Management of pulmonary hypertension in infants

ABSTRACT

Introduction

Pulmonary hypertension in infants is characterized by elevated pulmonary vascular resistance which leads to right-to-left shunting of deoxygenated blood across the ductus arteriosus and foramen ovale into the systemic circulation leading to hypoxemia. It is associated with significant morbidity and the mortality rate has been reported to be between 7 and 15%, which is increased particularly in infants with congenital diaphragmatic hernia (CDH) or bronchopulmonary dysplasia (BPD).

Areas covered

An update on the current pharmacological management strategies in PPHN is provided. A PubMed and Google Scholar search was performed using the keywords: ‘persistent pulmonary hypertension of the newborn,’ ‘PPHN,’ ‘inhaled nitric oxide,’ ‘sildenafil,’ ‘surfactant,’ ‘magnesium sulfate,’ ‘PDE 3 inhibitors,’ milrinone,’ ‘endothelin receptor antagonists,’ ‘prostacyclin,’ ‘inhaled Iloprost,’ ‘riociguat,’ ‘prostaglandin’ and ‘vasopressin’.

Expert opinion

Inhaled nitric oxide (iNO) is the mainstay of treating PPHN, but trials which are adequately powered should address whether use of iNO does reduce BPD and long-term respiratory morbidity. Importantly, it is essential that all other pharmacological treatments of PPHN are assessed in appropriately designed trials and future studies should incorporate detailed echocardiographic assessments to determine which infants are most likely to respond and biomarkers identified.

KEYWORDS:

• Endothelin receptor antagonists
• inhaled nitric oxide
• magnesium sulfate
• milrinone
• PDE 3 inhibitors
• persistent pulmonary hypertension of the newborn
• sildenafil
• surfactant

1. Introduction

Pulmonary hypertension (PH) in infants is characterized by elevated pulmonary vascular resistance which leads to right-to-left shunting of deoxygenated blood across the ductus arteriosus and foramen ovale into the systemic circulation leading to hypoxemia. Infants have swinging oxygen saturations, exacerbated by agitation or handling . It is associated with significant morbidity and the mortality rate has been reported to be between 7 and 15% [1], which is increased particularly in those infants with congenital diaphragmatic hernia (CDH) or bronchopulmonary dysplasia (BPD). The overall incidence of PH is 1.8 per 1000 births [2], with the incidence being higher in late preterm infants (5.4 per 1000 live births) rather than term infants (1.6 per 1000 live births) [1]. In the fetal circulation, the presence of circulating vasoconstrictors including endothelin-1 and thromboxane contribute to pulmonary vasoconstriction maintain a high pulmonary vascular resistance (PVR) in comparison to systemic vascular resistance [3]. After birth and the first breath the infant takes, there is significantly increased pulmonary blood flow. This is due to the rapid increase in the pulmonary vascular bed associated with the first breath, the effect of vasodilators such as oxygen, and the umbilical cord being clamped. This leads to pulmonary vascular vasodilation and a reduction in the PVR [4]. PH can be caused by impaired relaxation of the pulmonary vasculature [5], due to underdeveloped pulmonary vasculature associated with lung hypoplasia and decreased pulmonary artery number, maladaptation in which the pulmonary vessels have normal structure and number but have abnormal vasoreactivity or excessive muscularisation with increased smooth muscle cell thickness and distal extension of muscle into vessels not usually muscularised [6]. A number of pathological conditions give rise to pulmonary hypertension. For example, PH is common in infants with bronchopulmonary dysplasia (BPD) [7,8]. Affected infants have a high mortality between 14 and 36% [9] and survivors have increased neurodevelopmental delay. Risk factors in one series for early PH (EPH) in infants with BPD included intraventricular hemorrhage (IVH) and high frequency oscillatory ventilation (HFOV) and for late PH (LPH) were maternal diabetes, EPH, severe IVH and systemic corticosteroid use [8]. In another retrospective study which included 182 infants born at less than 33 weeks of gestational age who had developed BPD, 22 developed late PH and 100% were ventilated on day 28. They had increased mortality (36.4% versus 1.9%), a prolonged duration of stay (147 versus 109 days) and a higher cost of care (£113,494 versus 78,677) [10]. The PH relating to prematurity and BPD arises from different mechanisms compared to the PH that arises in infants with congenital diaphragmatic hernia (CDH), where the causative factor is related to pulmonary hypoplasia. Where causative factors differ, treatments can have different efficacies.

