Research in Developmental Disabilities 35 (2014) 2044–2052
Contents lists available at ScienceDirect
Research in Developmental Disabilities
Altered trunk movements during gait in children with spastic diplegia: Compensatory or underlying trunk control deficit? Lieve Heyrman a,*, Hilde Feys a, Guy Molenaers b,c, Ellen Jaspers d, Davide Monari e, Angela Nieuwenhuys a, Kaat Desloovere a,b a
KU Leuven, Faculty of Kinesiology and Rehabilitation Sciences, Department of Rehabilitation Sciences, Tervuursevest 101, 3001 Heverlee, Belgium University Hospital of Pellenberg, Clinical Motion Analysis Laboratory, Weligerveld 1, 3212 Pellenberg, Belgium c KU Leuven, Faculty of Medicine, Department of Development and Regeneration, Herestraat 49, 3000 Leuven, Belgium d ETH, Department of Health Sciences and Technology, Neural Control of Movement Lab, Ra¨mistrasse 101, 8006 Zurich, Switzerland e KU Leuven, Science, Engineering and Technology Group, Department of Mechanical Engineering, Celestijnenlaan 300, 3001 Heverlee, Belgium b
A R T I C L E I N F O
A B S T R A C T
Article history: Received 16 December 2013 Received in revised form 30 April 2014 Accepted 30 April 2014 Available online
Altered trunk movements during gait in children with CP are considered compensatory due to lower limb impairments, although scientific evidence for this assumption has not yet been provided. This study aimed to study the functional relation between trunk and lower limb movement deficits during gait in children with spastic diplegia. Therefore, the relationship between trunk control in sitting, and trunk and lower limb movements during gait was explored in 20 children with spastic diplegia (age 9.2 3 yrs; GMFCS level I n = 10, level II n = 10). Trunk control in sitting was assessed with the Trunk Control Measurement Scale (TCMS), a clinical measure that reflects the presence of an underlying trunk control deficit. Trunk movements during gait were measured with a recently developed trunk model including the pelvis, thorax, head, shoulder line and spine. Lower limb movements were assessed with the Plug-in-Gait model (Vicon1). Range of motion (ROM) of the different trunk segments was calculated, as well as the Trunk Profile Score (TPS) and Trunk Variable Scores (TVSs). Similarly, the Gait Profile Score (GPS) and Gait Variable Scores (GVSs) were calculated to describe altered lower limb movements during gait. Correlation analyses were performed between the presence of impaired trunk control in sitting (TCMS) and altered trunk movements during gait (ROM, TPS/TVSs) and between these altered trunk movements and lower limb movements (GPS/GVSs) during gait. A poorer performance on the TCMS correlated with increased ROM and TPS/TVSs, particularly for the thorax, indicating the presence of an underlying trunk control deficit. No significant correlation was found between the TPS and GPS, suggesting that overall trunk and lower limb movement deficits were not strongly associated. Only few correlations between specific lower limb deficits (GVSs for hip ab/ adduction, knee flexion/extension and ankle flexion/extension) and TVSs for thorax lateral bending and rotation were found. This study provided first evidence that the altered trunk movements observed during gait should not be solely considered compensatory due to lower limb impairments, but that these may also partially reflect an underlying trunk control deficit.
Keywords: Cerebral palsy Trunk control Clinical measurement scale Three-dimensional gait analysis
* Corresponding author at: KU Leuven – Faculty of Kinesiology and Rehabilitation Sciences, Department of Rehabilitation Sciences, Tervuursevest 101, B 3001 Heverlee (Leuven), Belgium. Tel.: +32 16 37 66 20; fax: +32 16 32 91 97. E-mail addresses: [email protected] (L. Heyrman), [email protected] (H. Feys), [email protected] (G. Molenaers), [email protected] (E. Jaspers), [email protected] (D. Monari), [email protected] (A. Nieuwenhuys), [email protected] (K. Desloovere). http://dx.doi.org/10.1016/j.ridd.2014.04.031 0891-4222/ß 2014 Elsevier Ltd. All rights reserved.
L. Heyrman et al. / Research in Developmental Disabilities 35 (2014) 2044–2052
2045
A better understanding of underlying trunk control deficits in children with CP may facilitate targeted therapy planning and ultimately can optimize a child’s functionality. ß 2014 Elsevier Ltd. All rights reserved.
1. Introduction Cerebral palsy (CP) describes a group of disorders of the development of movement and posture caused by a brain injury occurring early in life (Bax et al., 2005). Loss of selective control, muscle weakness, abnormal muscle tone and impaired postural control are primary impairments, resulting in the development of secondary problems such as contractures and bony deformities (Koman, Smith, & Shilt, 2004). Both primary and secondary impairments present in various degrees of severity, influencing the performance of functional activities such as gait (Gage, 2004). To reach maximum functionality, the child with CP finds his way to cope with these impairments, resulting in a pathological gait pattern that comprises primary, secondary and compensatory characteristics. For a clinician, the challenge is to distinguish pathological gait characteristics directly related to the neuromuscular disorder from those that are compensatory to cope with the biomechanical constraints caused by the underlying impairments (Davids, Foti, Dabelstein, & Bagley, 1999; Stebbins, Harrington, Thompson, Zavatsky, & Theologis, 2010). These impairments should be identified as the focus of the treatment plan, while compensatory movements are expected to diminish once they are no longer needed and therefore do not require specific treatment (Gage, 2004, 2009). Research on pathological gait in children with CP, based on three-dimensional (3D) movement analysis, is mainly focused on describing the impact of lower limb impairments on gait, while head and trunk movements are rarely addressed. Few studies that have analyzed head and trunk movements during gait found increased range of motion (ROM) in different planes (Heyrman, Feys, Molenaers, Jaspers, Monari, et al., 2013; Romkes et al., 2007). A recent study additionally suggested that the decreased gait efficiency found in children with spastic diplegia was mainly attributed to increased mechanical work of the head, trunk and arms (Van de Walle et al., 2012). Available literature thus underlines the importance of studying total body movements to gain further insights into the pathological gait patterns in children with CP. Remarkably, to date these altered trunk movements during gait in children with CP have been considered compensatory due to lower limb impairments, e.g. increased thorax lateral sway due to hip abductor weakness in patients with (Metaxiotis, Accles, Siebel, & Doederlein, 2000) or without hip deformities (Hadders-Algra & Brogren, 2008; Krautwurst et al., 2013); forward trunk lean to reduce the extension moment on the knee (Gage, 2004; Romkes et al., 2007), or backward trunk lean due to weak hip extensors (Perry, 1992). However, these studies failed to investigate the occurrence of an underlying trunk control deficit in their study population. As such, the reported compensatory role of the trunk appears mainly based on assumptions rather than on objective evidence (Stebbins et al., 2010). This might be partially explained by the lack of an available tool to measure an underlying trunk control deficit. Recently, we developed a valid clinical measurement for trunk control in children with CP, i.e. the Trunk Control Measurement Scale (TCMS) (Heyrman et al., 2011). The TCMS assesses static and dynamic aspects of trunk control in sitting position, thereby reducing the influence of lower limb impairments on trunk performance. While first evidence for the occurrence of an underlying deficit in trunk control in children with CP was demonstrated (Heyrman, Desloovere, et al., 2013), it remains unclear if and how this deficit is manifested during the performance of a functional activity such as gait. Therefore, the first aim of this study was to investigate whether the presence of altered trunk movements during gait was related to lower scores on the TCMS in sitting, thus pointing toward the presence of an underlying trunk control deficit. Secondly, we wanted to examine the relationship between altered trunk and lower limb movements during gait. This research represents a first step in investigating to what extent altered trunk movements during gait should be mainly considered as compensatory movements for lower limb impairments, or may also indicate the presence of an underlying trunk control deficit in children with CP. 2. Materials and methods 2.1. Participants Children and adolescents with CP (henceforth referred to as children with CP) were recruited from the database of the Clinical Motion Analysis Laboratory (University Hospital Pellenberg, Belgium) and included if they were diagnosed as spastic diplegia, aged between 5 and 15 years, able to walk without assistive devices (GMFCS levels I and II) (Palisano et al., 1997) and sufficiently cooperative to properly complete the test procedure. Children were excluded if they had received Botulinum Toxin-A injections within six months prior to testing, orthopedic surgery in the past year, or in case of an intrathecal baclofen pump implantation or other spinal interventions. Children scheduled for routine clinical gait analysis who met these criteria were invited to participate. Twenty children with spastic CP (GMFCS level I n = 10, level II n = 10; mean age 9.2 3 yrs) were eligible for the study. The hip statuses of all children were within normal values (Reimers’ index >66% coverage). Four children (20%) underwent a bilateral proximal derotation of the femur at least three years before assessment and sixteen children (80%) had
2046
L. Heyrman et al. / Research in Developmental Disabilities 35 (2014) 2044–2052
