Walk-Run Transition Speed and the Relevance of Loading, Muscular Fatigue, and Human Kinematics: A Comparison of Human Gait Patterns

摘要

INTRODUCTION: The walk-to-run transition (WRT) typically occurs at a preferred transition speed (PTS) of 1.9-2.1ms-1. Previous research has focused on potential triggers for this transition such as leg length, muscular fatigue, loading rates (LR), and vertical ground reaction forces (VGRFs). Rather than focus on mechanisms responsible for the transition, the purpose of the current study was to determine if basic anthropometrics or gait characteristics are predictive of the WRT. METHODS: Thirty participants were recruited for the current study (n = 13 male, 17 female; n = 11 normal weight, n = 10 overweight, n = 9 obese; age M = 26.3, SD = 5.5 years; height M = 68.8 SD = 3.8 inches; weight M = 182.6, SD = 41.0 lbs; BMI = 27 kg/ms). Participants‟ passive hip and ankle range of motion (ROM) was measured. Next, participants performed a minimum of three overground walking trials at their preferred walking speed (PWS), followed by three WRT trials and four tibialis anterior (TA) strength and endurance tests on a Biodex Isokinetic Machine (Biodex Medical Systems Inc, Shirley NY). TA Strength was the peak torque derived from three maximum voluntary contractions. TA endurance was defined as the graphical value that dropped below 60% of the peak torque for three consecutive trials. Kinematic data were collected with eight Vicon MX series cameras (VICON, Denver, CO, USA), and VGRFs were collected with four force platforms (Kistler, Amherst, MA, USA u26 Advanced Mechanical Technology, Inc., Watertown, MA, USA). The following variables were calculated from the overground trials: active ankle/hip ROM, foot progression, vertical LRs, stride length, stride frequency, VGRFs, and PWS. The PTS was assessed using a motorized treadmill with a velocity increasing by 0.10 mph every 10 s. STATISTICAL ANALYSIS: A Classification and Regression Tree (CART) analysis was used in MATLAB (Mathworks, Natick, MA, USA) to identify and assess variables‟ predictive ability of PTS. A series of t tests were also run on results from the CART. RESULTS u26 CONCLUSION: The CART analysis resulted in a tree with two splits and three terminal nodes. PWS was the primary splitter, creating a division at 1.61 ms-1. A PWS above 1.61 ms-1 resulted in a PTS of 2.28 +/- 0.21 ms-1 for three participants, creating the first terminal node. The second splitter was BMI, subdividing participants at 27 kg/m2, with 27 participants below 27 kg/m2 transitioning at 1.97 +/- 0.17 ms-1 and creating the second terminal node. Twelve participants were categorized above 27 kg/m2 and transitioned at 1.8 +/- 0.13 ms-1, creating the third terminal node. A cross-validation technique generated mean square errors of 0.0734, 0.0565, and 0.0456 for the first, second, and third terminal nodes, respectively. Independent t tests were run on the two BMI groups (u3c 27 kg/m2 and u3e 27 kg/m2) from the secondary split. Passive hip ROM was statistically significant between the participants above and below 27 kg/m2 (p = 0.009 u3c 0.05), at 136 +/- 13° degrees and 161 +/- 27°, respectively. Also, TA endurance (p = 0.043 u3c 0.05) and step width (p = 0.05) were statistically significant, with participants above 27 kg/m2 at TA endurance values of 32 +/- 2.48 repetitions and participants below 27 kg/m2 at 24 +/- 0.71 repetitions. Step width values were 0.14 +/- 0.02 m and 0.11 +/- 0.01 for participants above and below 27 kg/m2, respectively. According to the CART analysis, PWS and BMI were identified as the best predictors for PTS compared to the other measure variables. In general, it is likely that there are differences across multiple variables between these groups, and it is the collective nature of these differences that influence the PTS. Future research on PTS must examine diverse populations in order to gain further insight on transition speed.