Numerous management strategies have been proposed in all causes of PH, including gentle ventilation, techniques to optimize lung recruitment, improving cardiac function with inotropic support, targeting increased pulmonary vascular resistance with systemic and inhaled vasodilators [11] and extracorporeal membrane oxygenation (ECMO). This review provides an update on the current pharmacological management strategies in PH arising in the neonatal period,

2. Methods

A PubMed and Google Scholar search was performed using the following keywords: A PubMed and Google Scholar search was performed using the following keywords: ‘persistent pulmonary hypertension of the newborn,’ ‘PPHN,’ ‘inhaled nitric oxide,’ ‘sildenafil,’ ‘surfactant,’ ‘magnesium sulfate,’ ‘PDE 3 inhibitors,’ milrinone,’ ‘endothelin receptor antagonists,’ ‘prostacyclin,’ ‘inhaled Iloprost’ ‘riociguat,’ ‘prostaglandin’ and ‘vasopressin.’

2.1. Pharmacological strategies

2.1.1. Surfactant

Lung recruitment should improve oxygenation and hence reduce PVR. Potential strategies to improve lung recruitment are ventilation modes such as high frequency oscillatory ventilation and the administration of surfactant. Indeed, exogenous surfactant administration in a multi-center trial was found to significantly reduce the need for extracorporeal membrane oxygenation (ECMO) in neonates with PPHN, meconium aspiration syndrome (MAS) or sepsis, with the greatest benefit being noted in infants with relatively mild disease that is with an OI between 15 and 25 [12]. The use of surfactant as an adjunct to iNO has been postulated to improve the effectiveness of iNO and reduce the need for ECMO [5]. A randomized controlled trial (RCT) of 100 term neonates with hypoxemic respiratory failure demonstrated that surfactant combined with iNO improved oxygenation and prevented progression to severe hypoxemic respiratory failure compared to placebo with iNO. The cohort who were given two doses of surfactant as an adjuvant to iNO showed faster improvements in oxygenation leading to a significantly lower OI at 24 h and a lower proportion developed an oxygenation index (OI) greater than 40: 24% versus 50%. There was also a significant reduction in the combined outcome of death or ECMO in the cohort receiving surfactant with iNO (16% compared to 36%) [13].

2.1.2. Inhaled nitric oxide

Inhaled nitric oxide (iNO) is a potent selective pulmonary artery vasodilator. It increases cyclic guanosine (cGMP) monophosphate in vascular smooth muscle cells, which in turn activates cGMP-dependent protein kinase, opening calcium-sensitive potassium channels and thus causes membrane polarization and relaxation of the vascular smooth muscle. The OI has been used to determine when to initiate iNO treatment and monitor its efficacy. Infants who have not exhibited any improvement in the OI or an improvement was not seen within 60 min of treatment have been labeled ‘non-responders.’ The Clinical Inhaled Nitric Oxide Research Group Investigation (CINRGI) study, however, highlighted that the oxygenation response to iNO is complex as it is impacted by various clinical factors, hence the time to improvement in OI may not be a reasonable marker of the efficacy of treatment nor a predictive marker for the need for ECMO [14].

