Introduction

Inhaled nitric oxide (iNO) has been used in Europe for treatment of acute respiratory failure and pulmonary hypertension for several years, both in the operating room and the intensive care unit. In the middle 1980s Higenbotham and his group [1, 2] were the first to demonstrate that iNO selectively decreases pulmonary artery pressure (PAP) in a series of patients with primary pulmonary hypertension. Then it was demonstrated that iNO can selectively reverse experimental pulmonary arterial hypertension [3]. In the early 1990s it was shown that iNO selectively decreases pulmonary arterial hypertension and improves arterial oxygenation in patients with acute respiratory distress syndrome (ARDS) [4, 5, 6, 7, 8, 9]. The rationale for the treatment of critically ill patients with iNO was based on these studies [1, 2, 3, 4, 5, 6, 7, 8, 9]. However, subsequent randomised controlled trials (RCTs) failed to confirm an improvement in survival or morbidity in critically ill patients treated with iNO [10, 11, 12, 13, 14, 15]. In addition, to date there is no drug approval for these indications in adults, although iNO is still extensively used as an off-label drug, and many clinicians consider it an important treatment, combining effective selective pulmonary vasodilatation with a favourable pharmacological profile [16]. iNO has been approved for treatment of term and near-term neonates with hypoxic respiratory failure. The provision of a pharmaceutical product has led to high drug costs and an increased need for justification of the clinical use of iNO in adults in daily practice. This suggested to our group that recommendations should be established on the use of iNO in adults covering all aspects of current and potential applications based on expert opinion.

Methods

An Advisory Board was established under the auspices of the European Society of Intensive Care Medicine and European Association of Cardiothoracic Anaesthesiologists to coordinate the scientific program of the meeting. The board consisted of experts with proven scientific or clinical expertise relevant to the clinical use of iNO. The board identified a further panel of experts who were invited to act as section leaders whose role was to review the literature in their designated subject area, taking special care to ensure the presence of different opinions. Section leaders were asked to produce written summaries of their subject area, which were then circulated to delegates prior to the meeting and which formed the basis of the evidence presented to delegates at the expert meeting itself. A further panel of opinion leaders were invited to attend the meeting on the basis of their known interest in the use of iNO or their status as opinion leaders in the field of adult intensive care. The European Society of Intensive Care Medicine and the European Society of Cardiothoracic Anaesthesiologists were officially represented at the meeting. At the expert meeting each subject area was presented in summary by the section leader(s), following which open discussions led to the composition of draft expert recommendations statements. These were then edited and re-presented to delegates with further discussion and reading leading to final agreement on the individual recommendations.

The first part of this program was built upon discussions among a core group of experts, and this led to draft recommendations covering areas such as clinical pharmacology, toxicity, dosing, administration and various indications supported by appropriate literature and clinical data analysis. These draft recommendations were made available to a wider group of physicians through a dedicated restricted website. Following discussions based upon these statements revisions were published online. A 2-day conference was then organised, enabling 58 experts from different specialties and coming from 14 European Union countries to openly discuss all related issues and jointly agree on recommendations. During this conference an Editorial Committee was formed to summarise expert recommendations. These statements were then published anew on the dedicated, restricted website for final review and comments by all participants. Following a last round of online discussions the Editorial Committee prepared the final article which is presented in this contribution. The cost of this project, including hotel and accommodation, travel, online conferencing facilities, IT support and website, expenses for preparation work, was approx. €218,000 (€20,000 for the first part and €198,000 for the second part of the program) which was supported through an unrestricted grant from INO Therapeutics. The process of producing the present expert recommendations was entirely independent of the sponsoring company, and the contributors specified their potential conflicts of interest. The sponsor has no authorship or editorial control over the content of the meetings or any subsequent publication. Most of the expense for this effort has been time by the Committee.

