Botulinum toxin A injections and occupational therapy in children with unilateral spastic cerebral palsy: a randomized controlled trial

Hand function is central to the performance of many activities in daily life, involving the integration of the sensorimotor, visual, and cognitive systems. A child explores and interacts with their environment from early life and these systems progressively develop throughout childhood.1 Early damage to the brain, as in cerebral palsy (CP), may disturb the development of these skills.2 CP occurs in 2 per 1000 live births in western Sweden.3 More than one-third of the children have unilateral spastic CP (USCP).4

The performance of bimanual activities can be hampered by physical, cognitive, and perceptual factors. In USCP, one hand often has a reduction in the range and speed of movement, strength, coordination, sensation, and muscle tone, which reduces the ability to grasp and manipulate with the impaired hand.5 In the International Classification of Functioning, Disability and Health for Children and Youth (ICF-CY), these factors are considered in the domain of body function/structure.6 However, these prerequisites are not the only determinants of the performance of an activity. The child's ability to participate in life situations is influenced by the demands of the activity. This is captured in the activity/participation domain of the ICF-CY, where the execution of the task is defined as ‘activity’ and the involvement in life situations is ‘participation’.

Occupational therapy is often designed to span all ICF domains. Interventions targeting range of movement (ROM) and bimanual activities are based on the principles of motor learning and cognitive-based motor intervention, and are effective in different settings. Occupational therapy is also used in combination with other interventions.7

One such intervention that is used to change factors in the body function/structure domain is injecting botulinum toxin A (BoNT-A) into selected muscles, resulting in a temporary reduction in spasticity. As the negative impact of spasticity decreases, occupational therapy may be more effective. In a Cochrane Review in 2010, a high level of evidence was found to support the use of BoNT-A as an adjunct to managing the upper limbs in children with spastic CP.7 It concluded that BoNT-A should be accompanied by occupational therapy and that further research was needed to evaluate repeated injections. Only a few studies have investigated the effects of repeated BoNT-A injections in combination with occupational therapy.8, 9 The importance of choosing instruments that are sufficiently responsive, covering all ICF domains, has been pointed out,9 and the present study is designed with this in mind. The concept of the smallest detectable difference (SDD) was also considered in the interpretation of the results.

The aim of this study was to evaluate and compare the effectiveness of repeated BoNT-A plus occupational therapy (BoNT-A/OT) versus occupational therapy alone on hand function in children with USCP. This is the first-year result of a 4-year study.

Method

Interventions

Both groups received occupational therapy, including a splint, and one group was additionally treated with BoNT-A (Fig. 1). BoNT-A injections were carried out 2 weeks before the two blocks of therapy. After the 8-week blocks of intervention, families were instructed to resume their usual therapy, but refrain from other interventions for upper limb.

Botulinum toxin

BoNT-A injections (Botox, Allergan Norden AB, Upplands Väsby, Sweden) were made under general anaesthesia or nitrous oxide sedation after local anaesthetic EMLA cream (AstraZeneca, Södertälje, Sweden) was applied to injection sites. Injections were guided with neuromuscular electric stimulation, using Teflon-coated Botox needles (37mm, 27gauge) diluted to 100 units per millilitre (U/mL). Doses per muscle: biceps brachii 15–30U/mL, brachialis 10–15U/mL, m. brachioradialis 10–15U/mL, pronator teres 10–20U/mL, pronator quadratus 5–20U/mL, adductor pollicis 5–10U/mL, and flexor pollicis brevis 3U/mL. Doses were calculated according to size of the muscle, body weight, and degree of spasticity.10

Occupational therapy home programme

Both groups participated in two 8-week blocks of therapy involving the implementation of a home programme and a weekly session with an occupational therapist. Bimanual training (56h in 8wk) of repetitive task practice and goal-directed activities was conducted combined with manual stretching (once a day) and the use of a static circular splint (8h per night). The splint was individually made, with the elbow flexed, the forearm in supination, and the thumb between radial and volar abduction. When 90° of passive supination was not possible, the splint was adapted 10° less from the child's maximum ROM (Fig. S1, supporting information online). The approach was in accordance with the programme by Novak.11 A logbook was used to document the treatment.

Results

Outcomes

Primary outcome

ICF domain: activity/participation

The BoNT-A/OT group improved compared with the occupational therapy group at 12 months. The AHA score increased by a median of 7.5 AHA units from baseline to 12 months in the BoNT-A group (95% CI 2.5–12.0) and with a median of 0 AHA units in the occupational therapy group (95% CI −3.5 to 8.0) (Fig. 2, Fig. S2a, b [supporting information online], Table 2). There were no missing values. Improvement was found in 6 out of 10 participants by ≥5 AHA units in BoNT-A/OT group and in 1 out of 10 in the occupational therapy group (p<0.03). A 95% CI of the difference in the proportion of improved patients was given as 0.01 to 0.81.