In an early randomized trial including 58 full-term infants, iNO successfully doubled systemic oxygenation in 53% of infants and in 75% of those infants the improvement in oxygenation was sustained on longer term iNO [15] In addition, significantly fewer infants in the treatment compared to the control arm required ECMO. In a systematic review of 12 RCTs of iNO in infants born at or near term, the combined outcome of death or need for ECMO was reduced, but this was due to the lower ECMO requirement [16]. In a more recent systematic review of 17 RCTs of iNO in infants born at or near term, it was concluded that iNO was effective at an initial concentration of 20ppm, except in those infants with CDH [17] and it has been questioned whether iNO worsens outcomes in infants with CDH. Indeed, Noh et al. demonstrated that iNO in infants with CDH increased mortality (adjusted OR (aOR) 2.06) and increased ECMO use compared to infants with CDH who did not receive iNO (aOR 3.44) [18]. iNO, however, has been shown to improve OI in some infants with CDH [19,20]. A single-center, retrospective study found in a cohort of 95 infants with CDH treated with iNO, 38 showed a clinical improvement (PaO₂ >20 mmHg) and were less likely to need ECMO. Those who did respond had normal left ventricular systolic function [19]. A further single-center study showed infants with CDH who initially responded to iNO with a greater improvement in the PaO₂/FiO₂ ratio were more likely to survive. An increase of 20% post-initiation of iNO being a predictive marker for survival [20]. The likely reason for this variable effect is that the PH seen in CDH infants is complex and iNO only addresses certain components. Neonates with CDH and PH have pulmonary vasoconstriction and ventilation-perfusion mismatching, which should respond to iNO. They, however, are also at risk of left ventricular hypoplasia due to in utero compression of the left ventricle (LV) by abdominal contents as well as alteration in LV filling hemodynamics. In such patients, iNO may worsen pulmonary venous hypertension [18]. In a series of infants with CDH, left ventricular diastolic dysfunction was associated with greater respiratory morbidity [21]. iNO should not be started in CDH infants without echocardiographic evidence that it might be efficacious, the response carefully monitored and iNO stopped if improvements do not occur. A personalized approach needs to be taken to the management of PH in CDH infants [22].

A systematic review of 14 RCTs in preterm infants identified nine trials of early rescue therapy in which iNO had no significant effect on mortality or BPD. There were three studies with routine use of iNO in infants with pulmonary disease with no significant reduction in mortality or BPD and two trials of late iNO based on the risk of BPD with no significant effect on BPD [23]. The confidence intervals, however, were close to one and perhaps refining the criteria for starting iNO might yield better results. Despite the lack of significant differences iNO use in prematurely born infants has increased, one study reporting the largest increase in infants born between 23 and 26 weeks of gestation [24]. A more recent meta-analysis of 17 RCTs demonstrated iNO was not effective as rescue therapy for the very ill preterm infant and early use of iNO did not prevent serious brain injury or improve survival without BPD but concluded that its use to prevent BPD requires further study [17]. A large multicentre, international trial had explored whether low dose iNO (5ppm) in infants with moderate respiratory distress (those requiring surfactant or CPAP) would reduce BPD. Despite recruiting 800 infants, no significant differences were found in the immediate outcomes [25] or at long-term follow-up [26]. Essential for future studies is to determine the most appropriate iNO dose for such infants and how iNO can be delivered effectively to the growing numbers of prematurely born infants who are supported by noninvasive respiratory support.

2.1.3. Magnesium sulphate

Magnesium sulfate is a calcium antagonist and as such can cause vasodilation of pulmonary vasculature. In four observational studies involving a total of 40 term infants, a significant improvement in oxygenation measured by changes in partial oxygen pressure, alveolar‐arterial oxygen index and the oxygen index or changes in ventilatory requirements were seen following a loading dose of 200 mg/kg and subsequent continuous infusion of 20–150 mg/kg/hr of magnesium sulfate [27]. There are no large RCTs, however, assessing the effectiveness of magnesium sulfate in PPHN [28] and the associated side effects including hypotension, due to its nonspecific mechanism of vasodilation, can be problematic in PPHN.

2.1.4. Sildenafil

Sildenafil is a cyclic guanosine monophosphate-specific phosphodiesterase-5 inhibitor, which leads to increased cyclic guanosine monophosphate which acts as a pulmonary vasculature vasodilator. In a multi-center, dose-escalation trial in infants with PPHN, sildenafil was delivered by continuous infusion for at least 48 h and up to 7 days. A significant improvement (reduction) in OI (28.7 to 19.3; p = 0.0002) was observed after 4 h of the higher compared to the lower infusion dose [29]. In three small RCTs with a total of 77 infants, an improvement in oxygenation index compared to placebo was seen following sildenafil. Those studies were conducted in resource-limited settings which did not have facilities to provide iNO or high frequency ventilation [30]. In a systematic review of eight RCTs which in total included 216 infants demonstrated an improvement in the OI after 24 h with peak effects at 72 h with no significant side-effects, but no improvements in clinically meaningful outcomes [31].

