Gait and Posture 11 (2000) 224 – 232 www.elsevier.com/locate/gaitpost Common abnormal kinetic patterns of the knee in gait in spastic diplegia of cerebral palsy Chii-Jeng Lin a, Lan-Yuen Guo b,c, Fong-Chin Su b,*, You-Li Chou b, Rong-Ju Cherng d b a Department of Orthopaedic Surgery, National Cheng Kung Uni6ersity, Tainan, Taiwan Motion Analysis Laboratory, Institute of Biomedical Engineering, National Cheng Kung Uni6ersity, 1 Uni6ersity Road, Tainan 701, Taiwan c Department of Physical Therapy, Tzu Chi College of Technology, Hualian, Taiwan d Department of Physical Therapy, National Cheng Kung Uni6ersity, Tainan, Taiwan Received 13 May 1999; received in revised form 30 October 1999; accepted 29 December 1999 Abstract We studied the kinetic characteristics of the knee in patients with spastic diplegia. Twenty three children with spastic diplegia were recruited and had their 46 limbs categorised into the following four groups: jump (n= 7), crouch (n= 8), recurvatum (n =14) and mild (n=17). In the crouch pattern, the patients usually had a larger and longer lasting internal knee extensor moments in stance suggesting that rectus femoris had a relatively high activation. In the recurvatum pattern, the internal knee flexor moment was large and long lasting in stance. The biceps femoris showed less activity on EMG although the knee flexor moment was large and we concluded that the soft tissue behind the knee joint provided this flexor moment. In the jump knee pattern there was abnormal power generation at the knee and ankle joints in initial stance, which did not contribute to normal progression but aided upward body motion. In the mild group the kinetic data was similar to that seen in normal children. Knowledge of kinetic patterns in these patients may help in their subsequent management. © 2000 Elsevier Science B.V. All rights reserved. Keywords: Cerebral palsy; Spastic diplegia; Gait analysis; Kinetics; Moments; Powers 1. Introduction Acquisition of kinetic data requires more complicated procedures than for collection of kinematic data [1 – 8] but should provide a better understanding of pathological gait. A joint moment represents the body’s internal response to an external load [1 – 8]. During normal gait, the influence of soft tissues, other than muscles that are primary motion generators, on joint moment generation is minimal. However, in abnormal gait, the influence of soft tissue contracture on joint moments may become substantial and can give misleading information about moment and power curves. In such circumstances, dynamic EMG is important to differentiate the cause of the moment. If the joint agonist is not active on EMG, the moment could be produced by joint capsule and ligamentous structures * Corresponding author. Tel.: + 886-6-2757575, ext. 63422; fax: +886-6-2343270. E-mail address: [email protected] (F.-C. Su) [6,7]. Power is defined as the work performed per unit of time and may be used to document the net energy absorption or generation of the muscles [3–8]. More recently, joint kinetics, specifically joint moments and joint powers, have been available as an additional tool in the assessment of pathological gait [9–22]. In 1993, Sutherland and Davids classified the common gait abnormalities of the knee in cerebral palsy (CP) into four types: jump, crouch, recurvatum, and stiff . Jump knee gait was characterised by increased knee flexion in early stance phase, through initial double support, with correction of the knee wave to normal or near normal extension in mid-stance and late-stance. In crouch gait, there was increased knee flexion through the stance phase, with variable alignment in swing phase. Recurvatum knee gait described increased knee extension in mid-stance and late-stance phase, with variable knee motion in the swing phase. In comparison to the three gait patterns in stance phase they characterised stiff knee gait by a decreased dynamic range of motion of knee in swing phase. 