Abstract
Background: Intractable feeding intolerance in children with severe neurological impairment (SNI) is poorly defined and understood. Objectives: (1) To describe 9 children with SNI, where intractable feeding intolerance was thought to be a contributor to their deterioration or death. (2) To consider terminology to describe the severe end of the spectrum of feeding difficulties in children with SNI. Results: Mean age at death was 10.3 years (range: 5 – 15.6), and median time from palliative care referral to death was 3.1 months. Location of death was home (n = 3), hospice (n = 1), and hospital (n = 5) with 1 death in intensive care. Gastrointestinal “failure” or “dysfunction” were documented for 7 children, (median time between documentation and death was 3.9 months (range: .1 to 13.1)). All children were fed via a gastrostomy tube during their life (median age of insertion 2.5 years (range: 1.2 to 6.8 years)), and 7 via the jejunal route (median age of insertion 9.2 years (range 2.4 to 14.7 years)). Children lived a median of 9 percent of their lives after jejunal tube feeding was commenced. No child had home-based parenteral nutrition. Multiple symptom management medications were required. Conclusion: ‘Intractable feeding intolerance’ describes a clinical crossroads in a child’s life where there is an opportunity to consider the appropriateness of further interventions. Further work should explore predictors of intractable feeding intolerance and the delicate balance between cause or contributor to death. The importance of clinician-family prognostic conversations and goal-concordant care both during life and in the terminal phase is highlighted.
Intractable feeding intolerance is a potential contributor to deterioration and death for children with severe neurological impairment, with recognition vital for goal-concordant decision-making for children and families. Gastrointestinal challenges are common among children with severe neurological impairment, and a cohort of these children progress to intractable feeding intolerance where enteral feeding is not able to sustain life. Management for these children varies locally and internationally. Recognition and management of intractable feeding intolerance in children with severe neurological impairment is complex and poorly understood. Further research into predictors and management of how this may contribute to deterioration or death is needed to help optimize goal-concordant management and decision-making.Article Summary
What’s Known on This Topic
What This Study Adds
Introduction
An intimate interaction between the enteric and central nervous system (CNS) results in a predisposition to gastrointestinal (GI) problems in children with complex neurodisability. 1 Dysfunction may occur in the central, autonomic or peripheral enteric nervous system. 2 The umbrella term ‘severe neurological impairment’ (SNI) describes combined motor and cognitive impairment arising in childhood, that is associated with medical complexity and high daily care needs. 3 GI symptoms in children with SNI are common and may be non-specific, including vomiting, abdominal bloating, feeding-associated irritability, and pain.4-6 This can impair quality of life for the child and preclude tolerance of oral or enteral feeding. 6 In recent decades, as life expectancy for children with SNI has increased, so too has the potential for severe GI dysfunction which in some patients may progress to life-limiting or intractable feeding intolerance. 7
In 1999, Siden et al coined the term “pediatric progressive enteral feeding intolerance,” 8 and described a group of children with neurological impairment and an array of GI symptoms, whose symptoms only improved with cessation of feeds. This extreme end of the spectrum, where the enteral feeding difficulties are so severe that a child cannot be fed adequately via the enteral route to sustain life has been variably described as intractable feeding intolerance, 1 terminal GI dysfunction 9 and intestinal failure.10,11 Understanding the mechanisms of feeding failure in the setting of SNI is in its infancy, as is our understanding of its natural history. Children with SNI are at risk of dying from other comorbidities (eg, seizure disorders, respiratory failure) and it may be difficult to establish whether intractable feeding intolerance has been the primary cause or whether it has been a contributor to death. Variations in definitions used in the limited literature can further complicate communication and understanding about this patient population. In this study, we use the term ‘intractable feeding intolerance’ as it describes a clinical situation but does not indicate the expected impact on life expectancy as this can be influenced by other factors, such as initiation of parenteral nutrition (PN) and additional co-morbidities in children with SNI.
Children with severe enteral feeding difficulties and SNI are a clinically and ethically complex and diverse group. It can be difficult to predict which children in this group are at risk of life-limiting feeding intolerance, whether they might benefit from management options such as jejunal tube feeding or parenteral feeding, and the ideal timing of such interventions. Faced with the challenge of a child with deteriorating global health and quality of life, clinicians and families may find themselves having to consider the appropriateness, utility, and access to more intensive life-sustaining therapies such as parenteral nutrition (PN) vs withdrawing enteral feeds to allow a natural death when enteral feeds are unable to sustain life.
