Gait and Balance Assessment in Children with Genetic Neurodevelopmental Disorders with and without Autism Spectrum Disorders

Background: Neurodevelopmental disorders are frequently associated with motor impairments including locomotion.  The lack of objective measures combined with the challenges inherent in studying children with neurodevelopmental disorders, hinder valuable quantitative motor assessments.  The cerebellum controls gait and balance and has shown the most consistent neuroanatomical alterations in Autism Spectrum Disorders (ASD).  The basal ganglia has also been implicated in ASD and plays an important role in repetitive behaviors, control of movements, and is associated with aspects of social and emotions processing.  Gait and balance are amenable to quantitative methods and may help to refine the motor phenotype in neurodevelopmental disorders.  Furthermore, several genetically determined conditions such as 16p11.2 and 1q21.1 are known to be associated with ASD yet their motor phenotype has not been detailed.  Motoric measures help characterize locomotion and permit comparisons amongst neurodevelopmental conditions and as such can provide insight into neural pathway dysfunction.

Objectives: To determine the feasibility and sensitivity of quantitative gait assessments in children with neurodevelopmental disorders.

Methods:  Thirty eight children (23 probands with 16p11.2 or 1q21.1 mutation and 15 unaffected siblings) mean age: 8.5 years (range 3.2–15.4), and 55.3 % male were enrolled.  Among the probands, 23% carried a diagnosis of ASD and were all male.  Gait assessments included: six-minute walk test (6MWT), 10 meter run/walk test (10MRW), timed-up-and-go test (TUG) and spatio-temporal measurements of preferred and fast-paced walking.  The Pediatric Evaluation of Disability Inventory-Computer Adaptive Tests (PEDI-CAT), a caregiver-reported functional assessment was administered.  Measures of balance were calculated using Percent Time in Double Support and Base of Support.  Subsequently, the locomotion data from the 6 children diagnosed with ASD were compared with 6 age-matched non-ASD genetic probands and 6 non-ASD sibling controls.

Results:  All but 2 of the youngest children completed the protocol.  Probands had significantly lower scores than sibling controls on 6MWT (p=0.040), 10MRW (p=0.012), and TUG (p=0.005).  Group differences were identified in Base of Support (p=0.003).  Probands had significantly lower PEDI scores in all domains including the mobility scale (p<0.001).  In the fast-paced condition, all participants increased velocity on average from 114.7 to 189.2 cm/sec by increasing both stride length and cadence. As for balance measures, the ASD group did not demonstrate the expected concurrent reduction in Double Support time during the fast-paced condition which was observed in the two age-matched non-ASD groups. In addition, only the ASD group presented with upper limb stereotypies.

Conclusions:  Our motor assessment provides quantifiable results in children with 16p11.2 or 1q21.1 mutation as well as detailed characterization of locomotion performance.  Differences in Base of Support highlight a balance impairment during locomotion activities in children with neurodevelopmental disorders with and without ASD. However, only the children with ASD had a blunted response to Double Support time in the fast-paced condition. This finding may contribute to their greater functional impairment. Our results confirm motor dysfunction including balance in children with known genetic neurodevelopmental disorders. The ability to characterize the motor phenotype in all children with ASD may help identify the neurological underpinnings of the disorder.