Whether Sildenafil or iNO would be more effective in infants with CDH was to be explored in the CoDiNOS trial. This was a multicentre international study, but unfortunately due to problems including regulatory approvals failed to achieve the sample size [32]. International research communities need to learn from these problems and ensure that they can be addressed in the much needed large RCTs of the future

In a study of intravenous sildenafil in prematurely born infants with early PH, the echocardiographic severity of PH and right ventricular dysfunction decreased significantly and infants who responded to sildenafil were less likely to die [33]. A multicentre, RCT did not show increased effectiveness of intravenous sildenafil as an adjunct to iNO in infants with PPHN or hypoxic respiratory failure and highlighted common adverse effects including hypotension, hypokalemia, anemia, drug withdrawal syndrome and bradycardia [34]. In a retrospective study of 19 infants with BPD associated PH, sildenafil started at a median PMA of 40 weeks was associated with a reduction in the inspired oxygen concentration and the respiratory severity score approximately 12 weeks after starting treatment. Transthoracic echocardiography changes, however, did not correlate with clinical improvements [35].

2.1.5. PDE 3 inhibitors

Milrinone acts to inhibit cyclic guanosine monophosphate-specific phosphodiesterase-3 in cardiac myocytes and vascular smooth muscle, reducing PVR. A single-center retrospective study in nine term infants with PPHN found, when used as an adjunct to iNO, intravenous milrinone significantly improved the oxygenation index in term infants with a poor initial response to iNO, particularly in the immediate 24 h of treatment [36]. Despite this, the risk–benefit ratio should be considered when using milrinone due to its apparent risk of increasing intraventricular hemorrhage [37].

Subsequent studies have also assessed the use of milrinone as an addition to other therapies in the management of PPHN. A multi-center randomized double-blind study in neonates of greater than 34 weeks of gestation which compared the use of milrinone versus placebo (normal saline) as an adjuvant therapy to iNO in PPHN, however, was terminated after 4 years due to poor recruitment. This was due to multiple factors which the authors felt included the reduced incidence of incomplete or failed response to iNO due to earlier recognition and management of PPHN [38]. In a randomized double-blind trial in resource limited settings, dual therapy of milrinone and sildenafil were more effective than single therapy in improving oxygenation [39]. In later RCT, although both milrinone and sildenafil improved the oxygenation index in resource constrained settings, milrinone was superior in improving oxygenation without lowering the blood pressure and was associated with a lower length of hospital stay. The sample size, however, was small [40].

A recent meta analysis examining the use of milrinone in infants with PH of all causes in the neonatal period attempted to use data to describe the mode of action of milrinone by pooling physiological data collected from studies using milrinone [41]. In that study, the authors demonstrated that milrinone was associated with several systemic and side effects such as a higher left ventricular ejection fraction, a higher cardiac and stroke volume index, mean pulmonary arterial pressure and increased left ventricular output. These inotropic effects were balanced by there being no positive effects on right ventricular output, right ventricular myocardial performance index, pulmonary vascular resistance or isovolumetric relaxation time, suggesting surprisingly minimal lusitropic effect. Furthermore, there were no significant differences in many of the positive parameters when compared with other inotropes. The meta-analysis showed milrinone was associated with a lower oxygen index (−3.48, 95%CI −5.54 to −1.43, p ≤ 0.0009) and serum lactate (−0.59, 95%CI −1.15 to −0.02). The authors stated that these results suggested that the positive effects of milrinone were more likely due to its systemic effects with a smaller effect noted on the pulmonary vasculature [41]. Furthermore, they highlighted that there remain significant questions about the mechanisms of action of milrinone in infants with PH.

2.1.6. Vasopressin

Animal studies have shown that vasopressin causes dilatation of the pulmonary arteries reducing pulmonary artery pressures [42]. This is by inducing endothelium derived relaxing factor (EDRF), part of the nitric oxide pathway [43]. Vasopressin was shown in a study of 26 infants with PH to improve the mean arterial blood pressure and was reported to be an effective treatment in refractory PH [44]. It has also been reported to be a successful adjunct therapy in 27 infants with CDH [45]. Early therapy with vasopressin showed a reduction in the oxygen index and associated improvements in cerebral perfusion and heart rate. Both studies highlighted hyponatremia as a side effect [44,45].

