Secondary Logo

Journal Logo


Targeted Training in Managing Children With Poor Trunk Control: 4 Case Reports

Pin, Tamis W. PhD; Butler, Penelope B. PhD; Shum, Sandra Lai-Fong Prof Dip Physiotherapy

Author Information
doi: 10.1097/PEP.0000000000000499
  • Free


Trunk control is commonly delayed in children with motor disorders.1 Poor trunk control in children with cerebral palsy (CP) has been shown to contribute to limited functional skills such as sitting and walking and the ability to transfer.2,3 There is evidence that postural control of the trunk is essential to achieve independent sitting4 and reaching in sitting5 for infants developing typically and gross motor function in children with CP.2,6 A recent systematic review of strategies to promote trunk control in children with CP concluded that 5 clinical interventions had a moderate effect, including gross motor task training, hippotherapy, treadmill training without body support, specific trunk exercises with vibration therapy, and reactive balance training.7 However, all of these interventions consider the trunk as 1 unit, ignoring that the trunk consists of multiple segments.8 Good control of these multiple segments in the neutral vertical posture as in sitting and standing is essential to enable movement without loss of stability9 and is a fundamental challenge for children with motor disorders.

Therapists often begin trunk control training for children with motor disorders in a lying position where the control demands are reduced compared with the upright vertical posture, which requires simultaneous control of all trunk joints.10 The sensory and musculoskeletal demands and feedforward/feedback systems are also different in lying from the upright posture.9 Targeted training (TT)9 addresses these issues by training control in the neutral vertical posture from the outset. For children with poor head and trunk control, the Segmental Assessment of Trunk Control (SATCo)11 identifies the topmost segment at which control is no longer assured and this segment is “targeted” for control learning. Testing begins at the head and works downward (caudally), testing static, active, and reactive control at 7 defined segments (details below). This simplifies control learning to the few joints of the segment since the segments above have demonstrated control while those below are supported. Once voluntary control has been gained at the targeted segment, the support is moved caudally to the next segment to release the new “targeted” segment. Thus, trunk control is gained sequentially while eliminating the complexity of trying to control too many joints at one time.9 TT also uses the concept of “controlled open kinetic chains” in which the targeted segment is free of support. This support can be internal such as collapse of spinal joints by resting on the joint end stops (eg, extreme lumbar lordosis), or external support provided by the child's hands/arms resting down or from a seat back.8

TT generally uses specially designed equipment to stabilize the child's body below the targeted segment.9 These devices resemble a standard care stander or a saddle seat with trunk support but differ in ensuring firm encircling support, the height of which can be adjusted as control is gained. Once static control is present, the child's own movement encourages active control at the targeted segment. There is also a rocker base option to train reactive control. The device is generally used for 30 minutes, 5 to 7 days a week.12 However, manual support can also be used if the child is relatively small and can be combined with modification of “regular” baby seats as appropriate. Preliminary studies of TT using the specialized equipment showed promising results in improving sitting balance and walking performance in school-aged children with CP.12,13 However, no study has exclusively investigated the use of TT therapy without the specialized equipment on preschool-aged children with poor trunk control.

The present case reports examined whether TT therapy was feasible with children younger than 2 years with poor trunk control without the specialized equipment and change in trunk control and gross motor skills, with the addition of TT to a child's standard care physical therapy (PT) program.


A convenience sample of 4 young children, who were referred by their physical therapists or parents, was recruited. The inclusion criteria were children younger than 24 months, corrected for preterm infants; diagnosed with gross motor delay by their doctor or pediatrician; and whose parent or physical therapist had concern about the child's trunk control. Children with known congenital limb abnormalities or chromosome syndromes were excluded. Ethics were approved by The Hong Kong Polytechnic University. All parents signed an informed consent prior to the study.

