Effects of eHealth exercise on paediatric suppurative lung diseases: a systematic review and meta-analysis

Abstract

Background

Utilising information and communication technologies through eHealth in exercise programmes could support their delivery and improve clinical outcomes in children and adolescents with chronic suppurative lung diseases (CSLDs). This study aimed to systematically investigate the effects of eHealth exercise programmes on clinical outcomes for this population.

Methods

Five databases were searched from inception to 12 April 2024. Two researchers independently screened the retrieved results and rated the methodological quality of the included studies using the revised Cochrane Risk of Bias (RoB2) tool for randomised trials and the Risk Of Bias In Non-randomised Studies – of Interventions (ROBINS-I) tool. The quality of evidence was graded using the GRADE approach. A narrative synthesis of findings was performed, and a meta-analysis was conducted to evaluate the effects of eHealth exercise programmes on clinical outcomes that had data from at least two studies.

Results

Seven studies used eHealth exercise programmes through active video games (n=3), videoconferencing (n=3) and a digital spirometer (n=1) lasting from 3 to 12 weeks. Five studies had participants with cystic fibrosis (CF). Results showed a greater improvement in the 6-min walk test following the intervention compared to the control group (pooled estimate mean difference 37.2 m, 95% CI 7.91–66.48 m; p=0.013). Pulmonary function, exercise capacity, balance, peripheral and respiratory muscle strength, and health-related quality of life were also improved. Still, most studies involved a considerable risk of bias.

Conclusions

eHealth exercise programmes can improve clinical outcomes in children and adolescents with CF. Further research is needed for other paediatric populations with CSLDs and also for comparisons with conventional exercise programmes.

Shareable abstract

Children and adolescents with suppurative lung diseases can benefit from eHealth exercise, improving functional capacity. Research is needed for bronchiectasis and PCD, and comparison of eHealth to conventional programmes on long-term outcomes. https://bit.ly/4lhwgv2

Introduction

Chronic suppurative lung diseases (CSLDs), including cystic fibrosis (CF), localised-to-diffuse non-cystic fibrosis bronchiectasis (NCFB), primary ciliary dyskinesia (PCD) and protracted bacterial bronchitis, present abnormal sputum and persistent productive cough as their main symptoms [1]. Children with these conditions and their parents report poorer health-related quality of life (HRQoL) than healthy peers [2]. Furthermore, pulmonary function and HRQoL are inversely associated with sputum production and the severity of other symptoms [3]. According to a systematic review and meta-analysis in adults with bronchiectasis, HRQoL presents stronger correlation with subjective outcome measures for symptoms (cough, dyspnoea, fatigue) compared to objective outcome measures, such as pulmonary function and exercise capacity [4].

Using the treatable traits approach, the management of children with CSLDs can target multiple factors. Some extrapulmonary traits can be addressed by improving exercise capacity and HRQoL, while simultaneously reducing exacerbations and hospitalisation rates [5]. Children and adolescents with CSLDs experience reduced exercise capacity compared to healthy peers because of frequent respiratory infections and persistent inflammation of the airways, as well as airway damage [6]. Additionally, inadequate management of daily symptoms negatively impacts exercise capacity [7, 8], and low levels of physical activity are related to longer hospitalisation stay [9]. Thus, exercise programmes are core components of clinical management, aiming to reduce respiratory symptoms, increase exercise and functional capacity, prevent respiratory infections and improve HRQoL [10, 11]. In parallel, intrapulmonary traits can be targeted with management that includes breathing exercises [5].

The use of information and communication technologies to support the delivery of clinical management in healthcare services and improve health, also known as eHealth, can facilitate the utility of exercise programmes in centre-based or home-based settings [12]. Examples of eHealth technologies are active video games, videoconferencing and mobile apps [13]. Centre-based eHealth exercise programmes can be supervised by a healthcare professional in person, while videoconferencing allows for live-distant supervision. On the other hand, patients who follow eHealth exercise programmes in home-based settings have accessibility to healthcare services from distance (e.g., when circumstances require social distancing) and no requirement to travel [14].

