Gait & Posture 40 (2014) 633–639
Contents lists available at ScienceDirect
Gait & Posture journal homepage: www.elsevier.com/locate/gaitpost
Full length article
Mechanical work and energy consumption in children with cerebral palsy after single-event multilevel surgery Valeria Marconi a,*, He´le`n Hachez b, Anne Renders c, Pierre-Louis Docquier d, Chrisitine Detrembleur e a
Department of Neurological, and Movement Sciences, University of Verona, via Casorati, 43, Verona, Italy Service de Me´decine physique et Re´adaptation, Cliniques universitaires Saint-Luc, Avenue Hippocrate, 10–1200 Brussels, Belgium Service de Me´decine physique et Re´adaptation, Cliniques universitaires Saint-Luc, Avenue Hippocrate, 10–1200 Brussels, Belgium d Service de Chirurgie orthope´dique, Cliniques universitaires Saint-Luc, Avenue Hippocrate, 10–1200 Brussels, Belgium e Institute of Neurosciences, universite´ catholique de Louvain, Avenue Mounier 53 B1, 5304–1200 Brussels, Belgium b c
A R T I C L E I N F O
A B S T R A C T
Article history: Received 11 February 2014 Received in revised form 14 July 2014 Accepted 16 July 2014
Multilevel surgery is commonly performed to improve walking in children with cerebral palsy (CP). Classical gait analysis (kinetics, kinematics) demonstrated positive outcomes after this intervention, however it doesn’t give global indication about gait’s features. The assessment of energy cost and mechanical work of locomotion can provide an overall description of walking functionality. Therefore, we propose to describe the effects of multilevel surgery in children with CP, considering energetics, mechanical work, kinetic and kinematic of walking. We measured external, internal, total work, energy cost, recovery, efficiency, kinetic and kinematic of walking in 10 children with CP (4 girls, 6 boys; 13 years 2) before and 1 year after multilevel surgery. Kinetic and kinematic results are partially comparable to previous findings, energy cost of walking is significantly reduced (p < 0.05); external, internal, total work, recovery, efficiency are not significantly different (p = 0.129; p = 0.147; p = 0.795; p = 0.119; p = 0.21). The improvement of the walking’s energy consumption is not accompanied by a corresponding improvement of mechanical work. Therefore it is conceivable that the improvement of walking economy depend on a reduced effort of the muscle to maintain the posture, rather then to an improvement of the mechanism of energy recovery typical of human locomotion. ß 2014 Elsevier B.V. All rights reserved.
Keywords: Cerebral palsy Energetics Walking Multilevel surgery Mechanics
1. Introduction Children with cerebral palsy (CP) are characterized by many heterogeneous disorders that unavoidably influence the functions related to everyday activities. Walking is one of the most compromised functions in CP, therefore this condition affects participation and quality of life of this population. Nowadays single-event multilevel surgery (SEMS) is commonly performed to improve walking in children with CP [1–5]. The successful outcomes of this procedure have been demonstrated by several studies. For instance, it has been shown that it can improve the capability to achieve sagittal plane balance [5]; that the kinematic
* Corresponding author. Tel.: +0039 045 842 5139; fax: +0039 045 842 5131. E-mail addresses: [email protected], [email protected] (V. Marconi), [email protected] (H. Hachez), [email protected] (A. Renders), [email protected] (P.-L. Docquier), [email protected] (C. Detrembleur). http://dx.doi.org/10.1016/j.gaitpost.2014.07.014 0966-6362/ß 2014 Elsevier B.V. All rights reserved.
and kinetic of walking become more similar to the one of the healthy population, and therefore more advantageous [1]; the speed and the stride length can change significantly [1]; without improvement of the GMFM-66 [5]. In the last years, gait analysis has become a very common tool for the assessment of several clinical populations, and for the correct managing of the treatment. Indeed the information obtained by this evaluation procedure is considered necessary for planning the surgical interventions and it represents a good instrument to verify the treatment’s effectiveness [1–8]. Classical gait analysis focuses on the body segments, joint angles, muscular moments and powers during the gait cycle. Although this method provides detailed information about the walking’s mechanic, it does not provide directly evidences of treatment effectiveness in the everyday life. The measure of energy cost (C) and mechanical work (W) of locomotion can be useful since they consist of global index of walking’s functionality [9]. Many studies demonstrated that C tends to be larger in CP than in normal developing children [10–17]. One of the explanations of
V. Marconi et al. / Gait & Posture 40 (2014) 633–639
634
metabolic measurements. The patient was walking on a treadmill at spontaneous speed. Six CCD infrared cameras (Elite system V5, BTS; Milan, Italy) at 100 Hz, were used to collect the kinematic data, recording the coordinates in the 3 spatial planes of 19 reflective markers positioned on specific anatomical landmarks (according to the Davis’ protocol [19]. From kinematic data, the angular displacements of pelvis, hip, knee and ankle were calculated in the 3 planes. Spatiotemporal parameters were assessed using the 3D position data of foot (cadence, step length, and percentage of stance phase duration). The same speed was maintained in the first and in the second assessment.
