Home mechanical ventilation in children: evolving indications in an era of new treatment options

Abstract

Worldwide, there has been a dramatic increase in the use of paediatric home mechanical ventilation (HMV). In this review, we examine this rapid evolution in clinical practice through the prism of two distinct groups of children: those with neurodisability/medical complexity and patients with neuromuscular disease. We illustrate the changes in service provision for these two groups that are driven by a recognition that early intervention with HMV can enhance quality of life for these children and may complement the beneficial effects of novel disease-modifying medications to improve survival. Alongside this, we highlight the importance of balancing patient expectations with clinical need and discuss the ethical challenges that may be encountered when delivering HMV to this increasing population of children.

Shareable abstract

The emphasis of home mechanical ventilation in children with neurodisabilities/medical complexity should be on improving quality of life. https://bit.ly/4eOIN6q

Introduction

Long-term ventilation is generally considered to be required in “any child who, when medically stable, continues to require a mechanical aid for breathing, after an acknowledged failure to wean or a slow wean, 3 months after the institution of ventilation” [1]. In most cases this is now undertaken non-invasively, with the exception being in children who are deemed to be dependent on ventilation for life support and those who require support for >16 h per day [2].

The number of children on home mechanical ventilation (HMV) has increased significantly over the past few decades. A combination of factors has contributed to this increase, including improved survival in conditions such as chronic lung disease of prematurity and neuromuscular diseases (NMDs), advances in HMV technology, and a profound alteration in attitude towards HMV resulting in an expansion of its use in a far wider range of conditions, including those with neurodisability/medical complexity. A recent study of the international evolution of HMV in children by Toussaint et al. [3] performed a meta-analysis using data from 32 publications spanning 2000 to 2023. It confirms an increasing number of children on HMV worldwide, with an increasing proportion of those on non-invasive ventilation (NIV). The overall prevalence was estimated to be 7.4 per 100 000 children, with initiation at a mean of age of 6.7 years, consisting of a bimodal distribution with peaks at ages 1–2 and 14–15 years. Patients with NMD formed the largest group of children on HMV (37%), followed by cardiorespiratory conditions and central nervous system (CNS) conditions (both at 16%). Outside this, central hypoventilation (4%), thoracic (3%) and genetic/congenital disorders (1%) were identified as specific groups, although 10% fell into a catch-all of “other” disorders. Interestingly, the fastest growth in the number of patients using HMV was in the genetic/congenital, CNS and “other” populations, reflecting societal and parental expectations as well as a growing acceptance that HMV may have a beneficial role in the management of some patients with these conditions.

In this review, we aim to discuss the evolving indications of HMV, illustrating this with two groups of children: those with significant neurodisability (which would fall into the genetic/congenital and CNS groups classification in the Toussaint et al. [3] survey) and children with NMD, who currently form the largest group of patients on HMV and in whom novel disease-modifying therapies have entered clinical practice in the past few years.

Children with neurodisabilities (genetic/congenital and CNS disorders)

Respiratory problems are the commonest cause of morbidity and death in children with neurodisability [4]. Many children in this group have life-limiting conditions, where prognosis may be uncertain but gradual progressive weakness and complications of the condition such as scoliosis lead to eventual respiratory decline. In these children, mechanical ventilation with NIV has revolutionised care, resulting in an option that can reduce hospitalisation and lead to increased quality of life and survival [5]. NIV in this group therefore warrants special consideration, particularly as this population continues to increase [6]. Goals of therapy are often different to other groups of children, focusing on facilitating discharge, ensuring stability for home management and providing symptom relief, rather than full resolution of respiratory compromise.

Therapy may be initiated under many different circumstances: 1) in infancy following an inability to wean respiratory support; 2) acutely following respiratory deterioration with an infective trigger, persistent lobar collapse (which may result from progressive restrictive lung disease) or as a result of recurrent acute respiratory presentations, to reduce the frequency and severity of admissions; 3) as part of pre-operative evaluation, most commonly for spinal surgery; and 4) following evaluation undertaken as part of routine clinical respiratory/sleep monitoring.