We aimed to describe the clinical course and management of a cohort of children with SNI known to a pediatric palliative care service, where intractable feeding intolerance was thought to be a contributor to their death. In addition, we aimed to consider terminology used to describe the phenomenon of feeding intolerance in children in SNI, both in the literature, and in our study population.
Method
Study Design
A retrospective medical record review was conducted. Patients were included if they had SNI with severe GI dysfunction, intractable feeding intolerance and were seen by the palliative care service between 2015 and 2020. A child was deemed to have intractable feeding intolerance if enteral feed-related symptoms and/or suffering led to an inability to provide individual nutritional requirements and contributed to the overall deterioration or death of the child, for example, malnutrition leading to increased risk of infection and medical vulnerability, or death.
Nine patients were included. Given that the pediatric palliative care team in this study is state-wide, children were cared for across several healthcare institutions. In 1 of these hospitals, a Neurodevelopment and Disability team provides inpatient and outpatient multi-disciplinary care for children with SNI, in addition to the General Medicine team.
Data were collected from electronic medical records. Data were analyzed and summarized using descriptive statistics. Institutional ethics was obtained (reference number 86122).
Definitions
For this study, we have used the following definitions:
Severe Neurological Impairment
“Severe Neurological Impairment describes a group of disorders of the central nervous system which arise in childhood, resulting in motor impairment, cognitive impairment, and medical complexity, where much assistance is required with activities of daily living. The impairment is permanent but can be progressive or static.” 3
Intractable Feeding Intolerance
Intractable symptoms and suffering from enteral feeding which may lead to an inability to sustain life via the enteral route.
Gastrointestinal Dysfunction
A spectrum ranging from poor oro-motor coordination, gastro-esophageal reflux (GOR) and constipation at the mild end to more severe GI dysfunction associated with feed-induced GI dystonia and dysmotility.
Gastrointestinal Dystonia
“The clinical manifestations of pain behavior; hypertonicity; retching; vomiting; vagal phenomena; abdominal distension and straining attributable to the GI tract in severe neurodisability.” 12
Gross Motor Classification Scale (GMFCS)
A five-level classification scale that describes the gross motor function of an individual with cerebral palsy, with GMFCS V the most severe with severe limitations to mobility and head and neck control. 13
Results
Demographics, Medical Background, and End-of-Life Details
Patient Characteristics, and Details about Feeding and Palliative Care Involvement Relative to Death.
aFeed intolerance used rather than GI dysfunction or failure.
All patients were managed in a tertiary hospital, and had their overall care led by a pediatrician in either a General Medicine or Neurodevelopment and Disability team. Most children had multiple medical, nursing, and allied health clinicians involved in their care, with 1 patient supported by thirteen medical/surgical specialties. All were known to Gastroenterology, and all except 1 to Neurology. Other teams included Respiratory, Rehabilitation medicine, Orthopedics and Cardiology. Three children had home oxygen, 2 received home non-invasive ventilation and 5 received non-invasive ventilation during hospital admissions. Four were documented to have an aspiration risk or recurrent aspiration pneumonia.
The mean age at death was 10.3 years. The median time from palliative care referral to death was 3.1 months. Three children died at home, and 1 died in a pediatric hospice. Of the 5 children who died in hospital, 1 died in the intensive care unit.
Language and Timing of Description of GI Failure/dysfunction
Gut or GI “failure” or “dysfunction” was used in the notes for 7 out of the 9 children in our cohort. These descriptions/diagnoses were documented by General Medicine, Neurodevelopmental and Disability, Palliative care, Pediatric Surgery, Gastroenterology and Pediatric Intensive Care teams, at times during joint reviews. For the 7 children whose electronic medical records referenced GI failure or dysfunction, the median time between this documentation and death was 3.9 months (range: .1 to 13.1). For the 2 children whose notes did not include the terms GI dysfunction of failure, feed intolerance was documented 16.6 and 31.2 months prior to death.
