ERS statement on transition of care in childhood interstitial lung diseases

Question 1 (PICO): Does participation of children with a chronic respiratory disease in a care transition programme lead to better outcomes?

To assess the added value of participation in care transition programme we performed a systematic literature search. The systematic search identified 946 publications (MEDLINE/PubMed n=225, EMBASE n=332 and Cochrane n=389); 27 full-text articles were retrieved for further evaluation of eligibility. 10 studies were included in the qualitative synthesis. Results are depicted in table 2 and refer almost exclusively to patients with cystic fibrosis, although some studies also included nonrespiratory disorders such as inflammatory bowel disease and type-1 diabetes mellitus. This highlights the paucity of data about the transition of chILD patients [1, 15–23]. Based on the aforementioned studies, the participation of children with a chronic disease in a transition care programme enables a more successful transfer by lowering pre-transfer anxiety and significantly improving transition competence and readiness, self-efficacy and independence, as well as satisfaction with school/educational services. Overall quality of life and health-related outcomes such as lung function and body mass index do not worsen, but remain stable in the majority of studies, and seem to improve in a few. The benefits of the transition care programmes are observed not only in patients, but also in their parents.

Question 2: Which care transition programmes are currently used in clinical practice for children with ILD?

The literature search narrowed to chILD and transition programmes from paediatric to adult care revealed seven papers. Two papers addressed nonrelevant topics; two addressed only a narrow spectrum of diseases, mainly with regard to the long-term prognosis. Two others concentrated more on the diagnostic aspects and the need for development of guidelines. Therefore, we found only limited data on the transition process of patients with chILD from paediatric to adult care systems that are already used in clinical practice. There is one paper [25] which deals with the transition process of chILD in general. It reports outcomes from the COST Action 16125 (working group 5, dedicated to the transition process). In a survey among 39 centres from 19 European and two non-European countries, which was performed by this working group, there were no standardised transition protocols and no established transition system for children with ILD. They identified only three (11%) centres having a standard operation protocol for transition and nine (32%) centres using a standardised medical report form that accompanied the patient during the whole transition process. Nine (32%) centres organised a meeting between paediatric and adult teams without patient/parents; only two (18%) adult centres had an established system of long-term follow-up of patients with remitted chILD. The age of transition from paediatric to adult healthcare was mostly 18 years, ranging from 16 to 21 years. There was a high variability in the answer to the question about how long before the actual transition the transition process starts. This varied between 3 months to 4 years, with 1 year being most frequent. The availability of ILD multidisciplinary teams involved in care and transition were rather variable between countries. Only 20 (51%) of all responding centres reported availability of an established multidisciplinary team. Most frequently, pulmonologists/paediatric pulmonologists, radiologists and pathologists were involved. Only four (14%) paediatric centres routinely organised a pre-transition meeting with the patient, parents and both the paediatric and adult teams; nine (32%) centres had a meeting between paediatric and adult teams without patient/parents; five (18%) paediatric centres organised a meeting of the paediatric team with patient/parents. No centre reported an established system of long-term follow-up of children with chILD specific for early age (for example, neuroendocrine cell hyperplasia of infancy (NEHI), pulmonary interstitial glycogenosis).

Koucký et al. [25] suggested a model procedure for the transition of care. The patient would be re-assessed in detail during the last year of follow-up at the chILD centre. Based on the practices of the task force members, a thorough evaluation of clinical and functional status is useful, and radiology (chest high-resolution computed tomography (HRCT)) will usually be considered depending on the specific diagnosis and activity of the disease. Finally, a standardised transition report was suggested. There were no other articles found detailing individual chILD transition programmes and evaluating their efficacy with respect to patients’ disease management skills, patient access to and satisfaction with healthcare, self-efficacy, risk of loss to follow-up and risk of missing disease progression.

In the current clinical practice of the task force members, the timing and extent of diagnostic re-evaluation during the transition process is determined on a case-by-case basis, in the absence of any definitive evidence. Although some forms of ILD are common to both paediatric and adult pulmonary practice, dissimilarities between paediatric and adult disease classifications as well as evolution of the condition between the time of the original paediatric diagnosis and subsequent transition to adulthood are among the reasons why clinicians re-evaluate the patients on an ad hoc basis.