Discussions with families will differ depending on the specific circumstance but ultimately the focus for this group of children needs to be on maintaining quality of life rather than prolonging life and therefore acknowledging that, whilst possible, initiating mechanical ventilation may not always be the appropriate measure and may not be an avenue that all families wish to pursue.

Initiating NIV therapy in infancy following failure to wean respiratory support

In infants with neurodisability, respiratory insufficiency is often multifactorial, with abnormalities in tone, facial dysmorphism, anatomical airway issues, aspiration and bulbar dysfunction being some of the important contributors to failure to wean off respiratory support. As a result, the range of conditions in which infants may require consideration of long-term NIV therapy is diverse, including those with neuromuscular conditions, CNS disorders and a variety of airway disorders such as those with craniofacial disorders [7].

NIV is often the preferred approach in an infant with neurodisability who demonstrates continuing need for respiratory support because, at this early stage, prognosis may still be unclear, thereby leading to hesitation to consider tracheostomy and ventilation which cannot be discontinued as easily. The ethical issues relating to this decision making are beyond the scope of this review, but it is essential that those involved have significant expertise in undertaking these discussions in partnership with parents, ensuring that conversations are sensitive and centred around what is in the best interests of the child.

Long-term NIV has been used successfully in infants from the first weeks of life [8]. A comprehensive systematic review and meta-analysis of the use and outcomes of long-term NIV in infants has summarised the cohorts of infants in whom NIV therapy has been utilised and demonstrates that children with neurodisability conditions comprise a significant proportion of this population [9]. The review also highlights how, although evidence is still limited, there is significant benefit from NIV in this population, including improvements in respiratory parameters, decreased hospitalisations and prolonged survival. In comparison to older children, infants on NIV support, as shown in the meta-analysis, were more likely to be medically complex, with most having more than one comorbid condition and more commonly needing bilevel therapy rather than continuous positive airway pressure (CPAP) than older children [7]. There was also a wider spectrum of outcomes in this age range, with some infants improving and being able to discontinue support and others requiring escalation to invasive support. This suggests that infants may require closer monitoring than older children following commencement of therapy. More studies focusing specifically on including larger groups of infants with neurodisability are required to guide development of infant-specific treatment strategies, recognising that these children often have different challenges and trajectories to those who commence respiratory support later in life.

Initiating NIV acutely following an infective episode

Children with neurodisability may be extremely vulnerable if a significant respiratory infection occurs and adds stress to an already compromised respiratory system. Muscle weakness and abnormal muscle tone are common features in neurodisability conditions and may impair secretion clearance, result in fatigue especially when coughing, and exacerbate existing swallow dysfunction and aspiration risk during acute illness [4]. Spinal deformity is a common, almost universal complication in children with neurodisability. It may present at any age and in most cases progresses relentlessly, resulting in restrictive lung disease with progressive respiratory failure, which is exacerbated further in acute illness [10]. These factors often occur in the presence of blunted breathing control mechanisms, and coexisting upper airway obstruction increases the risk of developing acute respiratory failure with atelectasis and/or lobar collapse during infective episodes. In some cases, this may be isolated and respond to acute respiratory support which can be discontinued following resolution of the infection, but often, in children with neurodisability, these acute episodes may unmask background chronic respiratory compromise that has not been apparent and may be the point at which long-term NIV therapy is considered. Factors during clinical evaluation that may indicate that home NIV may be beneficial include a history of recurrent respiratory infections, particularly with severe presentations that require intensive care support, ineffective cough and clearance of secretions when well with report of excessive drooling, coughing and choking with any oral feeds and noisy breathing at night. Other management at this stage, prior to discussing NIV, may include organising a detailed speech pathology assessment to determine any deterioration in oromotor control and need for modification of feeding practices, review of airway clearance strategies, and the role of adding adjuncts such as a cough assist machine or physio vest therapy as well as the benefit of any anti-secretory medications such as glycopyrronium bromide. Discussions with families regarding NIV at this stage should include explaining the role of therapy in acute illness management, as well as the potential risks and benefits of long-term use. This includes explaining that whilst NIV may have benefits, side-effects such as uncontrolled secretions despite medical and surgical therapies, distress from face mask application, and an increased risk of gastro-oesophageal reflux and aspiration may render NIV less effective in some patients [11, 12]. The focus should always remain on balancing the acceptability of therapy to the individual child and family and aiming to improve the overall quality of life [13].