Outcome Measures

The gross motor abilities of the study children were tested by the first author using either the Alberta Infant Motor Scale (AIMS)14 or the Gross Motor Function Measure (GMFM 88)15 according to the published criteria. The AIMS is a standardized assessment of developmental milestones in infants from birth until the age of 18 months, or before independent walking and was used for the 3 children with developmental delay.14 The first author's reliability in using the AIMS has been previously established.16 The GMFM 88 is a standardized assessment examining gross motor function of children with CP in 5 dimensions—“A, lying and rolling,” “B, sitting,” “C, crawling and kneeling,” “D, standing,” and “E, walking, running, and jumping”—and was used for the child with CP.15

The SATCo was used, according to the published criteria, to assess the children's seated trunk control.11 Control was examined sequentially on each child giving firm manual support directly beneath each trunk segment in turn to determine control of the head, the upper thoracic, midthoracic and lower thoracic segments, the upper lumbar and lower lumbar segments, and, finally, no support to assess full trunk control in sitting. Trunk control at each segment was tested under 3 different conditions: maintaining the posture (static control), during voluntary head movements (active control), and recovery of the neutral posture after a disturbance of balance by a nudge (reactive control).11 The intra- and interrater reliability of the SATCo between the first and second authors has been established.17 Video recording was made of the outcome measures at start and at each 2-weekly review session for a period of 6 months.

Case Descriptions

Child 1. This girl was born prematurely (24.9 weeks' gestation, birth weight [BW] 700 g). Cranial ultrasound confirmed bilateral intraventricular hemorrhage grade 2. Child 1 was discharged home 3.5 months after birth and diagnosed with global developmental delay and very low tone. She received ongoing PT from 4 months' corrected age (CA), focused on attaining major motor milestones including lying prone propped on forearms with head lifted to 90° and rolling. Her parents' main concerns were slow weight gain, not using her hands for play, and slow motor development. Child 1 was enrolled in the study at 13 months 3 weeks' CA. At start, Child 1 could prop on her forearms in the prone position and roll from the supine to prone positions with some assistance. If placed, she could sit with both hands supporting in front (Figure 1A) and stood if supported around her chest.

Fig. 1.
Fig. 1.:
Functional change in infants over the course of TT. Child 1 sat with both hands supporting in the initial assessment, with thoracic spine in extension but lumbar spine in flexion. After 3 months of TT, she could sit without hand support and started to use her hands in play. Both thoracic and lumbar regions were in extension. Both figures are of consistent posture that lasted for a minimum of 5 seconds. Child 2 initially pulled to stand from the floor mainly using his upper limbs with supervision but no assistance. After 2.5 months of TT, he could pull to stand independently via half-kneeling. Both figures are from video recordings of 4 seconds. Child 3 initially crept on his tummy mainly using his right arm and leg. After 6 months of TT, he took weight on his left arm in a 4-point position and was able to move back to a heel-sit position. Both figures are of a consistent posture that lasted for a minimum of 3 seconds. Child 4 initially used her upper limbs to creep with her right lower limb in a total extension pattern. After 2 months of TT, she could achieve a 4-point position from the prone position independently and could maintain in the position. Both figures are of consistent posture that lasted for a minimum of 2 seconds. TT indicates targeted training.

Child 2. This boy was born late preterm (36 weeks' gestation, BW 2580 g) by emergency cesarean delivery due to fetal bradycardia. He was diagnosed with global developmental delay with hypotonia and dysphagia at 3 months' CA. Since discharged home at 5.5 months' CA, Child 2 received weekly PT focused on transfer in and out of standing and stepping with support in standing. His parents' main concerns were poor feeding and slow motor development. Child 2 was enrolled in the study at 19 months' CA. At start, Child 2 could crawl on hands and knees but with increased lumbar lordosis and could transfer between sitting and 4-point positions if strongly encouraged. He could pull to stand against furniture but could not sidestep or return to the floor independently (Figure 1A).