this low economy of walking, consist in an increased mechanical work in the CP population [17], likely due to an exaggerated vertical displacement of the body centre of mass (COMb). The evaluation of C has become quite common in clinical assessment of walking, for many types of pathologies [6–9]. Whereas, W is not a common evaluation parameter in clinical environment, maybe because requires more complicated computational procedures. The evaluation of W, might allow to understand whether the modification of the gait, due to SEMS, have actually a significant impact on the pendulum like mechanism of energy recovery, typical of human walking. An improvement of walking’s pendulum mechanism, would mean a reduction of the total work of the muscles to enable the body’s advance, and of the energy waste, and hence it should be reflected in reduction of C. This method could provide some new information about the efficacy of SEMS to improve walking’s functionality. The present study aims to describe the energy consumption and the mechanical work of locomotion, combined with kinetics and kinematics, in CP children, before and one year after SEMS.
2.2. Gait analysis
2.2.2. Mechanical data Total positive mechanical work (Wtot) performed by the muscles was calculated by the sum of external work (Wext) (performed to move the body centre of mass (COMb)), and internal work (Wint) (performed to move the body segments relative to the COMb). Wext was computed by the measurement of the ground reaction force in the three directions, by means of four strain gauges located under each the four corners of the treadmill [20,21]. From the ground reaction forces, by an integration of the COMb’s acceleration, it was possible to calculate the COMb’s speed (V), then kinetic energy in the three directions was calculated (forward: Ekf = m Vf2 21; vertical: Ekv = m Vv2 21; lateral: Ekl = m Vl2 21; where m = mass). By a second integration of Ekv it was possible to calculate the vertical displacement (Sv) of COMb, and then gravitational potential energy (Ep = m g Sv). The energy necessary to move COMb in the three directions (Wekf; Wekv; Wekl) was calculated by the sum of the increments of each curve Ekv, Ekl and Ev (Ev calculated by the sum of Ekf and Ep). Total Wext performed by the muscles was calculated by adding Ekf, Ekv, Ep curves, to obtain the total mechanical energy curve. Then by adding the increments of the total mechanical energy curve it was possible to calculate the total mechanical energy necessary to move COMb. Wint was calculated by the kinematic, according to the method of Willems [22]. The body was divided in 7 rigid segments: head– arm–trunk (H.A.T.), thighs, shanks, and feet. For each segment rotational and translational energy were calculated. Then the internal mechanical energy/time curve of thigh, shank and foot of each lower limb, were summed up together to take into account the transfer effect. Then the internal mechanical energy of H.A.T. and limbs were summed up. Percentage recovery was calculate by the following equation: 100 [(Wekf + Wekv + Wext)/(Wekf + Wekv), as proposed by Cavagna [20].
2.2.1. Kinematic data Gait was assessed using a 3D Gait Analysis motion system, which included synchronous kinematic, kinetic, mechanic and
2.2.3. Kinetic data By the four strain gauges it was possible to recorder the ground reactions forces [20] and deduced the force under each foot [23]. By
2. Material and methods The study was conducted in accordance with the 1964 Declaration of Helsinki. The ethical committee of the Clinique gave his approval to the aims, protocols and methods of the study. The children and their families were informed about the procedures and the relative risks about the assessment and surgical procedures, in clinical routine practice. This is a prospective study. Ten children with diplegic, hemiplegic and tetraplegic CP who had undergone a SEMS (4 girls, 6 boys; age: 13 years 2; height: 149.3 cm 8.8; weight: 39.4 kg 9.8) participated to this study (GMFCS between I and III). The patients were enrolled in the study between 2005 and 2010, and they processed to SEMS between 2005 and 2009. The anthropometrical and identifying data are reported in Table 1. All children were processed to a clinical examination and gait analysis, before and after SEMS (5.5 3.11 months before SEMS, and 12.4 1.6 months after). 2.1. Clinical assessments The clinical scale used for the walking evaluation was a translated Gillette Scale in French [18].
Table 1 Age, anthropometric and clinical data of the subjects before and after surgery. Subject
1 2 3 4 5 6 7 8 9 10 Average St. Dev.