Results

Clinical pharmacology

iNO acutely relaxes constricted vascular smooth muscle leading to vasodilatation of the pulmonary circulation with no measurable haemodynamic action outside the lung (‘selective pulmonary vasodilatation’). In addition, iNO potentially dilates constricted bronchial smooth muscle, and it may improve arterial oxygenation in hypoxaemic patients by reducing the intrapulmonary shunt leading to enhanced matching of ventilation and perfusion. The selective pulmonary vasodilator action of iNO has been confirmed in various animal models [3], in a human model of acute alveolar hypoxia [17] and in patients with pulmonary arterial hypertension resulting from pulmonary vascular constriction [1, 2, 18, 19]. However, due to its short half-life sustained vasodilatation requires the continuous delivery of iNO to the lungs. Sudden disruption of iNO therapy can therefore result in a severe withdrawal reaction with rebound and possibly severe vasoconstriction [20]. The bronchodilator effect of iNO is dose dependent in anaesthetised animals [21] and in volunteers or patients with active bronchoconstriction [22]. Even at the high doses of iNO used in these studies (80 ppm) the bronchodilator response of iNO was less effective than a subsequent inhalation of a standard β2-agonist [22].

Several clinical studies have tested the use of iNO for treatment of acute pulmonary hypertension or hypoxaemia employing doses between 3 and 80 ppm. However, these acute physiological effects did not alter clinical outcome parameters, such as mortality or morbidity, and a high proportion of patients do not respond to iNO therapy (non-responders). There is some evidence from experimental and human studies for potential pharmacodynamic effects outside the pulmonary circulation, mainly on diuresis and natriuresis [23, 24], platelet function [25, 26], and modulation of the immune response [27].

Expert recommendations:

  • It has been conclusively demonstrated in human experimentation and clinical studies that exposure to iNO causes a concentration-dependent and immediate selective pulmonary vasodilatation in the presence of pulmonary vasoconstriction in most patients.

  • Nitric oxide induces vasorelaxation in ventilated portions of the lung and redistributes pulmonary blood flow, thus reducing intrapulmonary shunting in most hypoxaemic patients, at concentrations ranging from 0.1–10 ppm iNO. However, the optimum dose may vary over time and between different subjects.

  • iNO is believed to have other pulmonary and extrapulmonary effects. Their clinical relevance and concentration-response relationships remain to be investigated.

Synergistic effects

The rationale for combining iNO with other therapeutics, either pharmacological or nonpharmacological, is to obtain a synergistic or additive effect on pulmonary vascular tone in patients with pulmonary arterial hypertension or hypoxaemia. Most proposed synergistic drugs are effective in influencing only one or the other of these two potential therapeutic aims. For example, prostacyclin [28] and adenosine [29] directly stimulate the synthesis of cyclic adenosine monophosphate (cAMP) whereas phosphodiesterases inhibitors inhibit the breakdown of cAMP and cyclic guanosine monophosphate (cGMP), thereby effecting pulmonary vascular relaxation through signalling pathways that are different from those which are directly brought about by NO.

Expert recommendations:

  • The rationale for combining iNO with other drugs is to obtain an additive (or synergistic) effect and to induce an additional reduction in pulmonary vascular tone and/or further optimisation of pulmonary gas exchange than is obtained by use of iNO alone.

  • There are clinical reports of the co-administration of ‘synergistic’ drugs with iNO. The majority of synergistic drugs are effective in influencing only one or another of these desired therapeutic aims [30, 31, 32, 33, 34, 35, 36, 37].

  • Only the association of iNO and inhaled nebulised prostacyclin has shown, in a limited number of patients, positive effects on both pulmonary hypertension and gas exchange [36]. The therapeutic benefit of this synergistic response has yet to be determined.

  • Published studies on the use of potentially synergistic drugs in association with iNO report effects on small populations or are methodologically inadequate.

  • Dose-response studies with both iNO and the associated drugs are incompletely defined.

  • The underlying molecular mechanisms of interaction between NO and potentially synergistic drugs and the cross-talk pathways of the two drugs acting together are only partially understood.

  • The interpretation of clinical data in individual patients and from small published experiences must therefore be made with caution.

  • On the basis of current evidence the clinical use of synergistic drugs in adults in association with iNO cannot be recommended outside the confines of clinical trials.