Child 3. This boy was born preterm (28.6 weeks' gestation, BW 1080 g). He had an intraventricular hemorrhage grade 4 right and grade 3 left side and a ventriculoperitoneal shunt was inserted for the resulting hydrocephalus. Child 3 was discharged home at 3 months' CA and diagnosed with left hemiplegia CP at 10 months' CA, Gross Motor Function Classification Scale level II.18 Since discharge, Child 3 received ongoing PT and occupational therapy (OT) to increase use of his affected limbs and to attain motor milestones including weight bearing on hands and knees, independent transfer to sitting from lying, and supported stepping. Child 3 was enrolled in the study at 16 months' CA. At start, the resting posture of Child 3's left arm was by the side of his body, elbow in 90° flexion, forearm fully pronated, and hand fisted due to increased tone over his left forearm and wrist (all grade 1 to 1+ on the modified Ashworth Scale)19 (Figure 1A). Functionally, Child 3 could forearm-prop in the prone position and pivot on his tummy (Figure 1A). He could commando-crawl or bottom hitch for a short distance if strongly encouraged and roll from the supine to prone positions mainly over his left side and vice versa. When placed, Child 3 sat independently and could get to the floor using his right arm but with both legs passive. He took weight through both feet in supported standing but used only his right upper limb for play.

Child 4. This girl was born prematurely (26.9 weeks' gestation, BW 950 g) by emergency cesarean delivery due to vaginal bleeding. Child 4 was discharged home at 1 month's CA and received ongoing PT from 2 months' CA, focused on rolling and sitting independently. Child 4 was diagnosed with global developmental delay and enrolled in the study at 16 months' CA. At start, Child 4 could pivot in the prone position and creep using mainly her upper limbs but with her right lower limb in total extensor pattern (Figure 1A). Child 4 could maintain sitting with both hands on the floor if placed but lost her balance as soon as she attempted to move. At 16 months, Child 4 increasingly showed dyskinetic-type movement patterns with uncontrolled trunk movements and total extensor pattern of her left leg, but her diagnosis remained global developmental delay.

Intervention-Targeted Training

With use of the SATCo we determined the trunk segment targeted for control training and enabled immediate recommendations for home TT therapy. Parents were shown how to provide firm manual support directly beneath the targeted segment with their child in a sitting position, in effect replicating that part of the SATCo. They were shown how to promote static, active, and reactive trunk control of their child through play while this manual support continued. Activities included remaining still (static), stretching to reach and play with a toy (active), and being bounced on the parent's knees (reactive). Parents were also shown ways to use existing “regular” infant/child seating and/or upright standing frame (if any) to enhance trunk control by ensuring a firm encircling support just beneath the targeted segment (eg, using a rolled towel within a Bumbo encircling type seat). Parents were encouraged to provide TT therapy to their child for a minimum of 20 minutes per session per day, in addition to the use of any “modified” regular child seat or standing frame. The latter lasted a minimum of 30 minutes per session per day. All parents responded readily to these requests with no difficulties in carrying out the TT program. All children continued their usual therapy intervention throughout the study. As the identification of the targeted segment for training was crucial for the TT program, the targeted segment and training recommendations were confirmed from video recordings by the second author within a week of each visit. Program changes were made at each home visit as required for each child during the 6-month intervention period.


All parents reported that they regularly carried out the TT program at home, although they were not asked to keep a log. All 4 children showed improvement in their outcome measures of trunk control (SATCo) and in their functional abilities (AIMS/GMFM) with their combined standard care and additional TT program (Table). Specific functional changes noted were shown in the next subsections.