no prior orthopedic surgery. Median scores on the Modified Ashworth Scale for lower limb spasticity were 0 for m. Rectus femoris and m. Tibialis posterior, 1 for hip flexors and hip adductors, 1.25 for hamstrings and 1.5 for plantar flexors. Written informed consent was obtained from all parents. The study was approved by the ethical committee of the University Hospitals Leuven. 2.2. Assessment and data analysis Trunk assessment was performed on one test occasion in two different conditions, i.e. in sitting using the TCMS and during gait using a 3D kinematic trunk model. The latter assessment also included lower limb data collection. All children were assessed by one trained physiotherapist. 2.2.1. Trunk control in sitting Trunk control in sitting was clinically evaluated with the Trunk Control Measurement Scale (TCMS) (Heyrman et al., 2011). This recently developed scale consists of three subscales (15 items in total) measuring both static and dynamic aspects of trunk control, with the ‘trunk’ including the thorax and the pelvis. The first subscale, static sitting balance, evaluates the ability of the child to maintain a stable trunk position during upper and lower limb movements. The second subscale, selective movement control, evaluates the performance of selective trunk movements within the base of support in three planes: sagittal (flexion/extension), frontal (lateral bending), and transverse (rotation). The third subscale, dynamic reaching, assesses the performance during three reaching tasks requiring active trunk movements beyond the base of support. Reliability and aspects of validity of the TCMS have been previously established (Heyrman et al., 2011). For test administration, children were seated on a table or bench without back, arm or feet support. No orthoses or shoes were worn during testing. The best of three performances on each item was considered for scoring. The TCMS total score ranges from 0 to 58, with a higher score indicating a better performance. Item scores of each subscale were summated, resulting in three subscale scores, and the sum of these subscale scores resulted in a total TCMS score. 2.2.2. Trunk and lower limb movements during gait Head and trunk movements during gait were evaluated with a custom-made trunk model based on 3D movement analysis (Heyrman, Feys, Molenaers, Jaspers, Van de Walle, et al., 2013). This model consists of five segments (head, thorax, pelvis, shoulder line, and spine), comprising a marker-set of 19 retro-reflective markers. In comparison with the Vicon Plugin-Gait model including a head and thorax segment, this trunk model further decomposed the trunk into several clinically relevant subparts. Also, the least-square optimization method of four thorax markers was used to better take into account the inter-segmental deformation and soft-tissue artifacts occurring at the thorax during motion (Soderkvist & Wedin, 1993). The reliability of this model was established in a previous study (Heyrman, Feys, Molenaers, Jaspers, Van de Walle, et al., 2013). A detailed description of the marker placement and definition of anatomical planes is provided elsewhere (Heyrman, Feys, Molenaers, Jaspers, Van de Walle, et al., 2013). Lower limb kinematics was measured with the conventional lower body Plug-In-Gait marker-set (VICON Oxford Metrics, Oxford, UK). Children walked barefoot over a 10 m walkway at self-selected speed. Marker trajectories were captured using a 15-camera VICON system (VICON Oxford Metrics, Oxford, UK) at a sampling rate of 100 Hz and filtered using spline interpolation (Woltring, 1995). Nexus software (VICON Oxford Metrics, Oxford, UK) was used for trajectory reconstruction and marker labeling, and for definition of gait cycle events (i.e. initial contact and toe-off). Three representative trials from left gait cycles were used for further data processing with custom-made Matlab routines (Mathworks, Inc.). Gait cycles were time-normalized and kinematic parameters were calculated for every cycle. For the pelvis, thorax and head, absolute angles (vs. the global laboratory frame) were calculated using Euler/Cardan decompositions (flexion/extension (tilt), lateral bending (obliquity), rotation) (Grood & Suntay, 1983). For the shoulder line, relative angles (shoulder line vs. thorax) were computed in the frontal and transverse plane. Spine movement was described as two projection angles, i.e. the angle of kyphosis (angle between T2–T6 and T10–L1) and the angle of lordosis (angle between L1–L3 and L3–L5) (Whittle & Levine, 1997). Range of motion (ROM), i.e. movement amplitude over the gait cycle, of each segment in the different planes was calculated for the three gait cycles per child, and then averaged. A higher value indicates increased movement amplitude over the gait cycle. The Gait Profile Score (GPS) (Baker et al., 2009) and Trunk Profile Score (TPS) (Heyrman, Feys, Molenaers, Jaspers, Monari, et al., 2013) were additionally calculated for each child. The GPS is a single index outcome measure that indicates the overall alteration of a child’s lower limb movements, based on kinematic data (pelvis, hip, knee and ankle/foot). The TPS is a newly developed outcome measure, similar to the GPS, providing a single index of overall alteration of trunk movements during gait, derived from head and trunk kinematics (pelvis, thorax, head, shoulder line, spine). Both measures are calculated from the root mean square difference between kinematic data of the individual child and averaged data from a reference database of typically developing (TD) children, based on three representative gait cycles. We further calculated a modified GPS (GPSmod) in which the pelvis segment was excluded to rule out the overlapping inclusion of the pelvis in both indices. The GPS, GPSmod and TPS were decomposed to individual Gait Variable Scores (GVSs) and Trunk Variable Scores (TVSs), providing an index of deviation of each component kinematic variable. All these measures are presented in degrees, with a higher value
L. Heyrman et al. / Research in Developmental Disabilities 35 (2014) 2044–2052
2047
indicating increased alteration of movements. Given that the left and right side of the head, thorax and shoulder line move concurrently in opposite directions during one gait cycle, only one side of data (left) was used for further analysis for the purpose of this study (Kiernan, Malone, O’Brien, & Simms, 2014; Krautwurst et al., 2013). 2.3. Statistical analysis Given the fact that several variables were not normally distributed (Kolmogorov–Smirnov test) and that the TCMS is an ordinal scale, non-parametric statistics were used. Descriptive statistics (median, IQR) were used to compute subscale and total TCMS scores, ROM, TPS and GPS/GPSmod. Spearman correlation coefficients (r) were calculated to investigate the relationship between ROM and TCMS scores (total TCMS and subscale scores), between the TPS and TCMS scores (total TCMS and subscale scores), between the TPS/TVSs and the GPS/GPSmod/GVSs and between total TCMS and GPS/GVSs. The magnitude of the coefficients were interpreted based on Hinkle’s guidelines, whereby r < 0.30 is considered a weak; 0.30– 0.50 a fair; 0.50–0.70 a good; and r > 0.70 a high correlation (Hinkle, Wiersma, & Jurs, 1998). The level of significance was set at p < 0.05. Multiple regression analyses were also performed using a forward selection procedure to further explore the relationship between trunk parameters (TVSs; dependent variables) and lower limb parameters (GVSs; independent variables) during gait. Inputs for the model were trunk parameters which showed a statistically significant linear relationship with lower limb parameters. The level of significance as criterion for an independent variable to enter the model was set at p < 0.15. Intercollinearity between all independent variables entering the model was verified using an a priori defined critical value of r 0.60. Statistical analyses were done with SAS Enterprise Guide 4.2 (SAS Institute Inc., Cary, NC, USA).
3. Results Descriptive data on total TCMS and subtotal scores (trunk control in sitting), and ROM, TPS/TVSs and GPS/GPSmod/GVSs (altered trunk and lower limb movements during gait) are presented in Table 1. A median total TCMS score of 42.5 out of 58 was found, corresponding to 72% of the maximal score. Children showed most deficits on the subscales selective movement control (64% of the total score) and dynamic reaching (45% of the total score). These scores correspond to previous findings in children with CP (Heyrman, Desloovere, et al., 2013). Increased ROM was found for pelvic tilt, thorax movements in all directions, and head lateral bending and rotation, compared to reference values of TD children (Heyrman, Feys, Molenaers, Jaspers, Monari, et al., 2013). The median TPS for the total group was 6.38. The median GPS and GPSmod were 9.38 and 10.88 respectively. These values indicate the overall presence of altered trunk and lower limb movements during gait in our study group, compared to TD children (Baker et al., 2012; Heyrman, Feys, Molenaers, Jaspers, Monari, et al., 2013). 3.1. Relationship between trunk parameters in sitting and during gait Table 2 summarizes Spearman correlation coefficients between ROM and TPS/TVSs (during gait) and the TCMS scores (in sitting). Significant correlations are indicated in bold. Significant correlations between ROM and TCMS scores were found only at the level of the thorax. A good correlation was found between the total TCMS score and thorax ROM for flexion/extension (r = 0.52) and lateral bending (r = 0.53), whereby larger thorax movements in forward/backward and sideward directions during gait were associated with lower total TCMS scores. Correlation analysis of the subscales showed fair significant correlations between thorax ROM for lateral bending and selective movement control (r = 0.43), and between thorax ROM for both flexion/extension and lateral bending and dynamic reaching (r = 0.43 and 0.44, respectively). The TPS was fairly correlated with the total TCMS score (r = 0.44) and the scores on the subscale dynamic reaching (r = 0.48), i.e. the occurrence of overall increased trunk movements during gait was associated with poorer trunk control in sitting. Good correlations were also found between the TVSs of thorax lateral bending and rotation, and pelvis rotation and the total TCMS (r = 0.63, 0.57 and 0.53, respectively). Correlation analysis between the TVSs and the TCMS subscale scores showed fair to good correlations between thorax lateral bending and rotation and all three subscales (r = 0.46 to 0.54), and fair correlations between pelvis rotation and selective movement control and dynamic reaching (r = 0.45 and 0.49, respectively). 3.2. Relationship between trunk parameters (sitting-gait) and lower limb parameters Table 3 summarizes Spearman correlation coefficients between the degree of impaired trunk control in sitting (total TCMS), altered trunk movements (TPS, TVSs) and altered lower limb movements (GPS/GPSmod, GVSs) during gait. Given the overlap between the GVSs and the TVSs for the pelvis, no GVSs for the pelvis were included in the correlation analyses with the TVSs, since the pelvis is considered a segment of the trunk model. As a consequence, only the GVSs for the hip, knee, ankle and foot represent lower limb parameters.
L. Heyrman et al. / Research in Developmental Disabilities 35 (2014) 2044–2052
2048
Table 1 Descriptive data of total and subtotal TCMS scores, segmental range of motion (ROM), and indices of altered trunk and gait movements. Median (IQR) Total TMCS (0–58) Static sitting balance (0–20) Selective movement control (0–28) Dynamic reaching (0–10) ROM (8)
Pelvis
Thorax
Head
Shoulder line Spine TPS (8) TVS (8)
Pelvis
Thorax
Head
Shoulder line Spine GPS (8) GPSmod (8) GVS (8)
Hip
Knee Ankle Foot
42.5 20 18 4.5 Tilt Obliquity Rotation Flex/ext Lat bend Rotation Flex/ext Lat bend Rotation Lat bend Rotation Kyphosis Lordosis Tilt Obliquity Rotation Flex/ext Lat bend Rotation Flex/ext Lat bend Rotation Lat bend Rotation Kyphosis Lordosis
Flex/ext Abd/add Rotation Flex/ext Dfl/pfl Rotation
(38.5–46.25) (19–20) (15.75–20.25) (3.75–6) 6.8 11.3 13.8 9.2 7.9 10.9 8.6 5.3 6.4 3.9 5.7 5.0 6.4 6.3 4.6 2.7 5.0 4.2 4.0 4.6 8.1 4.5 4.4 1.9 2.6 7.0 6.1 9.3 10.8 7.0 3.7 10.5 10.7 7.2 10.7
(4.4–7.8) (8.3–12.9) (12.6–18.6) (5.1–11.2) (4.8–12.9) (9.3–14.5) (6.2–11.3) (4.3–8.9) (4.8–8.3) (3.5–4.5) (4.3–8.1) (4.4–6.9) (5.9–9.4) (5.2–7.9) (3.9–6.3) (1.9–3.9) (4.2–5.6) (1.7–6.4) (2.8–5.4) (4.2–6.5) (6.8–12.7) (2.8–6.0) (2.4–5.7) (1.5–2.2) (1.9–4.0) (3.6–9.8) (4.6–11.3) (7.2–10.1) (8.1–11.7) (5.6–9.2) (2.7–4.5) (6.7–14.0) (8.0–14.4) (5.3–8.8) (6.8–15.7)
IQR, interquartile range; TCMS, Trunk Control Measurement Scale; TPS, Trunk Profile Score; TVSs, Trunk Variable Scores; GPS, Gait Profile Score; GPSmod, modified Gait Profile Score (pelvis not included); GVSs, Gait Variable Scores.