Toxicology, monitoring, delivery, transport

Over the course of the past decade iNO has been administered to numerous patients without any apparent major side effect [10, 11, 12, 13, 14, 15]. Although the use of iNO is considered to be safe, and there is no evidence of direct NO toxicity at clinically relevant doses, precautions and safety regulations must be taken into account, especially the risk of exposure to higher oxides of NO (i.e. nitric dioxide). Therefore care should be taken to use iNO in humans that has been manufactured according to agreed good manufacturing practice standards (medical grade iNO). These issues have been reviewed in detail previously, including toxicology, monitoring of iNO therapy, delivery and procedures for transport of patients on iNO together with environmental issues and considerations on staff training [38].

Toxicology and monitoring

Expert recommendations:

  • There is no evidence of direct NO toxicity at clinically relevant doses.

  • Methaemoglobin should be measured 4 h after commencing iNO and daily thereafter.

  • Clinically significant levels of methaemoglobin are unlikely to result unless iNO concentrations over 20 ppm are administered.

  • Administration of iNO is associated with NO2 formation which is potentially toxic.

  • Environmental exposure limits exposure to NO2 to a 2-ppm 8 h time-weighted average in non-intubated patients and staff.

  • Clinically significant levels of NO2 are unlikely to occur when iNO is delivered by an efficient delivery system at concentrations of 20 ppm or less.

  • If long-term iNO treatment is to be undertaken, attempts should be made to reduce the concentration of iNO to 10 ppm or less to further reduce exposure to potentially toxic NO2.

  • Use of iNO is associated with accumulation of nitrate and nitrite. The significance of these increases is uncertain.

  • There are no long-term follow-up studies from which freedom from late adverse effects following iNO therapy can be ascertained.

Delivery

Expert recommendations:

  • INO should be delivered by a system approved for clinical use, conforming to appropriate CE standards and capable of meeting the following specifications.

  • It should be able to deliver a constant concentration of iNO to the patient.

  • The design should minimise the generation of NO2 and should have continuous monitoring and alarms for inspired NO, NO2 and O2.

  • A backup system for hand ventilation should be immediately available to ensure continuous iNO delivery in the case of delivery device malfunction.

  • The delivery device should be compatible with the type(s) of ventilator(s) in use, which at present does not include closed-circuit, rebreathing systems during anaesthesia.

Transport

Expert recommendations:

  • iNO therapy must be delivered without interruption when patients are transferred within or between hospitals.

  • An iNO delivery and monitoring system which is of low weight and is designed and approved for use during transport in road and air ambulances is urgently needed.

Contraindication

  • Methaemoglobin reductase deficiency (congenital or acquired)

Diagnostic assessment

Heart failure

The frequently elevated PVR in patients with chronic left ventricular failure may be a result of dysregulation of vascular smooth muscle tone and structural remodelling [39]. There is growing evidence that the dysregulation of pulmonary vascular tone in disease states, such as chronic heart failure involves vascular endothelial dysfunction with impaired endogenous NO availability in the pulmonary circulation. Endothelial cell dysfunction predisposes the vessel wall to vasoconstriction, leucocyte adherence, platelet activation, mitogenesis, thrombosis, impaired coagulation and vascular inflammation [40]. In addition, endothelial function testing may serve as a useful biomarker of pulmonary circulatory function [41]. Bocchi and colleagues [42] reported sudden development of pulmonary oedema in patients with severe congestive heart failure treated with iNO, which was most probably due to a sudden increase in left atrial filling caused by pulmonary vasodilatation rather than a direct negative inotropic effect of iNO [43]. iNO may be used as a test for pulmonary vasoreactivity before cardiac transplantation.

Expert recommendations:

  • Response to iNO treatment may identify patients still suitable for heart or heart/lung transplantation or to help to identify patients with congenital heart disease suitable for further intervention.

  • iNO decreases PVR but potentially increases left ventricular preload which may be dangerous in left ventricular dysfunction. In the presence of left heart dysfunction it is increasingly recognised that iNO testing should be performed only after optimising heart failure therapy immediately prior to testing.