TABLE - Changes in Trunk Control and Functional Skills With Targeted Traininga
Child 1 Child 2 Child 3 Child 4
Age, mo Age, mo Age, mo Age, mo
Static control Start 13 Upper lumbar 19 Gained full trunk 16 Gained full trunk 16 Lower thoracic
End 16 Full trunk 21.5 Gained full trunk 18 Gained full trunk 18 Upper lumbar
Active control Start 13 Upper lumbar 19 Lower lumbar 16 Gained full trunk 16 Lower thoracic
End 16 Full trunk 21.5 Gained full trunk 18 Gained full trunk 18 Upper lumbar
Reactive control Start 13 Upper lumbar 19 Upper lumbar 16 Learning full trunk 16 Lower thoracic
End 16 Lower lumbar 21.5 Gained full trunk 18 Gained full trunk 18 Lower thoracic
Prone subscore (max score = 21) Start 13 7 19 18 NA 16 11
End 16 7 24 20 18 13
Supine subscore (max score = 9) Start 13 5 19 9 NA 16 8
End 16 8 24 9 18 8
Sit subscore (max score = 12) Start 13 5 19 11 NA 16 5
End 16 6 24 12 18 5
Stand subscore (max score = 16) Start 13 3 19 6 NA 16 3
End 16 3 24 10 18 3
AIMS total score (max score = 58) Start 13 20 19 44 NA 16 27
End 16 24 24 51 NA 18 29
Dim A: Lying and rolling Start 16 53%
End 20 67%
Dim B: Sitting Start 16 42%
End 20 55%
Dim C: Crawling and kneeling Start 16 7%
End 20 24%
Dim D: Standing Start 16 0%
End 20 15%
Dim E: Walking, running, jumping Start 16 0%
End 20 11%
aAll reported scores were raw scores. Trunk segments indicated are those where the infants were learning control. All ages are corrected for prematurity.
Abbreviations: AIMS, Alberta Infant Motor Scale; GMFM, Gross Motor Function Measure; NA, not applicable; SATCo, Segmental Assessment of Trunk Control.

Child 1

After 3 months of TT, Child 1 was able to roll from the supine to prone positions unaided and maintain sitting without hand support, enabling play with toys without loss of balance (Figure 1B).

Child 2

After 2.5 months of TT, a major difference was seen, with Child 2 moving easily in and out of sitting and sitting in a variety of postures. He could pull to stand through half-kneeling and started to sidestep (Figure 1B). By 5 months, Child 2 was cruising along furniture with trunk rotation and could walk for a few steps pushing a cart.

Child 3

At start, this child used his right upper limb almost exclusively. It was found that firm support at the lower thoracic segment resulted in increased left arm elevation up to 70°, contrasting with little/no elevation when support was given at the lumber segments (Figure 2A). Child 3's parents thus used this strategy while encouraging bilateral hand play and active left arm movements, such as reaching overhead. After 2 months of additional TT, Child 3 could creep, showing some activity in the left arm and the leg. He could rise from the supine through left side-lying positions to sitting. The greatest changes after 4 months of TT were Child 3's ability to maintain a 4-point position, to stand against furniture using support of both hands, and to sidestep along the furniture. At 6 months, Child 3 could transfer to a 4-point position from heel-sitting and maintain a 4-point position by taking weight through his left upper limb with normal trunk alignment (Figure 1B).

Fig. 2.
Fig. 2.:
The effect of the segmental support level on trunk posture and arm function. (A) (i) Child 3 initially showed little use or awareness of his left arm although he had gained full control of his trunk. (ii) With firm manual support at the upper lumbar segment, he showed he was aware of and could move his left hand a little. (iii) When this firm support was raised to a more appropriate segment for this child (ie, lower thoracic), Child 3 instantly and repeatedly elevated his left arm on request. His parents used this support to train his arm while also working to consolidate his full trunk control. (iv) After 4 months of TT, Child 3 elevated his left arm with ease on request and maintained a neutral vertical trunk posture during arm elevation without any manual support. Figures i to iii are of one frame from a video recording of the same therapy session. (B) (i) After 4 months of TT, Child 4's primary physical therapist supplied a box seat: this stabilized thighs and feet by straps but no trunk support. At that stage, Child 4 was learning to control her trunk at the lower thoracic level. She thus used the strategy of collapse into side flexion to the spinal joint end stops to use this passive bony/ligamentous control to compensate for the lack of active control through her trunk. (ii) When firm support was provided at the upper lumber level, there was near-full correction of trunk posture and some ability to use hands for play. This was the segment supported for training trunk posture with activities to fully correct the alignment. (iii) With firm segmental support given at a more appropriate segment to promote hand function for this child (ie, lower thoracic), Child 4's trunk posture above the support assumed neutral vertical. She was momentarily able to maintain this neutral vertical posture and use both hands for play (active control). All figures are of one frame from a video recording from the same therapy session. TT indicates targeted training.