The total TCMS score was fairly correlated with the GPS, indicating that children with poorer trunk control in sitting had an increased alteration of lower limb movements (r = 0.49). Fair to good correlations were also found between the total TCMS score and GVSs for hip ab/adduction and knee flexion/extension (r = 0.48 and 0.53, respectively). The TPS and GPS correlated fairly, though values were not significant (r = 0.35, p = 0.13). Also the GPSmod only showed a fair, not significant correlation with the TPS (r = 0.32, p = 0.17), indicating no association between the overall degree of altered trunk movements and lower limb movements. Correlation analysis between the different GVSs and the TPS revealed a good correlation only for GVS of hip ab/adduction (r = 0.53). Correlation coefficients between the TVSs and the GPS/GPSmod/GVSs were weak to fair (r < 0.50). The TVS for the pelvis showed fair to good correlations with the GVSs for hip flexion/extension and ab/adduction (r = 0.46–0.52). For the thorax, fair to good correlations were found between the TVS for thorax lateral bending and the GVSs for hip ab/adduction, knee flexion/ extension and ankle flexion/extension (r = 0.43–0.54). The TVS for thorax rotation fairly correlated with the overall GPS and the GVS for ankle flexion/extension (r = 0.47), and good correlations were found with the GVS for knee flexion/extension (r = 0.70). The TVSs for shoulder line rotation and kyphosis also showed good correlations with the GVS for knee flexion/ extension (r = 0.57). Following the multiple significant correlations that were found between several GVSs and the TVSs for thorax lateral bending and rotation (see Table 3), a forward selection multiple regression analysis was performed for these thorax parameters. No intercollinearity was found between variables entering the model. The GVS for knee flexion/extension and hip ab/adduction explained a total of 47% (F = 7.08, p = 0.006) of the variability found in the TVS for thorax lateral bending, with a partial explained variance of 29% by the GVS for knee flexion/extension (p = 0.02) and 18% additional variance by the GVS for hip ab/adduction (p = 0.03). The variability in the TVS for thorax rotation could only be explained for 39% by the GVS for knee flexion/extension (p = 0.004).
L. Heyrman et al. / Research in Developmental Disabilities 35 (2014) 2044–2052
2049
Table 2 Spearman rank correlation coefficients between head and trunk range of motion, TPS, and TVS, and the Trunk Control Measurement Scale (total score and subscales). Total TCMS ROM
Pelvis
Thorax
Head
Shoulder line Spine TPS TVSs
Pelvis
Thorax
Head
Shoulder line Spine
Tilt Obliquity Rotation Flex/ext Lat bend Rotation Flex/ext Lat bend Rotation Lat bend Rotation Kyphosis Lordosis Tilt Obliquity Rotation Flex/ext Lat bend Rotation Flex/ext Lat bend Rotation Lat bend Rotation Kyphosis Lordosis
0.30 0.23 0.05 0.52* 0.53* 0.33 0.29 0.34 0.30 0.19 0.16 0.38 0.31 0.44* 0.12 0.06 0.53* 0.38 0.63** 0.57** 0.02 0.32 0.13 0.22 0.39 0.11 0.02
Static sitting balance 0.06 0.18 0.17 0.26 0.34 0.02 0.05 0.22 0.09 0.11 0.21 0.03 0.15 0.02 0.06 0.23 0.08 0.27 0.46* 0.50* 0.22 0.14 0.15 0.22 0.09 0.03 0.03
Selective movement control 0.35 0.19 0.02 0.39 0.43* 0.25 0.41 0.42 0.29 0.28 0.08 0.34 0.20 0.34 0.10 0.08 0.45* 0.31 0.54* 0.52* 0.24 0.39 0.15 0.16 0.34 0.07 0.09
Dynamic reaching 0.16 0.19 0.14 0.44* 0.43* 0.39 0.09 0.12 0.14 0.04 0.31 0.42 0.43* 0.48* 0.27 0.05 0.49* 0.34 0.51* 0.35 0.30 0.11 0.03 0.09 0.30 0.28 0.12
TCMS, Trunk Control Measurement Scale; ROM, range of motion; TPS, Trunk Profile Score; TVSs, Trunk Variable Scores. * p 0.05. ** p 0.01.
4. Discussion This study aimed to gain more insights into the functional relation between altered trunk and lower limb movements during gait in children with spastic diplegia. In particular, we were interested if altered trunk movements occurring during gait would partially reflect the presence of an underlying trunk control deficit, or if they should be solely considered as compensatory movements for lower limb impairments. Compensatory trunk movements do not require direct specific treatment, while intrinsic trunk control deficits should be included in the treatment plan (Gage, 2004, 2009). The results of this study showed that increased altered trunk movements during gait (TPS) were related to a lower performance on the TCMS in sitting, indicating the presence of an underlying trunk control deficit. Further detailed correlation analyses between the segmental trunk movement parameters during gait (ROM, TVSs) and the TCMS (total and subscale scores) revealed most associations between impaired trunk control in sitting and the occurrence of altered movements during gait at the level of the thorax in all planes (ROM, TVSs) and the pelvis in the transverse plane (TVS). The majority of the significant correlations were found with the subscales selective movement control and dynamic reaching, pointing to the dynamic character of trunk control requirements during gait. These findings provide first and unique evidence for the presence of underlying impaired trunk control in children with spastic diplegia exhibiting difficulties with sustainment of trunk stability during gait, especially with regard to the thorax and pelvis. It should be noted that there were some differences found in the results of the correlation analysis for the two segmental parameters of trunk movements during gait, i.e. ROM and TVSs. The results of the current study indicated that the TPS and the TVSs were more strongly related with the performance on the TCMS, and thus with the presence of an underlying trunk control deficit, compared to ROM. These differences might be partially explained by the inherent difference between these parameters. Range of motion is a discrete parameter expressing the range between a segment’s minimum and maximum position within a gait cycle. Whilst it detects changes in amplitude, the pattern of change over the gait cycle is not taken into account. In contrast, the TPS and TVSs consider a segment’s movement trajectory throughout the gait cycle and value its deviation from a normal reference database. The inherent differences between both parameters were also apparent in a previous study comparing children with CP to typically developing children (Heyrman, Feys, Molenaers, Jaspers, Monari, et al., 2013). Therefore, the TPS and TVSs are considered more sensitive to capture alterations in movement patterns. The lack of significant correlation between overall altered trunk movements (TPS) and altered lower limb movements (GPS) during gait suggests that altered trunk movements during gait are not exclusively associated with the presence of lower limb impairments. This is further supported by the limited number of significant correlations between head and trunk TVSs and the different GVSs. Together, these findings appear to nuance the general assumption that trunk movements during
2050
TCMS
TPS
TVSs Pelvis Tilt
GPS GPSmod GVSs
Hip
Knee Ankle Foot
Fl/ext Ab/add Rotation Fl/ext Fl/ext Rotation
0.49* – 0.42 0.48* 0.30 0.53* 0.42 0.30
0.35 0.32 0.37 0.53* 0.31 0.38 0.01 0.26
– 0.20 0.46* 0.47* 0.20 0.06 0.06 0.33
Thorax Obliquity – 0.04 0.01 0.48* 0.24 0.12 0.08 0.10
Rotation – 0.39 0.52* 0.32 0.26 0.27 0.35 0.13
Fl/ext 0.19 0.22 0.03 0.36 0.11 0.40 0.31 0.20
Head Lat bend 0.33 0.26 0.28 0.48* 0.07 0.54* 0.43* 0.24
Rotation 0.47* 0.40 0.22 0.38 0.23 0.70** 0.47* 0.28
Fl/ext 0.47* 0.42 0.23 0.02 0.38 0.41 0.08 0.33
Shoulder line
Spine
Lat bend
Rotation
Lat bend
Kyphosis
Lordosis
0.18 0.20 0.36 0.24 0.16 0.02 0.11 0.24
0.20 0.17 0.36 0.24 0.09 0.02 0.03 0.33
0.07 0.02 0.06 0.31 0.05 0.31 0.03 0.19
0.38 0.31 0.07 0.20 0.06 0.44* 0.11 0.20
0.23 0.21 0.21 0.27 0.06 0.24 0.34 0.21
Rotation 0.33 0.34 0.34 0.17 0.07 0.57** 0.01 0.24
TCMS, Trunk Control Measurement Scale; TPS, Trunk Profile Score; TVSs, Trunk Variable Scores; GPS, Gait Profile Score; GPSmod, modified Gait Profile Score (with exclusion of pelvis); GVSs, Gait Variable Scores. * p 0.05. ** p 0.01.
L. Heyrman et al. / Research in Developmental Disabilities 35 (2014) 2044–2052
Table 3 Spearman rank correlation coefficients between parameters of trunk control in sitting (total TCMS score), trunk movement parameters during gait (TPS, TVS) and lower limb movement parameters during gait (GPS, GVS).