Sex
f m m f f f m m m m
Age
Height (m)
Weight (Kg)
yy
mo
Total months
Pre
Post
Pre
Post
14 15 16 11 12 14 13 13 9 16 13.30 2.21
9 3 5 1 11 1 1 7 3 8 4.90 3.67
177 183 197 133 155 169 157 163 111 200 164.50 27.57
1.63 1.61 1.53 1.44 1.52 1.45 1.49 1.40 1.37 1.72 1.52 0.11
1.63 1.65 1.61 1.46 1.57 1.45 1.58 1.50 1.43 178 1.57 0.11
63 45 38 41 38 35 34 33 29 58 41.28 10.97
61 50 38 45 53 37 40 38 32 63 45.65 10.56
GMFCS
Involvement
Gait Pattern
1 3 2 2 2 1 1 1 2 2
Bilateral Bilateral Unilateral Unilateral Bilateral Unilateral Unilateral Unilateral Bilateral Bilateral
Jump Crouch Jump Jump Jump Jump Jump Crouch Jump Crouch
V. Marconi et al. / Gait & Posture 40 (2014) 633–639
635
Table 2 Surgical procedures, and botulin toxin injection for each patient. Subject
Procedures (side)
1 2 3
Z-lengthening hamstring (R/L); rectus femoris transfer (R/L); Z-lengthening of psoas (R/L) Femoral derotation osteotomy (L); Z-lengthening hamstring (R/L); tibialis anterior transfer to the cuboid bone (L) Femoral derotation osteotomy (R); Z-lengthening hamstring (R); psoas tenotomy at the pelvic rim (R); tibialis anterior transfer to the cuboid bone (R); rectus femoris tenotomy (R) Femoral derotation osteotomy (R); psoas tenotomy at the pelvic rim (R); Z-lengthening of gastrocnemius (R); tibialis anterior transfer to the cuboid bone (R) Femoral derotation osteotomy (R); hamstring aponeurotomy (R/L); rectus femoris tenotomy (R/L); Z-lengthening of psoas (R/ L); gastrocnemius aponeurotomy (L); tibialis anterior transfer to the cuboid bone (L) Femoral derotation osteotomy (L); Z-lengthening hamstring (L); rectus femoris tenotomy (L); gastrocnemius aponeurotomy (L); tibialis anterior transfer to the cuboid bone (L) Femoral derotation osteotomy (R); Hamstring aponeurotomy (R); Gastrocnemius aponeurotomy (R); Femoral derotation osteotomy (R); Z-lengthening hamstring (R); gastrocnemius aponeurotomy (R) Femoral derotation osteotomy (R/L); Z-lengthening hamstring (R/L); gastrocnemius aponeurotomy (R/L); psoas tenotomy at the pelvic rim (R/L) Femoral derotation osteotomy (R/L); hamstring aponeurotomy (R/L); rectus femoris tenotomy (R); psoas tenotomy at the pelvic rim (R/L); gastrocnemius aponeurotomy (L); Z-lengthening of achiles tendon (R).
4 5 6 7 8 9 10
Botulin toxin
the method of the inverse dynamic, by the ground reactions force, the kinematic, and the anthropometric data it was possible to compute joint moment and power. The power of each joint was calculated as the product of the angular speed of the net moment. 2.2.4. Energetic data During exercise V?O2 was measured breath by breath (Quark b2, Cosmed; Rome, Italy). The energetic measure began with a rest period in the upright position. During exercise, V?O2 was measured until the steady state was reached and maintained for at least 3 min. C was calculated by subtracting V?O2r (measured during rest) from V?O2ss (measured during exercise at the steady of the cardiopulmonary parameters): V?O2net = V?O2ss V?O2r. Then the V?O2net was divided by the speed. C was then converted from O2 ml in Joule, by multiplying the O2 ml by the energetic equivalent of O2 [9,24]. The efficiency of work production was calculated by the ratio: Wtot C1.
Gastrocnemius (R/L)
Psoas Psoas Rectus femoris
Adductores
2.3. Surgery For five children the two lower limbs were operated on simultaneously. For 2 others, the right side and for 3 others, the left side was operated on. Nine of the children were operated on in the same hospital by the same surgeon. One was operated in another hospital by another surgeon. In total, 62 procedures were performed (see Table 2). 2.4. Data analysis For each patient, joint angles of ankle, knee, hip and pelvis were calculated in the three planes (sagittal, frontal, transverse) and plotted as function of percentages of the cycle. Moments and powers were calculated only in sagittal plane. On each curve, joint, moment and power parameters corresponding to precise percentage of the gait cycle, and to maximum and minimum values were calculated (see Table 3 and Fig. 1).
