OUP user menu

★ Controversies in cardiovascular medicine ★

More answers to the still unresolved question of nitrate tolerance

Thomas Münzel, Andreas Daiber, Tommaso Gori
DOI: http://dx.doi.org/10.1093/eurheartj/eht249 2666-2673 First published online: 17 July 2013


Organic nitrates are traditionally felt to be a safe adjuvant in the chronic therapy of patients with coronary artery disease. Despite their long use, progress in the understanding of the pharmacology and mechanism of action of these drugs has been achieved only in the last two decades, with the identification of the role of oxidative stress in the pathophysiology of nitrate tolerance, with, the discovery of the ancillary effects of nitrates, and with the demonstration that nitrate therapy has important chronic side effects that might modify patients' prognosis. These advances are however mostly confined to the molecular level or to studies in healthy volunteers, and the true impact of organic nitrates on clinical outcome remains unknown. Complicating this issue, evidence supports the existence of important differences among the different drugs belonging to the group, and there are reasons to believe that the nitrates should not be treated as a homogeneous class. As well, the understanding of the effects of alternative nitric oxide (NO) donors is currently being developed, and future studies will need to test whether the properties of these new medications may compensate and prevent the abnormalities imposed by chronic nitrate therapy. Intermittent therapy with nitroglycerin and isosorbide mononitrate is now established in clinical practice, but they should neither be considered a definitive solution to the problem of nitrate tolerance. Both these strategies are not deprived of complications, and should currently be seen as a compromise rather than a way fully to exploit the benefits of NO donor therapy.

  • Nitrate tolerance
  • Oxidative stress
  • Endothelial dysfunction
  • Peroxynitrite
  • Nitrite
  • Vasodilation

Introduction: short-term vs. long-term nitrate therapy

Organic nitrates, on the market since 1882, remain useful drugs in the treatment of acute ischaemia and heart failure and in preventing angina prior to physical stress; Sublingual nitroglycerin or isosorbide dinitrate dilate capacitance veins and conduit arteries, relieving symptoms of angina and decreasing cardiac oxygen demand. These effects are so consistent and reproducible that a symptomatic improvement immediately after the administration of sublingual nitroglycerin or isosorbide dinitrate is considered to represent valuable information in the differential diagnosis of chest pain.1 Similarly, the use of sublingual nitrates in settings where angina is anticipated (i.e. in the symptomatic prophylaxis) is also strongly recommended, and physicians should educate and encourage patients to self-administer nitrates before performing physical activity, or in cases of emotional stress.

During chronic therapy, however, these effects are less evident, and current guidelines only recommend nitrate therapy as a third-line approach to the treatment of patients whose angina symptoms persist despite the administration of beta-blockers and/or calcium antagonists (and/or coronary stenting). In the setting of coronary artery disease (CAD), the use of nitrates was however established well before the systematic introduction of double-blind randomized long-term studies, and, to date, concerns on the long-term effects of these drugs have been expressed based on the existence of controversial observations: (i) the evidence that nitrate tolerance reflects profound changes in vascular homeostasis; (ii) the suspect that therapy with organic nitrates might be associated with an increased incidence of coronary events in patients treated with nitrates (although these data need to be considered preliminary, for instance the study by Ishikawa et al.2 pooled different nitrates and no placebo control was used); and (iii) the results of the GRACE registry,3 suggesting a shift in the pattern of acute coronary syndromes from STEMI to NSTEMI in patients treated with nitrates. In contrast to ischaemic heart disease, the prognostic benefit of long-term therapy with organic nitrates is well documented in the setting of congestive heart failure:4,5 a number of well-designed studies have shown that the combination of hydralazine and isosorbide dinitrate improves cardiac haemodynamics and reduces mortality in patients with the left ventricular dysfunction particularly in black Americans making BiDil the first drug ever approved for one single ethnic group. Notably, a crucial difference between these two settings is that, in contrast to CAD, nitrates have historically been associated with other drugs in the therapy of congestive heart failure. For instance, the antioxidant properties of hydralazine are thought to have central importance in preventing nitrates side effects.6 In this sense, classical research on the clinical effects and side effects of nitrates in the era preceding the systematic use of angiotensin-converting enzyme inhibitors and statins should now be reviewed critically.

