Abstracts / Gait & Posture 30S (2009) S1–S153 S17 Fig. 1. Percentage of activity and cocontraction during gait: all parameters of activity and cocontraction were significantly higher in CP than in TD (p < 0.05). reference. Percentage activity of m. rectus femoris (REF), m.vastus lateralis (VAL), m. biceps femoris (BIF), medial hamstrings (MEH), m. gastrocnemius (GAS), m. soleus (SOL), m. tibialis anterior (TIA) and cocontraction between quadriceps and hamstrings (CC QH) and between plantar flexors and TIA (CC PFTIA) were calculated over the whole gait cycle using custom made Matlab ® software. On/off threshold was defined as 1.96 × S.D. of the linear envelope of the EMG normalised to the mean amplitude (after applying a 4th order Butterworthfilter) [2]. The differences between CP and TD were evaluated by one-way Anova. In patients spasticity was evaluated with Modified Ashworth Scale for Duncan Ely (REF), hamstrings and GAS. The correlation between CC and increased CC (>meanTD ± 1 S.D.) and spasticity was evaluated by spearman rank coefficient. Results Percentage activity of all muscles and CC QH and CC PFTIA were significantly higher in CP than TD (Fig. 1). There was a low but sig- nificant correlation between CC QH as well as increased CC QH and Ashworth for hamstrings ( = 0.25–0.29, p = 0.006–0.018) and REF ( = 0.38–0.42, p < 0.000). CC PFTIA correlated significantly with spasticity of GAS ( = 0.38, p < 0.000) but increased CC PFTIA did not ( = 0.18, p = 0.088). 43% of TD did not show any CC between plantar flexors and TIA at all where only 4/30 CP showed no CC PFTIA (13%). Discussion Results confirm that in children with CP there is more cocontrac- tion during walking then in age related TD children. The increased cocontraction is the result of an increase in percentage activity of both agonist and antagonist muscles. The low correlation between CC and spasticity indicates that spasticity is not the only cause of cocontraction. Damiano already quoted compensatory strategies to increase joint stability as a possible cause [3] but for example poor balance might also attribute to cocontraction. References [1] Desloovere, et al. Gait Posture 2006;(24):302–13. [2] Difabio, et al. Phys Ther 1987;67(1):43–8. [3] Damiano, et al. Arch Phys Med Rehabil 2000;(81). doi:10.1016/j.gaitpost.2009.08.027 O24 Is co-contraction of knee extensors and hamstrings a sign of poor neuromuscular control? Reinald Brunner 1,∗ , Carlo Frigo 2 1 Children’s University Hospital, Basel, Switzerland 2 Dept. of Bioengineering, Polytechnic, Milan, Italy Summary Contraction of knee extensors and hamstrings were tested sepa- rately and simultaneously using forward dynamics simulation. The test was performed with the foot attached to the floor (simulat- ing load condition) and the foot free to move but the pelvis fixed in space (simulating proximal constraint). In this latter condition the muscles produced a larger effect on the distal joints, while in contrary under load the larger movement was observed on the proximal joints. However, if the knee was in flexion, the hamstrings produced more knee flexion than hip extension. If at the same time the knee extensors contracted and hence the knee was locked, the hamstring activity shifted up to the hip joint. Conclusions Co-contraction of knee extensors and hamstrings are used to shift the effect of a biarticular muscle to the joint where the effect would be minor without such co-contraction. Introduction Some abnormalities found in dynamic EMG which are typical for cerebral palsy patients, were found also in neurologically normal subject with orthopaedic diseases [1]. Our aim in this work was to search for a possible reason for a co-contraction of knee extensors and knee flexors (quadriceps and hamstrings). Materials and methods Forward dynamics was used on a 13 segment rigid body model [2] to simulate the contraction of knee extensor (the model included vasti and rectus femoris) and hamstring (the semimem- branosus was modelled as representative of the whole group). The knee joint was controlled by two cruciate ligaments [3]. The effect of muscle contraction (isolated and in combination to simulate co- contraction) was studied in various joint position of hip (90 ◦ , 60 ◦ , 30 ◦ ,0 ◦ , -15 ◦ of flexion) and the knee (120 ◦ , 90 ◦ , 30 ◦ ,0 ◦ of flexion).