Table 3 Description of the kinetic and kinematic parameter. Angle
Pelvis
Mean pelvis in frontal plane
Hip
Mean pelvis in sagittal plane Mean pelvis in transverse plane Mean hip in frontal plane
Knee
Ankle
Moment and power
Hip
Knee
Ankle
Mean hip in transverse plane Maximal hip flexion in sagittal plane H1 Maximal hip extension in sagittal plane H2 Position of knee at initial contact (K1) Knee flexion in 20% of stride (K2) Knee in extension in 50% of stride (K3) Maximal knee flexion in swing phase (K4) Position of ankle in initial contact (A1) Maximal dorsiflexion of ankle in stance (A2) Maximal plantarflexion of ankle in stance (A3) Position of ankle in 85% of stride (A4) Mean of ankle in transverse plane Maximal extension moment of hip (Mz.H.Ext) Maximal flexion moment of hip (Mz.H.Flex) Maximal positive hip power (Pw.H1) maximal negative hip power (Pw.H2) Value of moment at 20% of stride (Mz.K1) Mz.K2 Pw.K1 Value of ankle moment at 20% of stride (Mz.A1) Maximal plantar flexion moment of ankle (Mz.A2) Maximal positive ankle power (Pw.A1)
Mean of the angle between the pelvis and in the three planes
Mean of the angle between the pelvis and in the frontal and transverse planes Maximal flexion Maximal extension Angle at the heel contact
Angle at the heel contact Corresponding to the maximal flexion Corresponding to 70% of the gait cycle Corresponding to 85% of the gait cycle Mean of the angle between the ankle in the transverse plane
Value of the moment at 20% gait cycle Maximal value Maximal negative power of knee
[(Fig._1)TD$IG]
636
V. Marconi et al. / Gait & Posture 40 (2014) 633–639
Fig. 1. Upper graphs show the evolution of hip, knee and ankle displacements in sagittal plane as function of stride (in%). Middle graphs show the evolution of hip, knee and ankle moment in sagittal plane as function of stride (in%). Lower graphs show the evolution of hip, knee and ankle power in sagittal plane as function of stride (in%). The arrows indicate the considered kinetic and kinematic parameters. The grey zone represents the reference values for healthy population of a comparable age (data collected on 6 voluntary children on the same laboratory).
2.4.1. Statistical analysis All parameters were compared before and after the operation by means of a Student paired t-test (SigmaStat 3.5, Systat Software; San Jose, California, USA). If they did not satisfy the assumption of normal distribution, they were analysed by the non-parametric Mann–Whitney rank sum test. To consider the family-wise error for the items that include more parameters, the significance level was modified according with the Bonferroni correction. 3. Results 3.1. Spatiotemporal parameters Spatiotemporal parameters were not significant different before and after SEMS: step length (p = 0.394), cadence (p = 0.073) and stance phase duration (p = 0.374) (see Table 4). 3.2. Kinematic The mean ankle displacement on the transverse plane was significantly different. Hip maximal flexion (H1) and extension (H2) in the sagittal plane, were significantly different after SEMS (p < 0.001). Three of the four parameters for the knee’s flexion-extension angles (K1–K3) decreased significantly postoperatively, becoming more similar to the gait cycle typical of typically developing children.
Considering the ankle, no one of the parameters was significantly different (see Table 4 and Fig. 2). 3.3. Kinetic Concerning joint moment, and power, no one of the parameters was significantly different after SEMS (see Table 4). 3.4. Mechanical work and cost External Work was not significantly different before and after the intervention (p = 0.129); even in the forward, vertical and lateral direction (p = 0.846; p = 0.75; p = 0.25). Although the difference did not reach the significance level, total Wext, and Wext in the vertical and lateral directions tended to decrease, and Wext in the forward direction tend to increase. After SEMS, Wint was not different (0.25 before and 0.25 after, p = 0.795); Wtot tended to decrease but it did not reach a significantly level (p = 0.147); C was significant reduced (p < 0.05); efficiency and recovery were larger but also in this case there were no significance (p = 0.21; p = 0.119). All the results are reported in Table 4.
4. Discussion The main aim of this study was to evaluate the outcome of SEMS in 10 CP children, considering kinetic, kinematic, energetic and
V. Marconi et al. / Gait & Posture 40 (2014) 633–639
637
Table 4 ** Means that the p-value is lower then 0.001, * means that the p-value is between 0.001 and 0.05. The p-value for the items that include more parameters, is corrected by the Bonferroni correction, the significance is indicated by means of y. Parameter
t-Test
Mann–Whitney Pre
Step length (m) Cadence (step min1) Stance phase (%) Mean position (8) Mean pelvis frontal Mean pelvis sagittal Mean pelvis transverse Mean hip frontal Mean hip transverse Mean ankle transverse Hip H1 Hip H2 Knee K1 Knee K2 Knee K3 Knee K4 Ankle A1 Ankle A2 Ankle A3 Ankle A4 H1–H2 K2–K1 K4–K3 A2–A1 A3–A2 Moment (N m kg1) Mz.H.Ext. Mz.H.Flex. Mz.K1 Mz.K2 Mz.A1 Mz.A2 Power (W kg1) Pw.H1 Pw.H2 Pw.K1 Pw.K.max Pw.A1 Mechanical work (J kg1 m1) Wext Wekf Wev Wekl Wint Wtot Recovery (%) Cost (J kg1 m1) Efficiency (%) Gillette scale score
** **y **y **y **y **y
pre
p
Mean
sd
Mean
sd
0.394 0.073 0.374
0.465 110.4 65.4
0.07 8.7 2.2
0.481 106.3 65.7
0.07 7.8 2.9
0.398 0.06 0.925 0.256 0.069
Contents lists available at ScienceDirect