There are a few promising clinical factors and studies for the use of eHealth technologies. Telerehabilitation is an easy and approachable way of communication between patients and healthcare professionals [15], and the COVID-19 pandemic forced many centre-based exercise or rehabilitation programmes to be replaced by home-based interventions to minimise the possibility of spread of infection [16]. A previous systematic review on the use of telehealth for monitoring symptoms, assessing adherence to pharmacological treatment, and delivering behavioural and nutritional intervention to children and adults with CF showed that there was insufficient evidence for the potential benefits of telehealth and suggested that research should also focus on other uses, such as exercise programme delivery [17]. Another systematic review focusing on children and adults with CF showed that video games lead to similar or higher cardiorespiratory demands compared to exercise programmes of conventional delivery [18]. Furthermore, patients showed a preference for active video games through a greater engagement in exercise activities after using video games [18]. Moreover, a recent systematic review for children and adolescents with chronic pulmonary diseases showed that telehealth may improve clinical outcomes such as symptom control, adherence to treatment, reduction of hospitalisations and HRQoL [19]. Other forms of eHealth, such as mobile applications (mHealth), seem efficient for symptom control and reducing hospital admissions in children and adolescents with asthma [19].

Digital technologies offer new opportunities to support home-based or centre-based exercise programmes by providing tools (such as smartphones and tablets) that children and their carers can use for instant access to online health information and supervision [20]. However, little is known about the effectiveness of eHealth exercise programmes in children and adolescents with CSLDs. The aim of this systematic review was to investigate the effects of home-based and centre-based eHealth exercise programmes on clinical outcomes in children and adolescents with some common CSLDs.

Methods

Study eligibility

This systematic review was conducted following the preferred reporting items for systematic reviews and meta-analysis (PRISMA) guidelines [21]. The Population, Intervention, Comparator, Outcomes, Time factor and Study Design (PICOTS) eligibility criteria were used (supplementary table S1). The protocol was registered in the International Prospective Register of Systematic Reviews (PROSPERO: CRD42024498403), and it was published [22]. In brief, study publications in English were included if they fulfilled the following criteria. They included children and adolescents (0–18 years) with some common diagnoses of CSLDs, namely, those with CF, NCFB and PCD. We did not include all the conditions that fall within this scope, and there were no restrictions on characteristics such as sex or ethnicity. Studies also assessed eHealth exercise programmes, where eHealth exercise programme was defined as the delivery of an exercise training programme, i.e., aerobic exercise, strengthening exercise and breathing exercise, or any of their combination, which is delivered via information and communication technologies (telecommunication) such as videoconferencing, active video games (with or without sensors for monitoring) and mobile applications. Finally, eligible for inclusion were the studies that evaluated the effect of eHealth on pulmonary function, functional capacity, exercise capacity, balance, peripheral and respiratory muscle strength, and HRQoL. Control groups could be any type of intervention, which is delivered in person in the hospital or at home, without using technological tools, as well as the usual care, placebo or no intervention.

Owing to the eHealth exercise interventions being in their primary stages of research and potentially limiting the number of available randomised controlled trials (RCTs), study designs could be RCTs, non-RCT interventional studies and pilot studies. For the mixed-method studies, only quantitative data were used. In the case of mixed samples, i.e. with different chronic diseases, including the aforementioned, the study was included only if data for the populations mentioned above were reported separately. Studies that included a mixture of eHealth exercise programmes and conventional in person exercise were included only if at least 50% of the intervention (e.g. number of sessions) was delivered via eHealth technology [23].

Systematic reviews and meta-analyses, narrative reviews, observational studies, duplicate studies, editorial or opinion articles, grey literature, guidelines, protocols and abstracts of conferences were excluded. Studies that included participants >18 years, had patients with another primary respiratory disease (such as asthma), involved other interventions (e.g. nutritional management) or used other methods of remote delivery (telephone and/or e-mail) were also excluded.

Search strategy and study selection

A systematic electronic search was performed from inception to 12 April 2024, in five databases: Scopus, PubMed, Medline (via EBSCOhost), Web of Science and ACM Digital Library (supplementary table S2), as these databases provide a broad selection of study designs. The concepts and key index terms used were adapted to the selected databases and the keywords were combined using Boolean logical operators (AND and OR).

All retrieved studies were imported into the Rayyan (Qatar Computing Research Institute) Web app and duplicates were automatically removed (https://www.rayyan.ai). Two researchers (A. Mavronasou and V. Sapouna) independently screened all titles and abstracts and then the full texts of the remaining studies against the eligibility criteria. Relevant systematic reviews that were identified during screening were used for a hand-held search of additional eligible studies for inclusion. Discrepancies at any stage of the study selection were resolved by discussion and a third independent researcher (E.A. Kortianou).