In practice, NIV initiation in this acute setting will likely occur in a paediatric intensive care or high dependency unit. Initial ventilator settings may need to be higher due to the need to re-recruit areas of collapse and are often guided by monitoring gas exchange. Supplemental oxygen may also be required during this acute period and may need to be weaned gradually prior to discharge home. Despite the acute presentation, a systematic approach to initiating therapy needs to be undertaken [14], ensuring choice of appropriate interface for the individual child, acquisition of home equipment, provision of education and explanation of the therapy to the child and family, and a clear plan for follow-up monitoring after discharge to review pressure requirements and ongoing tolerance of therapy after the acute period. Polysomnography (PSG) monitoring after resolution of the acute infection is ideal, to optimise the interface and formally titrate settings for ongoing use. However, in practice, limited channel studies that include oximetry combined with carbon dioxide monitoring is often more readily accessible. Sleep disordered breathing may not be evident at this point but the decision to continue NIV as a trial to determine the impact on hospital admissions and quality of life can be used to guide ongoing use. If continued, the follow-up schedule for monitoring of therapy will depend on the patient's age, diagnosis, local facilities and family support. A planned visit 1 month after NIV initiation followed by regular visits every 3–6 months is usually considered a minimum [15].

Initiating NIV prior to spinal surgery

Children with neurodisability are prone to development of kyphoscoliosis as a result of their underlying tone abnormalities and muscle imbalance. Kyphoscoliosis can lead to mechanical disadvantage of the respiratory muscles and decrease chest wall compliance, resulting in increased respiratory effort, decreased vital capacity and unequal lung ventilation, with progressive respiratory failure [16]. Respiratory evaluation prior to consideration of spinal surgery for children with neurodisability is therefore essential and should be provided as a standard of care to reduce the possibility of post-operative complications. Assessment should be bespoke, tailored to the understanding of the child, their spinal deformity (type, site, size and age of onset) and their proposed operation, whilst using investigations to aid the synthesis of post-operative recommendations [17]. A sleep study is often a particularly important part of the evaluation in this group of children, providing an idea of the child's baseline respiratory health. This may be the only objective measure possible in a child with neurodisability, who may not be able to undertake lung function testing. Results will assist the clinician in determining suitability for surgical intervention and whether initiation of NIV peri-operatively would be beneficial. An advantage of establishing NIV pre-operatively is that this can be done electively, to optimise the interface and ventilator settings, and ensuring the child has time to become familiar with the equipment to minimise any distress.

The decision making around spinal surgery in children with neurodisability is complex and requires the involvement of a skilled multidisciplinary team, who can perform a comprehensive assessment and undertake sensitive discussions with family members, balancing information with the need to address hopes, concerns and fears a parent/patient has about surgery [18].

Initiating NIV following routine clinical monitoring

As children with neurodisability are high risk for respiratory failure, sleep study monitoring may be requested following a routine clinical review that raises suspicion of declining respiratory health. Some families may elect not to proceed with investigation at this stage, expressing a desire to avoid additional therapies that may be distressing for their child and increase the burden of care. This is an important discussion to undertake prior to progressing with a sleep study so that a family's wishes are respected and care continues to focus on other measures to support symptom relief and quality of life. For those who do proceed, abnormal sleep study results may prompt discussion around initiation of NIV electively and should provide families with the potential risks and benefits of therapy. The inability of children with neurodisability to comprehend why NIV therapy is needed and the behavioural aspects of many of these conditions need to be carefully considered during the decision-making process as these features may present additional challenges for families when establishing therapy. For example, many children in this population have sensory issues that may limit toleration of masks and headgear. Increased secretions in the first few weeks of initiating NIV are common and may lead to a high burden of care for families, with discomfort and distress for the child that negates the beneficial effects of treatment. Families should be supported in their choice around NIV therapy, recognising that whilst NIV may help to reduce hospitalisation, increase stability and improve daytime function, it may not be appropriate for all children. If undertaken, the goals of therapy should be individualised for the child and clearly outlined, with the focus on supporting symptom management and improving quality of life.