Gait & Posture journal homepage: www.elsevier.com/locate/gaitpost
Full length article
Mechanical work and energy consumption in children with cerebral palsy after single-event multilevel surgery Valeria Marconi a,*, He´le`n Hachez b, Anne Renders c, Pierre-Louis Docquier d, Chrisitine Detrembleur e a
Department of Neurological, and Movement Sciences, University of Verona, via Casorati, 43, Verona, Italy Service de Me´decine physique et Re´adaptation, Cliniques universitaires Saint-Luc, Avenue Hippocrate, 10–1200 Brussels, Belgium Service de Me´decine physique et Re´adaptation, Cliniques universitaires Saint-Luc, Avenue Hippocrate, 10–1200 Brussels, Belgium d Service de Chirurgie orthope´dique, Cliniques universitaires Saint-Luc, Avenue Hippocrate, 10–1200 Brussels, Belgium e Institute of Neurosciences, universite´ catholique de Louvain, Avenue Mounier 53 B1, 5304–1200 Brussels, Belgium b c
A R T I C L E I N F O
A B S T R A C T
Article history: Received 11 February 2014 Received in revised form 14 July 2014 Accepted 16 July 2014
Multilevel surgery is commonly performed to improve walking in children with cerebral palsy (CP). Classical gait analysis (kinetics, kinematics) demonstrated positive outcomes after this intervention, however it doesn’t give global indication about gait’s features. The assessment of energy cost and mechanical work of locomotion can provide an overall description of walking functionality. Therefore, we propose to describe the effects of multilevel surgery in children with CP, considering energetics, mechanical work, kinetic and kinematic of walking. We measured external, internal, total work, energy cost, recovery, efficiency, kinetic and kinematic of walking in 10 children with CP (4 girls, 6 boys; 13 years 2) before and 1 year after multilevel surgery. Kinetic and kinematic results are partially comparable to previous findings, energy cost of walking is significantly reduced (p < 0.05); external, internal, total work, recovery, efficiency are not significantly different (p = 0.129; p = 0.147; p = 0.795; p = 0.119; p = 0.21). The improvement of the walking’s energy consumption is not accompanied by a corresponding improvement of mechanical work. Therefore it is conceivable that the improvement of walking economy depend on a reduced effort of the muscle to maintain the posture, rather then to an improvement of the mechanism of energy recovery typical of human locomotion. ß 2014 Elsevier B.V. All rights reserved.
Keywords: Cerebral palsy Energetics Walking Multilevel surgery Mechanics
1. Introduction Children with cerebral palsy (CP) are characterized by many heterogeneous disorders that unavoidably influence the functions related to everyday activities. Walking is one of the most compromised functions in CP, therefore this condition affects participation and quality of life of this population. Nowadays single-event multilevel surgery (SEMS) is commonly performed to improve walking in children with CP [1–5]. The successful outcomes of this procedure have been demonstrated by several studies. For instance, it has been shown that it can improve the capability to achieve sagittal plane balance [5]; that the kinematic
* Corresponding author. Tel.: +0039 045 842 5139; fax: +0039 045 842 5131. E-mail addresses: [email protected], [email protected] (V. Marconi), [email protected] (H. Hachez), [email protected] (A. Renders), [email protected] (P.-L. Docquier), [email protected] (C. Detrembleur). http://dx.doi.org/10.1016/j.gaitpost.2014.07.014 0966-6362/ß 2014 Elsevier B.V. All rights reserved.
and kinetic of walking become more similar to the one of the healthy population, and therefore more advantageous [1]; the speed and the stride length can change significantly [1]; without improvement of the GMFM-66 [5]. In the last years, gait analysis has become a very common tool for the assessment of several clinical populations, and for the correct managing of the treatment. Indeed the information obtained by this evaluation procedure is considered necessary for planning the surgical interventions and it represents a good instrument to verify the treatment’s effectiveness [1–8]. Classical gait analysis focuses on the body segments, joint angles, muscular moments and powers during the gait cycle. Although this method provides detailed information about the walking’s mechanic, it does not provide directly evidences of treatment effectiveness in the everyday life. The measure of energy cost (C) and mechanical work (W) of locomotion can be useful since they consist of global index of walking’s functionality [9]. Many studies demonstrated that C tends to be larger in CP than in normal developing children [10–17]. One of the explanations of
V. Marconi et al. / Gait & Posture 40 (2014) 633–639
634
metabolic measurements. The patient was walking on a treadmill at spontaneous speed. Six CCD infrared cameras (Elite system V5, BTS; Milan, Italy) at 100 Hz, were used to collect the kinematic data, recording the coordinates in the 3 spatial planes of 19 reflective markers positioned on specific anatomical landmarks (according to the Davis’ protocol [19]. From kinematic data, the angular displacements of pelvis, hip, knee and ankle were calculated in the 3 planes. Spatiotemporal parameters were assessed using the 3D position data of foot (cadence, step length, and percentage of stance phase duration). The same speed was maintained in the first and in the second assessment.