Gastrointestinal Details
Gastrointestinal problems were encountered in all 9 patients throughout their life and various investigations and changes to feeding regimens were performed to exclude treatable or reversible causes for their symptoms and attempt to improve enteral feeding tolerance (supplementary data). For some children, nuclear medicine studies described normal or relatively normal emptying times, yet despite this, documentation continued to raise concerns about slow gastric transit clinically.
All patients were fed via a gastrostomy tube (GT) during their life (median age of insertion 2.5 years (range: 1.2 to 6.8 years)). All except 1 child (patient #9) had increased weight for several months after initiating GT feeding. However, over time weight trended down, and 7 children were transitioned to jejunal tube feeding (median age 9.2 years (range 2.4 to 14.7 years)), with a median time of 4.9 years between GT to jejunal tube feeding (range: 1 month to 11.8 years). For 6 of 7 children, weight increased following the initiation of jejunal tube feeding. Figure 1 illustrates trends in weight (z-score) with the timing of initiation of GT and jejunal tube feeding for all 9 children. Figure 2 shows that when jejunal tube feeds were commenced, children had a median of 9 percent of their lives left to live. This equated to a median of 16.1 months (range: 1.1 – 31.2 months) left to live. Trends in weight corresponding to the percentage of life lived, with the timing of initiation of gastrostomy tube feeding (red arrow) and jejunal tube feeding (orange arrow) for all 9 children, (x-axis = proportion of child’s life lived, y-axis = z-score weight). Timeline depicting the median timing relative to death for commencement of gastrostomy and jejunal feeds (note x-axis = % of life lived).
Five children received a period of PN as an inpatient. Three children received short periods of PN, which allowed GI rest and recovery allowing enteral autonomy to be re-gained. This was in the context of intercurrent infection for 2 children, and a GI bleed for the third. For 2 children, PN was administered during what was ultimately an end-of-life admission. Of these children, 1 had previously had a period of PN with gut recovery and successful resumption of enteral autonomy. However, on the second occasion, PN was withdrawn in the context of overall progressive neurological deterioration, when PN was thought to be prolonging the dying process. PN was commenced as a treatment trial for the second child when enteral feeding was unable to sustain life and withdrawn once it was deemed likely that PN dependence would be lifelong following consultation with the hospital’s clinical ethics service.
Medications
Figure 3 illustrates medications used to manage symptoms or sequelae of GI problems at any timepoint in the child’s life. Opioid analgesics included morphine, oxycodone, hydromorphone, buprenorphine, and methadone. Non-opioid analgesics included paracetamol, ibuprofen, gabapentin, and amitriptyline. Tone management agents included trihexyphenidyl, gabapentin, clonidine, diazepam, and clonazepam, the latter 3 also with an anxiolytic effect. Prokinetic agents included erythromycin, domperidone and metoclopramide, the last 2 also having an anti-emetic effect. Other anti-emetics used included ondansetron, cyclizine, aprepitant and levomepromazine. Three children received alimemazine, and 1 received the anti-spasmodic hyoscine butylbromide. Although 6 children were recorded to have had a seizure disorder, only 3 were on anti-seizure medications when GI dysfunction or feeding failure was noted to be an issue. Specific details of constipation management were not included in this data however all patients were frequently assessed for the presence of constipation, and when constipation was likely or confirmed, patients were treated with regular aperients, titrated as needed. All children required opioid analgesia in the last week of life, but opioid and non-opioid analgesia were needed well before the terminal phase for all children. Medication classes administered for GI symptoms at any point in a child’s life (x-axis = number of children).
Discussion
Intractable Feeding Intolerance: Terminology and Role in Deterioration or Death
Although GI problems are well-recognized in children with SNI,4,5 intractable feeding intolerance has not typically been acknowledged as a cause or contributor to death. Intractable feed-related suffering and the inability to tolerate enteral feeding despite escalating feeding support (medications, tube feeding) are not typically seen as potentially irrecoverable organ dysfunction in the same way that escalating respiratory support is understood to represent deterioration in respiratory function. GI symptoms in children with SNI can be non-specific and fluctuate in intensity, both of which complicate diagnosis, management, and prognostication. Furthermore, feeding intolerance does not always result in death, for example, if PN is pursued as a life-prolonging intervention, or in the event of a terminal respiratory infection where the child dies with, rather than from, intractable feeding intolerance. In the latter example, feeding intolerance is likely to contribute to a child’s death, for example, through malnutrition, immunocompromise and frailty, even if feed withdrawal isn’t the ultimate cause of death. For some children however, a decision may be made to withdraw enteral feeding and allow a natural death. These examples highlight the challenge of assessing whether intractable feeding intolerance is the main cause vs a contributor to death; our opinion is that both can occur.
