Journal of Biomechanics 37 (2004) 1321���1327 Gastrocnemius and soleus lengths in cerebral palsy equinus gait��� differences between children with and without static contracture and effects of gastrocnemius recession Tishya A.L. Wrena,b,c,*, K. Patrick Doa, Robert M. Kaya,b a Childrens Orthopaedic Center, Childrens Hospital Los Angeles, 4650 Sunset Blvd., No. 69, Los Angeles, CA 90027, USA b Department of Orthopaedics, University of Southern California, Los Angeles, CA, USA c Departments of Radiology and Biomedical Engineering, University of Southern California, Los Angeles, CA, USA Accepted 28 December 2003 Abstract Equinus gait is one of the most common abnormalities in children with cerebral palsy. Although it is generally assumed that the calf muscles are abnormally short in equinus gait, no studies have been done to confirm that the muscles are short and that this shortness contributes to the equinus. This study used musculoskeletal modeling combined with computerized gait analysis to examine medial gastrocnemius (MGAS), lateral gastrocnemius (LGAS), and soleus (SOL) musculotendinous lengths during equinus gait in children with cerebral palsy. All three muscles were abnormally short during equinus gait whether or not the children had equinus contractures (Pp0.005). Children with static contractures had shorter maximum static MGAS and LGAS lengths than children with dynamic equinus (Pp0.002). The children with static contractures had ratios of peak dynamic length to maximum static length close to 1.0 for MGAS and LGAS (1.00570.015) but lower ratios for SOL (0.98470.024). For the children with static contracture, these ratios did not change significantly after gastrocnemius recession (PX0.14) because both static and dynamic lengths increased postoperatively (Pp0.04). These results support the current clinical understanding of the role of calf ������tightness������ in equinus gait, including the appropriateness and effectiveness of gastrocnemius recession for children with equinus contracture. r 2004 Elsevier Ltd. All rights reserved. Keywords: Muscle length Computer modeling Cerebral palsy 1. Introduction Equinus gait or ������toe walking������ is one of the most prevalent abnormalities in children with cerebral palsy (CP). Initially, the equinus is dynamic and can be managed with orthoses, physical therapy, stretching, strengthening of the dorsiflexors, and/or botulinium toxin injections. Even with treatment, however, a fixed contracture often develops, which may necessitate surgical correction. During surgery, the calf muscles are lengthened based on the assumption that short calf muscles play an important role in the equinus contrac- ture. However, no studies have been done to confirm that the muscles are indeed short and that this shortness contributes to equinus gait. During equinus gait, abnormal ankle plantarflexion may be accompanied by either knee hyperextension or by excessive knee flexion. Excessive knee flexion occurs in patients with static or dynamic hamstring contrac- ture, while knee hyperextension occurs in patients without hamstring contracture (Gage, 1991). This could result in normal gastrocnemius lengths since the gastro- cnemius crosses both the knee and the ankle. In contrast, the soleus crosses only the ankle and is therefore expected to be short during equinus gait regardless of knee position. Musculoskeletal modeling can be combined with computerized gait analysis to study musculotendinous lengths during gait. Several studies have used muscu- loskeletal modeling to examine hamstring lengths during crouch gait in children with CP. These studies have ARTICLE IN PRESS *Corresponding author. Childrens Orthopaedic Center, Childrens Hospital Los Angeles, 4650 Sunset Blvd., #69, Los Angeles, CA 90027, USA. Tel.: +1-323-660-2450 X433 fax: +1-323-666-4409. E-mail address: twren@chla.usc.edu (T.A.L. Wren). 0021-9290/$- see front matter r 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.jbiomech.2003.12.035