Supporting continued therapy with NIV in children with neurodisability

Adherence to NIV therapy is usually evaluated by regular downloads of data from the child's machine using built-in software. This is generally undertaken at the time of outpatient review, sleep study monitoring and if admitted acutely. There is no validated definition of good/optimal adherence in children, and it is recognised that numerous factors related to the patient and family may impact adherence [15]. In the past, presence of neurodisability may have precluded the decision to offer NIV therapy due to concerns around ability to tolerate therapy. However, there are now studies that demonstrate that NIV can be efficacious in this population of children and that, in fact, with the recognised increased caregiver engagement in this group, use and outcomes of NIV in this group of children may not differ from other children [19, 20]. The most well-studied group are those with Down syndrome in whom, whilst adherence may be lower than in other children, satisfactory NIV adherence has been shown to be achievable [21, 22]. Logistical factors such as excessive leak due to interface fit have been shown to be potential contributors to lower adherence rates in children with Down syndrome [22], in addition to the recognised psychosocial and behavioural challenges that can also impact toleration of therapy. Evaluating children with cerebral palsy, Grychtol and Chan [23] found that over half of their patients failed to establish on NIV either due to intolerance and/or ventilation pressure at the initial trial in hospital or to poor adherence during follow-up. However, more recently, in their study of children with complex neural disability, Morrison et al. [24] demonstrated equivalent NIV adherence rates to those in their general paediatric population, with the majority of their cohort (86%) achieving an average of >8 h use per night. Recognising the potential challenges, families of children with neurodisability may require significant support to sustain therapy and the approach to delivering treatment in this population of children may need to be adapted to achieve success. The involvement of specialised professionals such as experienced sleep clinical nurses, occupational therapists and psychologists early in the course of treatment may be beneficial. However, more studies are required to fully determine the impact of such measures on NIV adherence in children with neurodisability.

Weaning NIV therapy in children with neurodisability

As in other children, the possibility of weaning ventilation should be assessed on a regular basis in children with neurodisability. A proportion of children with neurodisability, particularly those who commence therapy as infants, may demonstrate improvement over time, making it possible to discontinue therapy [9]. Often this improvement relates to growth, improvements in tone and greater respiratory reserves.

Guidelines for when to commence weaning are lacking and timing is often reliant on the expertise of the clinicians involved. In their review, Khirani et al. [25] suggest criteria which may be useful in determining when weaning should be attempted, including taking into consideration the disappearance of symptoms and improvements in gas exchange and PSG parameters. However, as highlighted by Khirani et al. [25], in practice, individual variations between patients also need be taken into account and may influence decision making. Additionally, in children with neurodisability, it is recognised that sleep disordered breathing and respiratory failure can wax and wane, thereby necessitating long-term follow-up in this population to ensure prompt re-evaluation for any relapse or decline.

NIV therapy at the end of life

At the other end of the spectrum are those children with medical complexity in whom there is no improvement or even worsening in respiratory status over time. The role of NIV therapy in these children is recent but is being increasingly used to support those with life-limiting diseases [26]. NIV therapy offers support and stability, may help to alleviate distressing symptoms such as dyspnoea or hunger for air, and facilitates management at home for those who wish to avoid hospital for their child at the end of life. Involvement of clinicians with palliative care expertise is essential for children in this group, to ensure that other symptoms are relieved effectively, families are adequately supported through the end of life and that the focus remains on ensuring quality of life.