this low economy of walking, consist in an increased mechanical work in the CP population [17], likely due to an exaggerated vertical displacement of the body centre of mass (COMb). The evaluation of C has become quite common in clinical assessment of walking, for many types of pathologies [6–9]. Whereas, W is not a common evaluation parameter in clinical environment, maybe because requires more complicated computational procedures. The evaluation of W, might allow to understand whether the modification of the gait, due to SEMS, have actually a significant impact on the pendulum like mechanism of energy recovery, typical of human walking. An improvement of walking’s pendulum mechanism, would mean a reduction of the total work of the muscles to enable the body’s advance, and of the energy waste, and hence it should be reflected in reduction of C. This method could provide some new information about the efficacy of SEMS to improve walking’s functionality. The present study aims to describe the energy consumption and the mechanical work of locomotion, combined with kinetics and kinematics, in CP children, before and one year after SEMS.
2.2. Gait analysis
2.2.2. Mechanical data Total positive mechanical work (Wtot) performed by the muscles was calculated by the sum of external work (Wext) (performed to move the body centre of mass (COMb)), and internal work (Wint) (performed to move the body segments relative to the COMb). Wext was computed by the measurement of the ground reaction force in the three directions, by means of four strain gauges located under each the four corners of the treadmill [20,21]. From the ground reaction forces, by an integration of the COMb’s acceleration, it was possible to calculate the COMb’s speed (V), then kinetic energy in the three directions was calculated (forward: Ekf = m Vf2 21; vertical: Ekv = m Vv2 21; lateral: Ekl = m Vl2 21; where m = mass). By a second integration of Ekv it was possible to calculate the vertical displacement (Sv) of COMb, and then gravitational potential energy (Ep = m g Sv). The energy necessary to move COMb in the three directions (Wekf; Wekv; Wekl) was calculated by the sum of the increments of each curve Ekv, Ekl and Ev (Ev calculated by the sum of Ekf and Ep). Total Wext performed by the muscles was calculated by adding Ekf, Ekv, Ep curves, to obtain the total mechanical energy curve. Then by adding the increments of the total mechanical energy curve it was possible to calculate the total mechanical energy necessary to move COMb. Wint was calculated by the kinematic, according to the method of Willems [22]. The body was divided in 7 rigid segments: head– arm–trunk (H.A.T.), thighs, shanks, and feet. For each segment rotational and translational energy were calculated. Then the internal mechanical energy/time curve of thigh, shank and foot of each lower limb, were summed up together to take into account the transfer effect. Then the internal mechanical energy of H.A.T. and limbs were summed up. Percentage recovery was calculate by the following equation: 100 [(Wekf + Wekv + Wext)/(Wekf + Wekv), as proposed by Cavagna [20].
2.2.1. Kinematic data Gait was assessed using a 3D Gait Analysis motion system, which included synchronous kinematic, kinetic, mechanic and
2.2.3. Kinetic data By the four strain gauges it was possible to recorder the ground reactions forces [20] and deduced the force under each foot [23]. By
2. Material and methods The study was conducted in accordance with the 1964 Declaration of Helsinki. The ethical committee of the Clinique gave his approval to the aims, protocols and methods of the study. The children and their families were informed about the procedures and the relative risks about the assessment and surgical procedures, in clinical routine practice. This is a prospective study. Ten children with diplegic, hemiplegic and tetraplegic CP who had undergone a SEMS (4 girls, 6 boys; age: 13 years 2; height: 149.3 cm 8.8; weight: 39.4 kg 9.8) participated to this study (GMFCS between I and III). The patients were enrolled in the study between 2005 and 2010, and they processed to SEMS between 2005 and 2009. The anthropometrical and identifying data are reported in Table 1. All children were processed to a clinical examination and gait analysis, before and after SEMS (5.5 3.11 months before SEMS, and 12.4 1.6 months after). 2.1. Clinical assessments The clinical scale used for the walking evaluation was a translated Gillette Scale in French [18].
Table 1 Age, anthropometric and clinical data of the subjects before and after surgery. Subject
1 2 3 4 5 6 7 8 9 10 Average St. Dev.