Improved awareness of the spectrum of severe GI dysfunction will facilitate prognostic conversations and allow goal-concordant medical treatment decisions. Recognition that the enteral route is proving challenging as a means of sustaining life should trigger discussions about clinical and quality of life challenges, and appropriateness of additional medical interventions such as jejunal feeding, central venous access, a trial of PN, or a commitment to longer-term home-based PN.
Future work should explore factors that contribute to decision-making about feed withdrawal vs PN commencement in intractable feeding intolerance. Such factors may include associated global neurological deterioration or worsening quality of life. Regardless of causality, it is important to recognize intractable feeding intolerance as a risk factor for increased symptom burden for patients, expansion of medical interventions and impacts on quality of life for patient and carers, and premature death, in children with SNI.
Variable terminology has been used to describe deteriorating GI function in children with SNI, including intractable feeding intolerance, 1 terminal GI dysfunction 9 and intestinal failure.10,11 In our case series, the terms GI dysfunction or GI failure were used in 7 out of the 9 children’s files. Of the various terms in the literature, we propose that the term intractable feeding intolerance should be utilized more consistently, and we have used this terminology in our study. The term is helpful as it highlights a critical crossroads in a child’s life where the intractable nature of their GI problems should prompt prognostic and treatment conversations which are vital to consider and achieve goal-concordant care. The term describes a clinical situation but deliberately does not indicate the expected impact on life expectancy. This is because the child’s trajectory will be affected by other factors, for example, global neurological functioning, medical co-morbidities, and quality of life, which may in turn impact a decision about the appropriateness of life sustaining PN when life cannot be sustained via the enteral route. Additionally, we recognize that intractable feeding intolerance, even when enteral feeding is continued, contributes to malnutrition, immune dysfunction, and frailty, which are all likely contributors to death in children with SNI.
The Role of Tube Enteral Feeding and Parenteral Nutrition
Feeding is a primal parental responsibility, and the insertion of a feeding tube can be associated with a sense of parental loss and “giving in.” 14 For example, parents may perceive poor growth, nutritional challenges, and the suggestion of a feeding tube as a failure on their part to provide for their child. 15 However, the benefits of feeding tubes such as less time spent on feeding, ease of medication administration, and improved nutritional outcomes, are important too. 15 For some children with SNI, concerns about feeding tolerance may herald broader prognostic concerns and add further layers of grief and complexity to families.
All patients in our cohort had a GT inserted, and 7 transitioned to jejunal tube feeding. For the 7 children fed jejunally, this commenced at a median of 91% of life lived. This raises many discussion points, including whether there are children with SNI for whom failure of gastric feeding represents a pre-terminal change in condition, and therefore whether there some children may not actually benefit from transition to jejunal feeding. Alternatively, the question may be asked as to whether outcomes may have been different had jejunal feeding been initiated earlier. This is a complex area that needs further consideration with recognition that exclusive jejunal tube feeding carries its own risk and burden profile, including placement issues, radiation exposure, dislodgement, equipment concerns, and necessity for continuous feeding regimen by pump, which may limit a child’s ability to participate in activities such as physical therapies and add additional supervision burden.
While PN can be associated with increased morbidity and mortality, including death from overwhelming sepsis, metabolic complications and liver disease,16,17 and high financial costs and burden of care for families,6,16 the role of PN in feeding failure in SNI requires further exploration. One study found that 7/33 (21%) children with SNI and GI dysfunction/failure regained full enteral autonomy after PN. 11 Further details, such as preceding potentially sensitizing neurological events like an infection or operation, were not included.
