The incidence of cerebral palsy (CP) in India is three per thousand live births.1 The life expectancy of people with CP is shorter than the general population with minimal change over the past few decades.2 Respiratory complications are the most common cause of mortality and morbidity.3 Pneumonia is one of the main reasons for hospital admissions accounting for 40% of all deaths in youth and young adults with CP, highlighting the consequence of poor respiratory function on mortality and quality of life in this population.4

Children and adolescents with CP have poor respiratory function, secondary to the disease process.5 Impaired airway clearance, recurrent aspirations, chest infections, poor cough mechanism, impaired lung function, deformity of the spine and chest, and poor nutrition status influence the respiratory condition of children and adults with CP.6 Furthermore, neuromuscular impairment and poor biomechanics of the thorax contribute to lung damage and low cardiopulmonary function, resulting in reduced lung capacity in these individuals.7 Respiratory muscle strength and aerobic capacity, two inter-related physiologic components, essential to perform energy-efficient walking and functional activities, are markedly lower in children and adolescents with CP than their typically developing peers.8–10 Both lung function and respiratory muscle strength are positively correlated with mobility.9 Children with Gross Motor Function Classification System (GMFCS) level I and II have better respiratory function and mobility than those in GMFCS level III.11,12 The severity of impairment in progressing GMFCS levels entails higher energy expenditure even in seemingly simple functional movements such as sitting, standing, and walking.13 This results in a spiral of respiratory function deconditioning and reduced activity levels. Therefore, respiratory muscle strength training may improve lung function and mobility capacity in children and adolescents with CP.

In physical therapy treatment for CP, there is an often limited emphasis on respiratory muscle training. Past studies14,15 have investigated the effects of respiratory muscle training in children in GMFCS levels I–IV and measured their lung function parameters, but not mobility capacity. Keles et al.16 observed improvements in the six-minute walk test (6MWT) in independently unaided ambulatory children with CP, post inspiratory muscle training (IMT). Another study by Unnithan et al.17 reported beneficial changes in the standing and walking dimensions of gross motor function measures after strengthening and aerobic exercises in children in GMFCS level I-III. However; these two dimensions do not represent mobility capacity.

Considering the physical challenges children and adolescents with CP face in school environments, particularly those with GMFCS levels II-III, a regime consisting of only physical exercises may not be sufficient to improve mobility capacity.18 We, therefore, hypothesised that improving respiratory muscle strength by combined IMT and sensorimotor physical therapy might be a potential option for improving the mobility capacity in this population. The primary objective of the current study was to examine the effect of additional IMT to sensorimotor physical therapy on mobility capacity measured using the 6MWT in school children and adolescents with CP. The secondary objective was to investigate the benefits of IMT on respiratory muscle strength and pulmonary functioning in CP.

The inspiratory muscle trainer was the Threshold® IMT (Respironics Inc, Murrysville, PA, USA) with a flow-independent one-way valve with adjustable pressure setting from 9 to 41 cm H2O. The participants sat comfortably in a chair and had to close their mouth tightly and inhale deeply using the device. They were asked to inspire 10–15 breaths, rest for 30 s, and repeat the cycle. They performed IMT for 15 min, which is the amount of time they could complete the exercise in one session without fatigue (Fig. 1). One other study also recommended training for 15 min and found it beneficial.25 The initial pressure was set at 30% of MIP and gradually increased every week by 5% of initial MIP based on the participant's performance. A pediatric physical therapist with 10 years of clinical experience provided one practice session to the children to familiarize them to the device's use and later conducted the training sessions. Overall, everyone participated in 18 sessions of IMT, three days a week over six weeks.

Fig. 1.

A – Inspiratory muscle trainer; B – Child practices Inspiratory Muscle Training.

(0.23MB).

In addition to the IMT session, children and adolescents in the experimental group had sensorimotor physical therapy for 45 min per session, three times a week for six-weeks, for a total of 18 sessions. Participants in GMFCS level I and II performed the following activities: pushing a play tunnel; picking balls from a bucket on one end, crossing obstacles, and throwing it in a tub on the other end; jumping on a trampoline; climbing stairs; climbing ramps; running in the playground (in a set time). Those in GMFCS level III performed upper limb exercises in sitting (due to poor postural control) and balance and mobility exercises. The sensorimotor activities included picking balls from one end and throwing it in a tub kept at a distance of one meter by crawling, getting into a play tunnel picking balls and returning to the tunnel again. They underwent walking on an even surface, climbing up and down four steps, walking on a ramp, and wheelchair mobility. The exercises were individualized according to the GMFCS levels and the children's ability in keeping with sensorimotor activities (Fig. 2). Participants in the control group received 45 min of sensorimotor physical therapy that comprised similar physical exercises as the experimental group. This group also completed three training sessions a week, totaling 18 sessions over six weeks.

Fig. 2.

Sensorimotor Physical Therapy: A – Child passing through play tunnel; B – Child pushes play tunnel; C – Child jumps on trampoline; D – Child walks over and around obstacle course.

(0.43MB).

Data were analysed using the SPSS 16.0 version. Baseline descriptive statistics were reported, and Shapiro-Wilk was used to test for normality of data distribution. The outcome variables were tested for homogeneity of variance using Levene's test. A repeated measure analysis of variance was performed to test the intervention's effect on the outcome variables over time. The outcome measures collected at three-time points were reported as a within-subjects variable “time”, and the groups were reflected as a between-subjects factor “group”. A significant interaction of “time by group” means that the between-group change seen over time is statistically significant. The post hoc test with Bonferroni adjustment was used to test the differences in the outcomes between-groups at specific time intervals. We conducted the secondary analysis to observe the difference between groups in the primary measure (6MWT) for each of the GMFCS levels (I, II, and III) and the interaction between GMFCS levels and time. The level of significance was set at p < 0.05.