Child 4

After 2 months of TT, Child 4 showed marked improvement by pulling into and maintaining a 4-point position although with abducted hips (Figure 1B).


In TT therapy, the trunk consists of different segments and intervention should identify the appropriate segment for targeted learning of control of the neutral vertical posture. In this case report, 4 children under 2 years of CA added TT to their standard care PT program but without the use of specialized TT equipment. There were no difficulties with the program from a parental perspective demonstrating that TT is feasible in this age group and in this way.

At the end of 6 months, the children showed improvement in their trunk control (SATCo scores, the Table) with concurrent improvement in their gross motor skills (AIMS/GMFM, Table). Functional improvements were seen in all children within 3 months, such as independent rolling and sitting (Child 1), pulling into 4-point (Children 2, 3, and 4), and pulling to stand and returning to floor with control (Child 2). Whether these changes on the children are clinically significant is unknown, since there is no published minimal clinically significant difference for the AIMS. However, they were functionally valuable. For example, Child 1 was able to use her hands to play in sitting and Child 2 to pull to stand and step with support.

These cases do not determine whether changes in gross motor function can be attributed to the standard care therapy based on gross motor function, to the TT based on promoting trunk control or to the combination of both. Nevertheless, the functional changes seen were both apparent and relatively rapid. This justifies further investigation of TT therapy for young children to examine effectiveness.

The need for specific attention to which segment is supported and the change that this can bring about in hand function is of particular interest (Figure 2). This result applied to 2 children working toward greater use of the hand (Child 3) and increased bilateral hand use (Child 4) in play. Both children showed immediate improvement in hand function when trunk support was given to the targeted segment rather than directly beneath this segment, as is usual for promoting trunk control. The SATCo tests active trunk control by head turn.11 Research in infants aged 4 to 6 months has shown that the quality of the reaching movement improved if the infant had already gained the necessary segmental control in their trunk.5 We speculate that use of the hand for a child with hemiplegia or bilateral hand play places a greater demand on anticipatory postural trunk control (active trunk control). Thus, moving the support to 1 segment more cephalic enabled both children to show increased confidence in using their hands or bilateral hand play without loss of balance (Figure 2). The absence of a specific outcome measure to examine the hand function of these 2 children was a limitation of this study. However, there is improvement in hand function as shown in Figure 2. Based on these early findings, practitioners of TT may find it appropriate to train trunk control at the SATCo segment defined for promoting trunk control but also to provide support higher on the trunk to promote hand function if necessary.

The need to provide appropriate segmental support, as defined by the SATCo, is highlighted by Child 4 (Figure 2). If the trunk support is not directly beneath the true targeted segment (ie it is too caudal), then the child is required to attempt to control many joints at one time. This appears to present too much challenge with resulting use of compensatory strategies to maintain a neutral vertical posture. If the child uses internal support by taking some of the spinal joints to end range and gaining passive control through the ligaments (Figure 2B), the consequences can be more serious. This type of compensatory strategy may eventually cause irreversible spinal deformities, such as scoliosis.


This case report supports that TT therapy is feasible with children younger than 2 years with poor trunk control without the use of specialized equipment. This is at home with parents giving therapy. The 4 children showed improved trunk control and functional change within a short period, suggesting that TT should be further investigated.