PN was used in our cohort for time-limited periods to allow GI recovery. In 2 cases, PN was withdrawn because it was incongruent with the goals of care in the setting of ongoing global clinical deterioration. At our institution, home PN has not been offered as routine care for children with SNI and feeding failure; however, this is an evolving space internationally, with noted global variation in access to home-based PN services and resources. Ethical considerations need to be weighed when a child with SNI can no longer be fed via the enteral route, 11 but for some children and families, home PN will be ethically justified. 7
Burdens of Care and Support Needs
High medical complexity and symptom burden were reflected in our patient cohort. A high medication burden was present not just during the terminal phase; all patients required opioids alongside other medications to manage their symptoms before the terminal phase. Our review did not include aperients and given that constipation is a common problem for children with SNI, the medication burden is likely to be even greater than represented. The retrospective nature of our study did not allow for the collection of data on symptom scores. Future studies should consider multidimensional symptom assessment that explores severity, frequency, and distress, and not simply the presence of a symptom. This could be done with proxy reports by parents of children with SNI using a modified Memorial Symptom Assessment Scale. 18 This is important given that children with SNI experience pain that is frequent and severe but often difficult to assess.19,20
Investigations constitute an additional burden: GI dysfunction in SNI is often investigated with invasive tests and interventions that do not always yield diagnostic or prognostic information. Coupled with the high investigative burden experienced by the children in our cohort, there is a dearth of literature outlining how best to balance invasiveness, potential utility, and alternative investigation and management pathways in children with SNI and severe GI dysfunction. This may challenge health professionals caring for patients with SNI who are used to confirming organ dysfunction, such as heart or lung, with investigations. Communicating that episodes of severe intractable feeding intolerance may represent ‘dips’ in GI function might help health professionals and families understand that akin to other critical organs, the GI system is vulnerable to dysfunction and may be at risk of not being able to recover or sustain life in a small subset of patients. Further work is needed to guide an investigative pathway for children with SNI and intractable feeding intolerance, balancing the burdens of investigation with potential benefits, in the context of the child’s broader prognosis and goals of care.
An evolving understanding that GI dysfunction and symptoms may be refractory to treatment, highlights the importance of careful discussions about a palliative approach to care. 1 Distinguishing problems that are ‘fixable’ vs modifiable and issues that may be empirically managed or remain intractable 1 is a suggested approach to discussions with families about these complex issues. Management of children with SNI should encompass a multi-tiered perspective considering a child’s underlying GI function, nutritional status, symptom control, and priorities through a family-centered approach and using a coordinated multi-disciplinary approach.
In our study, death occurred mainly in the acute hospital setting. This raises the possibility that in our health and societal context, the needs of children with SNI and feeding failure in the pre-terminal and terminal phases are difficult to meet in the home environment. These needs may relate to challenges around diagnosis, prognosis, and management upstream of the terminal phase. Finally, disparate healthcare contexts and home-based support services locally and internationally are likely to contribute to the available management approaches for children with SNI who develop intractable feeding intolerance.
Palliative and End of Life Care
Given that our understanding of GI dysfunction and intractable feeding intolerance in SNI is in its infancy, we may be under-recognizing intractable feeding intolerance as a component and contributor to a child’s broader deterioration and death. The first mention of GI dysfunction or failure occurred within 2 months of death in 4 patients and 13 months in all children, possibly suggesting that recognition or articulation of the potentially life-limiting extent of GI problems occurs late. Jejunal feeding was typically commenced earlier than documentation of feeding failure, which may indicate that jejunal feeding was commenced for other reasons such as aspiration, or that for a cohort of patients this heralds global deterioration which might have been under-recognized. This is supported by Siden et al’s work in which he suggests that signs and symptoms of severe GI dysfunction may only become evident late in a child’s life. 8
The median time from palliative care referral to death of 3.1 months was similar to a median of 3.6 months for 67 children known to the same state-wide pediatric palliative care service who died from either a malignant or non-malignant condition over a 12 month period (July 2016 to June 2017) showing no significant delay in initial palliative care referral in the SNI and feeding failure subgroup. 21
Our cohort’s median age of death was 10.0 years (range 5.0 – 15.6 years). Although the life expectancy of children with SNI is increasing, 4 an increased risk of premature mortality persists. 22 Improved survival is likely related to medical advances and access to medical technologies and supports, 23 including non-invasive mechanical ventilation for children with chronic respiratory failure. 24 Therefore, we are witnessing a “new natural, or perhaps ‘unnatural’ history” for children with SNI whose lives are being extended by medical advances. 7 This only adds to the importance of understanding and progressing research to better understand this patient population.