  • iNO testing is useful to demonstrate the remaining reactivity of the precapillary component of postcapillary pulmonary hypertension. Reduction in PAP/PVR shown by iNO testing does not imply that long-term iNO therapy should be instituted

Pulmonary arterial hypertension

Pulmonary arterial hypertension, previously known as primary pulmonary hypertension, is a rapidly progressive disease of the pulmonary vasculature with consecutive right heart failure [44]. Prognosis may be improved in adult patients responding to calcium channel blocker and/or anticoagulation [45] and in patients treated with continuous prostacyclin [46]. The response to acute vasodilator testing has important implications both for the choice of therapy and for prognosis [28, 47, 48]. For example, only patients with a positive response to acute vasodilator testing remain suitable for long-term treatment with calcium channel blocker. Those who do not are treated with long-term intravenous epoprostenol. Today intravenous epoprostenol adenosine or iNO is recommended for acute vasodilator testing in adults, defined as a decrease in the mean PAP of at least 10 mmHg to less than 40 mmHg with an increased or unchanged cardiac output [49]. iNO has been shown to be superior to prostacyclin for this use [50] whereas aerosolised iloprost is more effective in improving oxygenation and haemodynamics in patients with primary pulmonary hypertension [51]. Combining oxygen and iNO can identify a greater number of appropriate candidates for corrective cardiac surgery or transplantation during preoperative testing [52].

Expert recommendations:

  • iNO is a potent selective pulmonary vasodilator which used alone or in combination with other vasodilators may be useful in revealing the extent of reversibility (if any) in selected patients with pulmonary arterial hypertension.

  • iNO clearly identifies responders suitable for long-term treatment with calcium channel blockers.

  • iNO dose recommended for acute vasodilator testing should be 10–20 ppm. iNO does not have relevant adverse effects during short-term acute testing.

  • iNO combined with additional O2 may lead to further pulmonary vasodilatation.

  • There is insufficient data to recommend iNO for long-term therapy of pulmonary arterial hypertension.

Medical conditions complicated by pulmonary arterial hypertension

Thromboembolism

iNO, which decreases PAP [2], is likely to unload the right ventricle in pulmonary embolism and chronic thromboembolic pulmonary hypertension. Furthermore, its platelet anti-aggregate property could prove beneficial [53, 54]. iNO for severe pulmonary embolism or chronic thromboembolic pulmonary hypertension has not been investigated in randomised controlled trials. Animal studies have shown that iNO decreases PVR [55] and platelet aggregation [54]. Use of iNO has been reported in case reports from patients with massive pulmonary embolism leading to cardiogenic shock. iNO decreased right ventricular afterload, improved cardiac output (CO) and increased arterial oxygen content [56, 57, 58]. After thrombendarterectomy iNO significantly improved arterial oxygenation but had a negligible effect on PAP. In one case postoperative hypotension progressively reversed with iNO [59].

Expert recommendations:

  • There are no controlled trials which support the routine use of iNO in patients with thromboembolic disease.

  • iNO might be of benefit in selected patients with thromboembolic disease who have severe right ventricular failure and/or severe hypoxaemia.

Sickle-cell disease

Stiffened red blood cells lead to impaired blood flow in the microcirculation, veno-occlusive phenomena, inflammation and haemolysis [60]. Given the depletion of endogenous NO by cell-free haemoglobin iNO may restore endothelial homeostasis by enhancing pulmonary vasodilatation and inactivation of cell-free haemoglobin. In transgenic sickle-cell mice iNO protects from hypoxia/reoxygenation induced lung injury, attenuates inflammatory response, modulates genes involved in ischaemic/reperfusion injury and improves survival [61]. Case reports in the acute chest syndrome have shown an improved oxygenation and decreased PAPm with iNO [62, 63]. A small prospective, double-blind, randomised, placebo-controlled paediatric study has shown that iNO is associated with a greater reduction in pain and less use of morphine over the first 6 h but not duration of hospitalisation [64].

Expert recommendations:

  • There is limited clinical experience suggesting that iNO improves oxygenation and decreases PVR in some patients with acute chest syndrome.

  • A single randomised, placebo-controlled trial in severe vaso-occlusive disease suggests that use of iNO is associated with improved pain control.

  • At present there are insufficient data to recommend the routine use of iNO to manage complications of sickle cell disease.

