A previously fit and well thirteen-year-old was admitted after a week of cough and fevers, followed by a twenty-four-hour history of worsening shortness of breath. She was initially treated as a possible asthma exacerbation with oxygen, nebulised salbutamol, ipratropium, intravenous magnesium, and aminophylline but she continued to deteriorate. She was intubated as an emergency when despite maximal therapy, she continued to desaturate and became increasingly obtunded.
On intubation, copious amounts of red-tinged froth erupted from her ET tube. This was likely pulmonary oedema. An urgent bedside echo showed a dilated left atrium and ventricle with severe mitral regurgitation.
She needed a high FiO2 (90%-100%) and a high PEEP (12-16 cmH20) to maintain oxygen saturations above 88%. They were transferred to the PICU on adrenaline for ventilatory and cardiac support. Her initial pro-BNP levels and troponin at the base hospital were 28,000 and 1832 respectively.
72 hours into her PICU stay her cardiac function worsened despite broad-spectrum antibiotics, nitric oxide for high pulmonary pressures, and inotropic support. The team elected to commence VA-ECMO support for suspected viral myocarditis.
Management
One day after starting ECMO: She needed two minutes of CPR with adrenaline boluses due to a catastrophic loss of flow in the ECMO circuit. This was probably due to the venous cannula abutting the posterior aspect of the right atrial wall. That same day, the patient underwent a balloon atrial septostomy. This was technically challenging and resulted in a small atrial communication due to the elasticity of the atrial septum.
Day 4 of ECMO: As her urine output tailed off, she was started on renal replacement therapy.
Day 5 of ECMO: She was taken for a CT brain with a view to ongoing management. It showed multifocal parenchymal infarcts. The transplant centres thought she would not be a candidate for a VAD given the multiple infarcts.
Day 6 of ECMO: After an unsuccessful attempt to wean ECMO flows she were started on levosimendan.
She remained on VA-ECMO, nitric oxide, ventilatory and inotropic support for eight days. She received steroids, antibiotics, and blood products and had close anticoagulation monitoring. Intravenous immunoglobulin was not felt to be of benefit. She was also discussed with two transplant centres often.
She was extensively investigated for possible viral myocarditis. This including genetic and infectious studies. She grew low positive samples of enterovirus and adenovirus, neither thought to be a definitive cause of myocarditis. ECMO complications included multifocal cerebral infarcts, loculated collections of intra-peritoneal fluid and bibasal nodular consolidation in both lungs.
Unfortunately, she passed away after 8 days of VA-ECMO, due to multiple complications with little improvement of cardiac function.
Levosimendan in weaning
Myocarditis remains a global clinical challenge. There are many definitions with no definitive clinical criteria that confirm the diagnosis [1]. The diagnosis of viral myocarditis is often made at post-mortem and the list of possible investigations is long.
Mechanical support for non-structural heart failure in children in the UK is uncommon. 10-15 children, on average, are referred to UK ECMO centres each year. Of these, approximately 80% will survive – half recovering and half needing transplantation [2].
VA-ECMO provides both respiratory and cardiac support for patients with refractory cardiogenic shock. It is associated with less organ failure than biventricular assist devices that require open-heart surgery. ECMO cannulas can be placed at the bedside. In patients with potentially reversible cardiac failure (for example, myocarditis, myocardial stunning post-myocardial infarction, post-cardiotomy or post-cardiac arrest), VA-ECMO is used as a bridge to recovery, and the patient weaned off after a few days of support.
Despite being the ultimate life-saving technology for refractory cardiac failure, veno-arterial ECMO is still associated with severe complications. Increased LV afterload and lack of LV unloading under VA-ECMO might induce LV stasis. This can lead to thrombus formation. The patient may also suffer from pulmonary oedema and myocardial ischaemia caused by ventricular distension. These increase mortality. ECMO support also exposes the patient to complications such as infection, haemorrhage and peripheral vascular embolism. These complications are more frequent with prolonged support and are responsible for significant morbidity and mortality, coupled with prolonged ICU and hospital stays and higher associated healthcare costs.
How does levosimendan work?
Levosimendan combines inotropic and vasodilating effects (inodilator) with a nearly unique (among inotropes) myocardial protective effect.
It increases the affinity of troponin C for calcium. This, in turn, prolongs the duration of actin/myosin cross-bridges providing an inotropic effect and increased contractility. This is not associated with a raise in intracytoplasmic calcium concentration and is independent of beta-receptors or cyclic AMP [3]. Unlike traditional inotropes, such as dobutamine, levosimendan does not impair diastolic function. Levosimendan also inhibits phosphodiesterase III. This may add to the positive inotropy via the c-AMP pathway, whilst increasing myocardial oxygen consumption [4].
It also has a vasodilatory effect, by opening ATP-sensitive potassium channels in vascular smooth muscles, leading to coronary, pulmonary, and peripheral vasodilation. This, in turn, leads to smooth muscle relaxation. There is increased force of contraction with decreased preload and afterload with possible anti-inflammatory, antioxidative, antiapoptotic, anti-stunning and cardioprotective effects. The clinical evidence supporting these experimental findings is limited in adults, however, and almost non-existent in children [3]. Levosimendan may be given as a single 24h perfusion as it has a long-lasting action (up to 7-9 days) due to its’ active metabolite,
Hypotension is the most common side effect of levosimendan. This might require a norepinephrine infusion to fix. Atrial fibrillation and hypokalaemia can also result though the mechanism not understood.[5]
Is there evidence for levosimendan for weaning VA-ECMO in kids?