Table 2. Outcome of the Assisting Hand Assessment (AHA) and range of movement (ROM) reported in medians (min–max) in botulinum toxin A plus occupational therapy (BoNT-A/OT) group (n=10) and occupational therapy group (n=10)

Measure	Group	Baseline	1mo	3mo	6mo	7mo	9mo	12mo	Median differencea
Primary outcome
AHA	BoNT-A/OT	51 (16 to 81)	51 (16 to 84)	59 (20 to 83)	57 (22 to 83)	52 (32 to 84)	59 (41 to 84)	61 (28 to 84)	7
	Occupational therapy	55 (39 to 76)		55 (39 to 76)	51 (40 to 76)		57 (42 to 73)	56 (42 to 73)	1
Secondary outcome
ROM
Active supination forearm	BoNT-A/OTb	−10 (−85 to 60)	0 (−85 to 85)	25 (−70 to 85)	10 (−70 to 75)	55 (−50 to 83)	40 (−20 to 90)	33 (−35 to 90)	22
	Occupational therapy	25 (−90 to 85)		55 (−70 to 75)	55 (−45 to 65)		63 (−60 to 72)	48 (−50 to 85)	15
Passive supination forearm	BoNT-A/OT	88 (55 to 90)	90 (50 to 90)	90 (60 to 90)	90 (45 to 90)	90 (75 to 90)	90 (70 to 90)	90 (50 to 90)	0
	Occupational therapy	88 (50 to 90)		90 (60 to 90)	85 (50 to 90)		0 (50 to 90)	90 (70 to 90)	0
Active extension elbow	BoNT-A/OTc	−8 (−30 to 0)	−10 (−28 to 0)	−14 (−25 to 0)	−8 (−22 to 0)	−5 (−20 to 0)	−5 (−50 to 0)	−3 (−15 to 0)	5
	Occupational therapyd	−28 (−75 to −5)		−15 (−33 to −5)	−15 (−45 to 0)		−13 (−40 to 0)	−13 (−35 to 0)	22
Passive extension elbow	BoNT-A/OT	0 (−5 to 0)	0 (−5 to 0)	0 (−10 to 0)	0 (−15 to 0)	0 (−5 to 0)	0 (−15 to 0)	0 (−10 to 0)	0
	Occupational therapy	0 (−25 to 0)		0 (−5 to 0)	−3 (−15 to 0)		0 (−12 to 0)	0 (−12 to 0)	0

• ^aMedian difference of 12mo scores minus baseline. ^bMissing value from one child at baseline, 1, 3, 9, and 12mo. ^c(n=4) and ^d(n=2) missing data at baseline.

The scores for AHA items ‘proceeds’, ‘changes strategy’, ‘reaches’, and ‘releases’ increased at 12 months in the BoNT-A/OT group compared with baseline and, in the occupational therapy group, the score for ‘orients objects’.

Five of the six children in the BoNT-A/OT group who improved ≥5 AHA unit, also improved in active supination and reached 0°, all increasing by 25° or more. In the occupational therapy group, 1 out of 10 improved, increasing by 35°, but did not reach 0° (Fig. S3a, b], supporting information online).

Secondary outcomes

ICF domain: body function/structure

In ROM active supination, both groups improved. At 12 months, 9 out of 9 participants in the BoNT-A/OT group and 6 out of 10 in the occupational therapy group had improved by >10°. The active supination increased by a median difference of 22° in the BoNT-A/OT group and 15° in the occupational therapy group from baseline to 12 months (Fig. 2, Table 2).

In ROM passive supination, at 12 months, 3 out of 10 participants in the BoNT-A/OT group and 1 out of 10 in the occupational therapy group had improved by >10°.

In ROM active extension of the elbow, at 12 months, 1 out of 6 in the BoNT-A/OT group and 6 out of 8 in the occupational therapy group had improved by >10°.

In ROM passive extension of the elbow, at 12 months, 0 out of 10 in the BoNT-A/OT group and 1 out of 10 in the occupational therapy group had improved by >10° (Table 2).

Missing baseline data for active elbow extension: BoNT-A/OT (n=4) and occupational therapy (n=2).

ICF domain: activity/participation

In COPM, both groups had improved after each treatment block at 3 and 9 months (Fig. 2).

In performance, both the BoNT-A/OT group and the occupational therapy group improved. At 3 months, the first goal improved by 7 out of 10 for the BoNT-A/OT group versus 8 out of 10 for the occupational therapy group; and the second goal by 10 out 10 versus 8 out of 10. At 9 months, the first goal improved by 7 out of 10 versus 7 out of 10; and the second by 8 out of 10 versus 7 out of 10.