Ideally, clear discussions regarding the goals of NIV therapy in children with life-limiting conditions will occur prior to commencement of therapy and family members should be actively involved in this palliative NIV decision-making process. Discussion should include explanation of the potential benefits of NIV initially but also how, with progressive respiratory decline, there are likely to be increasing challenges in delivering therapy, which may ultimately lead to the need for discontinuation. Sometimes families may choose to discontinue or limit use of therapy themselves during the last days of life, favouring the ability for their child to communicate, eat or engage in other activities that they consider important [27]. Other reasons to discontinue treatment at the end of life include mask discomfort, a feeling of claustrophobia and child/parent anxiety relating to therapy [26]. Shared decision making between families and their treating teams at these times is essential, ensuring that the families are provided with information sensitively but also ensuring that expectations are realistic and that NIV use does not prolong unnecessary suffering. As the role for NIV in this palliative setting increases, further studies evaluating the benefits and limitations of treatment in this population of children and addressing the ethical dilemmas that may arise in this context are required to provide clinicians with guidance around the optimal approach to management.

Children with NMD

The aim of HMV in children with NMD is the assistance of the respiratory muscles, to achieve/maintain adequate tidal volume, thus correcting alveolar hypoventilation. All the aforementioned indications for initiating HMV in neurodisability similarly apply to children with NMD, as do the common themes of multidisciplinary team input and working in partnership with the child and family with the child's best interests at the core of this work. There is now firm evidence that NIV is associated with an improvement in sleep-related breathing disorder symptoms, nocturnal and daytime gas exchange, sleep quality and architecture, and can potentially reduce the number of acute respiratory exacerbations and improve survival in certain NMDs whilst not compromising the child's quality of life [15, 28].

The French national paediatric NIV network survey found that at the time-point of 1 June 2019, 387 NMD patients were on NIV/CPAP, which made up 28% of the total paediatric NIV population [29]. In the prior survey in 2000, there had been 35 NMD patients on NIV [30], i.e. there had been a greater than 10-fold increase in two decades. 33% consisted of spinal muscular atrophy (SMA) patients (of whom 34% were SMA-I, 64% were SMA-II and 2% were SMA-III), 30% were congenital myopathy dystrophy patients, 20% were Duchenne muscular dystrophy (DMD) patients, 8% were Steinert myotonic dystrophy patients and 9% were “others”. This “others” group included arthrogryposis multiplex congenita (n=13), congenital myasthenic syndromes (n=7), Pompe disease (n=7), diaphragmatic disorders (n=6) and mitochondrial cytopathy (n=2). Adherence was excellent across groups at 8.0±3.1 h per night.

NIV was started electively in 85% of the patients in the French survey [29], reflecting effective screening for sleep disordered breathing in this high-risk population. As the major criterion to initiate NIV is alveolar hypoventilation, abnormal overnight gas exchange was unsurprisingly the most common indication for starting elective NIV. Other indications included abnormal blood gases, recurrent chest infections, optimisation in preparation for elective surgery, restrictive lung disease on lung function tests, thoracic deformity and inability to wean off NIV initiated during a respiratory exacerbation. NIV was more likely to be initiated during a respiratory exacerbation (33%) in SMA-I patients [29].

This highlights the importance of disease-specific guidelines for NIV establishment. SMA-I patients often do not show nocturnal hypoventilation before respiratory decompensations and needing NIV initiation. The standard-of-care recommendations for patients with SMA therefore suggest that NIV should not just be initiated in patients who have nocturnal hypoventilation, but also in patients with symptoms such as tachypnoea, increased work of breathing and orthopnoea, even before documented respiratory failure, to palliate dyspnoea and prevent chest wall deformity [31].

In DMD, the international DMD care considerations recommend that NIV should be considered when: 1) there are signs or symptoms of nocturnal hypoventilation or other sleep disordered breathing, e.g. fatigue, morning headaches, frequent nocturnal awakenings, hypersomnolence, difficulty concentrating, awakenings with dyspnoea, etc.; 2) abnormal sleep study results, e.g. carbon dioxide is >50 mmHg for ≥2% of sleep time, sleep-related increase in carbon dioxide of 10 mmHg above the awake baseline for ≥2% of sleep time, blood oxygen saturation by pulse oximetry (S_pO2) is ≤88% for ≥2% of sleep time or for ≥5 min continuously, or apnoea–hypopnoea index of >5 events·h⁻¹; and 3) significant impairment of lung function, e.g. forced vital capacity <50%, a maximal inspiratory pressure of <60 cmH₂O, or if the wake baseline S_pO2 is <95% or carbon dioxide tension is >45 mmHg [32].