Chronic obstructive pulmonary disease

In chronic obstructive pulmonary disease pulmonary hypertension responds poorly to oxygen therapy and has an adverse impact on prognosis. Pathophysiological studies in stable patients have consistently documented that iNO lowers PAPm and PVR [65]. Studies report both an improvement [65] and a worsening in arterial oxygenation [66]. The concomitant administration of O2 either prevents a decrease in [67] or increases PaO2 [33]. In intubated and mechanically ventilated patients neither arterial oxygenation nor cardiac function are influenced by iNO [68]. A recent prospective randomised trial comparing O2 alone and O2 plus iNO over a 3-month period in stable patients demonstrated a reduction in PAPm and PVR and an increase in cardiac output at the end of the trial period in the O2 plus iNO group only [69]. No study to date has explored the impact of either acute or chronic iNO on patient outcome.

Expert recommendations:

  • There is no evidence that iNO therapy is of clinical benefit in patients with chronic obstructive pulmonary disease.

Cardiac surgery

Perioperative pulmonary hypertension in adult cardiac surgery

Several strategies have been employed to avoid postoperative pulmonary hypertension, including pharmacological inhibitors of inflammation, improved cardioplegic solutions, and new surgical techniques avoiding the use of CPB. Treatment strategies include the use of iNO, prostaglandins and ultimately ventricular assist devices.

CPB has been shown to reduce NO production within the pulmonary vasculature, and replacement of endogenous NO by treatment with iNO lowers the increased vascular resistance [70] and reduces markers of CPB-induced inflammatory activation [71]. iNO was more effective than milrinone in lowering PVR in 45 adult cardiac surgery patients [72]. iNO is equally effective as standard intravenous vasodilators, without altering systemic haemodynamics [73]. Doses greater than 20 ppm offer no advantage [73, 74], and patients with a high preoperative PVR have a greater response to iNO [75]. Similar data have been reported for treatment of pulmonary hypertension in high-risk cardiac surgery patients [19], patients undergoing heart transplantation [76, 77] and insertion of left ventricular assist devices [78, 79].

Expert recommendations:

  • There are no randomised, placebo-controlled clinical trials that show that iNO improves clinical outcomes in adults with perioperative acute right ventricular dysfunction and elevated PVR

  • Clinical experience suggests that in patients with confirmed acute right ventricular dysfunction and elevated PVR, use of iNO may result in haemodynamic improvement when used during or after cardiac surgery

  • Prior to iNO administration right ventricular function should be optimised with conventional treatment (specifically, ensuring optimal ventilation, thoracic decompression, preload optimisation, attempting to lower PVR with standard measures, increasing systemic perfusion pressure to increase coronary perfusion, and reduction in myocardial oxygen consumption).

Left ventricular assist devices

Ventricular assist devices (VAD) can dramatically improve survival and morbidity of patients with severe acute or chronic heart failure, either as a bridge to transplantation, as a bridge to recovery or as a permanent therapy [80]. In the presence of pulmonary hypertension filling of the left-sided VAD is impaired and the right ventricle cannot be unloaded further worsening the right ventricular dysfunction. This situation may necessitate the implant of a biventricular assist device. The development of right ventricular dysfunction remains a serious clinical problem being associated with a high transfusion rate, multiple organ failure, increased length of stay and a high mortality rate [80]. Recent studies ranging from case reports, observational studies and RCTs have demonstrated beneficial effects of iNO therapy in such patients [43, 81, 82]. However, there are no data to suggest that this transient action of iNO has any lasting effect that favourably influences clinical outcomes. The doses used in theses studies varied between 10–40 ppm. Potential concerns arise from the possibility of abrupt rebound pulmonary hypertension from withdrawal of iNO, and sudden increases in left atrial filling, similar to the mechanisms occurring in patients with acute heart failure [43]. Currently the role of iNO in the treatment of heart failure patients undergoing left-sided VAD insertion is tested in an ongoing large, multicentre trial.