0966-6362/00/$ - see front matter © 2000 Elsevier Science B.V. All rights reserved. PII: S 0 9 6 6 - 6 3 6 2 ( 0 0 ) 0 0 0 4 9 - 7 C.-J. Lin et al. / Gait and Posture 11 (2000) 224–232 Though the grouping of such gait abnormalities by kinematics is useful [23 – 28], the kinetic patterns were not been described. The goal of this study was therefore to identify and investigate the kinetics of these abnormal gait patterns in stance phase in diplegic CP. 2. Method Twenty three patients (16 boys, seven girls, mean age 9 years) with spastic diplegia from CP were studied. None had been operated on previously. The laboratory was equipped with an Expert Vision motion analysis system (Motion Analysis Corp., Santa Rosa, CA, USA), integrated with three Kistler force plates Type 9281B, Kistler Instrument Corp., Winterhur, Switzerland) and a MA-100 electromyography system (Motion Lab System, Inc., L.A., USA). The motion analysis system included six CCD cameras, two VP320 video processors, SUN workstation, trigger and 21 pieces of 3/4–1 inch reflective markers. Three Kistler force plates were mounted on a 10.7 ×2.05-m walkway. Two of them were 60×40 cm in size, and the other was 50 × 50 cm. There was additional accessory equipment, including three preamplifiers, two AD converters (MP 280) and one 486 IBM compatible personal computer. The electromyography (EMG) system comprised of a backpack, interface unit and ten pre-amplified surface electrodes. Twenty one reflective markers were placed on each subject based on the Helen – Hays marker set. The motion analysis system was synchronised with three force plates and the EMG system during data collection. The subjects were asked to walk at self-selected speed after several trials. The force plates were concealed and the procedure was video taped for later review. At least five successful trials were collected and stored. The sampling rate of the cameras was 60 Hz. 225 The sampling rates of the EMG and force plates were 1000 Hz. The anthropometric data of each subject were also measured and used for calculation of joint angles, joint moments and joint powers. The reflective markers locations were used to define the co-ordinate system of linkage. OrthoTrak II software (Motion Analysis Corp., Santa Rosa, CA, USA) was used to calculate the joint angles, joint reaction forces, joint moments and joint powers of lower extremity in gait cycle. Three normalised (100%) gait cycles for each subject were averaged before gait parameters could be determined. We divided the gait cycle into five gait events: heel strike (HS), opposite toe off (OTO), opposite heel strike (OHS), toe off (TO) and next heel strike after the method of Sutherland . The kinematic and kinetic data at each gait event and their maximum and minimum, beside the curves, was analysed statistically. We separated the gait patterns of the knee into four groups: mild, crouch, recurvatum and jump based on the knee kinematic patterns. The mild group was defined as those who walked better than the other three groups and those without specific knee kinematic patterns. The normal data was elicited from the OrthoTrak II. A one-way ANOVA was used to determine the differences in ground reaction force, moment and power characteristics in the hip, knee and ankle between the four groups and Tukey’s post-hoc test was used to determine the significance between each pair of the four groups. 3. Results The gait patterns of 46 limbs in 23 patients were divided into four groups, mild knee (n = 17), crouch knee (n= 8), recurvatum knee (n = 14), and jump knee (n= 7) basd on the knee kinematic patterns. 