Data extraction

Two researchers (A. Mavronasou and V. Sapouna) independently extracted the data using a study-specific form, with the following domains: author, year, study design, diagnosis, population, professional responsible for delivering the programme, intervention (venue, type, delivery method, number of sessions, duration of sessions, duration of the programme), control group (where applicable) and outcome measures (primary and secondary). Pre–post-intervention differences between groups (when applicable) or post-intervention values of the outcomes measured were reported for each study. When outcome data were not reported, the corresponding authors were contacted to provide additional data. If these data were not returned within 2 weeks from the request communication, they were excluded from the analysis.

Risk of bias (quality) assessment

The revised Cochrane Risk of Bias (RoB2) tool for randomised trials [24] and the Risk Of Bias In Non-randomised Studies – of Interventions (ROBINS-I) tool [25] were used to assess the methodological quality of the included randomised and non-randomised studies, respectively. The RoB2 tool is used to assess and report the risk of bias for the following five domains: randomisation process, deviations from intended interventions, missing outcome data, measurement of the outcome and selection of the reported results. Each domain item is rated as “low risk”, “high risk” or “some concerns” and can be summarised with an overall risk-of-bias judgment [24]. The ROBINS-I tool was used to assess the risk of bias for seven domains: confounding, selection of participants, classification of interventions, deviations from intended interventions, missing data, measurement of outcomes and selection of the reported results. Each domain item is rated as “low risk”, “moderate risk”, “serious risk”, “critical risk” or “no information” and can be summed up in an overall risk-of-bias judgment, similar to the RoB2 tool. Two researchers (A. Mavronasou and E.A. Kortianou) independently evaluated the quality of all the included studies, and discrepancies were resolved by a third independent researcher (A. Spinou).

Synthesis of results

A narrative synthesis was conducted based on all relevant reported outcomes and the methodological characteristics of all included studies. Additionally, a meta-analysis to evaluate the effects of eHealth exercise programmes on clinical outcomes was performed when data were available from at least two studies, using SPSS for Mac, version 29.0 (SPSS Inc, Chicago, IL, USA). For studies reporting median and range or interquartile range values, the calculated mean±sd values were used based on relevant formulas [26]. Effect sizes were expressed as mean differences (MD) and their 95% confidence intervals (95% CI). Heterogeneity was statistically verified using the I² test; where I²>50% suggests significant heterogeneity. A fixed effect model was used to combine our data [27]. Statistical significance was set at p<0.05.

The overall quality of the evidence was assessed using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach by two independent researchers (A. Mavronasou and E.A. Kortianou) [28, 29]. Discrepancies between the two researchers were resolved by a third researcher (K. Douros). The ranking of the quality of the evidence was based on five criteria: risk of bias, inconsistency, indirectness of study results, imprecision and publication bias. The Summary of Findings tables were generated to summarise the certainty of the evidence among the outcomes, rating the quality of evidence as “high”, “moderate”, “low” or “very low”.

Results

Study selection

In total, 2182 publications were identified across the five databases, of which 358 duplicates were removed. After the title and abstract screening, 1785 studies were excluded. 39 full-text articles were reviewed and 32 articles were excluded along with reasons for exclusion (see PRISMA flowchart in figure 1). Finally, seven studies met the inclusion criteria and were included in the systematic review.

FIGURE 1.

PRISMA flowchart of the included studies.

Study characteristics

The characteristics of the included studies are presented in table 1. All studies were published between 2012 and 2022. Five studies included patients with CF [30–34], one study included patients with NCFB [35] and one study included patients with PCD [36]. The sample size ranged from 10 to 39 participants, with an age range of 6 to 20 years. Participants in the control groups of all the included studies received usual care (e.g., airway clearance techniques, pharmacological treatment), while the intervention groups additionally received the eHealth exercise programmes. Four studies compared the eHealth exercise programme with usual care, two of them in a home-based [31, 33] and two in a centre-based setting [35, 36]. One study was a randomised crossover pilot [30], while two studies did not include a comparator group [32, 34]. The description of the usual care referred to airway clearance techniques (ACTs) performed at home for three studies [33, 35, 36] and a combination of ACTs, inhaled medication and nutritional supplementation for one study [31]. Information regarding the duration of the usual care was lacking in most studies, except for two [35, 36] that reported similar duration to the intervention, i.e. 30 to 40 min per session.

TABLE 1.