Nitrate tolerance and intermittent therapy: the problem remains

Limiting the efficacy of chronic therapy, nitrate tolerance is the loss of effects, or the necessity to increase dosages to maintain the effects, of organic nitrates. Nitrate tolerance was first mentioned in 1889 in Brunton's Pharmacology and Therapeutics and in a contemporary report by Stewart, i.e. more than two decades before the first formal description of myocardial infarction by Herrick. Nitrate tolerance is a problem which invariably results in the loss of nitrates’ effects on anginal symptoms, exercise capacity, and haemodynamics.7 Despite a number of different approaches (reviewed in ref. 8), tolerance, with all its implications, remains a limitation to the therapy with nitrates. Pragmatically, in order to avoid tolerance, the most commonly employed strategy is to allow a minimum of 12-h daily nitrate-free interval.9 Such intermittent therapy allows maintaining the effects of nitrates and is widely accepted since at least 20 years. While effective in providing diurnal protection and symptomatic benefit, this therapy is however not associated with a reduction in the overall daily ischaemic burden: indeed, intermittent therapy prevents the development of tolerance, but at the same time it can be associated with a paradoxically increased frequency of ischaemic events during night hours. The clearest evidence of the risks associated with intermittent therapy was provided by Azevedo et al.,10 who demonstrated that during the nitrate-free interval coronary vasospastic responses to acetylcholine are increased in patients with mild CAD. Further, the preconditioning effects of nitroglycerin are lost upon long-term therapy, even when sufficient time between doses is allowed.11 Expanding this concept, a rebound increase in platelet aggregability has also been observed during nitrate-free hours.12 In the absence of effective strategies to prevent tolerance, intermittent therapy remains the only solution to administer nitrates, and one that is accepted by most patients and physicians. In view of the evidence that there are limitations to this strategy, and given the large clinical use of nitrates, this should however be seen as a compromise rather than a solution, and the search for other options remains prioritary.

New observations on the mechanisms of action of organic nitrates and implications for the understanding of nitrate tolerance

The mitochondrial aldehyde dehydrogenase (ALDH-2) has been proposed as the mediator of GTN biotransformation and activation, which lead to a ‘mitochondria-centred’ hypothesis of nitrate tolerance (reviewed in ref. 8); a number of issues regarding its role and implications however still remain unclear (Figure 1). Supporting the ALDH-2 hypothesis, the existence of the thionitrate intermediate that should be produced during the bioactivation of GTN by ALDH-2 was characterized at a molecular level in crystallized ALDH-2/GTN preparations.13 As well, recent data obtained with purified ALDH-2 also provide evidence that ALDH-2 could be a source of GTN-triggered reactive oxygen species (ROS) formation:14,15 the resulting oxidative inhibition of the enzyme and decreased GTN biotransformation has been advocated as one of the mechanisms of tolerance. In contrast with previous findings, however, recent data suggest that ALDH-2, and GTN biotransformation, might be located mainly in the cytoplasm, with only ∼5% of the enzyme localized in the mitochondria of smooth muscle cells. In this hypothesis, ROS production and tolerance would ensue only when (local) the concentrations of GTN >1 micromol/L are reached, beyond which mitochondrial biotransformation and oxidative reactions occur. This in vitro evidence is compatible with the recent observation that very low doses of GTN have similar haemodynamic effects when compared with standard ones, with the difference that they do not induce tolerance.16 A tighter spatial cytosolic colocalization of the ALDH-2 and the soluble guanylyl cyclase might also possibly explain why at least three separate groups failed to detect measurable nitric oxide (NO) formation in response to GTN administration, and be compatible with the fact that tolerance, oxidative stress and endothelial dysfunction also develop in response to isosorbide mononitrate, a compound, which does not undergo mitochondrial or ALDH-dependent biotransformation. In sum, recent data suggest that supraphysiological concentrations of nitrates might cause a (redox-based) activation of different sources (both mitochondrial and cytosolic) of oxygen-free radicals and trigger a feed-forward mechanism via enzyme-specific ‘redox switches’. Mitochondria might act as triggers for oxidative damage in the cardiovascular system but also as amplifiers in a series of cross-talk phenomena (reviewed in ref. 17). In this perspective, the use of drugs with a direct or indirect (e.g. ACE-inhibitors, statins, hydralazine) antioxidant effect would be expected to profoundly influence nitrate pharmacology. Downstream to the biotransformation of nitrates, the activation of the enzyme soluble guanylyl cyclase and of cGMP-dependent protein kinases triggers both Ca2+-dependent and -independent mechanisms. As well, recent studies demonstrate that GTN modulates vascular biology through epigenetic mechanisms (i.e. by modulating the processes, for instance histone acetylation and deacetylation, which control which and how much genes are expressed).18 Although research on the role of epigenetic regulation in cardiovascular pathophysiology is at its very beginning, it has been shown that the so-called ‘epi-drugs’ (i.e. drugs that act on the acetylation levels of histones, thereby modifying DNA expression), prevent tolerance in a murine model.18 The possible clinical applications of this observation remain to be explored.