3.1. Sagittal plane joint angles 3.1.1. Knee The sagittal-plane motion of the knee joint (Fig. 1) in the four groups demonstrated specific characteristics. In the initial stance phase, crouch and jump groups had increased flexion. Throughout the whole stance phase, only the crouch group had persistently increased knee flexion while the flexion angles returned to almost normal range in the jump group. The recurvatum knee group, on the contrary, had increased knee extension in mid-and late-stance phase. Fig. 1. Mean sagittal plane knee motion of four cerebral palsy groups. 3.1.2. Hip The hip joint angle curves are shown in Fig. 2. The crouch group showed excessive hip flexion during the entire gait cycle. 226 C.-J. Lin et al. / Gait and Posture 11 (2000) 224–232 Fig. 2. Mean sagittal plane hip motion of four cerebral palsy groups. 3.1.3. Ankle The curve patterns of ankle joint angle had their own characteristics (Fig. 3) in different groups. The crouch group showed excessive ankle dorsiflexion during the whole stance phase, comparing with the normal children. Both jump and crouch groups had excessive dorsiflexion during initial contact. The ankle motion shifted to plantarflexion pattern in the jump group while in the crouch group remained in a dorsiflexion pattern. The recurvatum group demonstrated excessive plantarflexion during the whole stance phase. 3.2. Vertical ground reaction force The curve patterns of ground reaction force (GRF) had their own characteristics (Fig. 4) in different groups. The difference could be observed in loading, first peak, valley, second peak and unloading. The jump knee group had the most exaggerated pattern, including rapid loading, high first peak (P B 0.0001), deep valley, moderate second peak and normal unloading phase. Fig. 3. Mean sagittal plane ankle motion of four cerebral palsy groups. Fig. 4. Mean vertical ground reaction forces of four cerebral palsy groups. The crouch group had the flattest pattern between the first and second peaks. The valley was not apparent during weight transmission period in this group. The mild and recurvatum group had GRF curves similar to the crouch group, though the changes were relatively moderate. 3.3. Sagittal plane joint moments 3.3.1. Hip The hip joint moments are shown in Fig. 5. All CP groups had excessive torque at the hip. The crouch group had the greatest extensor moment than the other three groups in initial stance and mid-stance. In addition, the jump group tended to have the highest flexor moment in terminal stance and pre-swing. 3.3.2. Knee The differences of joint moments between groups were the most significant at the knee joint (Fig. 6). The crouch group had the greatest extensor moments throughout the stance phase. The recurvatum group, on the contrary, exhibited excessive flexor moments Fig. 5. Mean hip joint moments of four cerebral palsy groups. C.-J. Lin et al. / Gait and Posture 11 (2000) 224–232 227 Table 1 Knee joint momentsa at specific gait events and its peak values Knee moment, N m/kg flexor (+)/ extensor (−) Mild (n= 17) Crouch (n = 8) Heel strike Opposite toe off Opposite HS Toe off Heel strike Maximum knee flexion Maximum knee extension 0.04 −0.39 −0.05 −0.02 0.09 0.27 −0.47 −0.14 −0.68 −0.58 −0.06 0.14 0.20 −0.84 (0.37) (0.26) (0.13) (0.04) (0.04) (0.13) (0.27) (0.38) (0.22) (0.39) (0.13) (0.06) (0.05) (0.23) Recurvatum (n = Jump (n = 7) 14) F valueb 0.07 0.08 0.08 −0.03 0.08 0.45 −0.21 1.91 18.64* 18.70* 0.88 8.72* 6.66** 16.13* (0.20) (0.27) (0.11) (0.06) (0.03) (0.20) (0.23) −0.28 −0.70 −0.22 0.00 0.17 0.20 −0.88 (0.44) (0.32) (0.17) (0.06) (0.03) (0.07) (0.17) Post-hocc R\M,C,J R\J,C M,J\C J\M,R C\R R\M,J,C R\M,C,J M\C,J a Means and S.D. in parenthesis. Statistic was done with one-way ANOVA. c Tukey’s post-hoc test was used to determine the significance between each pair of the four groups. * PB0.001; ** PB0.01. b throughout the stance phase with two peaks at OTO and OHS. The jump group had the greatest