Supervisors info:
Μπουντόλος Κωνσταντίνος, Καθηγητής, Σ.Ε.Φ.Α.Α., Ε.Κ.Π.Α.
Summary:
The walk-to-run transition (WRT) has been extensively studied on the treadmill but inadequately overground. The shortage of overground data might be responsible for the inability to determine the critical variable that triggers the WRT, even though there is evidence for the contribution of step length, a variable largely affected by the maximal angle between the thighs (ABTmax) (Minetti et al 1994). The critical variable should meet the criteria of Hreljac (1995a) and the predictions of the dynamic theory of gait transitions supported by Diedrich & Warren (1995), such as the critical fluctuations. Moreover, the effects of loading while performing the WRT are still unknown while very few studies have used inertial sensors. The purpose of this study was to examine the effects of trunk and limb loading while performing the WRT and the determination of the critical variables.
23 healthy males aged 18-26 years old performed the WRT in a laboratory environment. Five 3D inertial sensors Xsens type MTx (50 Hz) attached to the lumbar area and the center of mass of thighs and calves and a digital video camera Sony 25 Hz (deinterlacing in 50 fps) were used. Trunk loading (15% of the body mass) was achieved by an athletic vest of 10.2 kg and limb loading (same percentage) was achieved by four weights of 2.3-5.0 kg. Three valid trials were performed in each condition (Without Loading-WL, Trunk Loading-TL and Limb Loading-LL). The length of the transition step (SL0) and the WRT speed were calculated. Heel strike and toe-off times were defined in each step according to Aminian et al (2002) and ABTmax, time of ABTmax as a percentage of step duration (t% ABTmax), step frequency (SF) and duration of double support (for walking) or flight (for running) as a percentage of step duration (t% dble-fli) were calculated. Statistical analysis performed were: a) one-way ANOVA repeated measures (SL0 and WRT speed), two-way ANOVA repeated measures (ABTmax, t% ABTmax, SF and t% dble-fli) and trend analysis of the latter variables with the steps, b) t-test for dependent samples, one-way ANOVA repeated measures and F-test for confirming the criteria of Hreljac (1995a) and the predictions of the dynamic theory of gait transitions supported by Diedrich & Warren (1995) and c) stepwise analysis for the prediction of WRT speed.
WRT speed was significantly different (p<0.05) between the WL (2.56 ± 0.34 m/s) and TL (2.46 ± 0.36 m/s) conditions, but not significantly different between the WL and LL (2.49 ± 0.39 m/s) conditions. The same differences were found significant for the SL0. The effect of step was found significant (p<0.001) for all the variables while the effect of condition was found significant for SF (p<0.001) and t% dble-fli (p<0.01), but not for ABTmax which may indicate the same critical value across all the conditions. Indeed the significant decrease of ABTmax between the steps -1 and 0 in the WL condition and the non-significant differences of ABTmax across all the conditions, not only at the step -1 (critical value of 52.5 deg just before the WRT) but across all the steps, are strong evidence that ABTmax can be considered a critical variable according to the criteria of Hreljac (1995a).The variability of ABTmax in the WL condition increased within the steps -2 and -1 (presence of critical fluctuations) and decreased within the running steps, confirming the predictions of the dynamic theory of gait transitions supported by Diedrich & Warren (1995).
Taking into consideration the relatively small area of movement, ABTmax seems to be the critical value for the WRT. The comparison of people with different hip range of motion might possibly confirm the relationship of ABTmax with WRT speed while the incorporation of the step length into the analysis will isolate the pure contribution of ABTmax.
