of coronary blood flow to the surviving myocardium. Furthermore, the elevation in blood pressure that follows NOS inhibition most likely occurs due to increased systemic vascular resistance that reduces cardiac output.

Although tilarginine (L-NMMA) is devoid of vascular effects, high doses of this agent may result in severe vasoconstriction, end organ damage, and death,²¹ emphasizing the fact that the degree of NOS inhibition may be crucial for the outcome of treatment. The dosing of tilarginine in the TRIUMPH trial was supported by findings from 2 prior studies of patients with cardiogenic shock^{9 - 10} and by the SHOCK-2 trial.²² Considering that the TRIUMPH trial and these 2 prior trials came to contrary conclusions regarding the outcome of treatment and that the SHOCK-2 trial was inconclusive as a dose-ranging trial, dosing of tilarginine in TRIUMPH is not firmly established.

The interruption of the TRIUMPH trial due to futility brings ambiguity in the interpretation of the treatment outcome. Although the investigators are correct in their conclusion that tilarginine did not reduce mortality and that futility-driven interruption of the trial limited patient exposure to at best a futile treatment and reduced expenditures, the study power was markedly reduced; consequently, the possibility of increased mortality in patients who received the study drug remains an unresolved issue. The absolute 6% increase in 30-day mortality in the tilarginine group (48% vs 42% in the placebo group) is disturbing, even though this end point did not reach statistical significance.

Another clinical trial with a nearly identical study design testing the effect of NOS inhibition with L-NMMA hydrochloride on 28-day mortality in patients with septic shock was ended prematurely after detection of increased mortality among patients assigned to the study drug.²³ This trial found an absolute 10% difference in 28-day mortality in disfavor of the NOS inhibitor (59% vs 49%). The increased risk associated with the NOS inhibitor was more evident in patients with hypodynamic shock (cardiac index <3 L · min^-1 · m^-2) than in those with a higher cardiac index (≥3 L · min^-1 · m^-2). In addition, the increase in mortality reflected a higher proportion of cardiovascular deaths, leading the investigators to suggest that treatment with L-NMMA hydrochloride provoked a worsening of acute circulatory failure or, more specifically, exacerbated myocardial dysfunction.²³ Thus, it appears that patients with cardiogenic shock and extensive myocardial dysfunction may be particularly prone to exacerbation of myocardial dysfunction by NOS inhibition.

In demonstrating the futility of targeting inhibition of the nitric oxide pathway in patients with cardiogenic shock, the TRIUMPH trial represents an important contribution by clearly establishing that nonselective NOS inhibition should not be attempted in patients with cardiogenic shock complicating AMI. Early stabilization measures, urgent revascularization, inotropic and vasopressor support, and mechanical support by intra-aortic counterpulsation or left-ventricular assist devices should remain the mainstay of therapy for patients with cardiogenic shock after AMI.^{7 ,24}

An important consideration is whether NOS inhibition should be considered a closed issue in patients with cardiogenic shock. It is possible that the pathophysiology of the nitric oxide pathway in the setting of cardiogenic shock may not have been investigated thoroughly enough to allow a reliable therapeutic intervention directed at its inhibition. Investigators might develop specific inhibitors of inducible NOS that preserve nitric oxide production regulated by constitutive enzymes. Another consideration may involve nonselective NOS inhibition in conjunction with the use of external administration of nitric oxide titrated under hemodynamic guidance. As is clear from the TRIUMPH trial, cardiogenic shock outcomes are poor even with early revascularization. The failure of tilarginine to improve outcomes is a call for renewed basic, translational, and clinical research efforts for this difficult, perplexing, and highly fatal disease.