Characteristics and results of included studies

Author (year) [ref.]	Study design	Diagnosis	Population (n/age)	Professional responsible for delivering the programme	Intervention venue	Intervention group		Control group		Outcomes of the study
						Type/delivery (synchronous/asynchronous)	Programme duration/number of sessions (duration of session)	Type/delivery (synchronous/asynchronous)	Programme duration/number of sessions (duration of session)	Primary	Secondary
Bingham et al. (2012) [30]	Pilot RCT	CF	n=13/9.3±2.1 years	Respiratory therapist	Home-based	Biofeedback spirometry games in a “game mode” via non-clinical spirometer/asynchronous	3 weeks/21 (15 min)	Breath exercise games in a “control mode” software via non-clinical spirometer/asynchronous	3 weeks/21 (15 min)	Total number of HFE manoeuvres during the 21 sessions: no differences between the two training periods (p=0.52)	%FEV1 change/days used: 0.3±2.4 versus −2.5±5.2, p=0.01 %VC change: 4.3±10.6 versus −2.1±12.6, p=0.05 %VC change/days used: 0.4±1.7 versus −2.6±5.3, p=0.03
Del Corral et al. (2018)# [31]	Single-blind RCT	CF	IG: n=19/12.6±3.4 years CG: n=20/11±3 years	PT	Home-based	Active video game exercise at 70–80% HRmax (running, squats, lunges, bicep curls) via Nintendo Wii-ACTIVE 2 game Plus, usual care/asynchronous	6 weeks/30 (30–60 min)	Usual care (inhaled medication, chest physiotherapy, nutritional supplementation)/NR	NR	Modified SWD change in m: 58.95 versus −19.5, p<0.05	6MWD change in m: 30.95 versus −7.5, p<0.01 HJT change in cm: 9.18 versus 0.65, p<0.01 MBT change in cm: 34.1 versus 3.3, p<0.01 RHG change in kg: 6.5 versus 0.5, p<0.01 LHG change in kg: 6.1 versus −0.4, p<0.01 CFQ – adolescents/adults change in treatment burden (units): 55.56 versus 69.47, p<0.05
Chen et al. (2018) [32]	Single-arm pilot study	CF	n=10/6–20 years¶	Health instructor	Home-based	Supervised videoconferencing exercise (aerobic and plyometric) at moderate (60–75% HRmax)/vigorous intensity (>75% HRmax) via Vsee/synchronous	6 weeks/18 (30 min)	NA	NA	V′O2peak: no difference between pre- and post-exercise programme (p>0.05)	Pulmonary function: no difference between pre- and post-exercise programme (p>0.05)
Kenis-Coskun et al. (2022)+ [33]	Single-blind RCT	CF	IG: n=14/9.8±2.14 years CG: n=14/10±1.64 years	Researchers trained and supervised by PT experts	Home-based	Supervised videoconferencing exercise (HIIT and postural strengthening) via Zoom Plus, ACTs/synchronous	12 weeks/36 (NR)	ACTs/NR	NR	CFQ-R (units): no significant differences between groups (p>0.05) RCADS (units): major depressive disorder: 3.92 versus 7.78, p=0.006; generalised anxiety disorder: 3.42 versus 6.64, p=0.004	6MWD (m): 483.07 versus 405.66, p=0.05 FEV1% pred: no significant differences between groups (p>0.05)
Holmes et al. (2022) [34]	Single-arm pilot study	CF and pancreatic insufficiency	n=10/15.8±2.2 years	Personal trainer	Home-based	Supervised videoconferencing exercise (resistance training) via Zoom/synchronous	12 weeks/36 (max 60 min)	NA	NA	% Fat change: −1.3, p=0.03 FFM change in kg: 1.5, p=0.01 FFMI change in kg·m−2: 0.4, p=0.01	FEV1%: 109.1 versus 103, p=0.02 FVC%: 109.3 versus 104.8, p=0.05 Isokinetic 90°: average power change in watts: 16.2±17.7, p=0.02 V′O2peak change in L·min−1: 0.1±0.1, p=0.01 V′CO2peak change in L·min−1: 0.1±0.1, p=0.015 V′Epeak change in L·min−1: 0.3±6.8, p=0.04
Sonbahar-Ulu et al. (2022)+ [36]	RCT	PCD	IG: n=16/14.37±2.75 years CG: n=16/12.99±3.56 years	PT	Centre-based	Active video game exercise at moderate intensity (3–6 METs) via Xbox-Kinect 360/synchronous Plus ACTs (2 sessions per day)/asynchronous	8 weeks/24 (exercise: 40 min) Plus 112 (2 sessions per day) (30 min)	ACTs/asynchronous	8 weeks/112 (2 sessions per day) (30 min)	6MWD (m): 613.5 versus 503, p<0.001	ISWD (m): 735 versus 710, p<0.001 % FEV1: 81.77 versus 77.82, p=0.001 % FVC: 92 versus 85, p=0.003 % PEF: 78.5 versus 72.5, p=0.003 MIP (cmH2O): 98.68 versus 74.5, p<0.001 MEP (cmH2O): 110.56 versus 92.62, p<0.001 QMS (kg): 26.39 versus 23.78, p=0.006 Glittre ADL test duration (s): 144 versus 220.5, p<0.001 PCD-QoL Version-2 (units): physical functioning: 73.85 versus 43.37, p<0.001 social function: 66.6 versus 46.5, p=0.003 treatment burden: 64.64 versus 59.23, p<0.001 lower respiratory symptoms: 60.01 versus 51.53, p<0.001
Ucgun et al. (2022)+ [35]	RCT	NCFB	IGa: n=13/13±3.29 years IGb: n=13/11.77±2.83 years CG: n=13/ 13.46±3.52 years	PT	Centre-based	IGa: video game exercise (aerobic) via Nintendo Wii Fit/ synchronous Plus H-bCP (2 sessions per day, home-based)/asynchronous IGb: breathing video game exercises via Nintendo Wii Fit/synchronous Plus H-bCP (2 sessions per day, home-based)/asynchronous	8 weeks/16 (VGE: 50 min) Plus 112 (2 sessions per day) (40 min)	H-bCP/asynchronous	8 weeks/112 (2 sessions per day) (40 min)	Pulmonary function: no difference between groups (p>0.05) IGa versus CG MEP: 83.76 versus 65.53, p=0.038 IGb versus CG MIP: 67.61 versus 74.76, p=0.012 MEP: 66 versus 65.53, p=0.046	IGa versus CG QMS: 35.46 versus 32.28, p<0.001 6MWD: 639.07 versus 581.15, p<0.001 PST overall: 0.21 versus 0.25, p=0.035 LOST overall: 51.15 versus 45.38, p<0.001 CTSIB composite score: 1.52 versus 1.72, p=0.029 modified Borg scale: 2.61 versus 1.19, p=0.031 IGb versus CG PST overall: 0.25 versus 0.25, p=0.040 LOST overall: 49.38 versus 45.38, p<0.001 CTSIB composite score: 1.68 versus 1.72, p=0.037 IGa versus IGb QMS: 35.46 versus 27.93, p<0.001 6MWD: 639.07 versus 595.53, p=0.014 modified Borg scale: 2.61 versus 1.15, p=0.009