Expert recommendations:

  • There is a high prevalence of right ventricular dysfunction refractory to conventional clinical measures in patients in whom insertion of a left ventricular assist device is required

  • The expert panel believe that iNO therapy is effective in providing favourable pulmonary haemodynamics leading to improved right ventricular and left-sided VAD assisted cardiac output in patients with pulmonary hypertension and inadequate left-sided VAD flow refractory to conventional manoeuvres. On the basis of these improved critical physiological variables the expert panel recommend that it is reasonable to consider the use of iNO in this clinical situation among other vasodilator therapies.

  • Further studies are required to better define the indications for iNO in cardiac surgical patients and in particular to elucidate its effects on clinical outcomes.

Heart transplantation

Cardiac transplantation remains the surgical option for treatment of end-stage heart disease. In many cases severe pulmonary hypertension is present at time of transplantation, contributing to life-threatening right heart failure, a significant predictor of early postoperative mortality [83]. iNO reduced elevated pulmonary resistance in adult cardiac transplantation [84], was more effective in reducing PVR than intravenous prostaglandin E1 [85], reduced the occurrence of right ventricular failure [76] and increased survival irrespective of the preoperative condition in a large case series of patients undergoing cardiac transplantation [77]. Until now there are no randomised clinical trials determining the role of iNO for the treatment of pulmonary hypertension in cardiac transplant patients. However, iNO therapy is considered to be a part of a multimodal treatment strategy by expert opinion, including iNO, optimising right ventricular filling, increasing heart rate and reducing PVR [76].

Expert recommendations:

  • Several institutions with extensive experience in cardiac transplantation use and recommend iNO as a part of standard therapy for all cardiac transplant procedures associated with increased PVR.

  • Following weaning from the ventilator iNO therapy may be discontinued and be replaced by intravenous vasodilators.

Thoracic surgery

One-lung ventilation

Hypoxaemia frequently occurs during one-lung ventilation (OLV) and is due mainly to an increased pulmonary blood flow to the non-ventilated lung (intrapulmonary shunt) [86, 87]. High inspired oxygen concentrations, intermittent two-lung ventilation, continuous positive airway pressure and high frequency jet ventilation to the ventilated, dependent lung are common interventions to improve arterial oxygenation. Two studies suggest that in the absence of arterial hypoxaemia or pulmonary hypertension iNO does not modify oxygenation or pulmonary artery pressure [88, 89]. In theory iNO should increase pulmonary blood flow to the ventilated, dependent lung by dilating the pulmonary vasculature [90]. However, Fradj et al. [91] reported that iNO at 20 ppm was not superior to nitrogen in the treatment of arterial hypoxaemia during OLV. In another study iNO administration was effective only in a subgroup of patients with severe hypoxaemia, high intrapulmonary shunt and pulmonary hypertension during OLV [92].

Almitrine combined with iNO can attenuate arterial hypoxaemia during OLV or in patients with ARDS [30, 31, 34, 93, 94, 95]. However, as iNO during OLV is seldom effective when used alone, the role of iNO to improve the effects of almitrine has yet to be confirmed [94]. Therefore the expert panel cannot make any recommendations on the use of a combination of iNO with intravenous almitrine, taking also into consideration the long half-life time of the drug, its potential for systemic toxicity and its limited availability in most countries.

Expert recommendations:

  • There is no evidence to support the routine use of iNO for the prevention or reversal of hypoxaemia during OLV.

  • Some patients who develop very severe hypoxaemia during OLV which is refractory to conventional management may benefit from iNO.

Ischaemia-reperfusion injury

Postpneumonectomy pulmonary oedema is a potential complication of lung resection with an associated mortality of 50–100%. Age, side of lung resection, volume of fluids infused perioperatively, preoperative lung function, use of fresh-frozen plasma, mechanical ventilatory support and the extent of mediastinal lymphatic dissection all contribute to the risk of developing postpneumonectomy pulmonary oedema. During OLV relative ischaemia of the non-ventilated lung is followed by re-expansion and reperfusion of the remaining lung tissue leading to ischaemia-reperfusion with oxidative damage and lung injury [96]. iNO has been shown to be of therapeutic value in patients with postpneumonectomy lung injury by selectively dilating the pulmonary vasculature and improving ventilation/perfusion mismatch and oxygenation [97, 98]. However, the role of iNO in preventing ischaemia-reperfusion injury during clinical lung transplantation remains controversial [99]. Two uncontrolled clinical studies suggest that NO prevents ischaemia-reperfusion injury [77, 100]. These findings were not confirmed by a randomised, double-blinded, placebo-controlled trial studying the effect of iNO (20 ppm) starting 10 min after reperfusion on mortality and physiological variables in patients undergoing lung transplantation [101].

