Non-ambulatory children with cerebral palsy: effects of four months of static and dynamic standing exercise on passive range of motion and spasticity in the hip

Introduction

Non-ambulatory children with cerebral palsy cannot walk or sit without support, and their gross motor function is classified as level IV and V according to the five level classification system GMFCS-E&R, where level V implies the most severe function limitations (Palisano et al., 2008). Therefore, these children have difficulties being physical active. There are few high-quality exercise intervention studies on children with CP, and those that have been performed have mainly been on ambulatory children with CP (Ryan et al., 2017). Standard care in Sweden for Na-CP includes standing exercise training in standing frames for 45–90 min daily, in accordance with evidence-based recommendations (Paleg, Smith & Glickman, 2013). The standing exercise training in standing frames is a static standing (StS) exercise where the child is fixated in an individually casted frame or in another standing device. No lower body movements can be achieved but standing in an upright position is possible (Fig. 1A). In a feasibility study (Lauruschkus et al., 2017) about how to promote physical activity among children with CP, five children exercised with the motorised medical device Innowalk (Fig. 1B). Innowalk gives an opportunity to experience walking movements in an upright weight-bearing position, making dynamic standing (DyS) possible. The children enjoyed the exercise training and their parents reported improved gastrointestinal function, warmer feet, and improved motor function compared with StS. It seems as if StS and DyS represent two different exercise training modalities for non-ambulatory children with CP, and more knowledge about the two interventions are needed. The physiological effects of training in StS compared to DyS are not known.

Earlier studies on Passive Range Of Motion (PROM) have mainly been on the effects of CP severity and age. McDowell et al. (2012) showed that PROM of abduction, internal rotation and external rotation decreased with higher GMFCSE&R level (McDowell et al., 2012), and Nordmark et al. (2009) showed decrements in abduction and external rotation between the ages of two and fourteen (Nordmark et al., 2009). Fragala, Goodgold & Dumas (2003) performed a multiple case study on exercise showing that exercise improves PROM.

Spasticity has been defined as a velocity-dependent increase in the tonic stretch reflex as a result of hyper excitability of the reflex (Lance, 1980). Even at low velocities, an increase in the activation of the tonic stretch reflex has been seen (Thilmann, Fellows & Garms, 1991; Van den Noort et al., 2010; Bar-On et al., 2013). Additionally, the activation of the reflex continues after the movement has stopped (Sheean, 2002). It has also been reported that the size and number of muscle lengthening are related to the amount of tonic stretch reflex activation (Levin & Feldman, 1994; Malhotra et al., 2008).

To the best of our knowledge, no earlier studies comparing the long-term exercise effects from StS versus DyS on PROM and spasticity in children with CP GMFCS-E&R level and V have been performed.

Therefore, the aim of this study was to elucidate the effects of a four-month exercise regime with StS and a four-month exercise regime with DyS on PROM and spasticity in the hips among non-ambulatory children with CP GMFCS-E&R level IV and V.

Materials & Methods

Population

Twenty-four children were initially included in the study, and 20 children aged between 5 and 7 years (mean 11.6 ± 3.6 years; 9 female) with CP GMFCS-E&R level IV and V completed the study. The participants were recruited through the Child and Youth Habilitation Services in the Skane Region, Sweden. Seventeen of the 20 participants used augmentative and alternative communication. Participant characteristics are shown in Table 1 and CONSORT workflow in Fig. 2.

Table 1:

Characteristics of the participants.

Children (N = 20)	n
CP-subtype
Spastic unilateral	0
Spastic bilateral	14
Dyskinetic	6
Ataxic	0
GMFCS-E&R level
IV	11
V	9
Cognitive levela
No mental retardation	2
Mild mental retardation	5
Moderate to profound mental retardation	13

DOI: 10.7717/peerj.8561/table-1

Notes:

aWHO’s International Classification of Diseases (ICD): ICD-10 codes for mental retardation; the information was reported by parents.

Study protocol

We performed an exercise intervention study with a cross-over design (Fig. 3). Each participant performed four months of either StS or DyS with a wash-out period of a minimum two weeks between the two exercise interventions (Neufer, 1989; Mujika & Padilla, 2000; Coyle et al., 1984). Exercise tests were performed before and after each exercise intervention period.