L. Heyrman et al. / Research in Developmental Disabilities 35 (2014) 2044–2052
2051
gait are only compensatory to lower limb deficits (Gage, 2004, 2009; Krautwurst et al., 2013; Metaxiotis et al., 2000), but rather point toward a combination of both compensatory movements as well as the presence of an underlying trunk control deficit. Still, in-depth correlation analyses showed significant associations between altered movements of hip ab/adduction, knee flexion/extension and ankle dorsiflexion/plantarflexion and altered trunk movements mainly for the thorax (lateral bending and rotation) and pelvis (tilt and obliquity). The latter might be partially explained by the close kinematic relationship between hip and pelvis, characterized by correlated motions during gait (kinematic chain). Altered ankle and thorax movements during gait were related, though no such relation was found between altered ankle movements and performance on the TCMS. This favors the idea that the altered thorax lateral bending and rotation movements are compensatory for the altered ankle movements. In contrast, altered hip ab/adduction and knee flexion/extension movements, both elements of crouch gait, showed fair to good associations with altered thorax rotation and/or lateral bending movements during gait, as well as with the performance on the TCMS in sitting. The results of the regression analysis showed that a substantial proportion of the variability of the measured thorax lateral bending and rotation movements during gait could be explained by the occurrence of altered knee flexion and hip adduction movements, which corresponds with the findings of a recent review of Schmid, Schweizer, Romkes, Lorenzetti, and Brunner (2013) describing lateral trunk movements as compensatory for lower limb impairments. However, a large proportion of the measured variability of altered thorax movements still remained unexplained, which suggests that these trunk movements may reflect a combination of compensatory strategies for lower limb impairments as well as the presence of an underlying trunk control deficit. To the best of our knowledge, this is the first study that provided evidence for the presence of underlying impaired trunk control in children with CP that is related with altered trunk movements during gait. Krautwurst et al. (2013) studied the influence of hip abductor weakness on frontal plane motion of trunk and pelvis during gait in 375 ambulatory children with spastic diplegia. They found only a weak correlation (r = 0.28) between trunk lateral bending and hip abductor strength, suggesting that other factors influencing trunk deficits require further investigation. Also, recently, Schweizer, Brunner, and Romkes (2014) found no clinically relevant changes in trunk kinematics when walking with or without an ankle-foot orthosis in children with unilateral CP, suggesting that the observed increased ROM of the trunk should not be considered as compensatory for their toe walking. While this study provides a unique contribution to the understanding of impaired trunk control in children with CP, some critical reflections are warranted. First, correlation analyses were based on a limited number of children. Therefore, further study on a larger cohort of children with CP is recommended. Secondly, our study group only included children who were able to walk without assistive devices, i.e. GMFCS level I and II. Although children with GMFCS level III often show increased head and trunk movement deficits while walking, the use of an assistive device would have confounded their natural head, trunk and lower limb kinematics. Therefore these children should be subjected to a separate future study. Thirdly, in this study the TCMS is considered a measurement of an underlying trunk deficit, as the seated position minimizes the influence of lower limb impairments, compared to standing and walking. However, it must be recognized that the occurrence of lower limb spasticity and/or muscle weakness may still have influenced the performance on the TCMS, which necessitates further investigation. Fourthly, while interesting associations between altered thorax movements and several altered lower limb movements during gait were found, thorough interpretation of these individual correlations is hindered by the fact that a high interrelation in segmental movements is present during the execution of a functional activity. The challenge for future study is to further unravel the relation between trunk and lower limb movements, grouped into functional movement patterns. Moreover, additional information on trunk and lower limb kinetics and muscle activity will highly contribute to the understanding of this functional relationship, and will provide more in-depth insights into compensatory mechanisms of the trunk versus the presence of a primary trunk control deficit (Schmid et al., 2013). Finally, the presence of associations between parameters found in this study does not imply their causality as well. Therefore, in future research interventional strategies focusing on either the effect of a lower limb intervention (e.g. Botulinum Toxin-A injections) or a trunk intervention (e.g. trunk training program) and their impact on trunk movements during gait could enhance our understanding of underlying trunk control deficits in children with CP and the impact on functionality, as well as of compensatory strategies of the trunk due to lower limb impairments. Also, the impact of the cerebral damage on all trunk and lower limb parameters requires further study. Consequently, insights from such studies may provide a sound base for therapeutic interventions targeting impaired trunk control to improve gait in children with spastic diplegia, such as strengthening trunk and hip musculature to improve stability and coordination exercises. To date, interventional studies including the trunk are limited and describe a wide variety of interventions, for example hippotherapy, electrical stimulation of trunk muscles, and serious gaming (Barton, Hawken, Foster, Holmes, & Butler, 2013; Park, Park, Lee, & Cho, 2001; Zadnikar & Kastrin, 2011). Also, the effect on functional activities such as gait is poorly understood. Therefore, a patient-tailored trunk intervention program, based on a comprehensive trunk evaluation protocol including 3D-movement analysis and a clinical evaluation tool such as the TCMS, may have the potential to have a positive effect on a child’s gait pattern, which requires further study. In conclusion, the results of this study provide first evidence that the presence of an underlying trunk control deficit, measured in sitting with the TCMS, is related to altered trunk movements during gait in children with spastic diplegia, in particular with respect to the thorax. Altered thorax lateral bending and rotation movements were only partially related to altered hip ab/adduction and knee flexion/extension movements. The results thereby provide first evidence that these observed thorax movements during gait should not solely be considered as compensatory movements for lower limb impairments, but most likely are the resultant of both compensatory movements for lower limb deficits and an underlying
2052
L. Heyrman et al. / Research in Developmental Disabilities 35 (2014) 2044–2052
trunk control deficit. These findings imply opportunities for therapeutic interventions aiming to improve trunk control in these children. Conflict of interest statement The authors declare that the research is conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Acknowledgement Lieve Heyrman received a PhD fellowship of the Research Foundation Flanders (FWO), Belgium. References Baker, R., McGinley, J. L., Schwartz, M., Thomason, P., Rodda, J., & Graham, H. K. (2012). The minimal clinically important difference for the Gait Profile Score. Gait and Posture, 35, 612–615. Baker, R., McGinley, J. L., Schwartz, M. H., Beynon, S., Rozumalski, A., Graham, H. K., et al. (2009). The gait profile score and movement analysis profile. Gait and Posture, 30, 265–269. Barton, G. J., Hawken, M. B., Foster, R. J., Holmes, G., & Butler, P. B. (2013). The effects of virtual reality game training on trunk to pelvis coupling in a child with cerebral palsy. Journal of Neuroengineering and Rehabilitation, 10, 15. Bax, M., Goldstein, M., Rosenbaum, P., Leviton, A., Paneth, N., Dan, B., et al. (2005). Proposed definition and classification of cerebral palsy, April 2005. Developmental Medicine and Child Neurology, 47, 571–576. Davids, J. R., Foti, T., Dabelstein, J., & Bagley, A. (1999). Voluntary (normal) versus obligatory (cerebral palsy) toe-walking in children: A kinematic, kinetic, and electromyographic analysis. Journal of Pediatric Orthopedics, 19, 461–469. Gage, J. R. (2004). The treatment of gait problems in children with cerebral palsy. London: Mac Keith Press. Gage, J. R. (2009). The identification and treatment of gait problems in cerebral palsy. London: Mac Keith Press. Grood, E. S., & Suntay, W. J. (1983). A joint coordinate system for the clinical description of three-dimensional motions: Application to the knee. Journal of Biomechanical Engineering, 105, 136–144. Hadders-Algra, M., & Brogren, E. (2008). Postural control: A key issue in developmental disorders. London: Mac Keith Press. Heyrman, L., Molenaers, G., Desloovere, K., Verheyden, G., De Cat, J., Monbaliu, E., et al. (2011). A clinical tool to measure trunk control in children with cerebral palsy: The Trunk Control Measurement Scale. Research in Developmental Disabilities, 32, 2624–2635. Heyrman, L., Desloovere, K., Molenaers, G., Verheyden, G., Klingels, K., Monbaliu, E., et al. (2013). Clinical characteristics of impaired trunk control in children with spastic cerebral palsy. Research in Developmental Disabilities, 34, 327–334. Heyrman, L., Feys, H., Molenaers, G., Jaspers, E., Van de Walle, P., Monari, D., et al. (2013). Reliability of head and trunk kinematics during gait in children with spastic diplegia. Gait and Posture, 37, 424–429. Heyrman, L., Feys, H., Molenaers, G., Jaspers, E., Monari, D., Meyns, P., et al. (2013). Three-dimensional head and trunk characteristics during gait in children with spastic diplegia. Gait and Posture, 38, 770–776. Hinkle, D. E., Wiersma, W., & Jurs, S. G. (1998). Applied statistics for behavioral sciences. Boston: Houghton Mifflin. Kiernan, D., Malone, A., O’Brien, T., & Simms, C. K. (2014). A 3-dimensional rigid cluster thorax model for kinematic measurements during gait. Journal of Biomechanics, 47, 1499–1505. Koman, L. A., Smith, B. P., & Shilt, J. S. (2004). Cerebral palsy. The Lancet, 63, 1619–1631. Krautwurst, B. K., Wolf, S. I., Heitzmann, D. W., Gantz, S., Braatz, F., & Dreher, T. (2013). The influence of hip abductor weakness on frontal plane motion of the trunk and pelvis in patients with cerebral palsy. Research in Developmental Disabilities, 34, 1198–1203. Metaxiotis, D., Accles, W., Siebel, A., & Doederlein, L. (2000). Hip deformities in walking patients with cerebral palsy. Gait and Posture, 11, 86–91. Palisano, R., Rosenbaum, P., Walter, S., Russell, D., Wood, E., & Galuppi, B. (1997). Development and reliability of a system to classify gross motor function in children with cerebral palsy. Developmental Medicine and Child Neurology, 39, 214–223. Park, E. S., Park, C. I., Lee, H. J., & Cho, Y. S. (2001). The effect of electrical stimulation on the trunk control in young children with spastic diplegic cerebral palsy. Journal of Korean Medical Science, 16, 347–350. Perry, J. (1992). Gait analysis: Normal and pathological function. Thorofare, NJ: SLACK Incorporated. Romkes, J., Peeters, W., Oosterom, A. M., Molenaar, S., Bakels, I., & Brunner, R. (2007). Evaluating upper body movements during gait in healthy children and children with diplegic cerebral palsy. Journal of Pediatric Orthopedics Part B, 16, 175–180. Schmid, S., Schweizer, K., Romkes, J., Lorenzetti, S., & Brunner, R. (2013). Secondary gait deviations in patients with and without neurological involvement: A systematic review. Gait and Posture, 37, 480–493. Schweizer, K., Brunner, R., & Romkes, J. (2014). Upper body movements in children with hemiplegic cerebral palsy walking with and without an ankle-foot orthesis. Clinical Biomechanics. http://dx.doi.org/10.1016/j.clinbiomech.2014.02.005 Soderkvist, I., & Wedin, P. A. (1993). Determining the movements of the skeleton using well-configured markers. Journal of Biomechanics, 26, 1473–1477. Stebbins, J., Harrington, M., Thompson, N., Zavatsky, A., & Theologis, T. (2010). Gait compensations caused by foot deformity in cerebral palsy. Gait and Posture, 32, 226–230. Van de Walle, P., Hallemans, A., Truijen, S., Gosselink, R., Heyrman, L., Molenaers, G., et al. (2012). Increased mechanical cost of walking in children with diplegia: The role of the passenger unit cannot be neglected. Research in Developmental Disabilities, 33, 1996–2003. Whittle, M. W., & Levine, D. (1997). Measurement of lumbar lordosis as a component of clinical gait analysis. Gait and Posture, 5, 101–107. Woltring, H. J. (1995). Smoothing and differentiation techniques applied to 3-D data. In P. Allard, I. A. Stokes, J.-P. Blanchi (Eds.), Three-dimensional analysis of human movement. Champaign: Human Kinetics Publishers. Zadnikar, M., & Kastrin, A. (2011). Effects of hippotherapy and therapeutic horseback riding on postural control or balance in children with cerebral palsy: A metaanalysis. Developmental Medicine and Child Neurology, 53, 684–691.