RCT: randomised controlled trial; CF: cystic fibrosis; HFE: high-flow events; FEV₁: forced expiratory volume in 1 s; VC: vital capacity; IG: intervention group; CG: control group; PT: physiotherapist; HR_max: maximum heart rate; NR: not reported; modified SWD: modified shuttle walk test distance; 6MWD: 6-min walking distance; HJT: horizontal jump test; MBT: medicine ball throw; RHG: right handgrip; LHG: left handgrip; CFQ-R: cystic fibrosis questionnaire-revised; V′_O2peak: peak oxygen consumption; V′_CO2peak: peak carbon dioxide expiration; HIIT: high-intensity interval training; ACTs: airway clearance techniques; RCADS: anxiety and depression scale in children-revised; FFM: fat-free mass; FFMI: fat-free mass index; FVC: forced vital capacity; V′_E: ventilation; PCD: primary ciliary dyskinesia; METs: metabolic equivalent of tasks; ISWD: incremental shuttle walk test distance; PEF: peak expiratory flow; MIP: maximal inspiratory pressure; MEP: maximal expiratory pressure; QMS: quadriceps muscle strength; ADL: activities of daily living; QoL: quality of life; NCFB: non-cystic fibrosis bronchiectasis; H-bCP: home-based chest physiotherapy; VGE: video game-based exercise; LOST: limits of stability test; NA: not applicable; CTSIB: clinical test of sensory integration of balance; PST: postural stability test. ^#: study results are presented based on the intention-to-treat analysis approach; ^¶: study reports the age range (years); ⁺: study results are presented as post-intervention value.

Risk of bias of included studies

Five studies were assessed using the RoB2 and two using the ROBINS-I tool. Based on RoB2 (figure 2), only one study [35] was rated as low risk of bias. All five studies were rated as low risk of bias in the measurement of the outcomes and in the selection of the reported results (figure 2). The nature of the interventions made the blinding of participants or intervention providers impossible, but the outcome measurements for all studies were undertaken by assessors who were blinded to the treatment allocation. The risk of bias due to missing outcome data was rated as low for four studies. Only one study [30] was rated as high risk because it did not have outcome data available from at least 95% of the total number of participants. Concerning the randomisation, most studies did not follow allocation concealment and only one study was rated as low risk of bias [35].