Figure 1

Mechanisms of nitrate vasodilation: nitric oxide/cyclic GMP-mediated intracellular signalling leading to smooth muscle cell relaxation involve the activation of cGK-I and decrease in intracellular calcium levels (via inhibition of the IP3-receptor-regulated calcium channel, activation of potassium channels with subsequent inhibition of calcium channels, and activation of the calcium-pump) as well as epigenetic mechanisms. Recent data appear to suggest that the biotransformation of GTN occurs in the mitochondria only when higher local concentrations are reached. cGK-I, cGMP-dependent kinase; ALDH-2, aldehyde dehydrogenase; PDE, phosphodiesterase; MLCP, myosin light-chain phosphatase; MLCK, myosin light-chain kinase; GTN, nitroglycerin; PETN, pentaerythrityl tetranitrate; ISMN, ISDN, isosorbide mono- and dinitrate.

The mechanisms of nitrate tolerance

In contrast with the acute vasodilatory effects of nitrates, when administered chronically, organic nitrates also trigger counter-regulatory vasoconstrictor mechanisms, a phenomenon mediated by increased levels of the vasoconstrictors angiotensin II and noradrenalin, sympathetic activation, and oxidative stress.19 Beyond their pharmacological interest at the molecular level, these nitrate-induced changes might also have important implications in patients with CAD, hypertension, and heart failure, in which oxidative stress has been shown to have negative prognostic implications.20

While this has long been felt as a limitation to the benefit of these drugs, the knowledge that nitrate tolerance might also reflect potentially dangerous vascular abnormalities is much more recent. Every effort at identifying a single mechanisms responsible for tolerance has failed in the last 30 years: tolerance appears to consist of a number of phenomena both at the systemic level (neurohormonal activation and intravascular volume expansion, so-called pseudotolerance) as well as more specific vascular disturbances (so-called true vascular tolerance), such as the inhibition of nitrate biotransformation, desensitization of the soluble guanylyl cyclase, increase in phosphodiesterase activity, and uncoupling of the NO synthase leading to cross tolerance to other NO-donor substances (mechanisms reviewed in ref. 8). As previously described, the concept of GTN-induced oxidative stress provides the possibility to formulate a unifying hypothesis for several mechanisms proposed for the pathophysiology of nitrate tolerance.21 Importantly, the implications of mitochondrial ROS formation are not confined to the mitochondrial matrix, as ROS leaking into the cytoplasm activate a cross-talk with the vascular NADPH oxidase22 (Figure 2). It is now understood that the oxidation of thiol groups in the active site of the ALDH-2 observed during chronic GTN therapy may cause inhibition of the enzyme,23,24 and therefore both reduced GTN biotransformation and effectiveness.25 Similar changes might explain the desensitization of the soluble guanylyl cyclase (sGC), suggested as a mechanism of tolerance already 30 years ago26,27 but demonstrated only very recently: evidence from a murine model demonstrates that nitroglycerin-induced S-nitrosylation of the sGC results in decreased responsiveness to NO28 and that these processes—and tolerance—can be reversed by concomitant treatment with the sulphydryl donor N-acetylcysteine.29 Further, oxidative processes include among other the oxidization of the eNOS cofactor BH4, causing uncoupling of the enzyme, a mechanism that clinically translates into the evidence of endothelial dysfunction.30,31 In line with this, we recently demonstrated an increased expression and uncoupling of the eNOS in an animal model of GTN tolerance32,33 likely caused by oxidative depletion of the co-factor tetrahydrobiopterin and/or by post-translational changes in eNOS (decrease in Ser1177 and increase in Thr495 phosphorylation and/or S-glutathionylation). Interestingly, supplementation of tetrahydropbiopterin, folic acid, or telmisartan restored this abnormality and preserved GTN vasodilator properties.33,34 Importantly, evidence suggests that nitrate tolerance and nitrate-induced endothelial dysfunction, while possibly sharing common mechanisms are not the same thing: the persistence of a vasodilator effect at the level of the brachial and the coronary arteries has been indeed shown during GTN treatment despite the presence of endothelial dysfunction:31,35 these data might suggest that, while being an issue in the therapy of stable angina, tolerance should not be considered a limitation to the use of GTN in the setting of acute coronary syndromes, where the vasodilation of conduit coronaries is the principal mechanism of the benefit of nitrates.