The UK respiratory care guidelines are less didactic given the expert working group's concerns that starting NIV too early may adversely affect adherence [33]. They suggest that “decisions about initiation of ventilatory support are complex and depend on many factors, including patient wishes, and should be made by a specialist respiratory team after discussion with patient, carers and other relevant specialists including palliative care teams” [33].

Disease-modifying therapies using SMA as an illustrative example

The advent of novel disease-modifying therapies in NMDs is instituting a paradigm shift in the field. An exemplar is SMA, an autosomal recessive condition resulting from pathogenic variants in the survival of motor neuron 1 (SMN1) gene, with an incidence of approximately 1:10 000 live births. Patients develop progressive muscle weakness due to progressive degeneration and irreversible loss of the anterior horn cells in the spinal cord (i.e. lower motor neurons) and the brain stem nuclei. Humans have an almost completely homologous gene, SMN2, but this naturally harbours an exonic splicing enhancer variant that limits inclusion of exon 7 so that only extremely limited SMH2-derived mRNA produces a functional full-length SMN protein. SMN2 copy number is variable between individuals, with increasing copy number being inversely related to clinical severity. Historically, SMA has been classified clinically, with SMA-I being the most severe, infants becoming symptomatic by 6 months of age and never achieving the ability to sit unsupported. Indeed, until recently, SMA-I was one of the leading genetic causes of infant mortality. In SMA-II, onset is typically between 6 and 18 months of age with patients attaining the ability to sit independently. SMA-III patients usually develop symptoms after 18 months of age and patients with SMA-IV usually only present in early adulthood.

Three novel disease-modifying therapies have entered clinical practice in the past few years and have dramatically improved disease trajectories. In SMA-I patients, families and clinicians now frequently opt for active treatment instead of palliation [34].

Nusinersen and risdiplam

Nusinersen is an antisense oligonucleotide which facilitates the integration of exon 7 into the mRNA by alteration of the SMN2 pre-RMA splicing process [35]. It is given intrathecally, and trials have shown that infants who receive nusinersen have improved survival and motor function compared with controls [36–38].

Lavie et al. [39] reported their real-world experience of 20 SMA-I patients treated with nusinersen before and after 2 years of treatment. Nusinersen was initiated at a median age of 13.5 months. At baseline, eight were tracheostomy ventilated and eight were on NIV. There was no change in respiratory support among ventilated patients at the end of the follow-up period. However, four patients who did not need ventilatory support at baseline needed to be initiated on ventilatory support during the study period. Two patients died from acute respiratory failure, one sustained a severe brain injury and four had chronic and/or recurrent atelectasis.

Another multicentre study of 17 SMA-I patients on nusinersen found that 13 were started on NIV at a median of 12 months [40]. 11 had had recurrent hospital and intensive care unit admissions; nine had evidence of atelectasis and/or lower respiratory tract infection changes on chest radiography, yet their capillary gas and nocturnal gas exchange recordings prior to NIV initiation were all normal.

In SMA-II and SMA-III patients, nusinersen has been shown to slow the progression of deterioration in lung function [41]. Indeed, impaired diaphragmatic motility in adult SMA-II and SMA-III patients has also been shown to improve following nusinersen treatment [42]. In an Australian cohort of SMA patients treated with nusinersen, 21 were on nocturnal NIV (SMA-I n=3, SMA-II n=14 and SMA-III n=4), of whom six (SMA-II n=5 and SMA-III n=1) managed to be weaned off nocturnal NIV post-treatment [43]. Previously, weaning of NIV in this patient group was less likely as weakness tended to become more severe or at best plateaued with time. However, there may well be something else that changes with the advent of disease-modifying therapies.