FIGURE 2.

a) Risk of bias summary (RoB2). b) Risk of bias graph (RoB2).

Both studies that were assessed with the ROBINS-I tool were rated as having a serious risk of bias, due to failures in controlling confounders. Detailed presentation of the ROBINS-Ι assessment and the summary for these studies are presented in figure 3.

FIGURE 3.

a) Risk of bias summary (ROBINS-I). b) Risk of bias graph (ROBINS-I).

Types of eHealth

The delivery of the eHealth exercise programmes was through video games, biofeedback games and videoconferencing. Three studies used active video games in an asynchronous [31] or synchronous manner [35, 36] via the consoles Nintendo Wii, Nintendo Wii Fit Plus and Xbox-Kinect 360. One of them was conducted in a home-based environment without live supervision [31], while the other two studies were in a centre-based environment and exercise was supervised by a physiotherapist [35, 36]. Three studies used synchronous eHealth delivery through videoconferencing via the Zoom and Vsee platforms, and with live interaction between patients and healthcare professionals [32–34]. Parents or caregivers supervised the children during the synchronous delivery of the exercise programme without participation in the exercise programme, but this information was not available for the asynchronous delivery modes. One study reported the use of a biofeedback spirometry game through a non-clinical spirometry device, with asynchronous delivery of eHealth [30].

eHealth exercise programmes

The eHealth exercise programme features, i.e., programme venue, duration, frequency, intensity and type of exercise, were variable. Five studies implemented eHealth in a home-based [30–34] and two studies in a centre-based setting [35, 36]. Regarding the duration of the eHealth programme, one study had a 3-week duration [30], two studies had a 6-week [31, 32], two had an 8-week [35, 36] and another two had a 12-week duration [33, 34]. The frequency of the exercise sessions ranged from two to five sessions per week, with each session lasting 15 to 60 min. Moreover, just three studies based the intensity of their exercise intervention on the participant's exercise levels; all targeting a moderate intensity i.e. 60% to 80% of the maximum heart rate [31, 32] or 3 to 6 of metabolic equivalent of tasks (METs) [36].

The eHealth exercise programmes incorporated aerobic and strengthening exercises for the upper and lower extremities using minimal equipment, such as a ball or weight-adjustable dumbbells. Three studies reported an aerobic exercise programme that included basic exercises such as running, volleyball and hula hoop [31, 35, 36], and one study used an exercise programme that simulated “high-intensity interval aerobic training” through a letter game [33]. Two studies [30, 35] included breathing exercises of inspiratory and expiratory manoeuvres and one study reported an exercise programme with aerobic and plyometric exercises, without further information [32].

Three studies [31, 35, 36] had a physiotherapist deliver the programme, another three studies had other professionals delivering the programme under the supervision of a physiotherapist [32–34], and one study reported that a respiratory therapist delivered the programme [30]. Three studies reported monitoring compliance with the eHealth exercise programme [30, 31, 34]. Compliance was monitored through diaries, phone calls or e-mails [31], and a recording of the total number of sessions that children attended was also used [30, 34].

Clinical outcomes

Functional capacity, exercise capacity and balance

Functional capacity was assessed in four studies with the 6-min walk test (6MWT) [31, 33, 35, 36], while exercise capacity was assessed in two studies with the incremental shuttle walk test (ISWT) [36] and modified shuttle walk test (modified SWT) [31]. Two studies performed cardiopulmonary exercise tests to evaluate exercise capacity at baseline and following the eHealth intervention programmes [32, 34].

The studies that used exercise capacity outcomes reported a significant improvement in distance covered after the eHealth intervention programme compared to the control group (table 1) [31, 36]. One study [34] reported significant changes in favour of the intervention for the peak volume of oxygen uptake (V′_O2peak), the peak volume of carbon dioxide produced (V′_CO2peak) and ventilation after an eHealth 12-week resistance programme via videoconferencing. However, another study [32] found no differences in the aforementioned parameters after an eHealth 6-week aerobic and plyometrics exercise programme (table 1). Functional capacity was reported in at least two studies; thus, a meta-analysis was conducted for this outcome. Functional capacity was significantly improved in favour of the intervention group (MD: 37.20, 95% CI 7.91–66.48; I²=0%, p=0.013) (figure 4) [31, 33, 35, 36].