Task force members indicated that usually the process incorporates input from the paediatric and adult clinical teams including specialist nurses and respiratory support staff, as well as the patient and their family or carers. The task force members believe that clinical considerations, such as the history of the condition, its management up to that point and response to therapy may be important factors influencing the decision whether to re-evaluate at transition.

Discussion within the task force regarding the current practice identified the most commonly used components of a general clinical re-evaluation plan: 1) blood investigations including extended autoantibody panel, serology for hypersensitivity pneumonitis (IgG against a number of organic stimuli such as avian and fungal antigens) and repeat genetic testing for entities likely due to a known and not yet tested condition; 2) physiological testing such as detailed lung function testing (including static lung volumes, transfer factor of the lung for carbon monoxide), overnight oximetry, 6-min walk test and possibly oesophageal impedance–pH assessment; 3) thoracic radiological imaging with HRCT as the key modality; and 4) echocardiography (baseline cardiac chamber dimensions, biventricular function and estimate of pulmonary artery pressure). With respect to the radiation dose, the HRCT is done on a case-by-case basis. The cumulative radiation burden has to be taken into account; however, with modern imaging technologies (ultrasound, magnetic resonance imaging, ultra low-dose computed tomography) this may become less important [26]. A list of minimal content of the transition report is presented in table 3 [25]. No published evidence-based validated list could be found in the literature.

Question 3: In which other diseases are transition of care programmes used in clinical practice?

Most children and young people with chronic illnesses will survive into adulthood, and would optimally transition to an adult model of care. A literature search revealed 1748 papers with numerous transition protocols in different pathologies, such as rheumatological diseases, congenital heart disease, inflammatory bowel disease, chronic neurological disorders, diabetes mellitus and patients on long-term ventilation [8, 10, 12, 13, 27, 28]. Most of these patients have complex conditions requiring multidisciplinary medical care teams and are classified as “youth with special healthcare needs” [14].

Based on the literature and task force members’ personal practice, an important goal of transition is to empower patients and families, aiming at increasing autonomy, patient participation, awareness and consciousness, as well as the development of relevant psychosocial skills. It is also crucial to ensure an organised process to facilitate transition preparation, transfer of care and integration into adult-centred healthcare. Without such a process, there is a big risk that children and young people, even with complex ongoing treatment needs, will never transition, and be lost to follow-up [29–31]. Some individual centres and different organisations have developed transition programmes to achieve these goals. In 2011, updated in 2018, the American Academy of Pediatrics, with the American Academy of Family Physicians and the American College of Physicians, published a clinical report on healthcare transition that included a process for transition preparation, planning, tracking and follow-through [32, 33]. Afterwards, a structured clinical approach called the Six Core Elements of Health Care Transition was developed for all children and young people [34]. An important part of the process is talking to children and young people using the acronym HEADSS (home, education, activities/employment, drugs, suicidality, sex) [35].

The literature highlights the need for structured elements in a healthcare transition programme: a written transition policy, and a transition plan, which is a working document that records findings of assessments, describes the progress in transition and outlines planned actions. A very good example of this is the Ready Steady Go project in the United Kingdom (UK), which has been developed as a generic transition plan, and comprises assessment about knowledge and skills over time [36]. This programme is being used to develop increasing numbers of disease specific versions. Another example is the Got Transition (United States of America) Six Core Elements of Health Care Transition 2.0 programme (www.GotTransition.org). These programmes help to facilitate the transition process for the young person and try to involve the patient in the process and help in various settings and conditions. In addition, the UK National Institute for Health and Care Excellence has published excellent recommendations for transition to adult services [31].

Most transition programmes start in early adolescence and continue into early adulthood; however, the fact that transition will occur is usually mentioned much earlier, often at diagnosis. Usually, these programmes are provided in three phases: pre-transition, where the transition process and transfer policy is introduced to the patient and family; the transition phase, where the needs of the adolescent are assessed, he/she is enabled to manage his/her own health supported by education and counselling, and also includes the actual transfer to adult care; and the post-transition phase, when the continuity of care is monitored.