Results

Four children did not complete the intervention because of illness, surgery or pain (Table 2). None of the dropout reasons were related to the exercise intervention. Dynamic standing was performed on statistically higher number of occasions during the exercise intervention period than StS (Table 3). But the total time did not differ between DyS and StS. There was a higher number of non-exercising days during StS than DyS. The wash-out period between the two exercise interventions was 50.3 (24.5) days for all participants (Minimum 14 days; maximum 96 days). For the group starting with StS the wash-out period was 48.5 (26.7) days, (Minimum 18 days; maximum 96 days) and for the group staring with DyS 52.7 (22.8) days, (Minimum 14 days; maximum 76 days). No statistical difference (p = 0.71) was found between the two groups, tested with an un-paired students t-test.

Statistically significant differences were seen in PROM of the hip before and after exercise testing of StS and DyS analysed with a Friedman test (p = 0.001). Further analysis with paired Wilcoxon Signed Ranks Tests revealed statistical differences between before and after exercise test 1 of StS in abduction (p = 0.033), external rotation (p = 0.002), extension (p = 0.029) and before and after exercise test 2 in extension (p = 0.016) (Table 4). On the other hand, PROM was found to increase after exercise testing of DyS in all directions at both exercise tests (p < 0.001) (Table 4).

When comparing PROM before exercise testing at baseline and after the exercise intervention, statistically significant differences were found between abduction (p = 0.001), Ely’s test (p = 0.020) and external rotation (p = 0.042) at StS and Elys test (p = 0.031) and external rotation (p = 0.038) at DyS (Table 4).

Changes in PROM before and after exercise tests were statistically significant larger during DyS compared to StS in all movement directions (Table 5).

Statistically significant differences were seen in spasticity in muscles around the hip before and after exercise testing of StS and DyS analysed with a Friedman test (Fig. 4). Further analysis with paired Wilcoxon Signed Ranks Tests revealed spasticity to be significantly lower after DyS test 1 (Flexion Before: 2[1:3] vs Flexion After: 1[0:1.75], p < 0.001; Extension Before: 1[0:3] v. Hip Extension After: 0 [0:1], p < 0.001; Hip Adduction Before: 2[1:3.75] vs Adduction After: 1[0:1], p < 0.001) and after Dys test 2 (Flexion Before: 2[1:3] vs. Flexion After: 0.5[0:1], p < 0.001; Extension Before: 1[0:2] vs Extension After: 0[0:1] p < 0.001; Adduction Before 2[1:3] vs Adduction After: 1[0:1.75], p < 0.001) (Fig. 4). After 30 min of StS, the only statistically significant difference was found in Hip flexion (Flexion Before: 1.5[0:3] vs Flexion After: 1.5[1:2], p < 0.040) after test 1. Thirty minutes of DyS lowered the spasticity in the muscles around the hip to a statistically significantly higher degree than 30 min of StS (Flexion Test 1 p = 0.003; Flexion Test 2 p < 0.001; Extension Test 1 p < 0.001; Extension Test 2 p = 0.001; Adduction Test 1 p < 0.001; Adduction Test 2 p < 0.001) (Fig. 4). No statistically significant differences in spasticity were found after either DyS or StS, or the four months of exercise training.

Discussion

We have demonstrated an increased PROM and decreased spasticity in the hip after DyS, whereas almost no statistically significant differences were observed in PROM or spasticity after the Swedish standard care, StS. Dynamic standing was performed at a higher frequency than StS, but the total exercise time during the four months did not differ between DyS and StS on a group level, giving the same dose of exercise, and the two exercise periods were comparable to each other. We decided to have a minimum of two weeks wash-out period since it earlier has been reported that VO _2peak decline significantly among sedentary individuals (Neufer, 1989). Others have reported three weeks to be a valid break point, but these studies have been among elite athletes (Mujika & Padilla, 2000; Coyle et al., 1984). Since fitness level influences the detraining effect (Hyatt et al., 2019), we decided to have two weeks as a minimum.

Spasticity may imply a compensation for muscle weakness for some children with CP. However, spasticity may also contribute to muscle shortening, torsional deformities, hip dislocation and/or scoliosis, which often causes pain and can affect motor control, function and activity (Sheean, 2002; Franzen, Hagglund & Alriksson-Schmidt, 2017). Treatment with Botulinum toxin A in order to reduce spasticity has been used since 1993 (Franzen, Hagglund & Alriksson-Schmidt, 2017). StS is also recommended to reduce spasticity and to increase PROM (Paleg, Smith & Glickman, 2013). When planning this study, we chose, inspired from the experiences of our pilot study (Lauruschkus & Tornberg, 2018), to make all assessments in the children’s habitual environment, to ensure a calm, known and secure environment for the testing, and to enable an increased involvement of the children.