2.1.7. Endothelin receptor antagonists

Endothelin is a mediator in the development of PPHN due to its role as a potent pulmonary vascular vasoconstrictor, hence endothelin receptor antagonists (ETRA) have been proposed in the treatment of PPHN. Two small randomized controlled trials showed that there was no significant difference in morbidity or mortality compared to placebo when oral Bosentan (an ETRA agonist) was used as either a single treatment or as an adjuvant to iNO in PPHN [46]. A retrospective medical records review demonstrated that in 40 infants Bosentan improved oxygenation, including in those who were receiving iNO [47].

2.1.8. Prostacyclin & inhaled iloprost

Prostacyclin is an arachidonic acid metabolite that causes systemic and pulmonary vascular vasodilation by increasing circulating cAMP [48]. A study of eight patients more mature than 34 weeks of gestation demonstrated that escalating dose regimes of prostacyclin infusions were able to reverse shunting in PPHN and none of the infants required further support with ECMO or experienced severe neurological complications. Despite this, two infants went on to develop bronchopulmonary dysplasia [49]. lloprost is a prostacyclin analogue with several case reports showing the use of its inhaled form as a treatment for PPHN where iNO was not available. Most recently, in six infants with PPHN, inhaled iloprost at an escalating dose of 2–4 µg every 30 min to 8 h, was shown to reduce the OI and FIO₂ requirements in the whole cohort [50]. In a cohort of 20 neonates with PPHN, the use of endotracheal iloprost instillation significantly decreased the systolic pulmonary artery pressure, oxygen saturation index, mean airway pressure and fraction of inspired oxygen values [51]. A retrospective chart review of 22 patients who did not respond to iNO demonstrated inhaled iloprost was well tolerated and improved oxygenation [52]. Treprostinil, a prostacyclin analogue, was used in 51 CDH infants with improvements in right ventricular size and function. The majority of the infants had required ECMO and treprostinil was given to infants who had impending right ventricular failure or to aid weaning from ECMO. Those promising results require further investigation in a randomized trial [53].

2.1.9. Riociguat

Riociguate is a guanylate cyclase stimulator which cause pulmonary vasodilatation via the same pathway as nitric oxide, but downstream and this independently of nitric oxide levels [54]. In addition to this independent effect, it has been shown to increase the sensitivity of the endothelium to endogenous nitric oxide [55]. It was shown in a randomized, multicentre control trial in adults to significantly reduce PH and improve functional outcomes such as exercise tolerance [56]. Whilst studies in neonatal populations are limited, a case series of two patients who were born at term with severe refractory pulmonary hypertension, had improvement when riociguat was introduced at four to 6 months of age [57]. In both cases, who were responsive to iNO but poorly responsive to sildenafil, the introduction of riociguat was associated with improved pulmonary dynamics and eventual discontinuation of iNO. A case series of 10 infants with PH for more than 1 month showed improvements in right ventricular function by echocardiography assessment and weaning from iNO was reported [58]. That study included six infants born between 23 and 33 weeks of gestation as well as four term born infants.

2.1.10. Prostaglandin therapy

In infants with severe PH, opening or maintaining the patency of the ductus arteriosus has been suggested as a means to offload the right ventricle and improve right ventricular function [59]. In a study of 41 infants born at term who did not go onto require ECMO, the use of prostaglandin (PGE) was associated with within patient effects of improved blood lactate levels and right ventricular function on ECHO [60]. In that study, the PH was associated with either meconium aspiration or hypoxic ischemic encephalopathy and subsequent cooling therapy. The authors suggested that prostaglandin could be a useful adjunctive therapy in managing severe PH.

The benefit of using prostaglandin to maintain ductal patency in infants with PH related to CDH is unclear. A recent review highlighted that the studies investigating prostaglandin in that population are mostly limited to very heterogeneous cohort studies where doses, indications for starting and outcome measures varied [61]. Whilst some studies showed a benefit, a study comparing the use of PGE in combination with iNO was associated with a longer time to surgical repair and longer hospital stay [62].

2.1.11. Diuretics in infants with BPD related PH

Diuretic therapy is often given to infants with BPD, although without evidence of long- term improvement. In adults with pulmonary hypertension, diuretics are given to relieve symptoms and manage right ventricular congestion [63]. In an observational study of 27 infants born at mean gestational age of 27 weeks administration of 3 days of frusemide followed by maintenance therapy with a thiazide resulted in improvement with right and left ventricular function. Whether diuretic administration improves meaningful clinical outcomes requires testing in an appropriately designed randomized trial [64].