The task force members believe, based on their practice and current literature, that for most patients, the appropriate time for implementing specific stages of transition planning is determined by age, whereas for others, it may be modified based on developmental considerations [7, 8]. As this is a vulnerable time for young adults, and is also a challenging period for medical teams, patient care during transition would optimally be individualised.

Transition planning is a multidisciplinary process whereby different disciplines are involved in order to address different areas of need. Physicians and nurses usually discuss and assess medical aspects, disease knowledge, management tasks and health-related self-efficacy; psychologists are suited to target issues of nonadherence to treatment, and behavioural or emotional factors; social workers can help patients by connecting them with resources that may facilitate transfer. Recently, the crucial role of adult clinicians in accepting and partnering with young adults has been emphasised as a professional education and training challenge. For the most part, no published data has been found regarding diagnostic re-evaluation in the transition literature, but there are some (non-evidence-based) exceptions [37].

Transfer usually occurs at a time of disease stability. The task force members agreed that in their practice, supported by current literature, it was found essential to provide a report assessing readiness for transfer, as well as the important information that summarises all relevant data about the patient's medical history (an example being shown in table 3).

Additional data that are likely to be included in the transfer documentation include family history, bespoke treatment plans and contacts, and the patient's health education history and assessment of his/her understanding regarding health conditions, treatments, and prognosis with particular attention to entry into adult life [8, 12, 13, 28, 38].

The evidence on healthcare transition outcomes remains limited, but there are some evaluation studies that document beneficial outcomes of a structured transition approach in terms of quality of care and in terms of service use and patient and family experience. A systematic review identified 43 studies, of which two-thirds revealed statistically significant positive outcomes. The most commonly reported outcome was improvement in adherence to care, followed by improved perceived health status, quality of life and self-care skills [39].

In 2016, a Cochrane review [40] evaluated the effectiveness of interventions designed to improve the transition of care for adolescents from paediatric to adult health services. The few available pieces of evidence, which covered a limited range of interventions in a limited number of clinical conditions, showed improvement in patients’ knowledge of their condition, and improvement in self-efficacy and confidence. Nevertheless, it concluded that further research is needed for rigorous evaluation of models of transitional care, reporting on clinical outcomes and patient-reported outcomes with longer-term follow-up.

A recent study showed in a small sample that the common experiences included lack of transition preparation and planning, the need for learning how to adapt to adult services and a changing healthcare scene. Additionally, the subject of how to include the parent as a continued advocate for the young person was discussed. Among the weaknesses identified in the current transition process were the lack of a specific transition coordinator, the generally haphazard timing of first discussion about transition and the lack of involvement of psychologists and social workers [41].

Specific characteristics of the different transition programmes for well-studied chronic diseases of childhood other than chronic respiratory diseases are included in table 4 [8, 10, 12, 13, 27, 28]. Regarding chronic respiratory diseases, the nature and structure of the process are generally well described in cystic fibrosis care, where a relatively large number of patients transition each year. Cystic fibrosis is one of the diseases with best-studied transition pathway in the literature (table 1) [7, 43]. The task force members considered that some aspects could potentially be successfully transferred to chILD transition programmes. However, there are gaps in knowledge and recommendations regarding the structure of diagnostic re-evaluation for chILD which encompasses a more heterogeneous group of diseases than cystic fibrosis. There are also organisational issues: for a rare disease like chILD it will not be feasible for every centre to have a dedicated chILD disease coordinator, so a shared approach with other diseases is probably necessary. Lastly, there may be difficulties with regard to selection of the appropriate adult expert centre, especially for those entities of chILD not well-known to adult ILD experts, an issue better addressed in question 5 (Does the current definition of chILD versus adult ILD influence transition of care?).

Question 4: Is there evidence of long-term remission of chILD?

There is still a big knowledge gap on natural disease course, disease progression and treatment induced or spontaneous remission of chILD. We identified a limited number of studies on the long-term evolution of various chILDs in our literature search. Unfortunately, these papers provide no information on the transition process itself and do not evaluate its prognostic impact. There are only few studies reporting health status of patients transitioning from childhood to adulthood [41]. Moreover, these data are significantly biased, as those with unsuccessful transition are unlikely to be reported (loss to follow-up).