extensor moment in the loading response, a rapid decrease at mid-stance and second peak at OHS. Statistically, at OTO, there were significant differences between the four groups in knee joint moment (P B 0.05, Table 1). In this study, the moments described are internal moments. The recurvatum group had knee flexor moment, 0.08 N m/kg, while the crouch and jump groups had extensor moments, 0.68 and 0.70 N m/kg, respectively. At OHS, there were significant differences in knee joint moment between the four groups (P B 0.05). While the recurvatum group was dominated by knee flexors, the other three groups were knee extensor dominant. In addition, knee extensor moment of crouch group was significantly greater than the jump and mild groups (P B0.05). There were significant differences between the four groups in the maximum knee flexor moment (PB0.05). The joint moment of the recurvatum group, 0.45 N m/kg, was significantly greater than the other three Fig. 6. Mean knee joint moments of four cerebral palsy groups. groups (PB0.05). There were also significant differences between the four groups in the maximum knee extensor moment (P B 0.05). The maximum knee extensor moment of the crouch and jump groups were significantly greater than those of the other two groups (PB 0.05). 3.3.3. Ankle The moment curves at the ankle joint are shown in Fig. 7. The crouch group had a rapidly increasing and constantly large plantar–flexor moment throughout stance phase. The jump knee group showed a rapidly increasing plantar–flexor moment at loading response, a rapid decrease at mid-stance and an unusual second increase at terminal stance. The mild group demonstrated normal dorsiflexor/plantar –flexor moments in the whole of stance, except for an initial rapid plantar– flexor moment increase at loading. At OTO, there were statistically significant differences between the four groups in ankle joint moment (PB 0.05, Table 2). The plantar–flexor moment of the jump and crouch groups increased rapidly. During single limb stance, the crouch group maintained high Fig. 7. Mean ankle joint moments of four cerebral palsy groups. 228 C.-J. Lin et al. / Gait and Posture 11 (2000) 224–232 Table 2 Ankle joint momentsa at specific gait events and its peak values Ankle moment, N m/kg dorsiflexion (+)/plantar flexion (−) Mild (n = 17) Crouch (n = 8) Heel strike Opposite toe off Opposite HS Toe off Heel strike Maximum ankle dorsiflexion Maximum ankle plantar flexion 0.08 −0.34 −0.81 −0.11 0.01 0.13 −1.00 −0.12 −0.67 −0.71 −0.07 0.01 0.02 −0.90 (0.36) (0.28) (0.19) (0.11) (0.02) (0.34) (0.28) Recurvatum (n = 14) (0.08) (0.34) (0.25) (0.14) (0.00) (0.01) (0.26) −0.14 −0.44 −0.53 −0.06 0.01 0.02 −0.79 (0.12) (0.16) (0.31) (0.09) (0.00) (0.02) (0.21) F valueb Jump (n = 7) −0.16 −0.66 −0.48 −0.02 0.01 0.08 −0.93 (0.20) (0.39) (0.28) (0.05) (0.00) (0.13) (0.34) 2.68 3.30* 3.61* 1.13 0.43 0.78 1.39 Post-hocc R\M a Means and S.D. in parenthesis. Statistic was done with one-way ANOVA. c Tukey’s post-hoc test was used to determine the significance between each pair of the four groups. * PB0.05. b torque value in plateau shape, but the jump group demonstrated a sine wave phenomenon. At OHS, there were significant differences between the four groups (P B 0.05). The plantar – flexor moment of the mild group, 0.81 N m/kg, was significantly greater than that of the recurvatum group was (P B0.05). 