Contents lists available at ScienceDirect
Research in Developmental Disabilities
Altered trunk movements during gait in children with spastic diplegia: Compensatory or underlying trunk control deficit? Lieve Heyrman a,*, Hilde Feys a, Guy Molenaers b,c, Ellen Jaspers d, Davide Monari e, Angela Nieuwenhuys a, Kaat Desloovere a,b a
KU Leuven, Faculty of Kinesiology and Rehabilitation Sciences, Department of Rehabilitation Sciences, Tervuursevest 101, 3001 Heverlee, Belgium University Hospital of Pellenberg, Clinical Motion Analysis Laboratory, Weligerveld 1, 3212 Pellenberg, Belgium c KU Leuven, Faculty of Medicine, Department of Development and Regeneration, Herestraat 49, 3000 Leuven, Belgium d ETH, Department of Health Sciences and Technology, Neural Control of Movement Lab, Ra¨mistrasse 101, 8006 Zurich, Switzerland e KU Leuven, Science, Engineering and Technology Group, Department of Mechanical Engineering, Celestijnenlaan 300, 3001 Heverlee, Belgium b
A R T I C L E I N F O
A B S T R A C T
Article history: Received 16 December 2013 Received in revised form 30 April 2014 Accepted 30 April 2014 Available online
Altered trunk movements during gait in children with CP are considered compensatory due to lower limb impairments, although scientific evidence for this assumption has not yet been provided. This study aimed to study the functional relation between trunk and lower limb movement deficits during gait in children with spastic diplegia. Therefore, the relationship between trunk control in sitting, and trunk and lower limb movements during gait was explored in 20 children with spastic diplegia (age 9.2 3 yrs; GMFCS level I n = 10, level II n = 10). Trunk control in sitting was assessed with the Trunk Control Measurement Scale (TCMS), a clinical measure that reflects the presence of an underlying trunk control deficit. Trunk movements during gait were measured with a recently developed trunk model including the pelvis, thorax, head, shoulder line and spine. Lower limb movements were assessed with the Plug-in-Gait model (Vicon1). Range of motion (ROM) of the different trunk segments was calculated, as well as the Trunk Profile Score (TPS) and Trunk Variable Scores (TVSs). Similarly, the Gait Profile Score (GPS) and Gait Variable Scores (GVSs) were calculated to describe altered lower limb movements during gait. Correlation analyses were performed between the presence of impaired trunk control in sitting (TCMS) and altered trunk movements during gait (ROM, TPS/TVSs) and between these altered trunk movements and lower limb movements (GPS/GVSs) during gait. A poorer performance on the TCMS correlated with increased ROM and TPS/TVSs, particularly for the thorax, indicating the presence of an underlying trunk control deficit. No significant correlation was found between the TPS and GPS, suggesting that overall trunk and lower limb movement deficits were not strongly associated. Only few correlations between specific lower limb deficits (GVSs for hip ab/ adduction, knee flexion/extension and ankle flexion/extension) and TVSs for thorax lateral bending and rotation were found. This study provided first evidence that the altered trunk movements observed during gait should not be solely considered compensatory due to lower limb impairments, but that these may also partially reflect an underlying trunk control deficit.
Keywords: Cerebral palsy Trunk control Clinical measurement scale Three-dimensional gait analysis
* Corresponding author at: KU Leuven – Faculty of Kinesiology and Rehabilitation Sciences, Department of Rehabilitation Sciences, Tervuursevest 101, B 3001 Heverlee (Leuven), Belgium. Tel.: +32 16 37 66 20; fax: +32 16 32 91 97. E-mail addresses: [email protected] (L. Heyrman), [email protected] (H. Feys), [email protected] (G. Molenaers), [email protected] (E. Jaspers), [email protected] (D. Monari), [email protected] (A. Nieuwenhuys), [email protected] (K. Desloovere). http://dx.doi.org/10.1016/j.ridd.2014.04.031 0891-4222/ß 2014 Elsevier Ltd. All rights reserved.
L. Heyrman et al. / Research in Developmental Disabilities 35 (2014) 2044–2052
2045
A better understanding of underlying trunk control deficits in children with CP may facilitate targeted therapy planning and ultimately can optimize a child’s functionality. ß 2014 Elsevier Ltd. All rights reserved.
1. Introduction Cerebral palsy (CP) describes a group of disorders of the development of movement and posture caused by a brain injury occurring early in life (Bax et al., 2005). Loss of selective control, muscle weakness, abnormal muscle tone and impaired postural control are primary impairments, resulting in the development of secondary problems such as contractures and bony deformities (Koman, Smith, & Shilt, 2004). Both primary and secondary impairments present in various degrees of severity, influencing the performance of functional activities such as gait (Gage, 2004). To reach maximum functionality, the child with CP finds his way to cope with these impairments, resulting in a pathological gait pattern that comprises primary, secondary and compensatory characteristics. For a clinician, the challenge is to distinguish pathological gait characteristics directly related to the neuromuscular disorder from those that are compensatory to cope with the biomechanical constraints caused by the underlying impairments (Davids, Foti, Dabelstein, & Bagley, 1999; Stebbins, Harrington, Thompson, Zavatsky, & Theologis, 2010). These impairments should be identified as the focus of the treatment plan, while compensatory movements are expected to diminish once they are no longer needed and therefore do not require specific treatment (Gage, 2004, 2009). Research on pathological gait in children with CP, based on three-dimensional (3D) movement analysis, is mainly focused on describing the impact of lower limb impairments on gait, while head and trunk movements are rarely addressed. Few studies that have analyzed head and trunk movements during gait found increased range of motion (ROM) in different planes (Heyrman, Feys, Molenaers, Jaspers, Monari, et al., 2013; Romkes et al., 2007). A recent study additionally suggested that the decreased gait efficiency found in children with spastic diplegia was mainly attributed to increased mechanical work of the head, trunk and arms (Van de Walle et al., 2012). Available literature thus underlines the importance of studying total body movements to gain further insights into the pathological gait patterns in children with CP. Remarkably, to date these altered trunk movements during gait in children with CP have been considered compensatory due to lower limb impairments, e.g. increased thorax lateral sway due to hip abductor weakness in patients with (Metaxiotis, Accles, Siebel, & Doederlein, 2000) or without hip deformities (Hadders-Algra & Brogren, 2008; Krautwurst et al., 2013); forward trunk lean to reduce the extension moment on the knee (Gage, 2004; Romkes et al., 2007), or backward trunk lean due to weak hip extensors (Perry, 1992). However, these studies failed to investigate the occurrence of an underlying trunk control deficit in their study population. As such, the reported compensatory role of the trunk appears mainly based on assumptions rather than on objective evidence (Stebbins et al., 2010). This might be partially explained by the lack of an available tool to measure an underlying trunk control deficit. Recently, we developed a valid clinical measurement for trunk control in children with CP, i.e. the Trunk Control Measurement Scale (TCMS) (Heyrman et al., 2011). The TCMS assesses static and dynamic aspects of trunk control in sitting position, thereby reducing the influence of lower limb impairments on trunk performance. While first evidence for the occurrence of an underlying deficit in trunk control in children with CP was demonstrated (Heyrman, Desloovere, et al., 2013), it remains unclear if and how this deficit is manifested during the performance of a functional activity such as gait. Therefore, the first aim of this study was to investigate whether the presence of altered trunk movements during gait was related to lower scores on the TCMS in sitting, thus pointing toward the presence of an underlying trunk control deficit. Secondly, we wanted to examine the relationship between altered trunk and lower limb movements during gait. This research represents a first step in investigating to what extent altered trunk movements during gait should be mainly considered as compensatory movements for lower limb impairments, or may also indicate the presence of an underlying trunk control deficit in children with CP. 2. Materials and methods 2.1. Participants Children and adolescents with CP (henceforth referred to as children with CP) were recruited from the database of the Clinical Motion Analysis Laboratory (University Hospital Pellenberg, Belgium) and included if they were diagnosed as spastic diplegia, aged between 5 and 15 years, able to walk without assistive devices (GMFCS levels I and II) (Palisano et al., 1997) and sufficiently cooperative to properly complete the test procedure. Children were excluded if they had received Botulinum Toxin-A injections within six months prior to testing, orthopedic surgery in the past year, or in case of an intrathecal baclofen pump implantation or other spinal interventions. Children scheduled for routine clinical gait analysis who met these criteria were invited to participate. Twenty children with spastic CP (GMFCS level I n = 10, level II n = 10; mean age 9.2 3 yrs) were eligible for the study. The hip statuses of all children were within normal values (Reimers’ index >66% coverage). Four children (20%) underwent a bilateral proximal derotation of the femur at least three years before assessment and sixteen children (80%) had
2046
L. Heyrman et al. / Research in Developmental Disabilities 35 (2014) 2044–2052
