Petra Marsico ^1,2,3* Lea Meier ^1,2 Anke Buchmann ^1,2 Andrina Kläy ^1,2 Marietta Van Der Linden ³ Tom Mercer ³ Hubertus J. Van Hedel ^1,2,3

• ¹ Research Department, Swiss Children's Rehab, University Children's Hospital Zurich, Zurich, Switzerland
• ² Children's Research Center CRC, University Children's Hospital Zurich, University of Zurich, Zurich, Switzerland
• ³ Centre for Health, Activity and Rehabilitation Research, Queen Margaret University, Musselburgh, United Kingdom

Introduction: To investigate the feasibility, discriminative and convergent validity, and reliability of a lower limb sensor-based proprioception measure in children with upper motor neuron (UMN) lesions. Method: We assessed three proprioception modalities (joint movement, joint position, and dynamic position sense) of the lower limbs in 49 children with UMN lesions and 50 typically developing (TD) peers (5-19y). Forty-three children with UMN lesion had a congenital and six an acquired brain lesion and 82% are able to walk without a walking aid. We evaluated the feasibility, compared the test results between children with UMN lesions and TD peers, and calculated Spearman correlations (rs) between the modalities. We quantified relative reliability with Intra-Class Correlation Coefficients (ICC) and absolute reliability with Smallest Detectable Changes (SDC). Results: Most children with UMN lesions (>88%) found the tests easy to perform. The children with UMN lesions had significantly (p<0.001) lower proprioceptive function than the TD children. The correlation between the three proprioceptive modalities was moderate to high (0.50 ≤ rs ≤ 0.79). The relative reliability for test-retest and the inter-rater reliability was moderate to high (ICCs=0.65-0.97), and SDC between 2° and 15°. Discussion: The three tests are feasible, and discriminative and convergent validity and reliability were confirmed. Further studies should investigate the influence on motor function and performance in children with UMN lesions.