Biomechanical Analysis of Stroke Patterns and Functional Outcomes in Pediatric Manual Wheelchair Users

Approximately 88,0000 manual wheelchair (MWC) users are under the age of 21 in the United States (Kaye, LaPlante, & Kang, 2000). More than half of pediatric MWC users with spinal cord injury (SCI) report shoulder pain later in life and are at higher risk for developing upper extremity (UE) overuse injuries (Vogel & Zebracki, 2015). During MWC propulsion, a stroke pattern is the path the hand follows with respect to the pushrim (Boninger et al., 2002). Current clinical guidelines for adults recommend sole use of the semicircular stroke pattern as best practices because a MWC user is able to spend more time recovering and propel with less forceful, frequent pushes (Paralyzed Veteran’s Association, 2005). No clinical guidelines specific to pediatric MWC users exist. The biomechanical data from adult-based studies cannot be accurately scaled down to children because the size of their body segments is proportionally different (Schnorenberg et al., 2014). The characterization of stroke patterns amongst pediatric MWC users is needed in order to develop appropriate recommendations for this population. Twenty-two (22) participants, 13 males and 9 females between the ages of 7 - 21 years old (13.9 + 4.49) participated at the Motion Analysis Laboratory at Shriners Hospitals for Children – Chicago. The average positions of the 3rd metacarpal phalangeal reflective markers were tracked by a 14-camera Vicon MX Motion Analysis system and captured three-dimensional UE kinematics during each trial of steady-state MWC propulsion. Data was converted into a visual representation of each stroke cycle from the sagittal perspective via MATLAB. Three-hundred-eighty-three (383) stroke cycles were classified as one of six different types of stroke patterns. A two-stage protocol was used and confirmed good to very good (κ = 0.932) inter-rater agreement across all data. Most subjects alternated between two and four different types of stroke patterns. Among them were two novel pediatric stroke patterns that were identified in this study. A Multivariate Analysis of Covariance (MANOVA) tested for statistically significant differences in six glenohumeral (GH) joint dynamics between the different types of stroke patterns. The semicircular stroke pattern was characteristic of the largest average peak GH joint superior force (p < 0.01). The biomechanical characteristics of the two novel stroke patterns were significantly different than that of the other four types (p < 0.001). A One-Way T-Test was used to test for statistically significant differences between the scores of the sample and the normative population on the Short-Form Health Questionnaire for health-related quality of life. The average physical composite score (PCS) was significantly lower, while the mental health composite score (MCS) was significantly higher than that of the average healthy individual (p < 0.001). A MANOVA was conducted to determine if there were statistically significant differences in average PCS and MCS between groups stratified by use of 1 – 4 or more different types of stroke patterns. The PCS of subjects who used two stroke patterns during steady-state MWC propulsion was significantly less than those who performed one, three, or four (p < 0.05). The Visual Analog Scale was used to assess bodily pain; a small number of subjects indicated mild (6) and moderate (1) symptoms. This study is an important first step towards the development of pediatric-specific clinical guidelines for practitioners who are involved in life care planning for children and adolescents with SCI.