shown that most children with crouch gait walk with normal hamstring lengths despite excessive knee flexion throughout the gait cycle (Hoffinger et al., 1993 Delp et al., 1996 Thompson et al., 1998, 2001). However, children selected for hamstring lengthening surgery to correct crouch gait do have abnormally short dynamic hamstring lengths which become normal after surgery (Olsen et al., 2002). Botulinium toxin injections also increase dynamic hamstring length whether or not the hamstrings are short before injection (Thompson et al., 1998). These studies illustrate the importance of examining musculotendinous lengths since the lengthen- ing of muscles that are not short could possibly have deleterious consequences. To our knowledge, similar studies have not yet been conducted to study calf muscle length in equinus gait. The current study addressed this void by examining gastrocnemius and soleus lengths during equinus gait in children with CP. Children with dynamic equinus were studied, along with children who had static contractures. For the children with static contractures, muscle lengths were examined both before and after calf muscle lengthening via gastrocnemius recession. The purposes of the study were (1) to determine whether the gastrocnemius is abnormally short during equinus gait, (2) to examine the efficacy of gastrocnemius recession surgery in altering dynamic muscle lengths, and (3) to assess the degree to which gastrocnemius and soleus ������tightness������ contribute to equinus gait in children with and without static contracture. 2. Materials and methods The study protocol was approved by the Committee for Clinical Investigations (institutional review board) at Childrens Hospital Los Angeles. The study involved two groups of children with cerebral palsy and one control group (Table 1). The control group consisted of 10 limbs from 10 able-bodied children (Normal group). The first CP group consisted of 6 limbs from 4 children who had undergone isolated gastrocnemius recession to correct fixed contracture and equinus gait with no other simultaneous surgeries (GR group). The second CP group consisted of 10 limbs from 8 children who walked in equinus but did not have fixed contractures (Dynamic Equinus group). Fixed (static) contracture was defined as inability to dorsiflex the ankle to neutral with the knee maximally extended during physical examination. Equinus gait was defined as peak ankle dorsiflexion during the stance phase of gait more than 1 standard deviation (SD) below normal. Isolated foot drop during swing was not considered equinus gait. All subjects had passive knee extension within 710 of neutral. Children with bony deformities, ankle varus/valgus, or midfoot breaks were excluded from this study. All subjects underwent gait examinations which included comprehensive physical examination and computerized gait analysis. Subjects in the GR group had both pre- and postoperative gait examinations, while subjects in the Dynamic Equinus and Normal groups had only one examination. For the GR group, the average time between surgery and postoperative evaluation was 1.170.6 (mean7SD) years. During the physical examination, an experienced physical therapist dorsiflexed the ankle to the end of its range with the knee flexed at 90 to assess soleus contracture and with the knee extended to the end of its range to assess gastrocnemius contracture. The hindfoot was positioned in neutral varus/valgus during all measurements. Measurements were taken using a standard goniometer. One arm of the goniometer was placed laterally along the long axis of the shank (aligned from the lateral malleolus to the head of the fibula, with the axis just distal to the lateral malleolus), and the other along the lateral aspect of the calcaneus. The forefoot was not used for alignment of the goniometer so dorsiflexion could be measured at the ankle joint alone without any contribution from the mid-tarsal or tarso- metatarsal joints. ARTICLE IN PRESS Table 1 Subject demographics Gastrocnemius recession Dynamic equinus Normal (N=4 subjects) (N=8 subjects) (N=10 subjects) Age (yr) 5.171.5 6.772.3 7.471.1 Sex 4 male, 0 female 4 male, 4 female 5 male, 5 female CP subtype 1 hemiplegic 2 hemiplegic None 3 diplegic 6 diplegic Previous surgeries None Tendo-achilles lengthening (1 subject) Hip adductor & gastrocnemius lengthening (1 subject) Tendo-achilles & hamstring lengthening (1 subject) None Static dorsiflexion, knee flexed (deg) 1.376.6 (N=6 limbs) 19.277.3 (N=10 limbs) Not measured Static dorsiflexion, knee extended (deg) 12.274.2 (N=6 limbs) 10.275.3 (N=10 limbs) Not measured T.A.L. Wren et al. / Journal of Biomechanics 37 (2004) 1321���1327 1322