Figure 2

Mechanisms of nitrate tolerance: the combination of systemic phenomena and specific vascular abnormalities leading to a reduced responsiveness to nitrates and to important abnormalities in vascular homeostasis. Uncoupling of the endothelial NOS, production of ROS, oxidative inhibition of the ALDH-2, and desensitization of the sGC play an important role. Activation of the PDE and inhibition of the PGI-S also shift the balance towards vasoconstriction. sGC, soluble guanylyl cyclase; cGMP, cyclic GMP; RAAS, renin–angiotensin system; PKC, protein kinase C; NADPH-Ox, NADPH oxidase; ATII, angiotensin II; BH2 and BH4, di- and tetrahydrobiopterin; NOS, nitric oxide synthase; O2, superoxide anion; ONOO, peroxynitrite; ET-1, endothelin-1; PGI-S, prostaglandin I2 (prostacyclin) synthase; ALDH-2, mitochondrial aldehyde dehydrogenase; mtROS, mitochondrial reactive oxygen species; PDE, phosphodiesterase.

New insights into the negative vascular side effects of ISMN

To date, most of the research on the mechanisms of action and nitrate tolerance has been made using GTN. Isosorbide mononitrate (ISMN) is however since several years the most commonly used oral nitrate. While the important differences exist between ISMN and GTN (first of all the fact that the ALDH-2 is not involved in ISMN biotransformation), the complications associated with chronic therapy are similar, with the only exception of rebound phenomena, which are less manifest with ISMN probably due to the fact that changes in the bioavailability of the drug follow a shallower curve after oral administration of ISMN when compared with the transdermal on-off administration of GTN.36 As a matter of fact, however, chronic continuous therapy with ISDN and ISMN is also associated with nitrate tolerance, oxidative stress (of cytosolic, extra-mitochondrial origin), renin production, and plasma volume expansion, and intermittent therapy has been associated with endothelial dysfunction.3739 Further, probably because of the absence of changes in mitochondrial ROS production, ISMN is devoid of protective preconditioning-mimetic effects and of antiaggregant effects, two phenomena that are thought to concur with the benefit of GTN in the setting of acute coronary syndromes.37,40 We recently found that, like GTN, in vivo treatment with ISMN induces a marked degree of endothelial dysfunction and vascular superoxide production predominantly driven by the activation of the vascular NADPH oxidase and by the uncoupling of the eNOS41 (Figures 3 and 4). As well, similar to GTN, ISMN treatment was associated with a strong increase in the expression of endothelin-1, mainly within the endothelial cell layer and the adventitia, and by a subsequent increase in the sensitivity of the vasculature to vasoconstricting agents such as phenylephrine and angiotensin II.41 Nevertheless, there are fundamental differences between underlying mechanisms of the stimulation of autocrine, vascular ET-1 production by both nitrates: (i) in contrast to GTN, ISMN is not bioactivated by mitochondrial ALDH-2, and therefore mitochondrial oxidative stress (if any) plays a minor role, which emphasizes that this is not a necessary component for the adverse effects of nitrates; (ii) the NADPH oxidase activation in response to ISMN is not dependent on the cross-talk between ROS-producing mitochondria and the enzyme; and (iii) ISMN stimulates the phagocytic NADPH oxidase, a phenomenon which is completely blocked by the ET receptor blocker bosentan.

Figure 3

Mechanisms underlying isosorbide mononitrate-induced endothelial dysfunction. In animals, chronic isosorbide mononitrate therapy did not cause tolerance but induced a severe degree of endothelial dysfunction (A), as indicated by the marked right shift of the dose–response relationship for the endothelium-dependent vasodilator acetylcholine. Simultaneously, an increase in vasoconstrictor sensitivity of the vasculature to phenylephrine and angiotensin II was observed. Likewise, in healthy volunteers, isosorbide mononitrate therapy caused a decrease in basal and stimulated nitric oxide production, all of which was corrected by the intraarterial administration of the antioxidant vitamin C (B). Figure adapted from ref. 41 and 37 (with permission).