Risdiplam is the other novel disease-modifying medication which has entered clinical practice. Its mechanism of action also modifies SMN2 pre-mRNA splicing to promote inclusion of exon 7, increasing production of functional SMN. Its principal advantage over nusinersen is that it is an oral medication. Trials have shown that it improves survival in SMA-I patients. 35 out of 41 (85%) SMA-I infants treated with risdiplam survived to 12 months without the use of permanent ventilation compared with 42% reported in historical controls (p<0.001) [44].

No head-to-head randomised controlled trials have taken place between the two medications. Indirect treatment comparisons using matching adjusted indirect comparison methodology suggest the results of risdiplam may be slightly more favourable, with a reduction in the rate of death or permanent ventilation in children treated with risdiplam compared with those treated with nusinersen [45]. The caveats of indirect comparisons between treatments must of course to be taken into consideration.

Gene therapy

A big breakthrough was gene therapy, the SMN transgene delivered via an adeno-associated virus-9 vector, administered via a one-off intravenous infusion. In the US arm of the open-label, single-arm, multicentre phase 3 trial, 22 patients received gene therapy at a mean±sd age of 3.7±1.6 months [46]. One patient died at age 7.8 months because of respiratory failure and one family withdrew consent at age 11.9 months after the patient met the definition of requiring permanent NIV. 18 (82%) patients did not require ventilatory support at 18 months compared with none out of 23 patients in the natural history study (p<0.0001). 15 (68%) patients did not need ventilatory support during the trial period and of the seven patients who did, five had previously used it prior to receiving the gene therapy. In the European arm, 31 out of 32 (97%) patients in the intention-to-treat group survived free from permanent ventilatory support at 14 months compared with six out of 23 (26%) patients in the natural history cohort (p<0.0001) [47].

The 5-year extension results of the initial phase 1 trial found that of the six out of 10 patients who did not require regular ventilatory support at baseline, none needed to start NIV [37]. The four patients who required NIV at baseline continued to need ventilatory support but their ventilatory requirements remained stable and did not increase, as would have been historically expected over the study period.

Even more impressive were the results of the phase 3 trial for pre-symptomatic infants with two copies of SMN2, historically 80% of whom would have been in the SMA-I group [48]. 14 infants were recruited: five (36%) were diagnosed on pre-natal screening and nine (64%) were diagnosed at newborn screening. The median (range) age at delivery of gene therapy was 32 (9–43) days. All survived without the need for ventilatory support at 14 months. This marks a seismic shift from patients who survive to children who thrive. Furthermore, in a study of 11 patients (SMA-I n=9 and SMA-II n=2), all patients who were on ventilatory support pre-gene therapy (five SMA-I patients) managed to be weaned off NIV, with the exception of one tracheostomy-ventilated patient [49]. The annual hospitalisation rate halved and the average length of stay in the intensive care unit decreased from 18.2 to 0.9 days per patient per year after gene therapy.

More recently, real-world outcomes from the RESTORE registry of 168 patients who had up to 37 months of follow-up post gene therapy found that patients maintained or achieved new motor milestones, whilst the safety profile was consistent with that from the initial clinical trials [50].

Although these results have understandably generated great optimism, caution should be advised as gene therapy may not be the miracle cure that many families hope for. Long-term data are not yet available regarding the longevity of this “one-dose” treatment and repeat dosing may be challenging given the immunological constraints of adeno-associated virus vectors. Treatment effect may potentially wane over time as the episomal SMN cDNA within the vector will become diluted with cell division. When Barrois et al. [51] prospectively studied 15 infants who had received gene therapy at a median age of 8.6 months, following them up for 24 months, their inspiratory muscle strength was normal at baseline but slightly decreased over time, whilst expiratory muscle strength was low even at baseline; three required NIV. The cost is also prohibitively high, resulting in unequal access worldwide, leading to ethical issues, and some families have had to resort to crowd funding to be able to afford the treatment.