3.4. Joint powers The differences of the power curve at the hip between groups, similar to the moment curves, were not as significant as that at the knee joint (Fig. 8). For the sagittal plane hip power, the jump knee group had greater absorption at terminal stance. Its maximum absorption power, 1.09 W/kg, was significantly greater (P B 0.05) than those of the mild, crouch, and recurvatum groups, 0.36, 0.39, 0.48 W/kg, respectively (Table 3). For the knee power (Fig. 9), the jump knee group had a dramatically large generation at initial stance and excessive absorption power at pre-swing. Its maximum generation power is significantly greater (P B 0.05) than the other three groups with the value up to 1.88 W/kg (Table 4). Statistically, at OTO, there were significant Fig. 8. Mean hip joint powers of four cerebral palsy groups. differences between the four groups in knee joint power (PB 0.05). The joint power of the jump group, 1.24 W/kg, was significantly larger than the other three groups (PB 0.05). At OHS, there were also significant differences between the four groups in knee joint power (P B 0.05). The absorption power of the crouch group, 0.63 W/kg, was significantly greater than that of the recurvatum groups (P B 0.05). The ankle power is shown in Fig. 10. The jump group showed excessive absorption power in initial contact and a premature generation of power at the mid-stance phase. Except the mild group, the other three groups had larger absorption powers in early stance. Interestingly, all four CP groups were able to generate sufficient powers at push off. At push-off, the mild group had a larger power generation, 1.21 W/kg, than in the other three groups, 0.7–0.9 W/kg (Table 5). 3.5. EMG 3.5.1. Gastrocnemius All the abnormal knee groups, except the mild group, demonstrated excessive firing amplitude of gastrocnemius immediately after initial contact (Fig. 11). In contrast, the mild group showed slowly increasing firing Fig. 9. Mean knee joint powers of four cerebral palsy groups. C.-J. Lin et al. / Gait and Posture 11 (2000) 224–232 229 Table 3 Hip joint powersa at specific gait events and its peak values Hip power, W/kg generation (+)/absorption (−) Mild (n= 17) Crouch (n= 8) Recurvatum (n =14) Jump (n = 7) Heel strike Opposite toe off Opposite HS Toe off Heel strike Maximum hip generation Maximum hip absorption 0.57 0.44 −0.07 0.26 0.23 1.01 −0.36 0.22 0.59 −0.14 0.16 0.05 0.98 −0.39 0.67 1.01 −0.07 0.21 0.29 1.50 −0.48 0.01 0.24 −0.39 0.10 0.32 1.48 −1.09 (0.67) (0.38) (0.33) (0.20) (0.68) (0.65) (0.33) (0.39) (0.30) (0.30) (0.21) (0.07) (0.61) (0.24) (1.32) (1.16) (0.21) (0.26) (0.26) (1.10) (0.45) (0.87) (0.61) (0.50) (0.33) (0.25) (0.57) (0.48) F valueb Post-hocc 0.91 1.90 1.59 0.66 0.55 1.28 5.60* M,C,R\J a Means and S.D. in parenthesis. Statistic was done with one-way ANOVA. c Tukey’s post-hoc test was used to determine the significance between each pair of the four groups. * PB0.01. b amplitude, which reached its peak in mid-stance, similar to a normal subject. There was no muscle activity in the swing phase until terminal swing in all four groups. 3.5.2. Rectus femoris Except the recurvatum group, the other three groups had notable activity of rectus femoris immediately after initial contact (Fig. 12). The crouch group had sustained greater firing in the stance phase, while the jump group had the least activity. The mild group started the firing activity from terminal stance, while the other three groups had delayed onset of firing in mid-swing. 3.5.3. Biceps femoris All the four groups had the firing activities of biceps femoris in initial stance and terminal swing (Fig. 13). The biceps femoris of the recurvatum group had less activity than the other three groups in stance phase and an earlier onset of firing in swing phase. 