Latva et al. (2009) showed that levosimendan improved ventricular function and inhibited cardiomyocyte apoptosis in mice infected with the myocarditic Woodruff variant of coxsackievirus B3 [6]. Perhaps, then, levosimendan could be useful in paediatric myocarditis though it did not appear to influence areas of myocardial necrosis or inflammation, or the cardiac viral load.
The ability to wean off of VA-ECMO is dependent on the reason for starting it in adults. Evidence suggests that levosimendan may be more useful in postoperative low cardiac output states than after refractory cardiac arrest.
The 2020 update of a Cochrane Collaboration systematic review showed that the current dataset is inadequate to guide practice in adults in cardiogenic shock. The authors reported that there are “no convincing data supporting any specific inotropic or vasodilating therapy to reduce mortality in haemodynamically unstable patients with CS or LCOS” [7]. However, they felt that levosimendan has a good safety profile and may be useful for haemodynamic stabilization but the choice of agent should be individualized and based on haemodynamic response.
In 2021, Pan et al. performed a retrospective observational study comparing the outcomes of children who required VA-ECMO after cardiac surgery AND received levosimendan for weaning with those who did not receive the drug. 7% (4/54) of the group that received levosimendan (versus 19% (17/91) of those that didn’t) failed to wean off VA-ECMO. In the controlled analysis, levosimendan was associated with a decreased risk of weaning failure ARR 0.20 (0.07-0.57)] and lower in-hospital mortality ARR 0.45 (0.26-0.76) [8].
There have been four systematic reviews and meta-analyses to date looking at levosimendan in the paediatric population. The most recent being by Silvetti et al. [9] It identified 44 studies, published from 2004 to 2020, totaling 1,131 patients. Nine studies (547 patients) were RCTs performed in a paediatric cardiac surgery setting. In these RCTs, levosimendan use was associated with a statistically significant improvement in ScvO2 (p = 0.03) and a trend toward lower postoperative lactate levels (p = 0.08) as shown in Fig 1 below. However, this improvement in haemodynamic status did not have a significant impact on patient outcomes (longest follow-up mortality, ICU and hospital stays). The studies were heterogeneous so although there was some evidence that levosimendan is not associated with major side effects, its effect on major clinical outcomes remains unproven
Fig. 1. Taken from Silvetti et al. [9]
Lessons learnt
There is a scanty evidence base for the use of levosimendan in weaning from ECMO though it may be beneficial after cardiac surgery. We could have considered levosimendan earlier in the clinical course though it may not have changed the outcome.
There are adult studies that help us develop a clearer understanding of the efficacy of levosimendan within VA-ECMO weaning. These include the LEVOECMO trial and WEANILEVO study [10]. The algorithms suggested by Sangalli et al. in the 2022 literature update and the expert analysis by Giradis et al[. 11] may provide a starting point for when to include levosimendan in the weaning pathway for adults (fig 2 below). The literature base is still evolving within paediatrics.
Fig. 2 Figure copied from Giradis et al. [11]
Algorithms for weaning from a extracorporeal membrane oxygenation and b mechanical ventilation which include levosimendan. E/A, ratio of early to late (atrial) peak blood flow during diastole; DT, deceleration time; LVEDP, left ventricular end-diastolic pressure
PICSTAR is a trainee-led research network open to all doctor, nurses and allied health trainees within PICU. We are the trainee arm of Paediatric Critical Care Society- Study Group (PCCS-SG) and work together with them on research, audit and service evaluation.
If you would like to get involved in projects, join PICSTAR or have ideas you would like to propose/get advice or mentorship via PCCS-SG please contact us on picstar.network@gmail.com. Here is a link to our website: PICSTAR – Paediatric Critical Care Society (pccsociety.uk)
References
1. Law, Y.M., et al., Diagnosis and Management of Myocarditis in Children. Circulation, 2021. 144(6): p. e123-e135.
2. Robinson, S. and G. Peek, The role of ECMO in neonatal & paediatric patients. Paediatrics and Child Health, 2015. 25(5): p. 222-227.
3. Cholley, B., et al., Levosimendan in the light of the results of the recent randomized controlled trials: an expert opinion paper. Critical Care, 2019. 23(1): p. 385.
4. Maack, C., et al., Treatments targeting inotropy. Eur Heart J, 2019. 40(44): p. 3626-3644.
5. Mebazaa, A., et al., Levosimendan vs Dobutamine for Patients With Acute Decompensated Heart FailureThe SURVIVE Randomized Trial. JAMA, 2007. 297(17): p. 1883-1891.
6. Latva-Hirvelä, J., et al., Effects of levosimendan in experimental acute coxsackievirus myocarditis. Eur J Clin Invest, 2009. 39(10): p. 876-82.
7. Uhlig, K., et al., Inotropic agents and vasodilator strategies for the treatment of cardiogenic shock or low cardiac output syndrome. Cochrane Database of Systematic Reviews, 2020(11).
8. Pan, K.C., et al., Role of levosimendan in weaning children requiring veno-arterial extracorporeal membrane oxygenation after cardiac surgery. Eur J Cardiothorac Surg, 2021. 59(1): p. 262-268.
9. Silvetti, S., et al., Effect of Levosimendan Treatment in Pediatric Patients With Cardiac Dysfunction: An Update of a Systematic Review and Meta-Analysis of Randomized Controlled Trials. Journal of Cardiothoracic and Vascular Anesthesia, 2022. 36(3): p. 657-664.
10. Ellouze, O., et al., Levosimendan in venoarterial ECMO weaning. Rational and design of a randomized double blind multicentre trial. ESC Heart Failure, 2021. 8(4): p. 3339-3347.
11. Girardis, M., et al., Levosimendan in intensive care and emergency medicine: literature update and expert recommendations for optimal efficacy and safety. Journal of Anesthesia, Analgesia and Critical Care, 2022. 2(1): p. 4.