Sex
f m m f f f m m m m
Age
Height (m)
Weight (Kg)
yy
mo
Total months
Pre
Post
Pre
Post
14 15 16 11 12 14 13 13 9 16 13.30 2.21
9 3 5 1 11 1 1 7 3 8 4.90 3.67
177 183 197 133 155 169 157 163 111 200 164.50 27.57
1.63 1.61 1.53 1.44 1.52 1.45 1.49 1.40 1.37 1.72 1.52 0.11
1.63 1.65 1.61 1.46 1.57 1.45 1.58 1.50 1.43 178 1.57 0.11
63 45 38 41 38 35 34 33 29 58 41.28 10.97
61 50 38 45 53 37 40 38 32 63 45.65 10.56
GMFCS
Involvement
Gait Pattern
1 3 2 2 2 1 1 1 2 2
Bilateral Bilateral Unilateral Unilateral Bilateral Unilateral Unilateral Unilateral Bilateral Bilateral
Jump Crouch Jump Jump Jump Jump Jump Crouch Jump Crouch
V. Marconi et al. / Gait & Posture 40 (2014) 633–639
635
Table 2 Surgical procedures, and botulin toxin injection for each patient. Subject
Procedures (side)
1 2 3
Z-lengthening hamstring (R/L); rectus femoris transfer (R/L); Z-lengthening of psoas (R/L) Femoral derotation osteotomy (L); Z-lengthening hamstring (R/L); tibialis anterior transfer to the cuboid bone (L) Femoral derotation osteotomy (R); Z-lengthening hamstring (R); psoas tenotomy at the pelvic rim (R); tibialis anterior transfer to the cuboid bone (R); rectus femoris tenotomy (R) Femoral derotation osteotomy (R); psoas tenotomy at the pelvic rim (R); Z-lengthening of gastrocnemius (R); tibialis anterior transfer to the cuboid bone (R) Femoral derotation osteotomy (R); hamstring aponeurotomy (R/L); rectus femoris tenotomy (R/L); Z-lengthening of psoas (R/ L); gastrocnemius aponeurotomy (L); tibialis anterior transfer to the cuboid bone (L) Femoral derotation osteotomy (L); Z-lengthening hamstring (L); rectus femoris tenotomy (L); gastrocnemius aponeurotomy (L); tibialis anterior transfer to the cuboid bone (L) Femoral derotation osteotomy (R); Hamstring aponeurotomy (R); Gastrocnemius aponeurotomy (R); Femoral derotation osteotomy (R); Z-lengthening hamstring (R); gastrocnemius aponeurotomy (R) Femoral derotation osteotomy (R/L); Z-lengthening hamstring (R/L); gastrocnemius aponeurotomy (R/L); psoas tenotomy at the pelvic rim (R/L) Femoral derotation osteotomy (R/L); hamstring aponeurotomy (R/L); rectus femoris tenotomy (R); psoas tenotomy at the pelvic rim (R/L); gastrocnemius aponeurotomy (L); Z-lengthening of achiles tendon (R).
4 5 6 7 8 9 10
Botulin toxin
the method of the inverse dynamic, by the ground reactions force, the kinematic, and the anthropometric data it was possible to compute joint moment and power. The power of each joint was calculated as the product of the angular speed of the net moment. 2.2.4. Energetic data During exercise V?O2 was measured breath by breath (Quark b2, Cosmed; Rome, Italy). The energetic measure began with a rest period in the upright position. During exercise, V?O2 was measured until the steady state was reached and maintained for at least 3 min. C was calculated by subtracting V?O2r (measured during rest) from V?O2ss (measured during exercise at the steady of the cardiopulmonary parameters): V?O2net = V?O2ss V?O2r. Then the V?O2net was divided by the speed. C was then converted from O2 ml in Joule, by multiplying the O2 ml by the energetic equivalent of O2 [9,24]. The efficiency of work production was calculated by the ratio: Wtot C1.
Gastrocnemius (R/L)
Psoas Psoas Rectus femoris
Adductores
2.3. Surgery For five children the two lower limbs were operated on simultaneously. For 2 others, the right side and for 3 others, the left side was operated on. Nine of the children were operated on in the same hospital by the same surgeon. One was operated in another hospital by another surgeon. In total, 62 procedures were performed (see Table 2). 2.4. Data analysis For each patient, joint angles of ankle, knee, hip and pelvis were calculated in the three planes (sagittal, frontal, transverse) and plotted as function of percentages of the cycle. Moments and powers were calculated only in sagittal plane. On each curve, joint, moment and power parameters corresponding to precise percentage of the gait cycle, and to maximum and minimum values were calculated (see Table 3 and Fig. 1).