Respiratory causes have been recognized as the leading causes of death in children with SNI, for example, children with cerebral palsy.25-27 However, as children with SNI live longer, there is perhaps more time for GI deterioration to evolve, and significant intractable feeding intolerance to become a key additional end-of-life issue alongside respiratory failure. 7
We suggest that recurrent or progressive difficulty in sustaining life via the enteral route is a sentinel moment in the life of, and crossroads in care for, a child with SNI, and warrants careful consideration of the clinical presentation and appropriateness of interventions.
Strengths and Limitations
To our knowledge, this is 1 of the largest studies to describe the clinical course of pediatric palliative care patients with SNI who experienced intractable feeding intolerance that contributed to death. This is an important step in raising the profile of a medically and psychosocially complex patient population whose care will be improved by shared knowledge and consistent language.
The study is limited by its retrospective nature. Information can only be included if it was captured in the medical record. It is possible that not all diagnostic and prognostic conversations between health professionals, or with a family, were captured or concerns about GI dysfunction or deterioration may be under-recorded, highlighting the importance of improved understanding and knowledge in this area.
Future prospective studies should further explore the relationship between feeding intolerance and general neurological health status in children with SNI, the temporal relationship of neurologically sensitizing events such as infections and procedures with episodes of feeding intolerance, as well as longitudinal symptom assessment and quality of life, none of which could be reliably assessed in our study.
Conclusion and Future Directions
Children with SNI commonly experience a high burden of GI symptoms and associated investigation and interventional management approaches. As children with SNI live longer, GI dysfunction and feeding intolerance are likely to become more apparent as substantive contributing factors to deterioration and death in children at the most severe end of the spectrum. Episodes of intractable feeding intolerance in a child with SNI should be considered a potentially critical time point in decision-making, akin to significant respiratory deterioration triggering consideration about the appropriateness of escalating ventilatory support. We propose that the term ‘intractable feeding intolerance’ is helpful to describe this clinical crossroads in a child’s life where the appropriateness of further interventions should be considered.
Further research is needed to assess whether there are factors to help identify children at risk of developing intractable feeding intolerance and factors that might determine the degree to which this may contribute to or lead to a child’s death. Richer understanding in this area will improve content and timing of goals of care discussions and decision-making for families of children with SNI and intractable feeding intolerance. Finally, we recommend further exploration of the role and timing of referral for palliative care in children with SNI and GI problems. With such high degrees of medical and psychosocial burden and a lack of evidence-based practice in this area, holistic patient and family-centered care is paramount.
Supplemental Material
Supplemental Material - Intractable Feeding Intolerance in Children With Severe Neurological Impairment: A Retrospective Case Review of Nine Children Known to a Pediatric Palliative Care Service
Supplemental Material for Intractable Feeding Intolerance in Children With Severe Neurological Impairment: A Retrospective Case Review of Nine Children Known to a Pediatric Palliative Care Service by Naomi T Katz, Monica S Cooper, Aeshan Kularatne, Anna Prebble, Kathleen H McGrath, Zoe McCallum, Giuliana Antolovich, Ingrid Sutherland, and Bronwyn H Sacks in American Journal of Hospice and Palliative Medicine®
Footnotes
Acknowledgments
The authors wish to acknowledge Dr Julie Hauer MD, a pioneer in this field. We are grateful for the conversations we have had; these conversations have helped shape our thoughts and framing of a complex topic. Finally, and most importantly, we wish to acknowledge the young people, and their parents, who we look after. You motivate us to keep learning and researching, so that we can care for you in the best way possible.
Author Contributions
Naomi T Katz designed and directed the project, contributed to data collection, analysis and interpretation, and development of the manuscript. Bronwyn H Sacks designed and directed the project, contributed to data collection, analysis and interpretation, and development of the manuscript. Aeshan Kularatne and Anna Prebble contributed to data analysis and interpretation, as well as the initial development of the manuscript. Monica S Cooper contributed to data collection, and critically reviewed and revised the manuscript. Kathleen McGrath, Zoe McCallum, Giuliana Antolovich, and Ingrid Sutherland critically reviewed and revised the manuscript. All authors reviewed, and approved the final version, and agree to be accountable for all aspects of this work.
Declaration of Conflicting Interests
The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: KM is an advisory board member and speaker of Takeda for Revestive (teduglutide).
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Supplemental Material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