no prior orthopedic surgery. Median scores on the Modified Ashworth Scale for lower limb spasticity were 0 for m. Rectus femoris and m. Tibialis posterior, 1 for hip flexors and hip adductors, 1.25 for hamstrings and 1.5 for plantar flexors. Written informed consent was obtained from all parents. The study was approved by the ethical committee of the University Hospitals Leuven. 2.2. Assessment and data analysis Trunk assessment was performed on one test occasion in two different conditions, i.e. in sitting using the TCMS and during gait using a 3D kinematic trunk model. The latter assessment also included lower limb data collection. All children were assessed by one trained physiotherapist. 2.2.1. Trunk control in sitting Trunk control in sitting was clinically evaluated with the Trunk Control Measurement Scale (TCMS) (Heyrman et al., 2011). This recently developed scale consists of three subscales (15 items in total) measuring both static and dynamic aspects of trunk control, with the ‘trunk’ including the thorax and the pelvis. The first subscale, static sitting balance, evaluates the ability of the child to maintain a stable trunk position during upper and lower limb movements. The second subscale, selective movement control, evaluates the performance of selective trunk movements within the base of support in three planes: sagittal (flexion/extension), frontal (lateral bending), and transverse (rotation). The third subscale, dynamic reaching, assesses the performance during three reaching tasks requiring active trunk movements beyond the base of support. Reliability and aspects of validity of the TCMS have been previously established (Heyrman et al., 2011). For test administration, children were seated on a table or bench without back, arm or feet support. No orthoses or shoes were worn during testing. The best of three performances on each item was considered for scoring. The TCMS total score ranges from 0 to 58, with a higher score indicating a better performance. Item scores of each subscale were summated, resulting in three subscale scores, and the sum of these subscale scores resulted in a total TCMS score. 2.2.2. Trunk and lower limb movements during gait Head and trunk movements during gait were evaluated with a custom-made trunk model based on 3D movement analysis (Heyrman, Feys, Molenaers, Jaspers, Van de Walle, et al., 2013). This model consists of five segments (head, thorax, pelvis, shoulder line, and spine), comprising a marker-set of 19 retro-reflective markers. In comparison with the Vicon Plugin-Gait model including a head and thorax segment, this trunk model further decomposed the trunk into several clinically relevant subparts. Also, the least-square optimization method of four thorax markers was used to better take into account the inter-segmental deformation and soft-tissue artifacts occurring at the thorax during motion (Soderkvist & Wedin, 1993). The reliability of this model was established in a previous study (Heyrman, Feys, Molenaers, Jaspers, Van de Walle, et al., 2013). A detailed description of the marker placement and definition of anatomical planes is provided elsewhere (Heyrman, Feys, Molenaers, Jaspers, Van de Walle, et al., 2013). Lower limb kinematics was measured with the conventional lower body Plug-In-Gait marker-set (VICON Oxford Metrics, Oxford, UK). Children walked barefoot over a 10 m walkway at self-selected speed. Marker trajectories were captured using a 15-camera VICON system (VICON Oxford Metrics, Oxford, UK) at a sampling rate of 100 Hz and filtered using spline interpolation (Woltring, 1995). Nexus software (VICON Oxford Metrics, Oxford, UK) was used for trajectory reconstruction and marker labeling, and for definition of gait cycle events (i.e. initial contact and toe-off). Three representative trials from left gait cycles were used for further data processing with custom-made Matlab routines (Mathworks, Inc.). Gait cycles were time-normalized and kinematic parameters were calculated for every cycle. For the pelvis, thorax and head, absolute angles (vs. the global laboratory frame) were calculated using Euler/Cardan decompositions (flexion/extension (tilt), lateral bending (obliquity), rotation) (Grood & Suntay, 1983). For the shoulder line, relative angles (shoulder line vs. thorax) were computed in the frontal and transverse plane. Spine movement was described as two projection angles, i.e. the angle of kyphosis (angle between T2–T6 and T10–L1) and the angle of lordosis (angle between L1–L3 and L3–L5) (Whittle & Levine, 1997). Range of motion (ROM), i.e. movement amplitude over the gait cycle, of each segment in the different planes was calculated for the three gait cycles per child, and then averaged. A higher value indicates increased movement amplitude over the gait cycle. The Gait Profile Score (GPS) (Baker et al., 2009) and Trunk Profile Score (TPS) (Heyrman, Feys, Molenaers, Jaspers, Monari, et al., 2013) were additionally calculated for each child. The GPS is a single index outcome measure that indicates the overall alteration of a child’s lower limb movements, based on kinematic data (pelvis, hip, knee and ankle/foot). The TPS is a newly developed outcome measure, similar to the GPS, providing a single index of overall alteration of trunk movements during gait, derived from head and trunk kinematics (pelvis, thorax, head, shoulder line, spine). Both measures are calculated from the root mean square difference between kinematic data of the individual child and averaged data from a reference database of typically developing (TD) children, based on three representative gait cycles. We further calculated a modified GPS (GPSmod) in which the pelvis segment was excluded to rule out the overlapping inclusion of the pelvis in both indices. The GPS, GPSmod and TPS were decomposed to individual Gait Variable Scores (GVSs) and Trunk Variable Scores (TVSs), providing an index of deviation of each component kinematic variable. All these measures are presented in degrees, with a higher value
L. Heyrman et al. / Research in Developmental Disabilities 35 (2014) 2044–2052
2047
indicating increased alteration of movements. Given that the left and right side of the head, thorax and shoulder line move concurrently in opposite directions during one gait cycle, only one side of data (left) was used for further analysis for the purpose of this study (Kiernan, Malone, O’Brien, & Simms, 2014; Krautwurst et al., 2013). 2.3. Statistical analysis Given the fact that several variables were not normally distributed (Kolmogorov–Smirnov test) and that the TCMS is an ordinal scale, non-parametric statistics were used. Descriptive statistics (median, IQR) were used to compute subscale and total TCMS scores, ROM, TPS and GPS/GPSmod. Spearman correlation coefficients (r) were calculated to investigate the relationship between ROM and TCMS scores (total TCMS and subscale scores), between the TPS and TCMS scores (total TCMS and subscale scores), between the TPS/TVSs and the GPS/GPSmod/GVSs and between total TCMS and GPS/GVSs. The magnitude of the coefficients were interpreted based on Hinkle’s guidelines, whereby r < 0.30 is considered a weak; 0.30– 0.50 a fair; 0.50–0.70 a good; and r > 0.70 a high correlation (Hinkle, Wiersma, & Jurs, 1998). The level of significance was set at p < 0.05. Multiple regression analyses were also performed using a forward selection procedure to further explore the relationship between trunk parameters (TVSs; dependent variables) and lower limb parameters (GVSs; independent variables) during gait. Inputs for the model were trunk parameters which showed a statistically significant linear relationship with lower limb parameters. The level of significance as criterion for an independent variable to enter the model was set at p < 0.15. Intercollinearity between all independent variables entering the model was verified using an a priori defined critical value of r 0.60. Statistical analyses were done with SAS Enterprise Guide 4.2 (SAS Institute Inc., Cary, NC, USA).
3. Results Descriptive data on total TCMS and subtotal scores (trunk control in sitting), and ROM, TPS/TVSs and GPS/GPSmod/GVSs (altered trunk and lower limb movements during gait) are presented in Table 1. A median total TCMS score of 42.5 out of 58 was found, corresponding to 72% of the maximal score. Children showed most deficits on the subscales selective movement control (64% of the total score) and dynamic reaching (45% of the total score). These scores correspond to previous findings in children with CP (Heyrman, Desloovere, et al., 2013). Increased ROM was found for pelvic tilt, thorax movements in all directions, and head lateral bending and rotation, compared to reference values of TD children (Heyrman, Feys, Molenaers, Jaspers, Monari, et al., 2013). The median TPS for the total group was 6.38. The median GPS and GPSmod were 9.38 and 10.88 respectively. These values indicate the overall presence of altered trunk and lower limb movements during gait in our study group, compared to TD children (Baker et al., 2012; Heyrman, Feys, Molenaers, Jaspers, Monari, et al., 2013). 3.1. Relationship between trunk parameters in sitting and during gait Table 2 summarizes Spearman correlation coefficients between ROM and TPS/TVSs (during gait) and the TCMS scores (in sitting). Significant correlations are indicated in bold. Significant correlations between ROM and TCMS scores were found only at the level of the thorax. A good correlation was found between the total TCMS score and thorax ROM for flexion/extension (r = 0.52) and lateral bending (r = 0.53), whereby larger thorax movements in forward/backward and sideward directions during gait were associated with lower total TCMS scores. Correlation analysis of the subscales showed fair significant correlations between thorax ROM for lateral bending and selective movement control (r = 0.43), and between thorax ROM for both flexion/extension and lateral bending and dynamic reaching (r = 0.43 and 0.44, respectively). The TPS was fairly correlated with the total TCMS score (r = 0.44) and the scores on the subscale dynamic reaching (r = 0.48), i.e. the occurrence of overall increased trunk movements during gait was associated with poorer trunk control in sitting. Good correlations were also found between the TVSs of thorax lateral bending and rotation, and pelvis rotation and the total TCMS (r = 0.63, 0.57 and 0.53, respectively). Correlation analysis between the TVSs and the TCMS subscale scores showed fair to good correlations between thorax lateral bending and rotation and all three subscales (r = 0.46 to 0.54), and fair correlations between pelvis rotation and selective movement control and dynamic reaching (r = 0.45 and 0.49, respectively). 3.2. Relationship between trunk parameters (sitting-gait) and lower limb parameters Table 3 summarizes Spearman correlation coefficients between the degree of impaired trunk control in sitting (total TCMS), altered trunk movements (TPS, TVSs) and altered lower limb movements (GPS/GPSmod, GVSs) during gait. Given the overlap between the GVSs and the TVSs for the pelvis, no GVSs for the pelvis were included in the correlation analyses with the TVSs, since the pelvis is considered a segment of the trunk model. As a consequence, only the GVSs for the hip, knee, ankle and foot represent lower limb parameters.
L. Heyrman et al. / Research in Developmental Disabilities 35 (2014) 2044–2052
2048
Table 1 Descriptive data of total and subtotal TCMS scores, segmental range of motion (ROM), and indices of altered trunk and gait movements. Median (IQR) Total TMCS (0–58) Static sitting balance (0–20) Selective movement control (0–28) Dynamic reaching (0–10) ROM (8)
Pelvis
Thorax
Head
Shoulder line Spine TPS (8) TVS (8)
Pelvis
Thorax
Head
Shoulder line Spine GPS (8) GPSmod (8) GVS (8)
Hip
Knee Ankle Foot
42.5 20 18 4.5 Tilt Obliquity Rotation Flex/ext Lat bend Rotation Flex/ext Lat bend Rotation Lat bend Rotation Kyphosis Lordosis Tilt Obliquity Rotation Flex/ext Lat bend Rotation Flex/ext Lat bend Rotation Lat bend Rotation Kyphosis Lordosis
Flex/ext Abd/add Rotation Flex/ext Dfl/pfl Rotation
(38.5–46.25) (19–20) (15.75–20.25) (3.75–6) 6.8 11.3 13.8 9.2 7.9 10.9 8.6 5.3 6.4 3.9 5.7 5.0 6.4 6.3 4.6 2.7 5.0 4.2 4.0 4.6 8.1 4.5 4.4 1.9 2.6 7.0 6.1 9.3 10.8 7.0 3.7 10.5 10.7 7.2 10.7
(4.4–7.8) (8.3–12.9) (12.6–18.6) (5.1–11.2) (4.8–12.9) (9.3–14.5) (6.2–11.3) (4.3–8.9) (4.8–8.3) (3.5–4.5) (4.3–8.1) (4.4–6.9) (5.9–9.4) (5.2–7.9) (3.9–6.3) (1.9–3.9) (4.2–5.6) (1.7–6.4) (2.8–5.4) (4.2–6.5) (6.8–12.7) (2.8–6.0) (2.4–5.7) (1.5–2.2) (1.9–4.0) (3.6–9.8) (4.6–11.3) (7.2–10.1) (8.1–11.7) (5.6–9.2) (2.7–4.5) (6.7–14.0) (8.0–14.4) (5.3–8.8) (6.8–15.7)
IQR, interquartile range; TCMS, Trunk Control Measurement Scale; TPS, Trunk Profile Score; TVSs, Trunk Variable Scores; GPS, Gait Profile Score; GPSmod, modified Gait Profile Score (pelvis not included); GVSs, Gait Variable Scores.