FIGURE 4.

Forest plot for the effects of the eHealth exercise programme on functional capacity. The measurements for the functional capacity used 6-min walk test. Vertical zero line indicates the direction of the effectiveness; dashed line represents the overall meta-analysis mean difference; solid lines indicate confidence interval limit.

Only one study [35] reported balance as an outcome measure. Specifically, the Postural Stability Test (PST), Limits of Stability Test (LOST) and Clinical Test of Sensory Integration of Balance (CTSIB) were assessed with the Biodex Balance System (BBS) before and after the 8-week intervention programme implemented through active video games (AVG) use. eHealth intervention groups, i.e. aerobic AVG and breathing AVG, were significantly improved in static and dynamic postural stability compared to usual care [35].

Pulmonary function

Pulmonary function was evaluated in six studies. Three studies [32, 33, 35] reported no changes in the pulmonary function of the participants in the intervention arm compared to the control group, two studies reported improvements in the per cent predicted of the forced expiratory volume in 1 s (FEV₁), forced vital capacity (FVC), peak expiratory flow rate (PEF) and vital capacity (VC) values following a 3- to 8-week duration eHealth programme [30, 36], and one study [34] reported a decrease in FEV₁ % and FVC % predicted after a 12-week intervention programme.

Peripheral and respiratory muscle strength

The upper and lower limb muscle strength was evaluated in four studies [31, 34–36], and respiratory muscle strength was evaluated in two studies [35, 36]. Peripheral muscle strength measurements used devices such as handheld dynamometer, handgrip and isokinetic dynamometer or simple clinical tests such as horizontal jump test or medicine ball throw. All relevant studies reported an improvement in peripheral muscle strength in the eHealth group compared to the control group [31, 35, 36]. Furthermore, peripheral muscle strength was also improved compared to pre-intervention measurements when the study did not include a control [34].

Significant improvements in the eHealth group compared to the control group were also presented for respiratory muscle strength [35, 36]. Respiratory muscle strength increased significantly in the group that received either the breathing video game [35] or the aerobic video game [35, 36] when compared to the control group (table 1).

Health-related quality of life

Only three studies evaluated HRQoL. One study reported improvements in the eHealth group in six out of seven domains of the Quality of Life questionnaire for Primary Ciliary Dyskinesia (PCD-QOL Version-2) [36]. However, the study by Kenis-Coskun et al. [33] (2022) demonstrated no differences in the Cystic Fibrosis Questionnaire-Revised (CFQ-R) between the eHealth and control groups, while another study reported improvements in the treatment burden domain of the CFQ-R for the control group [31].

Other clinical outcomes

None of the included studies reported on the number of hospitalisations, non-scheduled healthcare visits or mortality rate as an outcome measure.

GRADE evidence: effects of eHealth interventions

Using the GRADE approach, the quality of evidence for clinical outcomes in studies comparing eHealth exercise programmes utilising AVG to control groups was rated as moderate to low confidence. The major reasons for downgrading were the serious risk of bias and the imprecision of the effect size in the included studies. Only one study [33] met the criteria for GRADE evaluation of the eHealth exercise programme delivered via videoconferencing compared to the control group. Therefore, the quality of evidence was rated very low. The GRADE recommendations for the selected outcomes of all studies are presented in supplementary tables S3 and S4.

Discussion

This systematic review presents the effects of the eHealth exercise programmes on children and adolescents with CSLDs, i.e., CF, NCFB and PCD. Our results showed that home-based and centre-based eHealth exercise programmes are delivered synchronously and asynchronously through videoconferencing and active video games. These interventions can improve functional and exercise capacity, balance, and peripheral and respiratory muscle strength, although improvements in pulmonary function and some domains of the HRQoL are unclear.

Limited exercise tolerance in children and adolescents with NCFB has a negative impact on their HRQoL [37]. In this systematic review, the 6MWT, ISWT and modified SWT demonstrated significant improvements following eHealth exercise programmes. Previous evidence in a paediatric population with CF who followed home-based rehabilitation reported similar improvements in functional capacity, attributing them to the combination of aerobic and resistance exercises for the upper and lower extremities [38]. This systematic review reported a mean difference of 37.20 m in the 6MWT after the eHealth exercise programme. Although there are no values for clinically important differences in populations with CSLDs, a recent study reported a minimal detectable change of 53 to 63 m in children and adolescents with CF when performing the test twice [39]. The results of our meta-analysis highlight the need for further optimisation of the eHealth interventions to achieve higher changes in distance covered during the 6MWT than those previously detected [39].