Of the total 337 school children and adolescents screened, 114 were diagnosed with other neurological disorders or no movement dysfunction. Only 45 met the study inclusion criteria. Parents of five children did not consent to participation in the study. Thus, 40 children and adolescents with CP were included in the study. All participants were available for the two months follow-up except one child in the control group. We used the last observed value carried forward method for imputing the missing data for that participant. Fig. 3 shows the CONSORT diagram. The baseline demographics and outcome variables between both groups were similar, except for GMFCS levels (Table 1). There were 12 participants in the control group and five in the experimental group belonging to GMFCS level I. In the GMFCS level II, five individuals were in the experimental intervention group and four underwent sensorimotor training. There were ten and four individuals with CP and GMFCS level III in the experimental and control groups, respectively. Of the 20 participants in the experimental group, two had a diagnosis of unilateral spastic CP (GMFCS I, n = 2), 10 had bilateral spastic CP (GMFCS II, n = 2; GMFCS III, n = 8), three individuals showed clinical signs of spasticity in all four limbs (GMFCS I, n = 3) and five participants belonged to the categories of dyskinesia (GMFCS II, n = 2; GMFCS III, n = 2) and ataxia (GMFCS II, n = 1). In the control group (n = 20), there were seven children and adolescents who had a diagnosis of unilateral spastic CP (GMFCS I, n = 7), five individuals with bilateral spastic CP (GMFCS II, n = 1; GMFCS III, n = 4), six had spasticity features in all four extremities (GMFCS I, n = 5; GMFCS II, n = 1), and dyskinesia and ataxic signs were present in two adolescents (GMFCS II, n = 2).

Fig. 3.

CONSORT diagram.

(0.75MB).

There was no between groups difference in performance on the 6MWT at post-training and follow-up. Only performance on the PEF showed substantial between group difference post-training.

The overall distance covered by study participants in the present trial were less for both the control and experimental groups than distances reported by previous studies.12,20,21,26 On interviewing parents and teachers about their children's mobility capacity, regardless of their GMFCS levels, most of them walked less owing to their family and environmental constraints. Impaired biomechanics, movement dysfunction, and high energy requirement might have potentially contributed to reduced walking distance. In the sub-group analysis of GMFCS, all three strata showed statistical insignificance due to heterogeneous diagnostic groups and wide CI in the 6MWT. The large CIs reflect the reality of diversity of diagnostic classification of people with CP. Sensorimotor physical therapy did not improve the walking distance. To date, there are no published data on the minimal detectable change for the 6MWT in individuals with GMFCS level III. When considering individuals in GMFCS level III, the increase in walking distance for the experimental group (n = 10, MD: 19.8 m) and control group (n = 4, MD: 23 m) is encouraging, warranting future studies with a longer duration of training intensity.

The improvement in MIP post-training and at two months follow-up might be related to practice and carry-over benefits of IMT. The mean difference of 9.6 cm H2O in MIP from baseline to 2-month follow-up for the experimental group is encouraging and clinically useful improvements. The substantial changes across time in MIP values of the current study is consistent with results in previous trials.14-16,27 The increased residual volume may be due to neuromotor dysfunction, affecting the respiratory muscles to force the thorax back to normal size voluntarily.7,9 Inspiring against set resistance might have resulted in increased lung expansion, causing the chest to push air out at higher flow rates, thereby increased PEF. The between-group change of 19.4 points in the% of predicted PEF variable from baseline to 6 weeks supports the experimental intervention.

There was mixed result in MEP and lung function test with no discernable pattern in the values of FEV1 and FVC. As we had to accommodate the training session within the school timetable, the frequency of training was shorter than the other studies that showed significant changes in the FEV1 and FVC.14,15,25 This study is consistent with the results by Keles et al.16 who observed no significant difference in pulmonary function following 42 sessions of IMT. During the study, we noticed that the participants were not wholly engaged in the IMT and found it monotonous, thereby taking frequent breaks during the training, requiring constant encouragement to continue. The problems such as dyskinesia, drooling, oromotor tone variations and difficulty holding onto the mouthpiece might have limited their ability to execute an optimal inspiration, leading to sub-optimal performance of the IMT. In addition to oromotor issues, the children and adolescents with CP had difficulty performing the PFT and respiratory muscle strength test correctly and required several attempts. Despite this, there was high variability in results and we had to consider the best of the readings, therefore potentially affecting the outcomes. Considering the challenges we encountered, a couple of questions arises as to whether PFT and respiratory muscle strength tests in this population reflect their actual values? Do we need to measure these if other accessible and functional tests are available? If improvement in mobility-related tasks is the primary focus, can other more fun and visual feedback interventions with less strain on the already compromised musculoskeletal system be designed?

This study's major limitation is the heterogeneity of individuals with CP and their GMFCS levels across both groups. Children and adolescents in GMFCS level III have more pronounced restrictions in mobility than those in levels I and II, resulting in a variety in the study sample. We would have prevented this by implementing stratified randomization based on GMFCS levels to create study arms instead of simple randomization. However, due to recruitment challenges and time constraints, we included participants from all three GMFCS levels. The participants from two special-schools may not represent the total population. The intervention duration could have been longer, but we had to complete it within the available school timetable. The sub-optimal performance of IMT might be one of the potential drawbacks as the children did not enjoy the sessions and required constant encouragement and supervision.

The study concludes that adding IMT to sensorimotor physical therapy did not impact mobility as measured with the 6MWT. The IMT did not change pulmonary function.