4. Discussion problems in diplegic CP into jump, crouch, recurvatum and stiff knee gait is widely accepted. Their description was based on kinematics but our study has provided kinetic information in a similar patient group. We observed the same crouch, recurvatum and jump groups but did not observe the stiff knee pattern described by Sutherland and Davids but did observe another group where the kinetic patterns were not very dissimilar from normals. There may be two explanations for this difference between the two studies. Firstly, the populations studied may have differed and secondly the additional information that we obtained from kinetics may account for this difference. We found that the jump knee group had the largest peak vertical force, necessitating extremely large push up effect of the supporting leg. Excessive knee extensor torque in the initial stance started from the extreme flexion position. This might explain the eccentric contraction of hamstring with observed over-activity in EMG. In initial stance, the sagittal ankle kinematics revealed a slight increase in dorsiflexion, often without heel strike, because of the increased knee flexion. How- Sutherland and David’s classification  of knee Fig. 10. Mean ankle joint power of four cerebral palsy groups. Fig. 11. Mean linear envelopes of gastrocnemius of four cerebral palsy groups. 230 C.-J. Lin et al. / Gait and Posture 11 (2000) 224–232 Table 4 Knee joint powersa at specific gait events and its peak values Knee power, W/kg generation (+)/absorption (−) Mild (n = 17) Crouch (n =8) Recurvatum (n =14) Heel strike Opposite toe off Opposite HS Toe off Heel strike Maximum generation Maximum absorption 0.16 −0.02 −0.16 −0.12 0.09 0.60 −0.82 0.06 0.22 −0.63 −0.10 0.06 0.65 −1.13 −0.01 −0.32 0.10 −0.12 −0.05 0.64 −0.79 (0.38) (0.29) (0.37) (0.22) (0.11) (0.28) (0.37) (0.32) (0.40) (0.70) (0.23) (0.12) (0.20) (0.54) (0.62) (0.48) (0.49) (0.21) (0.20) (0.33) (0.43) F valueb Jump (n =7) 0.20 1.24 −0.52 0.00 −0.04 1.88 −0.98 (0.43) (1.05) (0.37) (0.15) (0.23) (0.91) (0.33) 0.43 12.17* 4.40** 0.58 2.0 14.28* 1.25 Post-hocc J\C,M,R R\C J\C,R,M a Means and S.D. in parenthesis. Statistic was done with one-way ANOVA. c Tukey’s post-hoc test was used to determine the significance between each pair of the four groups. * PB0.001; ** PB0.01. b ever, the plantar flexion moment increased rapidly and might be considered to arise from triceps surae activity as seen on the EMG. Ankle joint power revealed shock absorption at this stage and we also observed on video a mild toe strike and consider that the triceps surae was contracting eccentrically at this stage. This was then followed by a concentric contraction of triceps surae, which created a period of excessive plantar flexion, and generation of joint power in early second rocker. This might explain why we observed a premature plantarflexion during second rocker, which was not mentioned in Sutherland and David report . The vertical force curve had a deep mid-stance trough created by concentric contraction of quadriceps femoris, signifying the characteristic pattern of transmitting body weight. In mid-stance, the knee moved back to the normal extension position so the extensor torque decreased rapidly, because it is no longer necessary. Due to the mid-stance plantar flexion of ankle still being resisted by a passive dorsiflexor moment, the generation of ankle power decreased gradually. In late second rocker, when the body weight was ahead of the foot, premature plantar flexion of triceps surae occurred. The second peak of power generation preceded the third rocker and might explain the appearance of a double bump in the power pattern at the ankle. Whilst the changes at the ankle and knee occurred mostly in early and mid stance, the changes at the hip appeared mainly in late stance. When the hip was extended and the GRF passed behind the joint, the hip flexors contracted eccentrically producing power absorption and a flexor moment. The hip extensors did not appear to have a significant role in this pattern. The crouch group showed a decreased and practically absent mid-stance trough in the vertical force curve. The decreased mid-stance trough reveals a poor efficiency of weight transmission pattern. When the knee is held excessively flexed in stance and the GRF passes behind the knee a net internal extension moment is generated, which we observed in this group. In addition, there was a double-bump pattern in the knee Fig. 12. Mean linear envelopes of rectus femoris of four cerebral palsy groups. Fig. 13. Mean linear envelopes of biceps femoris of four cerebral palsy groups. C.