Expert recommendations:

  • There is no evidence that iNO administered during reperfusion prevents or attenuates the development of ischaemia/reperfusion injury after lung transplantation or thrombendarterectomy.

  • It remains to be determined whether iNO applied alone or together with anti-inflammatory agents administered prior to or at reperfusion modulates reperfusion injury in humans.

  • There is evidence from one RCT that iNO may reduce the need of cardiopulmonary bypass in sequential bilateral lung transplantation by improving haemodynamic stability or oxygenation during clamping of the recipients pulmonary artery or following reperfusion of the firstly implanted lung [102].

ARDS/ALI

In acute lung injury (ALI) and ARDS there is a marked maldistribution of pulmonary perfusion in favour of poorly or non-ventilated lung areas. iNO therapy offers the possibility to selectively modulate the pulmonary blood flow, reduce pulmonary hypertension and improve matching of ventilation to perfusion [4]. Early clinical trials of iNO demonstrated improved arterial oxygenation and pulmonary haemodynamics in patients with ARDS [4, 5, 6, 7, 8, 9, 103]. Further clinical studies demonstrated that combining iNO with positive end-expiratory pressure or prone positioning augments the beneficial effects on arterial oxygenation [33, 104, 105], thus pointing to a valuable role for iNO in combination with recruitment manoeuvres for treatment of severe hypoxaemia in patients with ARDS.

Subsequent RCTs involving over 900 patients with ALI/ARDS confirmed a significant increase in arterial oxygenation in the majority of patients without any sign of clinically relevant side effects. However, this increase in PaO2 was only transient, and no improvement in important outcome parameters such as mortality and ventilator-free days was found [10, 11, 12, 13, 15, 106].

With respect to the available data the expert panel felt it reasonable to use iNO as a rescue treatment in patients with severe refractory hypoxaemia. This interpretation is essentially suggested by a recent meta-analysis of RCTs on iNO therapy [107]. Aerosolised prostaglandins have been shown to exert similar effects as iNO therapy in improving arterial oxygenation and/or pulmonary haemodynamics and may offer an alternative form of therapy [108, 109].

Expert recommendations:

  • iNO improves arterial oxygenation and haemodynamics in most ARDS patients in the acute phase. However, there is no evidence of any beneficial effect of iNO beyond the first 24–72 h of therapy, and no benefit on clinical outcomes has been demonstrated by at least four published randomised trials of ARDS.

  • However, RCTs provide no evidence that iNO affects mortality in ALI/ARDS, and its routine use in these conditions cannot be recommended

  • Available RCTs do not resolve the question of whether iNO leads to any clinically significant benefits (such as improved survival, increased ventilator-free days and reduced use of extracorporeal membrane oxygenation) in certain subgroups of patients, such as those with severe hypoxaemia not responding to conventional treatment.

  • The expert group considered that, based on the known physiological effects of iNO in ARDS and whilst awaiting evidence from further clinical trials, it is reasonable to use iNO as a rescue treatment in patients with severe refractory hypoxaemia

Conclusion

These recommendations were compiled by an interdisciplinary panel of European experts in the field of iNO therapy. They are designed to promote the safe use of this therapy and to suggest areas in which iNO therapy may be of benefit in adults. In summary, the expert panel identified a variety of medical conditions in which the use of iNO as a rescue treatment in patients with severe acute pulmonary arterial hypertension and/or severe refractory arterial hypoxaemia is reasonable. In addition iNO is a useful drug for testing pulmonary vasoreactivity in patients undergoing heart transplantation or in patients with pulmonary arterial hypertension. It is hoped that these recommendations will encourage evidence-based practice and further clinical trials on the use of iNO therapy in adults.