Figure 4

Chronic therapy with isosorbide mononitrate causes a marked increase in vascular superoxide production throughout the vessel wall and stimulates the expression of endothelin-1 and big-endothelin mainly in the endothelial cell layer and the adventitia. Figure adapted from ref. 41 (with permission).

Taken together, these observations emphasize the fact that ISMN, although probably deprived of rebound ischaemia effects, is also not a solution to the problems associated with organic nitrates. Further, the fact that ISMN is not metabolized by the ALDH-2 points out that a dysfunction in this enzyme is only a component of the abnormalities induced by GTN, and that tolerance, oxidative stress, and nitrate induced endothelial dysfunction cannot be reduced to ALDH-2 or mitochondrial dysfunction.

PETN: an old nitrate experiencing a new life

The above evidence suggests that, despite similar side effects, nitrates should not be considered a homogeneous class of drugs with similar mechanisms, effects, and complications. Supporting this view, the nitrate PETN might be a remarkable exception to this regard: in contrast to other long-acting nitrates, studies in healthy volunteers showed preserved vasodilator potency as well as the absence of oxidative stress and endothelial dysfunction42,43 during continuous treatment. In patients with CAD, an 8-week treatment with t.i.d. PETN did not cause endothelial dysfunction and it actually increased responses to sublingual GTN, which might also represent an obvious benefit for patients with effort angina.44 In line with this, pharmacological inhibition (including GTN-induced inhibition) or gene deletion of ALDH-2 attenuates vasodilation to PETN, but exposure to PETN alone does not result in a reduced activity of the enzyme.45 As well, PETN does not modify the expression and activity of the sGC,46 and in animals, PETN was reported to prevent the endothelial dysfunction as well as atherogenesis.47 Notably, in further support of a role of oxidative modifications in the mechanisms of nitrate-induced abnormalities, the absence of these effects in response to PETN might be mediated by its capacity to induce the antioxidant defense protein haeme oxygenase-1, thereby increasing the expression and formation of ferritin (which binds iron and therefore prevents hydroxyl radical formation), of the antioxidant molecule bilirubin, and the vasodilator carbon monoxide.48,49 While these data provide a strong biological rationale for the superiority of PETN against other nitrates, the lack of clinical data on the antianginal effects of this compound has limited its use. PETN was marketed until the early 1990s in North America and Europe, but it was then removed from the pharmacopeia essentially due to the absence of efficacy data. For the same reason, PETN was removed in 2012 from the list of reimbursable drugs in Germany, the major market for this drug. The recently concluded Cleopatra study addressed this issue by randomizing >600 CAD patients to receive placebo or PETN 80 mg b.i.d., demonstrating a benefit of PETN on exercise tolerance during a 12-weeks treatment (T Münzel, unpublished data). These data will hopefully lead to reconsider the use of (and further research on) this drug in clinical practice worldwide.

Nitrite therapy, an alternative to classical organic nitrates?

While GTN was the very first drug ever synthesized (Ascanio Sobrero, 1846), amyl nitrite was the first NO donor to be used in CAD (1867, Murrell and Brunton). Due to unease of administration, longer duration of action and lack of commercial interests, nitrite therapy was until recently abandoned. Long from being, as until recently thought, a simple by-product of endothelial activity, recent studies provide support for the concept that the inorganic nitrite anion represents a storage form of NO that may have important therapeutic potential (reviewed in ref. 50) and some advantages compared with organic nitrates.

The ability of sodium nitrite to relax isolated arteries was studied in detail in 1953 by Furchgott51 and the effects of nitrite on GC activity and cGMP levels were documented by Ignarro and coworkers.52 The mechanism by which nitrite is reduced to NO is uncertain, and a number of mechanisms have been proposed, including deoxyhaemoglobin, xanthine, and aldehyde oxidases as well as a non-enzymatic disproportionation at low pH5356 (Figure 5). Whatever these mechanisms, bioactivation of nitrite appears to be maximal in hypoxic conditions, which would obviously represent an advantage, as it results in a larger pharmacological vasodilator effect at the sites where perfusion is lower.57 The generation of NO by deoxygenated myoglobin might thus re-direct blood flow specifically in hypoxic vascular beds, making nitrite an endocrine vehicle for both exogenous and endogenous NO. Further, in addition to direct vasodilatory effects, nitrite has been reported to have preconditioning-like effects in settings of ischaemia–reperfusion injury in different tissues58 through NO- and cGMP-independent mechanisms,59 and orally administered nitrate has been shown to improve revascularization in chronically ischaemic limbs and decrease platelet activity.60,61 At the cellular level, nitrite has been shown to inhibit mitochondrial ROS generation and decrease leukocyte recruitment at sites of inflammation.62