3. Conclusion

PPHN is a complex condition due to its numerous causative factors. It results in mortality and significant morbidity, particularly in infants with CDH or those with BPD. iNO remains the mainstay of pharmacological treatment due to numerous trials showing it can reduce the oxygenation index and in term born infants reduces the need for ECMO. The evidence for its efficacy in improving the long-term outcomes in infants with CDH and BPD, however, is lacking. Surfactant as an adjuvant therapy to iNO has yielded promising results as it can reduce the oxygenation index and requires further exploration. Other agents such as sildenafil, milrinone and prostacyclin have been shown in small studies to be beneficial with or without iNO, but appropriately designed RCTs are required to robustly assess their clinical impact appreciating, however, that these will be difficult to accomplish because proper stratification will severely limit sample sizes.

4. Expert opinion

Inhaled NO is the mainstay of treating PPHN, but it is essential that this is used to result in meaningful clinical outcomes, not only an improvement in the oxygenation index in prematurely born infants. Trials which are adequately powered should address whether use of iNO does reduce bronchopulmonary dysplasia and more importantly long-term respiratory morbidity. Before performing such a large RCT, it is essential that the most appropriate dose and population are identified. Nowadays, the majority of very prematurely born infants are supported by noninvasive ventilation, and thus it is important to ensure iNO is being delivered effectively under such circumstances. The pulmonary hypertension in infants with CDH is complex and it is clear that not all infants will respond favorably to iNO. A personalized approach should be undertaken following detailed echocardiographic assessments and CDH infants should be delivered in institutions which have appropriate expertise available. Clinicians should cease having a knee jerk response to starting iNO in CDH patients, it might make them worse. Inhaled NO is currently not the answer for developing countries Although other agents in such settings have been shown to improve short-term outcomes, which are the most cost-effective with regard to long-term outcomes that need to be appropriately explored.

Currently, there are no treatments that are approved by the FDA for infants with PH-related BPD, yet these are a growing population of such patients who are at risk of increased mortality and morbidity. Trials in this population are urgently required and set in the context of the overall management of affected infants. For example, which diuretics and when, what is the most effective and least damaging method of respiratory support. How long should affected infants stay on pulmonary vasodilator therapy and what is the optimum schedule for post discharge from the neonatal unit follow-up assessments and when it the correct time to stop such therapies and what would be the criteria.

There are a number of potential promising vasodilators, but evidence for their efficacy comes from retrospective, observational or small randomized trials. It is essential that appropriately designed RCTs are carried out powered to detect differences in important long-term outcomes. The difficulty in undertaking such studies, however, should not be underestimated, particularly in infants with PH related to CDH who may be uncommon in single institutions. As a consequence, it is essential there are multicentre studies and if these are to be successful, they are supported by the international communities.

Future studies need to incorporate detailed echocardiographic assessments to determine why there may be responders and non-responders to specific treatments and biomarkers identified. In this way, therapies can be more appropriately directed. In addition, the effectiveness of a combination of therapies needs to be assessed providing the sample size is adequate to determine the impact on clinical outcomes that are meaningful to clinicians and most importantly infants and their parents. In the future, following appropriately designed trials, algorithms will have been developed that will allow infants to follow a personalized pathway to appropriate treatment.

Article highlights

• In a systematic review of 12 RCTs of iNO in infants born at or near term, the combined outcome of death or need for ECMO was significantly reduced, but this was due to the lower ECMO requirement. In a more recent review of 17 RCTs of iNO in infants born at or near term, it was concluded that iNO was effective at an initial concentration of 20ppm, except in those infants with CDH.

• iNO should not be started in CDH infants without echocardiographic evidence that it might be efficacious, the response then carefully monitored and iNO stopped if improvements do not occur.

• A meta-analysis of 17 RCTs demonstrated iNO was not effective as rescue therapy for the very ill preterm infant and early use of iNO did not prevent serious brain injury or improve survival without BPD.

• In resource poor settings, that is those without access to iNO or ECMO, sildenafil has been shown to improve oxygenation, but the total number of infants included in the trials was small.

• Other therapies have only been investigated in observational studies or small RCTs, so clinically meaningful conclusions cannot be drawn.

Declaration of interest

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or material discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or mending, or royalties.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Data availability statement

Data available on reasonable request.

Additional information

Funding

This paper was not funded.