Table 3 Description of the kinetic and kinematic parameter. Angle
Pelvis
Mean pelvis in frontal plane
Hip
Mean pelvis in sagittal plane Mean pelvis in transverse plane Mean hip in frontal plane
Knee
Ankle
Moment and power
Hip
Knee
Ankle
Mean hip in transverse plane Maximal hip flexion in sagittal plane H1 Maximal hip extension in sagittal plane H2 Position of knee at initial contact (K1) Knee flexion in 20% of stride (K2) Knee in extension in 50% of stride (K3) Maximal knee flexion in swing phase (K4) Position of ankle in initial contact (A1) Maximal dorsiflexion of ankle in stance (A2) Maximal plantarflexion of ankle in stance (A3) Position of ankle in 85% of stride (A4) Mean of ankle in transverse plane Maximal extension moment of hip (Mz.H.Ext) Maximal flexion moment of hip (Mz.H.Flex) Maximal positive hip power (Pw.H1) maximal negative hip power (Pw.H2) Value of moment at 20% of stride (Mz.K1) Mz.K2 Pw.K1 Value of ankle moment at 20% of stride (Mz.A1) Maximal plantar flexion moment of ankle (Mz.A2) Maximal positive ankle power (Pw.A1)
Mean of the angle between the pelvis and in the three planes
Mean of the angle between the pelvis and in the frontal and transverse planes Maximal flexion Maximal extension Angle at the heel contact
Angle at the heel contact Corresponding to the maximal flexion Corresponding to 70% of the gait cycle Corresponding to 85% of the gait cycle Mean of the angle between the ankle in the transverse plane
Value of the moment at 20% gait cycle Maximal value Maximal negative power of knee
[(Fig._1)TD$IG]
636
V. Marconi et al. / Gait & Posture 40 (2014) 633–639
Fig. 1. Upper graphs show the evolution of hip, knee and ankle displacements in sagittal plane as function of stride (in%). Middle graphs show the evolution of hip, knee and ankle moment in sagittal plane as function of stride (in%). Lower graphs show the evolution of hip, knee and ankle power in sagittal plane as function of stride (in%). The arrows indicate the considered kinetic and kinematic parameters. The grey zone represents the reference values for healthy population of a comparable age (data collected on 6 voluntary children on the same laboratory).
2.4.1. Statistical analysis All parameters were compared before and after the operation by means of a Student paired t-test (SigmaStat 3.5, Systat Software; San Jose, California, USA). If they did not satisfy the assumption of normal distribution, they were analysed by the non-parametric Mann–Whitney rank sum test. To consider the family-wise error for the items that include more parameters, the significance level was modified according with the Bonferroni correction. 3. Results 3.1. Spatiotemporal parameters Spatiotemporal parameters were not significant different before and after SEMS: step length (p = 0.394), cadence (p = 0.073) and stance phase duration (p = 0.374) (see Table 4). 3.2. Kinematic The mean ankle displacement on the transverse plane was significantly different. Hip maximal flexion (H1) and extension (H2) in the sagittal plane, were significantly different after SEMS (p < 0.001). Three of the four parameters for the knee’s flexion-extension angles (K1–K3) decreased significantly postoperatively, becoming more similar to the gait cycle typical of typically developing children.
Considering the ankle, no one of the parameters was significantly different (see Table 4 and Fig. 2). 3.3. Kinetic Concerning joint moment, and power, no one of the parameters was significantly different after SEMS (see Table 4). 3.4. Mechanical work and cost External Work was not significantly different before and after the intervention (p = 0.129); even in the forward, vertical and lateral direction (p = 0.846; p = 0.75; p = 0.25). Although the difference did not reach the significance level, total Wext, and Wext in the vertical and lateral directions tended to decrease, and Wext in the forward direction tend to increase. After SEMS, Wint was not different (0.25 before and 0.25 after, p = 0.795); Wtot tended to decrease but it did not reach a significantly level (p = 0.147); C was significant reduced (p < 0.05); efficiency and recovery were larger but also in this case there were no significance (p = 0.21; p = 0.119). All the results are reported in Table 4.
4. Discussion The main aim of this study was to evaluate the outcome of SEMS in 10 CP children, considering kinetic, kinematic, energetic and
V. Marconi et al. / Gait & Posture 40 (2014) 633–639
637
Table 4 ** Means that the p-value is lower then 0.001, * means that the p-value is between 0.001 and 0.05. The p-value for the items that include more parameters, is corrected by the Bonferroni correction, the significance is indicated by means of y. Parameter
t-Test
Mann–Whitney Pre
Step length (m) Cadence (step min1) Stance phase (%) Mean position (8) Mean pelvis frontal Mean pelvis sagittal Mean pelvis transverse Mean hip frontal Mean hip transverse Mean ankle transverse Hip H1 Hip H2 Knee K1 Knee K2 Knee K3 Knee K4 Ankle A1 Ankle A2 Ankle A3 Ankle A4 H1–H2 K2–K1 K4–K3 A2–A1 A3–A2 Moment (N m kg1) Mz.H.Ext. Mz.H.Flex. Mz.K1 Mz.K2 Mz.A1 Mz.A2 Power (W kg1) Pw.H1 Pw.H2 Pw.K1 Pw.K.max Pw.A1 Mechanical work (J kg1 m1) Wext Wekf Wev Wekl Wint Wtot Recovery (%) Cost (J kg1 m1) Efficiency (%) Gillette scale score
** **y **y **y **y **y
pre
p
Mean
sd
Mean
sd
0.394 0.073 0.374
0.465 110.4 65.4
0.07 8.7 2.2
0.481 106.3 65.7
0.07 7.8 2.9
0.398 0.06 0.925 0.256 0.069