Figure 5

(A) Nitrite reduction to nitric oxide is favoured by decreasing physiological oxygen tensions and low pH, via non-enzymatic pathways (in the presence of acid or reducing substrates) or enzymatic pathways catalysed by metal-containing enzymes. The nitric oxide generated modulates critical signal transduction processes inducing cytoprotection, vasodilation, and inhibiting SMC proliferation. SMC, smooth muscle cell (adapted from ref. 55 with permission). (B) Nitrite is not subject to tolerance formation in non-human primates. Nitrite was infused constantly over a period of 14 days. Every day, nitrite boluses were administered, and before and after the bolus, blood pressure was determined. Nitrite-induced drops in the blood pressure were similar along the 14 days of continuous nitrite infusion. Figure adapted from ref. 63 (with permission).

More recently, Dejam et al. found that nitrite is a relatively potent and fast vasodilator at near-physiological concentrations and, interestingly, that nitrite functions as an endocrine reservoir of NO by producing remote vasodilation, with a selective preference for oxygen-deprived territories. Importantly, the non-enzymatic nature of the biotransformation of nitrite to vasodilatory compounds might turn into an advantage in terms of sustainability of the vasodilatory effects of this molecule since it would imply that enzymatic tolerance is by-passed. In studies testing prolonged infusions of nitrite in humans and primates it did not lead to a sustained blunting in the arterial blood pressure (likely reflecting the activation of compensatory mechanisms), but the haemodynamic effect of small bolus administrations of nitrite was not changed, thus reflecting the absence of tolerance-like phenomena63(Figure 5).

While these effects appear to suggest that nitrite could be an alternative to organic nitrates, problems with the administration pathway obviously exist: for instance, the easiest administration route of nitrite would be under the form of nitrate which can be converted by commensal symbiotic oral flora through nitrate reductase enzyme systems, but the use of antiseptic mouthwashes abolishes these reactions.64

Conclusions and clinical implications

More than a century after their first clinical use, nitrates remain interesting drugs. The improved understanding of the pharmacology of these drugs, however, does not compensate the lack of clinical data on their long-term effects. A number of very basic concepts remain to be investigated: for instance, recent data demonstrate that low doses of GTN have similar haemodynamic effects as standard ones, with the advantage that they do not induce tolerance.16 As stated by the authors of this observation, it might well be that nitrate tolerance and toxicity are the result of supraphysiological (and suprapharmacological) administered doses, a hypothesis that appears particularly senseful when one considers that a clear dose–response relationship between different dosages of GTN and exercise capacity improvement has never been presented. In the setting of long-term therapy, non-randomized studies suggest that the therapy with long-term nitrates increases the incidence of acute coronary syndromes.65 At the same time, data from the GRACE registry suggest that therapy with nitrates causes a shift from ST-elevation to non-ST elevation myocardial infarctions,3 which would imply a positive impact on the pathophysiology of plaque rupture and/or a preconditioning-like effect. As well, patients should be educated on the use of short-term sublingual formulations in the situational prophylaxis of angina: self-administration of nitroglycerin or isosorbide dinitrate before physical or emotional stress may indeed improve both exercise tolerance and quality of life. Finally, more research is necessary on alternative NO-based therapies, for instance the administration of nitrite/nitrate or PETN. While basic science has provided more answers to the still unresolved question of nitrate tolerance, the most important question, i.e. how does this therapy change patient prognosis, remains unanswered.


This research was supported by the Stiftung Mainzer Herz to all authors, research grants from Actavis Deutschland GmbH to A.D. and a research grant from the Center of Thrombosis and Hemostasis Mainz to T.G. (BMBF 01EO1003). The authors are responsible for the contents of this publication.

Conflict of interest: Authors have received lecture honoraria from Actavis Deutschland GmbH, 3M and Pohl Boskamp.


View Abstract