Although there is no specific definition of remission of chILD, based on studies and expert opinion it could be postulated that remission consists of four dimensions [44]:

• clinical signs and symptoms;

• pulmonary function (spirometry, body plethysmography, diffusion capacity of the lung for carbon monoxide (D_LCO), exercise tolerance tests);

• imaging (HRCT scan of the chest);

• pathology, cytology (results of tissue biopsy and bronchoalveolar lavage abnormalities).

Complete remission is defined as an absence of clinical signs and symptoms, normal lung function and exercise tests and resolution of imaging abnormalities. However, it should be noted that this does not capture the dimension of risk of relapse.

Partial remission can consist of improvement of one or more of the aforementioned dimensions.

Remission has been reported spontaneously or during treatment and depends on the underlying disease. Diseases with a chance of spontaneous remission are nonfibrotic hypersensitivity pneumonitis after removal of the causative agent, NEHI, sarcoidosis and organising pneumonia [45–49]. For other chILD entities, partial remission can be achieved depending on underlying cause and treatment response. We summarise studies on chILD entities with long-term follow-up and (partial) remission.

These data confirm the necessity of continuous patient follow-up from childhood to adulthood. The risk of relapses later in life remains to be elucidated, especially in patients with apparently stable disease and prescribed no long-term treatment at the time of the transition; we do not have the tools for assessment of risk. Up to 50% of children with sarcoidosis still required treatment at the end of paediatric follow-up, the rest were stable off treatment. However, 19% of those stable patients relapsed later in adulthood. The authors concluded that childhood-onset sarcoidosis is characterised by a multivisceral involvement and a protracted course requiring follow-up into adulthood. Juvenile cases are characterised by severe organ manifestation and the severity cannot be predicted at onset [50].

Sisman et al. [51] followed 22 children with hypersensitivity pneumonitis for median of 3.3 years (age at end of follow-up ranging from 11.3 to 26.9 years) and most were stable off treatment. Six (27.3%) relapsed. Spirometry was normal in 90.9% of children at follow-up, while lung clearance index was increased in 47.4% cases, indicating persistent small airway disease. Abnormal total lung capacity, D_LCO and peak oxygen uptake were noted in 13.6%, 40.9% and 11.1%, respectively. Children with post-infectious bronchiolitis obliterans have been shown to suffer from severe airway obstruction, in some cases with need of supplemental oxygen. During a 12-year follow-up period, pulmonary function remained severely impaired, showing an obstructive pattern with air trapping. Although most of the patients stop needing supplemental oxygen around the age of 10 years, airway obstruction persisted with only slow if any improvement. Patients required frequent readmission due to recurrent respiratory infections; 19.6% developed chest deformity; and 15.2% underwent lobectomy because of treatment-resistant bronchiectasis, a higher figure than in most centres [52].

Diffuse alveolar haemorrhage (DAH) represents a heterogeneous group of conditions with variable onset and prognosis. In a large multicentric study of retrospective character Ring et al. [53] studied 117 children with DAH with a median (interquartile range (IQR)) age at presentation of 5 (2–12.9) years. There was a median (IQR) diagnostic delay of 2 (0–12) months. Outcome data were available for 90%, of whom 13% died (15 children); 86% were still alive: 19% were considered healthy, 16% were chronic clinically stable, 31% were active clinically stable and 12% had ongoing medical treatment and were unstable. While this study does not provide direct data on long-term follow-up into adulthood, the analysis of outcomes suggests that many such patients would be surviving into adulthood with sequelae of the disease and risk of a relapse.

ILD may complicate various rheumatological diseases and significantly impacts morbidity. Juvenile forms of rheumatological diseases are rare and poorly investigated with limited data on long-term follow-up. In a longitudinal follow-up of a cohort of 60 children with juvenile-onset systemic sclerosis from the European League Against Rheumatism (EULAR) Scleroderma Trials and Research (EUSTAR) database, the median age at disease onset was 12.4 years. Pulmonary fibrosis was present in 23.3% of patients and pulmonary hypertension in 13.3%. Median disease duration at last follow-up was 17.6 years and 26.6% of patients had active disease at that time. There were no data on the relapse rate in previously inactive disease [54].