The total TCMS score was fairly correlated with the GPS, indicating that children with poorer trunk control in sitting had an increased alteration of lower limb movements (r = 0.49). Fair to good correlations were also found between the total TCMS score and GVSs for hip ab/adduction and knee flexion/extension (r = 0.48 and 0.53, respectively). The TPS and GPS correlated fairly, though values were not significant (r = 0.35, p = 0.13). Also the GPSmod only showed a fair, not significant correlation with the TPS (r = 0.32, p = 0.17), indicating no association between the overall degree of altered trunk movements and lower limb movements. Correlation analysis between the different GVSs and the TPS revealed a good correlation only for GVS of hip ab/adduction (r = 0.53). Correlation coefficients between the TVSs and the GPS/GPSmod/GVSs were weak to fair (r < 0.50). The TVS for the pelvis showed fair to good correlations with the GVSs for hip flexion/extension and ab/adduction (r = 0.46–0.52). For the thorax, fair to good correlations were found between the TVS for thorax lateral bending and the GVSs for hip ab/adduction, knee flexion/ extension and ankle flexion/extension (r = 0.43–0.54). The TVS for thorax rotation fairly correlated with the overall GPS and the GVS for ankle flexion/extension (r = 0.47), and good correlations were found with the GVS for knee flexion/extension (r = 0.70). The TVSs for shoulder line rotation and kyphosis also showed good correlations with the GVS for knee flexion/ extension (r = 0.57). Following the multiple significant correlations that were found between several GVSs and the TVSs for thorax lateral bending and rotation (see Table 3), a forward selection multiple regression analysis was performed for these thorax parameters. No intercollinearity was found between variables entering the model. The GVS for knee flexion/extension and hip ab/adduction explained a total of 47% (F = 7.08, p = 0.006) of the variability found in the TVS for thorax lateral bending, with a partial explained variance of 29% by the GVS for knee flexion/extension (p = 0.02) and 18% additional variance by the GVS for hip ab/adduction (p = 0.03). The variability in the TVS for thorax rotation could only be explained for 39% by the GVS for knee flexion/extension (p = 0.004).
L. Heyrman et al. / Research in Developmental Disabilities 35 (2014) 2044–2052
2049
Table 2 Spearman rank correlation coefficients between head and trunk range of motion, TPS, and TVS, and the Trunk Control Measurement Scale (total score and subscales). Total TCMS ROM
Pelvis
Thorax
Head
Shoulder line Spine TPS TVSs
Pelvis
Thorax
Head
Shoulder line Spine
Tilt Obliquity Rotation Flex/ext Lat bend Rotation Flex/ext Lat bend Rotation Lat bend Rotation Kyphosis Lordosis Tilt Obliquity Rotation Flex/ext Lat bend Rotation Flex/ext Lat bend Rotation Lat bend Rotation Kyphosis Lordosis
0.30 0.23 0.05 0.52* 0.53* 0.33 0.29 0.34 0.30 0.19 0.16 0.38 0.31 0.44* 0.12 0.06 0.53* 0.38 0.63** 0.57** 0.02 0.32 0.13 0.22 0.39 0.11 0.02
Static sitting balance 0.06 0.18 0.17 0.26 0.34 0.02 0.05 0.22 0.09 0.11 0.21 0.03 0.15 0.02 0.06 0.23 0.08 0.27 0.46* 0.50* 0.22 0.14 0.15 0.22 0.09 0.03 0.03
Selective movement control 0.35 0.19 0.02 0.39 0.43* 0.25 0.41 0.42 0.29 0.28 0.08 0.34 0.20 0.34 0.10 0.08 0.45* 0.31 0.54* 0.52* 0.24 0.39 0.15 0.16 0.34 0.07 0.09
Dynamic reaching 0.16 0.19 0.14 0.44* 0.43* 0.39 0.09 0.12 0.14 0.04 0.31 0.42 0.43* 0.48* 0.27 0.05 0.49* 0.34 0.51* 0.35 0.30 0.11 0.03 0.09 0.30 0.28 0.12
TCMS, Trunk Control Measurement Scale; ROM, range of motion; TPS, Trunk Profile Score; TVSs, Trunk Variable Scores. * p 0.05. ** p 0.01.
4. Discussion This study aimed to gain more insights into the functional relation between altered trunk and lower limb movements during gait in children with spastic diplegia. In particular, we were interested if altered trunk movements occurring during gait would partially reflect the presence of an underlying trunk control deficit, or if they should be solely considered as compensatory movements for lower limb impairments. Compensatory trunk movements do not require direct specific treatment, while intrinsic trunk control deficits should be included in the treatment plan (Gage, 2004, 2009). The results of this study showed that increased altered trunk movements during gait (TPS) were related to a lower performance on the TCMS in sitting, indicating the presence of an underlying trunk control deficit. Further detailed correlation analyses between the segmental trunk movement parameters during gait (ROM, TVSs) and the TCMS (total and subscale scores) revealed most associations between impaired trunk control in sitting and the occurrence of altered movements during gait at the level of the thorax in all planes (ROM, TVSs) and the pelvis in the transverse plane (TVS). The majority of the significant correlations were found with the subscales selective movement control and dynamic reaching, pointing to the dynamic character of trunk control requirements during gait. These findings provide first and unique evidence for the presence of underlying impaired trunk control in children with spastic diplegia exhibiting difficulties with sustainment of trunk stability during gait, especially with regard to the thorax and pelvis. It should be noted that there were some differences found in the results of the correlation analysis for the two segmental parameters of trunk movements during gait, i.e. ROM and TVSs. The results of the current study indicated that the TPS and the TVSs were more strongly related with the performance on the TCMS, and thus with the presence of an underlying trunk control deficit, compared to ROM. These differences might be partially explained by the inherent difference between these parameters. Range of motion is a discrete parameter expressing the range between a segment’s minimum and maximum position within a gait cycle. Whilst it detects changes in amplitude, the pattern of change over the gait cycle is not taken into account. In contrast, the TPS and TVSs consider a segment’s movement trajectory throughout the gait cycle and value its deviation from a normal reference database. The inherent differences between both parameters were also apparent in a previous study comparing children with CP to typically developing children (Heyrman, Feys, Molenaers, Jaspers, Monari, et al., 2013). Therefore, the TPS and TVSs are considered more sensitive to capture alterations in movement patterns. The lack of significant correlation between overall altered trunk movements (TPS) and altered lower limb movements (GPS) during gait suggests that altered trunk movements during gait are not exclusively associated with the presence of lower limb impairments. This is further supported by the limited number of significant correlations between head and trunk TVSs and the different GVSs. Together, these findings appear to nuance the general assumption that trunk movements during
2050
TCMS
TPS
TVSs Pelvis Tilt
GPS GPSmod GVSs
Hip
Knee Ankle Foot
Fl/ext Ab/add Rotation Fl/ext Fl/ext Rotation
0.49* – 0.42 0.48* 0.30 0.53* 0.42 0.30
0.35 0.32 0.37 0.53* 0.31 0.38 0.01 0.26
– 0.20 0.46* 0.47* 0.20 0.06 0.06 0.33
Thorax Obliquity – 0.04 0.01 0.48* 0.24 0.12 0.08 0.10
Rotation – 0.39 0.52* 0.32 0.26 0.27 0.35 0.13
Fl/ext 0.19 0.22 0.03 0.36 0.11 0.40 0.31 0.20
Head Lat bend 0.33 0.26 0.28 0.48* 0.07 0.54* 0.43* 0.24
Rotation 0.47* 0.40 0.22 0.38 0.23 0.70** 0.47* 0.28
Fl/ext 0.47* 0.42 0.23 0.02 0.38 0.41 0.08 0.33
Shoulder line
Spine
Lat bend
Rotation
Lat bend
Kyphosis
Lordosis
0.18 0.20 0.36 0.24 0.16 0.02 0.11 0.24
0.20 0.17 0.36 0.24 0.09 0.02 0.03 0.33
0.07 0.02 0.06 0.31 0.05 0.31 0.03 0.19
0.38 0.31 0.07 0.20 0.06 0.44* 0.11 0.20
0.23 0.21 0.21 0.27 0.06 0.24 0.34 0.21
Rotation 0.33 0.34 0.34 0.17 0.07 0.57** 0.01 0.24
TCMS, Trunk Control Measurement Scale; TPS, Trunk Profile Score; TVSs, Trunk Variable Scores; GPS, Gait Profile Score; GPSmod, modified Gait Profile Score (with exclusion of pelvis); GVSs, Gait Variable Scores. * p 0.05. ** p 0.01.
L. Heyrman et al. / Research in Developmental Disabilities 35 (2014) 2044–2052
Table 3 Spearman rank correlation coefficients between parameters of trunk control in sitting (total TCMS score), trunk movement parameters during gait (TPS, TVS) and lower limb movement parameters during gait (GPS, GVS).