In satisfaction, both the BoNT-A/OT group and the occupational therapy group improved. At 3 months, the first goal improved by 7 out of 10 for the BoNT-A/OT group versus 7 out of 10 for the occupational therapy group; and the second by 10 out of 10 versus 6 out of 10. At 9 months, the first goal improved by 7 out of 10 versus 6 out of 10 and the second by 8 out of 10 versus 6 out of 10 (Fig. 2).

The most frequent goals, 64 out of 80, prioritised by parents in the COPM were related to self-care, followed by play, mobility, and leisure. The goals were similar for both groups; however, play-related goals were only found in the BoNT-A/OT group (9/40).

Discussion

This population-based randomized controlled study demonstrated that BoNT-A/OT, as well as occupational therapy alone, result in improvement in active ROM and goal performance in young children with USCP. The patients receiving BoNT-A/OT had superior outcomes compared with occupational therapy alone for bimanual performance. Thus, only the combined treatment model resulted in improvements in all ICF domains. The results were analysed to reflect the SDD, the smallest amount of change that can be seen above the threshold of error expected in the measurement. A change larger than the SDD level was considered relevant in the current study. Measuring the minimal clinically important difference is desirable, but also more complicated. The SDD of the AHA19 was established by the author. Active supination was considered to be a change of more than 10°, according to Armstrong et al.20 A difference of two points is considered to be a clinically relevant change for the COPM according to Law et al.18 and Carswell et al.21 This concept provided an opportunity to detect a relevant change in all ICF domains.

Repeated injections combined with training appeared to have a positive effect. Lowe et al. reported similar findings.8 Active supination was increased immediately after the first block of intervention, in both groups, to some extent maintained at the start of block two. Starting at a higher point in the second block of intervention, a maximum effect was seen at 9 months, mostly maintained at 12 months. No similar finding of increased active supination has been reported previously. The use of a night splint, designed to stretch the pronator muscles and used several hours per night may have contributed to the beneficial effect, especially in those who did not receive BoNT-A. A similar result was seen in the AHA in the BoNT-A/OT group in terms of increased bimanual ability. The AHA scores increased in the BoNT-A/OT group and were maintained after 12 months. At 6 months, the scores were higher compared with baseline and, consequently, the participants started at a higher level in the second block. Their score still increased in the last block, indicating increasing performance over time and results lasting longer than the effects of the drug. The child with CP needs time to learn effective strategies finally to reach the autonomous stage.1 It can be speculated that the decrease in muscle tone after BoNT-A provided the window of opportunity the child needed to facilitate the use of the impaired hand and perform bimanual activities of enhanced quality, while, in the occupational therapy group, the absence of improvement was the result of sustained muscle tone. Other studies show unchanged results22 after occupational therapy, but also improvement.23 The study design often differs, making comparisons difficult.9 The majority had a maximum passive ROM at baseline, and a ceiling effect therefore occurred.

Five of the six participants in the BoNT-A/OT group, who maintained their improvements in the AHA score at 12 months, also improved in active supination. This is consistent with the statement by Braendvik et al., suggesting that active supination is a contributory factor in the performance of bimanual activity.24 Six participants in the BoNT-A/OT group and three in the occupational therapy group were unable to actively supinate the forearm to 0° at baseline. Therefore, three more children had the opportunity to reach 0° in the BoNT-A/OT group. This may be one of many reasons explaining improvements in the AHA score in this group.

Both groups improved in individual goal achievement measured by the COPM. Goal-setting was essential in the clinical decision-making.25 The parents prioritized the majority of their goals in the self-care category, consistent with other studies.9 The children were too small to be able to select goals themselves that were sustainable over time.

Goal-directed training increased performance, regardless of the therapy model. Increased ability in a larger number of activities was also enhanced by improving the overall bimanual skills. In the current study, this occurred only when BoNT-A was added.

Limitations of this study were the small sample, the simple randomization, and the partially unblinded design, where participants were aware of their group allocation. Only 20% of the total group with USCP in this geographical area were eligible to be part of the study, according to our detailed knowledge also of those who were not included. The small study size limits generalization of the results. The improvement in active ROM and goal performance in the occupational therapy group is difficult to interpret because of the lack of a non-treated control group; however, this was not considered to be ethical. Moreover, it is important to note that the difference in active supination between the groups at baseline may have been of clinical importance.

To study the effects of early interventions on a long-term basis, the participants will be monitored for 3 years.

Everyday life includes a variety of activities requiring bimanual ability. This small study compared the effectiveness of BoNT-A/OT versus occupational therapy alone for young children with unilateral CP. Both interventions demonstrated improvements in active ROM and goal performance, but BoNT-A, as an adjunct to occupational therapy, demonstrated a superior effect on bimanual performance. BoNT-A/OT may be considered when the objective is to improve bimanual ability in daily life.