ILD in surfactant protein C (SP-C) disease is very heterogeneous. Disease may manifest both in children and adults [55]. Follow-up of 17 children with SP-C disorder for a median 3 (range 0.9–19) years showed disease progression in three (17.6%) cases despite treatment, stability in seven (41.1%) patients and improvement in seven (41.1%) patients, one apparently becoming completely healthy. In a study of five children with open lung biopsy diagnosed ILD were followed long-term (mean 27.2 years) and all subsequently had confirmed SFTPC gene mutations. Three of the patients felt well after follow-up and had normal lung function; two females conducted normal activities of daily living, had healthy children but had evidence of restrictive lung disease. A positive effect of hydroxychloroquine seemed likely [56]. There are other papers on the evolution of health status in children with SP-C mutations [57–60]. It does appear that after years of relative stability there may be relapse and progression to refractory pulmonary fibrosis. In adults, SP-C mutations may present at any age often with usual interstitial pneumonia and progressive pulmonary fibrosis (often no pulmonary disease is detected in childhood). Very recently a longitudinal study of ABCA3-related ILD in patients surviving beyond infancy demonstrated that this is a severe condition even in survivors beyond the first year of life and is characterised by radiological progression and functional decline [60].

Neuroendocrine cell hyperplasia of infancy and pulmonary interstitial glycogenosis are specific conditions of undefined aetiology usually presenting in infancy. Current data suggest favourable prognosis with symptom regression during early life [45, 61–64]. While 50% of patients with NEHI had no symptoms by age 3 years in one study [65], acute exacerbations have been documented in NEHI [63], and children with NEHI may also suffer from growth and developmental delay or asthma-like symptoms at school age which are not steroid-responsive [45, 65]. As these entities were not described until after 2000, we still lack data from adults regarding their long-term prognosis and risk of relapse.

In some other disease entities, a chronic progressive course has been seen with no documented remission. Mutations in MARS (methionyl-tRNA synthetase) manifest in infancy with pulmonary alveolar proteinosis and progress to fibrotic lung disease with chronic respiratory failure with a median survival of ∼1 year [66, 67]. Hermansky–Pudlak syndrome type 2 manifests with fibrosing lung disease early in childhood (onset of respiratory symptoms on average at 5.5 years (range 0.8–12 years). At last follow-up at age 12 years (4–19 years) all patients had clinical and computed tomography abnormalities [68].

Question 5: Does the current definition of chILD versus adult ILD influence transition of care?

Literature search revealed that no study has been performed so far; therefore, only expert opinion dictates that lack of harmonisation between the current adult ILD and chILD classifications on transition of care affects transition negatively. A common understanding of terminology and disease states is a prerequisite for successful transfer. Based on the most recent classification, ILDs in adults can be broadly subgrouped into idiopathic, autoimmune-related, caused by external agents (e.g. exposure to organic and inorganic inhaled antigens or treatments, such as drugs or radiation), ILDs with cysts or airspace filling, sarcoidosis and orphan disorders [44]. The classification of chILD includes almost all the subgroups of the adult ILD classification in addition to specific entities that are absent from adult classifications including developmental disorders, such as alveolar capillary dysplasia, genetic surfactant disorders (of both synthesis and catabolism) and neuroendocrine cell hyperplasia of infancy. Furthermore, it embraces entities such as diffuse parenchymal lung disease (DPLD) related to lung vascular structural disease (e.g. pulmonary hypertension) and DPLD in the immunocompromised host or transplanted (bronchiolitis obliterans related to rejection) [69, 70]. There is no accepted age-overarching classification covering both paediatric and adult patients that could be routinely used in everyday care for the entire spectrum of chILD. A recent perspective, reviewing the history of ILD classifications, may be a good starting point for an international consensus to harmonise ILD disease classification [71].

In summary, important barriers based on differences between disease categorisations of children and adults could be addressed, to facilitate successful transition of chILD into adult ILD care.