This randomized controlled trial (RCT) was conducted in a tertiary-level pediatric teaching hospital and its associated rehabilitation centre from July, 2014 to July, 2016 after obtaining approval from the institutional ethics committee.

Children between 5 and 12 years of age, clinically diagnosed with spastic diplegic CP with Gross Motor Function Classification System (GMFCS) Score II/ III were eligible. Spastic diplegic CP was defined as ‘CP with predominant bilateral lower limb involvement with hyperreflexia, spasticity and relative non-involvement of upper limbs’. Enrolled children required a minimum visual acuity of 6/60 and the ability to follow single-step commands. They were enrolled if their parents/primary caregivers were willing and capable of following instructions and maintaining an activity log. Informed consent from the parent(s)/primary caregiver(s) and assent (as and when applicable, from a child) were obtained before enrolment. Children with fixed lower-limb contractures or deformities affecting stance and gait, chronic systemic/acute illnesses interfering with physiotherapy and children who received botulinum toxin injection or underwent corrective orthopedic surgery up to one year prior to the day of screening were excluded.

A detailed general and systemic examination of every child was performed and findings recorded on proforma. Each child was scored for 6-minute-walk-test (6MWT) (in meters), 10-metre-fast-walk-test (10MFWT)(in seconds), modified Ashworth scale (MAS), modified Tardieu scale (MTS), Gross Motor Function Classification System (GMFCS), Gross Motor Function Measure-88 (GMFM-88) (D & E) and Cerebral Palsy Quality of Life (CP-QoL) (Primary-caregiver).

Randomization was done by another person who was otherwise not involved in any other aspect of the trial. Block randomization of varying sizes was prepared with an open access randomization software (www. randomizer.org). Allocation concealment was achieved using sealed opaque envelopes. Envelopes were kept in custody of a person not involved in the study.

Scoring for 6MWT: Test arena comprised of a marked, flat, non-slippery, rectangular cemented area with a perimeter of 34 meters. Children were made to walk along the outer boundary of the area after an initial demonstration. If at any point, the child had difficulty walking (due to any reason) or did not want to walk, he or she could stop. In that case, the test would be postponed to a later date. Practice trial immediately preceding test was avoided to reduce fatigue [8]. The parents/primary caregivers were instructed to encourage the child throughout the process. In case a child needed one-hand support, the parent/primary caregiver walked alongside the child holding his/her hand. Distance covered in 6 minutes was recorded in meters (up to nearest centimeter). The test was videographed for future reference.

Scoring for 10MFWT: Test area comprised of a cemented, flat, non-slippery, rectangular floor. Start and finish lines were marked with parallel white lines 10m apart. An explanation and demonstration preceded the test. To start, the child was made to stand with toes touching the start-line. On being told to start, the child had to start walking. The stop command was given after the child had walked 5m past the finish line so that he/she may not decelerate until after reaching the 10 m mark [9]. Time to walk 10m was recorded with a stopwatch with the least count of 0.01 seconds. The test was video graphed for future reference.

Eight activities were recommended in the ‘Intervention Arm,’ viz., walking on plain surface, standing from squatting position and squatting from standing position on level floor, climbing up and down a flight of stairs, kicking a football while standing, dancing on level floor, riding a tricycle/bicycle (with additional support wheels) depending on child’s age and functional capability. Every child was expected to perform at least seven out of eight activities (exception being cycling) to the best of his/her ability. Parents/primary caregivers were given a brief overview of the interventions in the language they understood. They were taught how to keep the activities interesting by incorporating them within play activities. Activities were tailored according to the age and functional capacity of the child and modified periodically to avoid monotony. Strict compliance was emphasized upon. At the outset, videographic demonstrations of the activities were displayed to the parents. Videos were prepared in the same centre previously and approved by all the researchers involved in the study. The parents/primary caregivers were asked to make their children perform the activities under the researcher’s supervision. Doubts were clarified and modifications suggested. A logbook was issued to each parent/primary caregiver with simple instructions written in English and Hindi and sketches depicting the activities. The logbook had an earmarked space for every day ,in which the parent/primary caregiver was instructed to record whether the pre-assigned activity was performed as advised or not. There was a ‘remarks’ column for noting any deviations.The logbook was designed to serve as the main mechanism to ensure compliance. At the completion of the study, percentage compliance was calculated from the entries in the logbooks. In addition, home videos taken by parents acted as a reinforcement of compliance. The activities were designed to be repeated three times per session, with three sessions per day for five days per week. In case of any acute illness, pain, injury, compelling domestic issues or in the case of a child strongly resenting therapy, a break could be allowed but the parent/primary caregiver was supposed to record it in the logbook. These activities were in addition to activities that were already being performed as a part of an ongoing physiotherapy program. Parents/primary caregivers of children in the Control arm were advised to continue ongoing physiotherapy i.e. certain sets of active and passive exercises that were administered by certified physiotherapists after individualized assessment as per guidelines. Three physiotherapists offered conventional physiotherapy in the study centre. These manoeuvres were tailored to the needs of a child with periodic reassessment and modification as per the disability, tolerability of intervention and clinical response (spasticity, stability of gait, etc). Techniques of conventional physiotherapy are described in Web Box I.

Besides filling the logbook, the parents/primary caregivers also made intervention home videos periodically, which were perused by one investigator at follow-up. Telephonic follow-up was done at 2 weeks of enrolment. The telephone number of the primary investigator was shared with parents/primary caregivers for addressing queries. In addition, parents who expressed difficulty in understanding the procedures were contacted telephonically to reinforce the prescribed procedures. Follow-up visits were planned at 3 months (+ 1 week), 6 months and (+ 2 weeks) after enrolment. Doses of drugs (such as baclofen or antiepileptics) were not modified for study purposes. If any child developed seizures during the course of study, he/she was managed as per standard protocol.

The primary study outcome was change in 6 MWT score at 6 months. The secondary outcomes included change in 10MFWT score (in seconds), MAS, MTS, GMFM-88 (D & E), CP-QoL (Primary caregiver version) scores at 3 and 6 months, and change in 6MWT score at 3 months. A single investigator assessed all the outcomes.

Two hundred and sixty-two children with CP were screened, among whom 169 met inclusion criteria and 59 were enrolled (Fig. 1). Thirty children were randomized to the Intervention arm and 29 to the Control arm. One child allocated to the control arm withdrew from the study.

Fig. 1 CONSORT flow diagram for the study.

The mean (SD) ages of children in the study was 115 (23.1) months. Anthropometric and clinical character-istics of the two groups were comparable at baseline (Table I). Five children (intervention arm: 3, control arm: 2) had fresh onset seizures during the study period in all of which oral sodium valproate was initiated. Good seizure control was achieved in four of them with monotherapy of sodium valproate (dose range 15-40mg/kg/day) while one child (intervention arm) required additional therapy with oral clobazam. Except for a brief interruption of their respective intervention schedules, there was no long-term interruption of the rehabilitation program in any of these children. The majority of children in both groups achieved a compliance rate of 80-100%, There were no differences in the compliance rates between the children in the two groups, as derived from logbooks filled by the primary caregivers, with 93.3% and 93.1% children in intervention and control group, respectively having rates >60%.