In addition, two studies utilised maximal cardiopulmonary exercise testing using a cycle ergometer [32, 34], and although one study showed no difference in V′_O2peak [32], the other one was in favour of the eHealth programme [34]. The longest duration of the eHealth exercise programme in terms of a single session (60 min versus 30 min) and intervention programme (12 weeks versus 6 weeks) may explain the improvement of the V′_O2peak in the latest study.

Both in childhood and adult life, peripheral muscle strength is lower in patients with CF and NCFB compared with healthy individuals [40, 41]. A recent meta-analysis of children and adolescents with CF concluded that a combination of aerobic and strengthening exercise programmes could increase upper and lower muscle strength, emphasising the importance of exercise training [42]. Our results corroborate the findings of this review, as the included studies performed similar exercise programmes through eHealth technology such as active video games [31, 35, 36] and videoconferencing [34].

Furthermore, aerobic exercise can benefit respiratory muscle strength. Through active video games [35, 36], aerobic exercise seems to increase the maximal inspiratory (MIP) and expiratory pressure (MEP), as it subjects patients to greater respiratory demands. In children and adolescents with CF who undertook aerobic physical activity at least three times per week for 12 weeks, MIP and MEP were higher compared to those who did not participate in any physical activity programme [43].

Pulmonary function is important in respiratory disease progression [44]. Decreases in lung function indices, e.g., FEV₁, are associated with poorer quality of life [45]. Our included studies showed some variances, from significant improvement in FEV₁ after the eHealth [30, 36] to no differences [32, 33, 35] or reduction [34]. It seems that improvements in FEV₁ in children with mild disease severity (FEV₁ % predicted ranging from 74.85 to 85.5) require an eHealth exercise programme over three sessions for 8 weeks.

Children with CF and NCFB who have impaired HRQoL are those who require frequent antibiotics, present more severe symptoms and have lower lung function [3]. Our work showed that children could potentially improve their HRQoL through eHealth interventions that are implemented using remote technology (active video games) [31, 36]. This is likely due to using an enjoyable and satisfying way of exercise for children [18].

An international consensus of a core outcome set for our population suggests that studies should focus on patient-oriented outcomes such as quality of life, disease symptoms, exacerbation frequency, number of hospitalisations, number of non-scheduled healthcare visits and mortality [46]. None of the included studies in this systematic review include the aforementioned parameters as outcome measures except for the HRQoL in studies for CF, and research on eHealth exercise programmes for children with chronic respiratory diseases in the future should incorporate these clinical outcomes. Moreover, most of our included studies incorporated a considerable risk of bias.

Strengths and limitations

With the rapid development of technology throughout the area of rehabilitation, this systematic review investigated both eHealth exercise programmes at home and in community settings, supporting its generalisability. Our findings are based on studies that used synchronous or asynchronous methods for delivering the exercise programmes, such as videoconferencing and active video games, and are relevant to people with CSLDs who frequently present with similar issues in this age group.

However, since our review included non-RCTs and pilot studies, it might have introduced potential biases due to the lack of randomisation and confounding variables. Unlike RCTs, which minimise bias through random allocation, these study designs rely on small sample sizes, nonrandom allocations of participants into groups or one-group interventions, which could influence the validity and generalisability of the findings. Nevertheless, assessing the risk of bias using Cochrane tools such as the RoB2 and ROBINS-I tools and rating the evidence with the GRADE classification system makes our review transparent and reliable. Furthermore, due to the scarcity of publications for the paediatric population with NCFB, our results might be more applicable to those with CF. Future studies should explore the effects of eHealth exercise interventions in RCT for all populations with CSLDs.

Conclusion

Our study supports the theory that eHealth exercise programmes could improve functional capacity, and results are also promising for other clinical outcomes, such as pulmonary function, exercise capacity, balance, peripheral and respiratory muscle strength, and HRQoL, in paediatric patients with CF. This systematic review presents a variety of eHealth settings, including videoconferencing, video games and digital spirometer, mainly in home settings with live supervision. Still, our results were based on moderate-to-very low-quality evidence, demonstrating a need for larger, high-quality RCTs assessing the effectiveness of eHealth. Future work is needed to confirm and expand our results through assessing the effects of eHealth exercise programmes, especially in other paediatric populations with CSLDs.

Supplementary material

Please note: supplementary material is not edited by the Editorial Office, and is uploaded as it has been supplied by the author.

Supplementary material

00977-2024.SUPPLEMENT