-J. Lin et al. / Gait and Posture 11 (2000) 224–232 231 Table 5 Ankle joint powersa at specific gait events and its peak values Ankle power, W/kg generation (+)/generation (−) Mild (n = 17) Crouch (n =8) Heel strike Opposite toe off Opposite HS Toe off Heel strike Max generation Max generation 0.25 −0.33 0.36 0.19 −0.00 1.21 −0.54 −0.20 −0.35 0.24 0.13 0.02 0.74 −0.79 (1.00) (0.32) (0.55) (0.26) (0.02) (0.95) (0.28) (0.22) (0.29) (0.46) (0.27) (0.02) (0.22) (0.33) Recurvatum (n =14) −0.26 −0.62 0.49 0.07 0.01 0.87 −0.78 (0.29) (0.66) (0.78) (0.12) (0.00) (0.67) (0.58) Jump (n = 7) −0.32 −0.32 0.11 0.01 0.02 0.89 −1.18 (0.38) (1.03) (0.32) (0.07) (0.02) (0.19) (0.71) F valueb 1.86 0.72 0.60 1.42 2.95* 0.97 2.56 a Means and S.D. in parenthesis. Statistic was done with one-way ANOVA. * PB0.05. b moment, similar to the findings of Ounpuu . As the ankle was held in extreme dorsiflexion position, the GRF passes anteriorly producing an external dorsiflexor moment and an internal plantarflexor moment. The loss of the mid-stance trough of the vertical ground reaction force produces a flat moment pattern at the ankle. Crouch gait produced increased power absorption at the knee in terminal stance as the quadriceps contracted eccentrically in the presence of knee flexion. Thus in crouch gait, energy transfers about the knee is inefficient. We noted that the hip moment curve in the crouch group had a relatively larger extensor torque because of the hip flexion pattern in stance. The knee moment in the recurvatum group was the opposite to that seen in the crouch group and was associated with an external extension moment and an absorption power pattern that was not seen in the other groups. The knee flexors were silent on EMG and the internal flexor moment might originate from the passive resistance of the posterior soft tissue structures of the knee, instead of knee flexors themselves. It would be clinically significant if this joint were at risk of damage due to the absence of muscular support [6,7]. We also noted a decreased period of second ankle rocker and premature plantarflexion in this group. There was a trend towards an increased ankle plantarflexor moment that was also seen in the jump pattern group. However the ankle moments did not show significant differences between the four groups, though great variations in kinematics were observed. The knee moment patterns in the mild group were very similar to those seen in normal children. The mild group was the only one to show a net dorsiflexor moment at the ankle at initial contact, implying that the GRF was posterior to the ankle at this stage. The mild group also had a normal first rocker and normal ankle kinematics. Our study has demonstrated kinetics in the three knee patterns of jump, crouch and recurvatum gait seen in diplegic CP. The jump group was characterised excessive power generation, whilst this was lost in the crouch and recurvatum groups. An understanding of kinetics in this group may help in their clinical management. Acknowledgements This work was supported by National Health Research Institute, DOH-HR-410 and DOH-HR-821, Taiwan, ROC. References  Seliiktar R, Bo L. The theory of kinetic analysis in human gait. In: Craik RL, Oatis CA, editors. Gait Analysis: Theory and Application. St. Louis, MO: Mosby, 1994:223 – 38.  Barnes SZ, Berme N. Measurement of kinetic parameters technology. In: Craik RL, Oatis CA, editors. Gait Analysis: Theory and Application. St. Louis, MO: Mosby, 1994:239 – 51.  Meglan D, Todd F. Kinetics of human locomotion. In: Rose J, Gamble JG, editors. Human Walking. Baltimore, MD: Williams and Wilkins, 1994:73 – 99.  