L. Heyrman et al. / Research in Developmental Disabilities 35 (2014) 2044–2052
2051
gait are only compensatory to lower limb deficits (Gage, 2004, 2009; Krautwurst et al., 2013; Metaxiotis et al., 2000), but rather point toward a combination of both compensatory movements as well as the presence of an underlying trunk control deficit. Still, in-depth correlation analyses showed significant associations between altered movements of hip ab/adduction, knee flexion/extension and ankle dorsiflexion/plantarflexion and altered trunk movements mainly for the thorax (lateral bending and rotation) and pelvis (tilt and obliquity). The latter might be partially explained by the close kinematic relationship between hip and pelvis, characterized by correlated motions during gait (kinematic chain). Altered ankle and thorax movements during gait were related, though no such relation was found between altered ankle movements and performance on the TCMS. This favors the idea that the altered thorax lateral bending and rotation movements are compensatory for the altered ankle movements. In contrast, altered hip ab/adduction and knee flexion/extension movements, both elements of crouch gait, showed fair to good associations with altered thorax rotation and/or lateral bending movements during gait, as well as with the performance on the TCMS in sitting. The results of the regression analysis showed that a substantial proportion of the variability of the measured thorax lateral bending and rotation movements during gait could be explained by the occurrence of altered knee flexion and hip adduction movements, which corresponds with the findings of a recent review of Schmid, Schweizer, Romkes, Lorenzetti, and Brunner (2013) describing lateral trunk movements as compensatory for lower limb impairments. However, a large proportion of the measured variability of altered thorax movements still remained unexplained, which suggests that these trunk movements may reflect a combination of compensatory strategies for lower limb impairments as well as the presence of an underlying trunk control deficit. To the best of our knowledge, this is the first study that provided evidence for the presence of underlying impaired trunk control in children with CP that is related with altered trunk movements during gait. Krautwurst et al. (2013) studied the influence of hip abductor weakness on frontal plane motion of trunk and pelvis during gait in 375 ambulatory children with spastic diplegia. They found only a weak correlation (r = 0.28) between trunk lateral bending and hip abductor strength, suggesting that other factors influencing trunk deficits require further investigation. Also, recently, Schweizer, Brunner, and Romkes (2014) found no clinically relevant changes in trunk kinematics when walking with or without an ankle-foot orthosis in children with unilateral CP, suggesting that the observed increased ROM of the trunk should not be considered as compensatory for their toe walking. While this study provides a unique contribution to the understanding of impaired trunk control in children with CP, some critical reflections are warranted. First, correlation analyses were based on a limited number of children. Therefore, further study on a larger cohort of children with CP is recommended. Secondly, our study group only included children who were able to walk without assistive devices, i.e. GMFCS level I and II. Although children with GMFCS level III often show increased head and trunk movement deficits while walking, the use of an assistive device would have confounded their natural head, trunk and lower limb kinematics. Therefore these children should be subjected to a separate future study. Thirdly, in this study the TCMS is considered a measurement of an underlying trunk deficit, as the seated position minimizes the influence of lower limb impairments, compared to standing and walking. However, it must be recognized that the occurrence of lower limb spasticity and/or muscle weakness may still have influenced the performance on the TCMS, which necessitates further investigation. Fourthly, while interesting associations between altered thorax movements and several altered lower limb movements during gait were found, thorough interpretation of these individual correlations is hindered by the fact that a high interrelation in segmental movements is present during the execution of a functional activity. The challenge for future study is to further unravel the relation between trunk and lower limb movements, grouped into functional movement patterns. Moreover, additional information on trunk and lower limb kinetics and muscle activity will highly contribute to the understanding of this functional relationship, and will provide more in-depth insights into compensatory mechanisms of the trunk versus the presence of a primary trunk control deficit (Schmid et al., 2013). Finally, the presence of associations between parameters found in this study does not imply their causality as well. Therefore, in future research interventional strategies focusing on either the effect of a lower limb intervention (e.g. Botulinum Toxin-A injections) or a trunk intervention (e.g. trunk training program) and their impact on trunk movements during gait could enhance our understanding of underlying trunk control deficits in children with CP and the impact on functionality, as well as of compensatory strategies of the trunk due to lower limb impairments. Also, the impact of the cerebral damage on all trunk and lower limb parameters requires further study. Consequently, insights from such studies may provide a sound base for therapeutic interventions targeting impaired trunk control to improve gait in children with spastic diplegia, such as strengthening trunk and hip musculature to improve stability and coordination exercises. To date, interventional studies including the trunk are limited and describe a wide variety of interventions, for example hippotherapy, electrical stimulation of trunk muscles, and serious gaming (Barton, Hawken, Foster, Holmes, & Butler, 2013; Park, Park, Lee, & Cho, 2001; Zadnikar & Kastrin, 2011). Also, the effect on functional activities such as gait is poorly understood. Therefore, a patient-tailored trunk intervention program, based on a comprehensive trunk evaluation protocol including 3D-movement analysis and a clinical evaluation tool such as the TCMS, may have the potential to have a positive effect on a child’s gait pattern, which requires further study. In conclusion, the results of this study provide first evidence that the presence of an underlying trunk control deficit, measured in sitting with the TCMS, is related to altered trunk movements during gait in children with spastic diplegia, in particular with respect to the thorax. Altered thorax lateral bending and rotation movements were only partially related to altered hip ab/adduction and knee flexion/extension movements. The results thereby provide first evidence that these observed thorax movements during gait should not solely be considered as compensatory movements for lower limb impairments, but most likely are the resultant of both compensatory movements for lower limb deficits and an underlying
2052
L. Heyrman et al. / Research in Developmental Disabilities 35 (2014) 2044–2052
trunk control deficit. These findings imply opportunities for therapeutic interventions aiming to improve trunk control in these children. Conflict of interest statement The authors declare that the research is conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Acknowledgement Lieve Heyrman received a PhD fellowship of the Research Foundation Flanders (FWO), Belgium. References Baker, R., McGinley, J. L., Schwartz, M., Thomason, P., Rodda, J., & Graham, H. K. (2012). The minimal clinically important difference for the Gait Profile Score. Gait and Posture, 35, 612–615. Baker, R., McGinley, J. L., Schwartz, M. H., Beynon, S., Rozumalski, A., Graham, H. K., et al. (2009). The gait profile score and movement analysis profile. Gait and Posture, 30, 265–269. Barton, G. J., Hawken, M. B., Foster, R. J., Holmes, G., & Butler, P. B. (2013). The effects of virtual reality game training on trunk to pelvis coupling in a child with cerebral palsy. Journal of Neuroengineering and Rehabilitation, 10, 15. Bax, M., Goldstein, M., Rosenbaum, P., Leviton, A., Paneth, N., Dan, B., et al. (2005). Proposed definition and classification of cerebral palsy, April 2005. Developmental Medicine and Child Neurology, 47, 571–576. Davids, J. R., Foti, T., Dabelstein, J., & Bagley, A. (1999). Voluntary (normal) versus obligatory (cerebral palsy) toe-walking in children: A kinematic, kinetic, and electromyographic analysis. Journal of Pediatric Orthopedics, 19, 461–469. Gage, J. R. (2004). The treatment of gait problems in children with cerebral palsy. London: Mac Keith Press. Gage, J. R. (2009). The identification and treatment of gait problems in cerebral palsy. London: Mac Keith Press. Grood, E. S., & Suntay, W. J. (1983). A joint coordinate system for the clinical description of three-dimensional motions: Application to the knee. Journal of Biomechanical Engineering, 105, 136–144. Hadders-Algra, M., & Brogren, E. (2008). Postural control: A key issue in developmental disorders. London: Mac Keith Press. Heyrman, L., Molenaers, G., Desloovere, K., Verheyden, G., De Cat, J., Monbaliu, E., et al. (2011). A clinical tool to measure trunk control in children with cerebral palsy: The Trunk Control Measurement Scale. Research in Developmental Disabilities, 32, 2624–2635. Heyrman, L., Desloovere, K., Molenaers, G., Verheyden, G., Klingels, K., Monbaliu, E., et al. (2013). Clinical characteristics of impaired trunk control in children with spastic cerebral palsy. Research in Developmental Disabilities, 34, 327–334. Heyrman, L., Feys, H., Molenaers, G., Jaspers, E., Van de Walle, P., Monari, D., et al. (2013). Reliability of head and trunk kinematics during gait in children with spastic diplegia. Gait and Posture, 37, 424–429. Heyrman, L., Feys, H., Molenaers, G., Jaspers, E., Monari, D., Meyns, P., et al. (2013). Three-dimensional head and trunk characteristics during gait in children with spastic diplegia. Gait and Posture, 38, 770–776. Hinkle, D. E., Wiersma, W., & Jurs, S. G. (1998). Applied statistics for behavioral sciences. Boston: Houghton Mifflin. Kiernan, D., Malone, A., O’Brien, T., & Simms, C. K. (2014). A 3-dimensional rigid cluster thorax model for kinematic measurements during gait. Journal of Biomechanics, 47, 1499–1505. Koman, L. A., Smith, B. P., & Shilt, J. S. (2004). Cerebral palsy. The Lancet, 63, 1619–1631. Krautwurst, B. K., Wolf, S. I., Heitzmann, D. W., Gantz, S., Braatz, F., & Dreher, T. (2013). The influence of hip abductor weakness on frontal plane motion of the trunk and pelvis in patients with cerebral palsy. Research in Developmental Disabilities, 34, 1198–1203. Metaxiotis, D., Accles, W., Siebel, A., & Doederlein, L. (2000). Hip deformities in walking patients with cerebral palsy. Gait and Posture, 11, 86–91. Palisano, R., Rosenbaum, P., Walter, S., Russell, D., Wood, E., & Galuppi, B. (1997). Development and reliability of a system to classify gross motor function in children with cerebral palsy. Developmental Medicine and Child Neurology, 39, 214–223. Park, E. S., Park, C. I., Lee, H. J., & Cho, Y. S. (2001). The effect of electrical stimulation on the trunk control in young children with spastic diplegic cerebral palsy. Journal of Korean Medical Science, 16, 347–350. Perry, J. (1992). Gait analysis: Normal and pathological function. Thorofare, NJ: SLACK Incorporated. Romkes, J., Peeters, W., Oosterom, A. M., Molenaar, S., Bakels, I., & Brunner, R. (2007). Evaluating upper body movements during gait in healthy children and children with diplegic cerebral palsy. Journal of Pediatric Orthopedics Part B, 16, 175–180. Schmid, S., Schweizer, K., Romkes, J., Lorenzetti, S., & Brunner, R. (2013). Secondary gait deviations in patients with and without neurological involvement: A systematic review. Gait and Posture, 37, 480–493. Schweizer, K., Brunner, R., & Romkes, J. (2014). Upper body movements in children with hemiplegic cerebral palsy walking with and without an ankle-foot orthesis. Clinical Biomechanics. http://dx.doi.org/10.1016/j.clinbiomech.2014.02.005 Soderkvist, I., & Wedin, P. A. (1993). Determining the movements of the skeleton using well-configured markers. Journal of Biomechanics, 26, 1473–1477. Stebbins, J., Harrington, M., Thompson, N., Zavatsky, A., & Theologis, T. (2010). Gait compensations caused by foot deformity in cerebral palsy. Gait and Posture, 32, 226–230. Van de Walle, P., Hallemans, A., Truijen, S., Gosselink, R., Heyrman, L., Molenaers, G., et al. (2012). Increased mechanical cost of walking in children with diplegia: The role of the passenger unit cannot be neglected. Research in Developmental Disabilities, 33, 1996–2003. Whittle, M. W., & Levine, D. (1997). Measurement of lumbar lordosis as a component of clinical gait analysis. Gait and Posture, 5, 101–107. Woltring, H. J. (1995). Smoothing and differentiation techniques applied to 3-D data. In P. Allard, I. A. Stokes, J.-P. Blanchi (Eds.), Three-dimensional analysis of human movement. Champaign: Human Kinetics Publishers. Zadnikar, M., & Kastrin, A. (2011). Effects of hippotherapy and therapeutic horseback riding on postural control or balance in children with cerebral palsy: A metaanalysis. Developmental Medicine and Child Neurology, 53, 684–691.
