Bronchopulmonary dysplasia (BPD) is the most frequent severe respiratory disorder during the neonatal period among premature children.1 In Europe, the incidence is approximately 20% among new born with low birth weight (BW) (<2500g) and 30% in extremely preterm (<28 week).2 BPD incidence is related to gestational age (GA) and low BW. Indeed, in Spain, it affects 53% of children who were born before 28 weeks of GA compared to only 0.3% of children with a GA greater than 30 weeks, and 67% of children with a BW lower than 800g compared to 1% of children with a BW around 1250–1500g.3

Extremely preterm infants and, particularly those with BPD, have frequently impaired lung function in infancy, childhood and adulthood. It increases the risk of respiratory disease. Reduced functional capacity was also demonstrated in BPD preterm children and adolescents compared to children born full-term.1,4–6 These children have an important risk of limited ventilatory reserve and/or a need for adaptation during exercise, both resulting in limitations in daily life.7

Exercises programme are highly recommended in many lung diseases.8,9 Indeed, in chronic obstructive pulmonary disease (COPD), pulmonary rehabilitation including a physical exercise program is highly effective (Evidence 1A)8. In paediatric chronic diseases such as congenital heart diseases or cystic fibrosis, exercise training was also recommended as an important part of the treatment.8 However, the effects of such programme have never been documented in BPD preterm children, despite similar physical limitations. We hypothesized that BPD preterm children would benefit from a physical activity programme.

The aim of this study was to evaluate the effects of a physical activity programme on functional capacity (exercise tolerance and exercise capacity) (main outcome), flexibility and lung function in BPD extremely preterm children.

Fifty-eight subjects (27 females, 31 males) were eligible. Twenty-six subjects fulfilled the inclusion criteria (gestational age=25.9±1.2 week, birth weight=747.2±131.5g). Six children were excluded due to lack of interest (n=2), travel (n=2) and language difficulties (n=2). Twenty children were randomized (6 females, 4 males in IG; 4 females, 6 males in CG) and completed the study (Fig. 1). No adverse events were observed neither during the evaluations nor during the programme.

Fig. 1.

CONSORT 2010 flow diagram.

(0.23MB).

Characteristics of the subjects are presented in Table 1. Lung function was assessed in 10 and only 6 children in CG and IG respectively, due to the impossibility to obtain valid manoeuvres. Both groups were similar at inclusion for all analyzed outcomes (P>.05).

Effects of the intervention are illustrated in Table 2. After the 4-week programme, exercise tolerance (P=.002), exercise capacity (P=.013), flexibility (P=.003) and airways obstruction (P=.004) were significantly improved in the IG. No significant difference was observed for all outcomes between pre- and post-intervention in CG.

Cardiorespiratory response (pulsed oxygen saturation (SpO2) and heart rate (HR)) during 6MWT and ISWT are shown in Table 2. SpO2 did not show clinically relevant changes. HR at the end of ISWT was higher after the programme in IG (P=.016).

Lung function was similar at baseline in both groups. No changes in lung function were observed in CG. However, airways obstruction was significantly improved in IG (P=.004).

The improvement in 6MWT (gain in metres) (P=.017), ISWT (gain in metres) (P=.013) and MSRT (gain in centimetres) (P=.002) were significantly higher in IG than in CG and clinically relevant only in IG at the end of the programme (Table 3). It means that the improvement was higher than minimal clinically important difference (MCID) for these outcomes (15m for 6MWT, 40m for ISWT, 2.2cm for MSRT).

The aim of this study was to evaluate the effect of a physical activity programme on functional exercise capacity (exercise tolerance and exercise capacity) (main outcome), flexibility and lung function in BPD extremely premature children (from 4 to 6 years old). After a 4-week exercise programme, we highlighted a clinically relevant and statistically significant improvement in functional exercise capacity and flexibility. Airway obstruction level was also improved even if it was not clinically relevant.

Functional exercise capacity was considered as the main outcome. The ATS/ERS propose assessing this endpoint by two different validated, inexpensive and safe tests (6MWT and ISWT).12 6MWT has been shown to reflect daily life activities limitation,23 to measure the response to therapeutic interventions and to be strongly correlated to cardiopulmonary exercise test.24 ISWT is incremental and progressive test, stressing the individual symptom limiting maximal performance.18

In adults, it is well know that a learning effect exists in field tests (6MWT and ISWT).12 Then, to decrease this effect a training test has to be performed.14 However, Martins et al. evaluated the repeatability of the 6MWT in children with 30min between both tests and they observed no significant difference in the walked distance between both tests.15 They suggested that there is no learning effect in children. Li demonstrated also a good reproducibility in children with no significant difference in the walked distance when 6MWT was repeated after 18 days.13 This study concerned non-Caucasian children but we postulated that the learning effect (or reproducibility) on walking distance is not related to the ethnic origin. For these reasons, we did not perform a training test.