Assessments of PROM with a handheld goniometer have been shown to be reliable for abduction (±2.5°) and internal rotation (±2.4°) with errors less than 5 degrees described as standard errors of the measurement (SEM) (McDowell et al., 2000). Ashton, Pickles & Roll (1978) demonstrated an absolute reliability of goniometer assessment of PROM in the hip made by trained physiotherapists among children with CP to vary between ±2.1–4.2° (abduction ±2.5–2.8°; extension ±2.1–3.5°; externa rotation ± 3.5–4.2°) described as SEM. SEM reported in these studies support that the differences found in our study were lager than the expected measurement error (Table 5). After StS no changes in PROM were found (Table 5). On the other hand, PROM increased after DyS exercise testing in median between 5–10° in all motions except for internal rotation at Test 1 where an increase of 2.5° was found (Table 5), meaning that our findings were larger that SEM earlier reported (McDowell et al., 2000; Ashton, Pickles & Roll, 1978) and thereby not explained by random error. This was also seen when comparing StS with DyS, where the differences varied between 2.5−10° (Table 5) which is larger than SEM (McDowell et al., 2000; Ashton, Pickles & Roll, 1978).

Differences in PROM were seen between before and after exercise testing in both StS and DyS. For StS, an increase in PROM was seen in abduction, external rotation, extension after exercise test 1, and in extension after exercise test 2. For DyS, an increase was seen in all movement directions after both test 1 and test 2. When comparing before and after each exercise training period, increases in PROM were found in abduction, extension and external rotation after StS and extension and external rotation after DyS. Changes in PROM were statistically larger after DyS compared to StS. No adverse events were seen, which is an important finding since clinicians have expressed worries that DyS would decrease PROM. Earlier studies have shown a decrease in PROM with age and GMFCS-E&R level (Nordmark et al., 2009), which makes these current findings of clinical interest.

More objective methods to assess spasticity have been suggested (Bar-On et al., 2014b), but we chose to use the Modified Ashworth Scale (Bohannon & Smith, 1987), since this is widely used clinically and within the National Quality Registry CPUP (http://www.cpup.se). It has been shown that spasticity in children with CP increases up to the age of 5 years, and then decreases over time (Nordmark et al., 2009), and that PROM decreases from 2–14 years of age (Nordmark et al., 2009). This information is important to take into consideration for our study. When PROM is expected to decrease, even small changes may have an impact for the children. Since spasticity can have an impact on PROM an interesting finding in our study was that spasticity decreased in all motion directions after both test sessions of DyS but only in flexion after test 1 of StS. DyS lowered spasticity to a statistically higher degree after each test session compared with StS. No differences in spasticity were found after four months of either DyS or StS exercise training. The effects on spasticity seems to be acute, but not adaptable to the exercise. These findings were somewhat surprising, since spasticity has been reported to be velocity dependent (Falisse et al., 2018). In earlier studies, only a few repetitions of the movement have been performed (Bar-On et al., 2014a). In our study, the movements were repeated for 30 min. One can speculate that the decrease in spasticity in our study is a reflection of fatigue in the systems stimulated by the repeated movements during DyS (Allen, Lamb & Westerblad, 2008). This speculation needs to be tested in future studies.

Limitations of our study were the difficulties of standardising the exercise interventions. The length of each StS session varied because the children use the standard care differently due to individual reasons. Likewise, it was hard to standardise walking speed and exercise time during DyS due to the individual child’s preferences. After analysing the exercise diaries, we saw that the children statistically got the same dose of exercise both during StS and DyS, which made us decide that a comparison between the two interventions was possible. Additionally, the daily delivery of the intervention was performed by parents, school personnel and personal assistants of the children, which might have introduced variation in the delivery. All, parents, personnel and assistants got the same information and education to minimise this. The follow up half time during DyS, also supported the delivery of the intervention. The wide ranges of days in the wash-out period could also be a problem. In the analysis of the number of days within the wash-out period we couldn’t find any statistical differences between the children starting with StS intervention and those starting with the DyS intervention, which made us decide that this most likely didn’t influence our results on a group level. We also found that no child had a shorter period than 14 days, which was our criteria for the wash-out. Indeed, the average child had a wash-out of 50 day, well within the limits that have been suggested for stabilisation on a new activity level (Coyle et al., 1984). Another limitation of our study is that it wasn’t possible to blind the participants to their intervention and to the assessors. Our results have to be interpreted in light of this.

Conclusions

Thirty minutes of dynamic standing increased PROM and decreased spasticity, and four months of DyS increased PROM but did not decrease spasticity among non-ambulatory children with CP on a group level. These results can help develop individualised standing recommendations.

Supplemental Information