These issues have led to the European consensus statement on gene therapy for SMA which was updated in 2024 [52]. In patients who have symptoms, age of symptom onset, disease duration and motor function status at the time of gene therapy are the most important factors that predict response to treatment. In patients who are asymptomatic, SMN2 copy number is the most important predictor of disease severity, so treatment decisions should be based on this. For patients who are already clinically severely affected, parents need to be counselled that despite gene therapy, there is a high risk of living with severe disability and palliative care should be discussed as an alternative treatment option. The dose of gene therapy administered is proportional to weight and risks increase with increased dose, so heavier and older children should be treated cautiously and treatment should only be performed under a rigorous protocol with continuous monitoring of safety and efficacy. Clinicians need to discuss with families that the risk/benefit ratio is still unclear and manage their expectations. Treatment of patients weighing >21 kg cannot be recommended. There is no convincing evidence that duo therapy with gene therapy and nusinersen or risdiplam is superior and clinical trials are needed. Early initiation of any disease-modifying treatment, ideally in the pre-symptomatic stage, is associated with significantly better outcome and therefore SMA should be included in newborn screening programmes in countries where at least one disease-modifying treatment is available.

What is still unknown?

There is now a realisation that SMA is not just a motor neuron disease; the SMN protein is ubiquitously expressed and its depletion is thus likely to affect peripheral organs, including muscle, liver, heart and pancreas. We have yet to discover what new treated phenotypes will emerge. Interestingly, cognitive impairment has been described in some treated SMA-I patients [53, 54]. The cause of this impairment is still unclear. SMN protein is expressed in cortical neurons, and whether there is ongoing neuronal degeneration due to cerebral SMN protein deficiency despite treatment or it is purely due to factors such as frequent illness or social isolation, or a combination thereof, remains to be elucidated.

Overall, as NMD patients survive longer with better proactive medical management, we already have an inkling of new medical issues that have arisen. Examples of these which have been described in patients with SMA-II, congenital muscular dystrophy, congenital myopathy and DMD include cardiac complications such as arrythmias and cardiomyopathy, bowel problems including constipation, volvulus and intermittent pseudo-obstruction, renal problems such as renal calculi and urinary retention, and problems with autonomic dysfunction such as aberrant temperature control and postural hypotension [55].

Invasive ventilation versus NIV

Patients with NMD are typically first initiated on NIV if they need HMV. However, as disease progresses, should they become ventilator dependent requiring ventilatory support for >16 h per day, if effective ventilation is no longer able to be achieved with NIV or if they are no longer able to protect their airway, tracheostomy ventilation may be considered. This is in contrast to children with neurodisabilities where the higher incidence of CNS involvement has historically meant that often invasive ventilation is considered not to be in the child's best interests, albeit it does depend on the individual child.

Conclusions

In summary, the circumstances in which HMV may be initiated include inability to wean respiratory support in infancy or following respiratory deterioration, in preparation for surgery, in particular spinal surgery, and electively following abnormal results from routine respiratory/sleep monitoring or symptoms. One group of children in which HMV previously would not have been considered but now is increasingly being offered is children with neurodisability/medical complexity. The goals of HMV in this group of patients differ from that of others, with the focus often being on facilitating discharge, ensuring stability for home management and providing symptom relief, not on complete normalisation of sleep study results. It is important to emphasise that the goal of treatment is improving quality of life, not merely treating parameters on the sleep studies to prolong life. HMV may not be appropriate for all. It can be trialled and must be stopped if it causes more distress to the patient than benefit. Patients with NMD have always formed the largest proportion of children on HMV. This is likely to increase further with the advent of novel disease-modifying therapies as patients are surviving longer and those with severe genotypes who were previously being offered palliation are now more likely to opt for active treatment including ventilatory support. Long-term outcomes and treatment-emergent phenotypes are still unknown. However, the ultimate goal of HMV is to improve the overall quality of life of patients so that they achieve their full potential.

Points for clinical practice

• The goals of HMV in patients with neurodisabilities/medical complexity differ from those of others, with the focus often being on facilitating discharge, ensuring stability for home management and providing symptom relief. The emphasis should be on improving quality of life, not merely prolonging life.

• With the advent of novel disease-modifying therapies such as nusinersen, risdiplam and gene therapy for SMA, and more in development, the number of NMD patients on HMV is likely to increase further. Long-term outcomes and treatment-emergent phenotypes are still unknown.