The authors thank Miss Joyce HL Choi for her assistance in data collection and all the children and their parents who participated in this study.


1. Hadders-Algra M, Carlberg EB, eds. Postural Control:AKey Issue in Developmental Disorders. London, England: Mac Keith Press; 2008.
2. Sæther R, Helbostad JL, Adde L, Brændvik S, Lydersen S, Vik T. The relationship between trunk control in sitting and during gait in children and adolescents with cerebral palsy. Dev Med Child Neurol. 2015;57(4):344–350.
3. Woollacott MH, Crenna P. Postural control in standing and walking in children with cerebral palsy. In: Hadders-Algra M, Carlberg EB, eds. Postural Control: A Key Issue in Developmental Disorders. London, England: Mac Keith; 2008.
4. Saavedra SL, van Donkelaar P, Woollacott MH. Learning about gravity: segmental assessment of upright control as infants develop independent sitting. J Neurophysiol. 2012;108(8):2215–2229.
5. Rachwani J, Santamaria V, Saavedra SL, Wood S, Porter F, Woollacott MH. Segmental trunk control acquisition and reaching in typically developing infants. Exp Brain Res. 2013;228(1):131–139.
6. Curtis DJ, Butler P, Saavedra S, et al. The central role of trunk control in the gross motor function of children with cerebral palsy: a retrospective cross-sectional study. Dev Med Child Neurol. 2015;57(4):351–357.
7. Dewar R, Love S, Johnston LM. Exercise interventions improve postural control in children with cerebral palsy: a systematic review. Dev Med Child Neurol. 2015;57(6):504–520.
8. Major RE, Johnson GR, Butler PB. Learning motor control in the upright position: a mechanical engineering approach. Proc Inst Mech Eng H. 2001;215(3):315–323.
9. Butler PB, Major RE. The learning of motor control: biomechanical considerations. Physiotherapy. 1992;78(1):6–11.
10. Bernstein N. The Coordination and Regulation of Movement. London, England: Pergamon; 1967.
11. Butler PB, Saavedra S, Sofranac M, Jarvis SE, Woollacott MH. Refinement, reliability, and validity of the Segmental Assessment of Trunk Control. Pediatr Phys Ther. 2010;22(3):246–257.
12. Butler PB. A preliminary report on the effectiveness of trunk targeting in achieving independent sitting balance in children with cerebral palsy. Clin Rehabil. 1998;12(4):281–293.
13. Butler PB, Thompson N, Major RE. Improvement in walking performance of children with cerebral palsy: preliminary results. Dev Med Child Neurol. 1992;34(7):567–576.
14. Piper MC, Darrah J. Motor Assessment of the Developing Infant. Philadelphia, PA: W.B. Saunders; 1994.
15. Russell D, Rosenbaum P, Wright M, Avery LM. Gross Motor Function Measure (GMFM-66 & GMFM-88) Use's Manual. 2nd ed. Ontario, Canada: Mac Keith Press; 2013.
16. Pin TW, de Valle K, Eldridge B, Galea MP. Clinimetric properties of the Alberta Infant Motor Scale in infants born preterm. Pediatr Phys Ther. 2010;22(3):278–286.
17. Pin TW, Butler PB, Cheung H-M, Shum S. Psychometric study on Segmental Assessment of Trunk Control in infants from 4 to 9 months of age. Dev Med Child Neurol. 2017;59(suppl 2):5.
18. Palisano RJ, Rosenbaum P, Walter S, Russell D, Wood E, Galuppi B. Gross Motor Function Classification System (GMFCS). Ontario, Canada: CanChild; 2007.
19. Bohannon RW, Smith MB. Interrater Reliability of a Modified Ashworth Scale of Muscle Spasticity. Phys Ther. 1987;67(2):206–207.

child; motor development; physical therapy; segmental trunk control; targeted training

© 2018 Academy of Pediatric Physical Therapy of the American Physical Therapy Association