Our programme improved functional exercise capacity in the recruited children. As previously demonstrated,1,4,5,25,26 BPD children showed lower functional exercise capacity compared to healthy children. Indeed, in our study, walked distance in 6MWT was only 61% and 64% of the predicted values at baseline in IG and CG, respectively.16 Spanish children have one of the most sedentary behaviour in Europe27 and it can contribute to the reduced walked distance. We hypothesize that such a programme could reduce the risk of sedentary and inactivity associated with being born extremely prematurely. After the exercise programme, the distance improved only in IG (+60m). It reinforces this hypothesis. Despite MCID has never been determined for 6MWT in children, this improvement can be considered as clinically relevant. Indeed, it was largely higher than the error related to the repeatability (15m) of this test in children.13 The distance walked during ISWT was also dramatically lower in both groups than in healthy children (45% and 31% of predicted value in CG and IG respectively). IG improved significantly the walked distance (217m) at the end of the programme. This improvement was clinically relevant.

Flexibility is a health-related fitness quality included in several fitness batteries.28 The most used evaluation tools are the SRT and the MSRT. SRT is validated for estimating hamstring extensibility in children.29 However, MSRT protocol decreases the limb length bias establishing a relative zero point for each person. Even if Hoeger et al. demonstrated no significant differences between SRT and MSRT,30 it can be particularly important in paediatric population to perform MSRT due to anthropometric characteristics of children. After our 4-week programme, we observed a significant improvement (+8.3cm) in flexibility in IG (14.5±7.7 vs 22.8±6.9; P=.003). De Miguel-Etayo et al.31 proposed European reference values of pre-pubertal children of fitness measures (6–11 years-old). Comparing their 6-years age group (19.8cm in boys and 21.9cm in girls) with our baseline values, we found a reduced flexibility among BPD children in CG (14.8cm in boys and 12.0cm in girls) and IG (12.1cm in boys and 16.6cm in girls). Mayorga-Vega et al.32 considered 2.2cm as a minimal detectable change value of the SRT. In our study, we found a greater improvement in IG (+6.1cm). As previously demonstrated,33 our results corroborated that extremely low birth weight (ELBW) BPD children have less flexibility compared to term-born children (P=.002).

Cardiorespiratory response during exercise are similar between preterm and full-term children5,34 and between preterm children with and without BPD.4 As expected, in our study, this response at exercise (SpO2 and HR) was normal in all included children.

In general, lung function tends to be reduced in BPD children. Contrary to previous publications,4,26,35 it was normal in both groups at inclusion. It can be explained by the impairment in lung function related to different BPD definitions,36 the BPD severity37 and the choice of Global Lung Function Initiative equation as reference values.22 BPD children tend to present airway obstruction, especially in adolescence and adulthood9,35,38,39 and their lung function may deteriorate at a more rapid rate compared to healthy children.36 For this reason, Rosenfeld40 suggested that spirometry provides a useful longitudinal measurement in young children with BPD. Despite that, there is a paucity of data on preschool spirometry in children with BPD. Spirometric tests can be performed in children from 3 years.22 Despite the difficulty to perform lung function test in preschool children (4–6 years old), only four children from our sample failed to perform successful spirometry manoeuvres in our study. A similar success rate was classically observed in children under 6 years old.41,42 Surprisingly to what it is classically expected after exercises programme, we observed a significant difference in FEV1 in IG (P=.004). However, this improvement is clinically non-relevant (lower than 8% predicted value or 150mL).43 It can be explained by the gain in peripheral and respiratory muscle strength associated to exercise training44 even if we did not evaluate them. Indeed, a relationship between FEV1 and peripheral45 and respiratory46 muscles strength was previously found in cystic fibrosis children.

To the best of our knowledge, no specific exercise programme has been proposed for BPD preterm children. This is the first study evaluating the effects of a short and supervised outpatient exercise programme in these patients. This programme included all the elements classically associated to exercise training programme in other diseases such as endurance (aerobic training), muscular strength training (resistance) and flexibility.8 Each week the intensity of the exercises and duration of the sessions were increased. Programme duration, structure, sessions duration and frequency, and training intensity were extrapolated and adapted from pulmonary rehabilitation programme for COPD and Cystic Fibrosis (CF) patients.8,47,48 We performed 2 sessions per week and the duration of each session varied from 40 to 60min based on ATS/ERS statement requiring a minimum of 30min.8 Our study confirmed the feasibility and the efficacy of such a programme in BPD children. Based on our results, we postulate that this programme is sufficient to obtain a benefit even though longer programme should be also considered in the future.

Some limitations need to be addressed. Even if the sample size was previously determined, extrapolation to larger population should be verified. We investigated only ELBW children with moderate BPD and we cannot extrapolate our results to less premature children. Children with neurocognitive sequels were excluded while they are usually associated with more severe BPD.49 However, it is the first time that a specific programme was proposed to BPD children. The duration of our programme was short. Indeed, the general recommendations consider that exercise programmes should be proposed during 6–12 week and 2–3 times per week.8,48 Nevertheless, similar 4-week programmes were already proposed with success in the literature.48,50,51

We highlighted for the first time that a 4-week exercise programme was beneficial for BPD preterm children. This intervention at early age is feasible, leads to an improved exercise capacity. Indeed, our supervised programme improves exercise tolerance, exercise capacity and flexibility. As for all rehabilitation programmes, long-term benefit must be determined. Further studies will be necessary to evaluate the maintenance of the observed benefits and if children change their behaviour.

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

NM and GR conceptualized and designed the study, drafted the initial manuscript, and approved the final manuscript as submitted.

AP, VM and GC carried out the initial analyses, reviewed and revised the manuscript, and approved the final manuscript as submitted.

No conflicts of interest.