Winter DA, Eng JJ, Ishac MG. A review of kinetic parameters in human walking. In: Craik RL, Oatis CA, editors. Gait Analysis: Theory and Application. St. Louis, MO: Mosby, 1994:252 – 70.  Ounpuu S, Gage JR, Davis RB. Three-dimensional lower extremity joint kinetics in normal pediatric gait. J Pediatr Orthop 1991;11:341 – 9.  Ounpuu S, Davis RB, Deluca PA. Joint kinetics: method, interpretation and treatment decision-making in children with cerebral palsy and myelomeningocele. Gait Posture 1996;4:62–78.  Ounpuu S. Joint kinetic patterns in pathological gait. In: Instruction Course Lectures: Clinical Gait Analysis. Connecticut: Newington Children Hospital, 1991:57 – 64.  Gage JR. Gait Analysis in Cerebral Palsy. Oxford: MacKeith, 1991.  Gage JR. Gait analysis: an essential tool in the treatment of cerebral palsy. Clin Orthop Relat Res 1993;288:126 – 34.  DeLuca PA, Davis RB 3rd, Ounpuu S, Rose S, Sirkin R. Alterations in surgical decision making in patients with cerebral palsy based on three-dimensional gait analysis. J Pediatr Orthop 1997;17:608 – 14.  Stefko RM, de Swart RJ, Dodgin DA, Wyatt MP, Kaufman KR, Sutherland DH. Kinematic and kinetic analysis of distal 232         C.-J. Lin et al. / Gait and Posture 11 (2000) 224–232 derotational osteotomy of the leg in children with cerebral palsy. J Pediatr Orthop 1998;18:81–7. Rose SA, Deluca PA, Davis RB. Kinematic and kinetic evaluation of the ankle after lengthening of the gastrocnemius fascia in children with cerebral palsy. J Pediatr Orthop 1993;13:727 – 32. Yngve DA, Chambers C. Vulpius and Z-lengthening. J Pediatr Orthop 1996;16:759 –64. Ounpuu S, Bell KJ, Davis RB 3rd, DeLuca PA. An evaluation of the posterior leaf spring orthosis using joint kinematics and kinetics. J Pediatr Orthop 1996;16:378–84. Rose SA, Ounpuu S, Deluca PA. Strategies for the assessment of pediatric gait in the clinical setting. Phys Ther 1991;71:961 – 80. Judge J, Davis RB, Ounpuu S. Age-associated reduction in step length: testing the importance of hip and ankle kinetics. Gait Posture 1995;3:81. Olney SJ, MacPhail HA, Hedden DM, Boyce WF. Work and Power in hemiplegic cerebral palsy gait. Phys Ther 1990;70:431 – 8. Colborne GR, Wright FV, Naumann S. Feedback of triceps surae EMG in gait of children with cerebral palsy: a controlled study. Arch Phys Med Rehabil 1994;75:40–5. Lai KA, Kuo KN, Andriacchi TP. Relationship between dynamic deformities and joint moments in children with cerebral palsy. J Pediatr Orthop 1988;8:690–5. .  Davis RB, Deluca PA. Second rocker ankle joint stiffness during gait. Gait Posture 1995;3:79.  Crenna P. Spasticity and ’spastic’ gait in children with cerebral palsy. Neurosci Biobehav Rev 1998;22:571 – 8.  Davids JR, Bagley AM, Bryan M. Kinematic and kinetic analysis of running in children with cerebral palsy. Dev Med Child Neurol 1998;40:528 – 35.  Sutherland DH, Davids JR. Common gait abnormalities of the knee in cerebral palsy. Clin Orthop Relat Res 1993;288:139–47.  Winters TF, Hicks R, Gage JR. Gait patterns in spastic hemiplegia in children and young adults. J Bone Joint Surg Am 1987;69:437– 41.  Kadaba MP, Ramakrishnan HK, Jacobs D. Gait patterns in spastic diplegia. Dev Med Child Neurol 1991;33:28.  Hullin MG, Robb JE, Loudon IR. Gait patterns in children with hemiplegic spastic cerebral palsy. J Pediatr Orthop B 1996;5:247– 51.  O’Byrne JM, Jenkinson A, O’Brien TM. Quantitative analysis and classification of gait patterns in cerebral palsy using a three-dimensional motion analyzer. J Child Neurol 1998;13:101 – 8.  Sirkin R. Rotational abnormalities in cerebral palsy. In: Instruction Course Lectures: Clinical Gait Analysis. Connecticut: Newington Children Hospital, 1991:43 – 5.  Sutherland DH. The Development of Mature Walking. Oxford: MacKeith, 1988.
© Copyright 2018